PROCEEDINGS of The 10th International Conference On Mine Closure PDF

June 1−3, 2015  Vancouver, Canada
Andy Fourie, The University of Western Australia, Australia

Mark Tibbett, University of Reading, United Kingdom
Les Sawatsky, Senior Consultant, Golder Associates Ltd., Canada
Dirk van Zyl, Norman B. Keevil Institute of Mining Engineering,
University of British Columbia, Canada
© 2015, InfoMine Inc.
All rights reserved. No part of this publication may be reproduced, stored or transmitted in any form or by any means,
electronic, mechanical, by photocopying, scanning, recording or otherwise, without the prior written permission of InfoMine
Inc.
Production and Editorial: Joe Goodwill, InfoMine Inc.

Cover design: Michael Cohen, InfoMine Inc.
Cover illustrations courtesy of Syncrude Energy Ltd.
Any views and opinions expressed in the articles published in these proceedings are solely those of the authors and do not
necessarily represent those of InfoMine Inc. or of the panel of editors. The authors take full and exclusive responsibility for
technical content and accuracy of the information published herein. This information is not intended nor implied to be a
substitute for professional advice. InfoMine Inc. and the panel of editors are not responsible for any damage to property or
persons that may occur as a result of the use of the information contained in this volume.
ISBN: 978-0-9917905-9-3
InfoMine Inc.
Suite 900, 580 Hornby Street, Vancouver, British Columbia, Canada V6C 3B6
Tel: +1 604 683 2037; Toll free: +1 888 683 2037; (Canada and USA only); Fax: +1 604 681 4166
Email: info-ca@infomine.com; Web: www.infomine.com
ii
Organisers .............................................................................................................................................. ix
Committees ............................................................................................................................................xii
Technical Reviewers .............................................................................................................................. xiii
Foreword │ L. Sawatsky and M. Davies .................................................................................................xv
Acknowledgements............................................................................................................................. xviii
Sponsors ................................................................................................................................................ xix
Institutional and Media Partners ........................................................................................................... xx
Chapter 1: Keynotes and Plenaries ........................................................................................................ 1
Opening Keynote: Some histories of mine closure, the idea │ D.E. Hockley and L.C. Hockley .............. 3
Theme Keynote 1: Teck – 100 years of sustainability focus: water, reclamation and closure │
M. Davies ................................................................................................................................. 12
Plenary Presentation 1: Ten years on: what have we learned from the Mine Closure conference
series? A review of the conference proceedings from 2006 to 2014 │ C. Digby and
D. Limpitlaw ............................................................................................................................. 13
Plenary Presentation 2: Managing long-term orphaned mine liabilities: exploring the
relinquishment of closed sites │ E.J. Gardiner, W.R. Cowan, G.A. Tremblay
and C.M. Hogan ....................................................................................................................... 14
Plenary Presentation 3: Post-closure stewardship of mine sites: institutional control in
Saskatchewan – a case history │ D.G. Hovdebo, K.E. Cunningham, D.M. Kristoff
and M.S. Webster .................................................................................................................... 23
Theme Keynote 3: Changing water treatment standards and the impact on mine closure │
R. Knapp ................................................................................................................................... 37
Plenary Presentation 4: Now and then: eight decades of lessons in closure planning – case
studies from West Africa │ P.A. Varris and M.B. Thorpe ......................................................... 38
Plenary Presentation 5: Mine closure in Canada: a perspective on the past, present and future │
C. Pelletier, E. Prelypchan and K. Etches ................................................................................. 52
Plenary Presentation 6: Field study of biobased technologies for mine rehabilitation purposes │
C. Ortiz, M.A. Wilkens, D.V. Barros and J.K. Pizarro ................................................................ 53
Theme Keynote 4: Mine closure – past, present and perpetuity │ R.S. Siwik and M.D. Clemens........ 59
Theme Keynote 5: Updating Australia’s leading practice sustainable development mine closure
handbook for 2015: closing the gaps and understanding the mineral resource
legacy │ H. Lacy and K. Bennett ............................................................................................... 71
Plenary Presentation 7: Closure planning for the largest copper producer in the world │
E. Sanzana, B. Weeks and S. Fernandez .................................................................................. 81
Plenary Presentation 8: Integrated closure planning: case studies on changing operational
strategies to reduce closure liabilities │ C.D. Grant and P.R. Botha ........................................ 92
Plenary Presentation 9: Using genomics in mine reclamation │ L.H. Fraser, H.W. Garris,
S.A. Baldwin, J.D. Van Hamme and W.C. Gardner ................................................................. 106
iii
Chapter 2: Regulatory ........................................................................................................................ 119
A framework for standardised, performance-based completion criteria for mine closure and
mine site relinquishment │ R. Holmes, M. Flynn and M.B. Thorpe....................................... 121
British Columbia’s Crown Contaminated Sites Program – a ten year review │ G. Stewart ................ 131
How to reconcile more than a century of mining history with today’s rehabilitation standards: a
case study in Cobalt, Ontario │ J. Noël, J. Cayouette, J.-F. Doyon and M.R. Julien ............... 143
Development of international standards for mine reclamation management │ P.A. Steenhof ........ 157
Closure drainage design for oil sands mining – robustness for dam delicensing │ R.J.M. Wirtz
and M.E. Urquhart ................................................................................................................. 169
Chapter 3: Landform .......................................................................................................................... 183
Highwall blasting: a cost effective way to reshape and reduce closure liabilities │ C.D. Grant
and L. Hantler ........................................................................................................................ 185
Evolution of the Bige Stockpile closure design for the Ok Tedi Mine │ K. Brand, M.J. Ind,
M. Rid, M. Thompson and L. Murray ..................................................................................... 193
Assessing the evolution of a post-mining landscape using landform evolution models at
millennial time scales │ J.B.C. Lowry, G.R. Hancock and T.J. Coulthard ................................ 207
Five fundamentals for successful mined land rehabilitation │ M.S. Theisen ..................................... 221
End land use as a guide for integrated mine planning and closure │ N. Slingerland
and G.W. Wilson .................................................................................................................... 235
Chapter 4: Case Studies...................................................................................................................... 249
Remediation of Wismut´s Uranium mill tailings pond Culmitzsch – progress achieved and
challenges ahead │ U. Barnekow, T. Metschies, G. Merkel and M. Paul .............................. 251
Decommissioning the Cluff Lake Project │ C.B. Brown, D.M. Huffman and B.J. Schmid .................... 263
In situ neutralisation of a lake impacted by legacy uranium tailings in northern Saskatchewan,
Canada │ L.M. Christoffersen, S.R. Jensen, T.R. Sharp, M.W. Liskowich and D.E. Allen ....... 275
Sustainable closure of two former uranium mines │ L.K. Holder, F. Shuri, B. Nielsen and R. Jones .. 289
Evapotranspiration control on groundwater migration at a uranium mine in the high desert │
C. Ardito, J.M. Sigda and R. Blickwedel.................................................................................. 301
Adaptive risk management and lessons learned: post-closure at the “mighty” Sullivan mine │
R.A. Peterson, S.A. Humphries and M.L. Unger..................................................................... 315
Century Mine closure │ M.T. Lord, M.H. Adams and T. J. Shearman ................................................. 327
Groundwater contamination due to historical magnetite mine tailings deposited in a
near-shore, marine environment │ R.C. Dickin, S.E. Dickin, L. Groat, K. Jia, U. Mayer,
R. Mills and M. Sanborn......................................................................................................... 339
Treatment of mine drainage with high topographical constraints: case study of the Bodennec
site (France) │ J.C. Jacob and M. Save ................................................................................... 353
Final closure at the Georgetown Canyon phosphate ore processing facility — case study │
P. Kos, M.J. Hart and J.B. Williams ......................................................................................... 363
iv
Ekati Long Lake containment facility reclamation research │ L. Novy, G. Koop, M.E. Wen
and W. Petherbridge ............................................................................................................. 375
Engineered flow-through closure of an acid pit lake: a case study │ C.D. McCullough
and C. Harkin ......................................................................................................................... 377
From tailings basin to aquatic ecosystem: the ecological recovery of two waterbodies in
Kirkland Lake, Ontario, Canada │ A. Chute and N. Dombrowski ........................................... 389
Optimising application rates of waste residuals in mine soil reclamation programs using
response surface methodologies │ G.S. Piorkowski, G. Price and N. Tashe .......................... 403
The rehabilitation of the Bicapa–Tarnaveni waste storage facility │ M. Dey, C. Williams,
C. Brough, R. Bowell, D. Runciman and V. Turea................................................................... 413
Giant Mine Remediation Project │ C. Wells ........................................................................................ 423
Innovative process for stabilising the subsurface at the Giant Mine Site │ D. Kennard
and C. MacInnis ..................................................................................................................... 433
Decontamination and deconstruction of the Giant Mine roaster complex: risks, challenges, lessons
learned and successes │ M. Larivière, D. Hango and C. Corrigan.......................................... 449
Beyond ML/ARD: the many faces of neutral mine drainage in the context of mine closure │
D.A. Bright and N. Sandys ...................................................................................................... 461
Developing the Hollinger open pit mine – leaving a positive legacy │ T. Sulatycky, D. Bucar
and P. Buttineau .................................................................................................................... 471
Using applied research to reclaim a northern Canadian uranium mine │ B.H. Cranston
and L.A. Leskiw ...................................................................................................................... 485
Opportunities to incorporate mine closure into mine planning │ S.E. Ames ..................................... 497
Passive biological treatment of acid mine drainage with recovery of copper │ A. Schwarz,
N. Pérez and G. Chaparro ...................................................................................................... 507
Optimisation – a rationalised approach to developing mine closure objectives │ A.R. Butler,
T.R. Anderson and A. Sexton ................................................................................................. 517
Key factors in developing and implementing a successful mine closure plan – Denison sites
20 years after closure │ R. Pedlar-Hobbs, I. Ludgate and S. Lanteigne ................................. 531
Process-driven design and piloting of a site-specific constructed wetland for copper and
selenium treatment in the Yukon, Canada │ M. Haakensen, V. Pittet, J. Spencer,
J.H. Rodgers Jr. and J.W. Castle ............................................................................................. 545
Evaluation of acid generation in mine rock at an iron mine 25 years after closure │ M. Venhuis,
S. Shaw, R.V. Nicholson, S. Bartle and J. Stirling .................................................................... 561
Utilising novel technology for residue management and sustainable mine closure │ P. Littlejohn,
D. Kratochvil and A. Consigny ................................................................................................ 575
Waste dumping optimisation for final closure profile: a South African colliery case study │
F.A. Maimela and T. Zvarivadza ............................................................................................. 585
Chapter 5: Risk Assessment ............................................................................................................... 597
Using environmental risk assessment to meet changing expectations for mine closure │
B.A. Power and R. Baker ........................................................................................................ 599
v
A risk based approach to planning for progressive reclamation in the Canadian oil sands │
D.J. Theberge and E. Ferster .................................................................................................. 609
The role of aquatic ecological risk assessment to guide effluent management from mine
waste │ N. Lower, M. Quach and R. Marsland ...................................................................... 621
An environmental risk assessment framework for developing financial liability estimates for
closing five abandoned metal mines in northern Saskatchewan, Canada │ N.J. Healey,
B.A. Power and D. Kristoff ..................................................................................................... 633
Worker certification: a tool for responsible mine closure and worker redeployment │
R. Montpellier ........................................................................................................................ 645
Quantitative risk assessment tools to assist with waste management and placement
guidelines │ S.R. Pearce and S. Lehane.................................................................................. 657
Chapter 6: Landforms and Evolution in Reclamation ....................................................................... 671
Effective, long-term monitoring of ground deformation throughout mine closure phases │
M.D. Henschel, S. Sato, R. Ahola and M.A. McParland ......................................................... 673
From mines to mesas: a rehabilitation journey at a gold mine in the Pilbara, Western Australia │
C.D. Tiemann and M. Wealleans ........................................................................................... 681
Reclaiming mined-out pits at industrial sands mines in Zimbabwe: a case of cost minimisation
and community benefit maximisation │ T. Chirinda and T. Zvarivadza ................................ 697
Experiences with mine closure in the European coal mining industry: an overview of the
situation in Germany and adjacent regions │ C. Melchers, P. Goerke-Mallet, L. Henkel
and M. Hegemann ................................................................................................................. 711
Identifying, mapping and modelling infiltration into and through mine workings │ V.O. Kofoed,
M.L. Jessop and D.F. Bierman ................................................................................................ 725
Earthworks/Mineworks: creating shared visions for postmined landscapes/landshapes │
P. Kirsch, A. Hine and I. Amizlev ............................................................................................ 737
Chapter 7: First Nations and Communities ....................................................................................... 751
Mine closure and First Nations – social licence strategies for effective community engagement │
R.E. Breadmore and G.J. Lafferty ........................................................................................... 753
First Nations engagement in mine closure: Sä Dena Hes mine decommissioning and
reclamation │ C.M. Jeffrey, M.L. Unger and J.C. Pugh .......................................................... 763
Making the most out of Zimbabwe’s Marange diamonds: leaving a lasting positive legacy for
distressed communities │ T. Zvarivadza ................................................................................ 775
Best practices for wetland reclamation for Alberta oil sand mines │ T. Charette and J. Martin........ 787
Social impacts of mine closure: engaging employees and host communities in planning for
closure │ S. Costa ................................................................................................................... 797
Challenges of integrating mine closure plans mid-way through the life of mine in Mongolia │
M.B. Dagva, G.L. Ainsworth, Ts. Davaatseren, Kh. Vladimir and O. Erdenetuya ................... 805
Managing the socio-economic impact of tin mining in Bangka Island, Indonesia – preparation
for closure │ E. Nurtjahya and F. Agustina ............................................................................ 817
vi
Chapter 8: Vegetation ........................................................................................................................ 827
Improving geotechnical stability of saturated mine tailings by dewatering with deep-rooted
vegetation │ M.P. Dobrowolski, S.A. Rowlands and R. Hattingh ........................................... 829
A plant ecophysiological approach to assess the performance and potential success of mine
revegetation │ S. Ebbs, L. Glaeser, M. House and D. Vitt ...................................................... 837
Selecting graminoids to increase substrate macroporosity for the revegetation of milled
tailings of a gold mine │ M. Guittonny-Larchevêque ............................................................ 851
Achieving biodiversity conservation goals in mine development, operation and closure │
S. Hilts, W. Franklin and T. Gullison ....................................................................................... 863
Chapter 9: Modelling.......................................................................................................................... 875
Developing a geo-environmental model for long-term post-mining management: the former
Salsigne Gold Mine area (France) │ F. Le Loher, F. Cottard and P. Chartier.......................... 877
Integrated surface water and groundwater modelling for oil sands reclamation │
R.M. Nagare, Y.J. Park and J. Pal ............................................................................................ 891
Use of contaminant dispersion modeling to guide close-out of the former Beaverlodge Lake
mine site, Canada │ C.E. Lucas, B.E. Halbert and M.S. Webster ............................................ 901
Chapter 10: Covers ............................................................................................................................. 915
Instrumentation and early monitoring results of the south waste rock dump at Century Mine │
P.L. Defferrard, T.K. Rohde and M.T. Lord ............................................................................. 917
Performance evaluation of reclamation soil cover systems at Cluff Lake mine in northern
Saskatchewan │ L.A. Barber, B.K. Ayres and B.J. Schmid ...................................................... 929
Evaluation of cover systems for closure of tailings storage facilities at North American
Tungsten Corporation’s Cantung mine │ G.C. Allen, L.K. Tallon, D. Flemming, S. Shaw
and B.K. Ayres ........................................................................................................................ 939
Stainless steel covers – an alternative closure method for remote abandoned mines │
D. Sanscartier, J. Lambert, C. Reid and D.E. Allen.................................................................. 951
Closure of legacy waste rock piles: can we achieve passive treatment to manage residual
seepage in the short term? │ G. Meiers, M. O’Kane, D. Mayich, P. Weber, C. Bradley
and J. Shea ............................................................................................................................. 965
Chapter 11: Ecosystem Reconstruction ............................................................................................. 981
Does the concept of novel ecosystems have a place in mine closure and rehabilitation? │
R.N. Humphries and M. Tibbett ............................................................................................. 983
Phytoremediation to improve pit lake water quality │ M.E. Wen, C.A. Pelletier, K. Norlund,
G.W.R. Wolff and D. Berthelot .............................................................................................. 995
Potential impact of improved rehabilitation methods on the water balance of an opencast coal
mine │ M.W. Palmer, A. van Zyl and C. Smith ..................................................................... 1011
Mine reclamation and surface water balances: an ecohydrologic classification system for
mine-affected watersheds │ J. Straker, T. Baker, S.L. Barbour, M. O’Kane, S. Carey
and D. Charest ..................................................................................................................... 1023
vii
Ecosystem reconstruction: a global assessment of methods of estimating soil water regimes
for mine reclamation and closure │ J. Straker, T. Baker, M. O’Kane, R. Shurniak,
S.L. Barbour and S. Carey..................................................................................................... 1035
The influence of coal mine spoil physical properties on the spatial distribution of lichen-rich
communities │ D.S. Whitlock, J. Rickson, R.N. Humphries, R. Thompson and M. Tibbett .. 1047
The triad plant-soil-microbes, an ecofriendly way to reconstruct healthy ecosystems on mining
sites: a global perspective │ M.B. Nadeau and D.P. Khasa .................................................. 1061
Effect of soil cover system design on cover system performance and early tree establishment │
L.A. Barber, J. Bockstette, D.O. Christensen, L.K. Tallon and S.M. Landhäusser ................. 1071
Land reclamation in Canada’s mining sandbox: do novel ecosystems require novel
approaches? │ P. Audet, B.D. Pinno and E. Thiffault ........................................................... 1081
Effective strategies for the reclamation of large mines │ D. Polster ................................................ 1093
Chapter 12: Miscellaneous ............................................................................................................... 1101
Sustainability appraisal for mine closure │ E. Petelina, D. Sanscartier, S. MacWilliam and
R. Ridsdale ........................................................................................................................... 1103
Teck’s Elkview Operations: visualising an integrated approach to conceptual closure and
reclamation planning │ D. Harrison, D. Crockett and L. Eykamp......................................... 1117
Assessing the level of difficulty of vegetation establishment on reclaimed sites │ B.H. Cranston
and L.P. Waterman .............................................................................................................. 1125
Update on mine closure, sealing and abandonment practices │ T.A. Gray, R.E. Gray and
W.P. Balaz, Jr........................................................................................................................ 1137
A case study of cost effective detailed waste characterisation techniques for closure of
abandoned uranium mines │ A.S. Orechwa and D.C. Workman ......................................... 1151
Chapter 13: Author Index................................................................................................................. 1165
viii
InfoMine Inc. is the world leader in providing mining knowledge online, delivering content via our
websites, through corporate intranets, and by email.
The InfoMine websites provide focused, in-depth information and functionality encompassing most
aspects of mining and mineral exploration activities worldwide. The websites are organised as a series
of InfoMine Editions which collectively provide access to the largest, most fully integrated source of
worldwide mining and mineral exploration information. Each InfoMine Edition has its own specific
audience and content provided in the local language.
Whether you are developing mineral properties, investing in mining, selling products and services to
the mining industry, or you just want to find a job, industry research and knowledge is an important
requisite. Reliable, focused information is necessary to keep a competitive advantage, and InfoMine
is equipped to conveniently provide the knowledge you need.
For more information, visit our website at http://www.infomine.com/.
ConferenceMine, the InfoMine Conferences division, organises highly technical seminars and
conferences, addressing specialised topics for the purpose of technology and knowledge transfer.
InfoMine Conferences provide an opportunity for specialists and technology users to exchange
information, discuss latest research and propose innovative solutions to technical challenges. Physical
conferences include technical sessions, trade shows and social functions, enabling personal exchanges
in formal and informal settings.
ConferenceMine conferences are a very important medium for fostering discussion, developing
technology, building relationships and gathering information. Since its creation ConferenceMine has
organised the following conferences:
 Tailings and Mine Waste (2011, Vancouver, Canada)

 Mine Water Solutions in Extreme Environments (2013, Lima, Peru; 2015, Vancouver,
Canada)
 Paste and Thickened Tailings (2013, Belo Horizonte, Brazil; 2014, Vancouver, Canada)
 Heap Leach Solutions (2013, Vancouver, Canada; 2014, Lima, Peru; 2015, Reno, Nevada)
 Cold Covers Practice Seminar (2014, Whistler, Canada)
 Mine Closure Solutions (2014, Ouro Preto, Brazil)
 Geosynthetics Mining Solutions (2014, Vancouver, Canada)
 Mine Closure (2015, Vancouver, Canada)
Please see http://www.infomine.com/conferences/ for a list of upcoming conferences.
ix
The Australian Centre for Geomechanics was formally established in 1992 as a University of Western
Australia research centre in order to promote research excellence and continuing education in
geomechanics, with particular emphasis on its application to the mineral and energy extraction
sections of Australia’s resource industry. The Australian Centre for Geomechanics is an
unincorporated Joint Venture involving:
 CSIRO Earth Science and Resource Engineering
 The University of Western Australia – School of Civil and Resource Engineering
The Centre draws together its staff knowledge and experiences with the expertise within the three
groups forming the Centre and facilitates a multi-disciplinary approach to research and education in
geomechanics. Research undertaken by the ACG attracts both national and global support and the
outcomes of the projects are utilised to promote safer mining and environmental geomechanics
practices, operating efficiencies and to meeting community expectations for sustainable mining
practices.
With the guidance of strong industry representation on the Board of Management, and close
collaboration with senior representatives of the mining industry, research, training and further
education activities are tailored directly to the needs of industry. The ACG Board expects the
Australian Centre for Geomechanics to be the focal point for industry on geomechanics issues and to
address the needs of industry through a collaborative interdisciplinary approach.
One of the greatest and continual challenges facing mining companies, regulators and local
communities is how to effectively and sustainably close a mine with positive social, economic and
environmental outcomes. The series of International Conferences on Mine Closure, founded by the
Australian Centre for Geomechanics in 2006, provides an ongoing global forum for this universal issue.
The event series allows for scientists, consultants, regulators, policy makers, company executives and
environment managers to meet, share ideas, concerns and solutions to some of the most intractable
problems in the primary sector.
This ACG hub provides mine closure practitioners with access to our mine closure further education
and training platform. It lists most of the mine closure event proceedings from 2006, specialist
publications, mine closure events, new products and much more.
x
The British Columbia Technical and Research Committee on Reclamation (TRCR) originated in the early
1970s, in response to a demonstrated need in the province BC for greater government-industry
communications in the area of environmental protection and reclamation associated with mining.
Membership is drawn from the corporate sector (several of the operating mines are represented); the
Ministry of Energy and Mines (MEM); the Ministry of Environment (MOE); CanmetMINING; the Mining
Association of British Columbia (MABC), the Mining Association of Canada, the Association of Mineral
Exploration BC and British Columbia universities and colleges such as UBC and TRU. The Committee
meets monthly to discuss matters of joint concern and interest, exchange experience, plan Symposium
activities and prioritise research needs.
UBC’s Department of Mining Engineering is known for being a small, close-knit family.
The department is exemplified by the dedication of the faculty and staff who provide a dynamic,
hands-on learning experience for both undergraduate and graduate students.
In addition to teaching, the faculty works with graduate students and staff to undertake research in
all aspects of mining in order to study and improve the industry for future generations. Gifts from
alumni, corporations, foundations, students, parents and other friends assist the Keevil Institute in
conducting leading edge research, providing outstanding education and contributing to social and
economic development.
The University of Reading is ranked in the top 1% of universities in the world. It is a global university
that enjoys a world-class reputation for teaching, research and enterprise. The University was
established in 1892, received its Royal Charter in 1926, and has developed into a leading force in British
and international higher education. The University delivers a world-class student experience,
research-led teaching and the graduate employability record is excellent. The number of students
going on to higher level study is well above the national average.
The University continues to evolve, reflecting an ever-changing world, which drives the development
of the areas of research excellence and strength. They are leading the way in multidisciplinary work in
biomedical and pharmaceutical sciences, housed in a new £17 million centre, for example.
Environmental sciences, and in particular soil science, ecology and agriculture are major strengths of
the university.
The University is committed to maintaining a supportive, challenging and high-quality experience for
students and staff alike and to preserving the heritage and beauty of some of the most beautiful
university campuses in the UK.
xi
Chair
Les Sawatsky, Senior Consultant, Golder Associates Ltd., Canada
Co-chair
Michael Davies, Vice President, Environment, Teck Resources Limited, Canada
Vice-chairs
Michael Nahir, Director of Engineering, Aboriginal Affairs and Northern Development
Canada
Dirk van Zyl, Professor, Norman B. Keevil Institute of Mining Engineering, University of
British Columbia, Canada
Technical Advisor
Andy Robertson, President, InfoMine and Robertson GeoConsultants, Canada
Members, BC TRCR Representatives

Jonathan Buchanan, Director, Communication and Public Affairs, AMEBC, Canada
Nicole Pesonen, Environmental Manager, Nyrstar, Canada
Conference Manager
Olga Cherepanova, InfoMine, Canada
Technical Publisher
Joe Goodwill, InfoMine, Canada
Marketing and Sales Manager

Marcia Mendonca, InfoMine, Canada
xii
Anil Beersing, Golder Associates, Canada
Bill Biggs, Biggs and Associates, Australia
Ken Bocking, Golder Associates, Canada
Jack Caldwell, Robertson GeoConsultants, Canada
Deborah Chan-Yan, Golder Associates, Canada
Sonia Finucane, Bioscope Environmental, Australia
Andy Fourie, The University of Western Australia, Australia
Mark Freberg, Teck Resources Limited, Canada
David Jasper, MWH Global, Australia
Walter Kuit, Independent Mining and Metals Professional (retired), Canada
Harley Lacy, MWH Global, Australia
Jennifer Leinart, InfoMine, USA
Gordon McPhail, SLR Consulting, Australia
Ken Mercer, Australian Centre for Geomechanics, Australia
Claudia Ortiz, University of Santiago, Chile
Mark Phillip, O’Kane Consultants, USA
Caius Priscu, Anglo American, Chile
Thierry Regnier, Tshwane University of Technology, South Africa
Les Sawatsky, Golder Associates Ltd., Canada
J. Ashley Scott, Northern Ontario School of Medicine, and Laurentian University, Canada
Alister Spain, School of Earth and Environment, The University of Western Australia, Australia
Mark Tibbett, University of Reading, United Kingdom
David Tongway, Retired (formerly with CSIRO), Australia
Bryan Ulrich, Knight Piésold, USA
Dirk van Zyl, The University of British Columbia, Canada
Angela Waterman, Mining Association of British Columbia, Canada
Björn Weeks, Golder Associates, Canada
Barbara Wernick, Golder Associates, Canada
Andre Xavier, The University of British Columbia, Canada
xiii
Bruce Anderson, Suncor, Canada
Jim Cassie, BGC Engineering, Canada
Ben Chalmers, Mining Association of Canada, Canada
Caroline Digby, Centre for Sustainability in Mining and Industry, University of the
Witwatersrand, South Africa
Steven Hilts, Environmental Legacies, Teck Resources Limited, Canada
Randy Knapp, SENES Consultants, Canada
Darrell Martindale, Shell, Canada
Mike O’Kane, O’Kane Consultants, Canada
Jeff Parshley, SRK Consulting, USA
Caius Priscu, Anglo American, Chile
Doug Ramsey, Copper North Mining Corp., Canada
Adam Whitman, Yamana Gold, Canada
Dennis Wilson, Newgold, Canada
xiv
This portfolio of conference proceedings presents peer-reviewed papers that were accepted
for presentation at Mine Closure 2015, an international conference being held in Vancouver,
BC on June 1 to 3 and organised by InfoMine. This conference marks the 10th international
mine closure conference in a series that was initiated by the Australian Centre for
Geomechanics (ACG) and the Centre for Land Rehabilitation in 2006 but, impressively, is also
the 39th Annual BC Mine Reclamation Symposium. It is fitting that the 10th international
conference is being held in British Columbia where the British Columbia Technical and
Research Committee on Reclamation (BC TRCR) originated in the early 1970s in response to a
demonstrated need in BC for greater government-industry communications in the area of
environmental protection and reclamation associated with mining. For the second time, the
BC TRCR will collaborate with the International Conference on Mine Closure. The previous
collaboration was at Lake Louise, Alberta in 2011.
Like the preceding conferences, Mine Closure 2015 assembles mine closure practitioners,
researchers, planners and operators and follows a successful tradition with presentations that
cover state-of-the-art research findings, case histories of effective mine closure practices,
recommended mine closure planning guidelines, analysis of effective regulatory systems and
sharing of best practices. The initial enthusiasm for this type of international collaboration in
2006, stemming from regional initiatives like the BC TRCR, has matured as the mine closure
community shares successes and failures, and as it identifies key challenges and even
obstacles to sustainable mine closure. The conference series was envisaged to become a
powerful force for improving mine closure and reclamation practices at mines all over the
world. Ten years following the initial International Mine Closure Conference held in Perth
Australia and 39 years after commencing BC TRCR, it appears that this vision is becoming a
reality. The successes include widespread adoption of best practices related to mine closure,
improved sustainability and reduced long term liability. But the mine closure storyline is not
entirely good news as there are some pockets of resistance, disappointing compromises that
may be related to economic conditions, inconsistent bonding mechanisms and inconsistent
regulatory leadership in some countries. There is more work to be done and the proceedings
of Mine Closure 2015 as reported herein, represent a significant contribution to assist in
guiding that work.
With over 350 delegates, significant interest in conference presentations (over 200 abstracts
submitted) and 90 papers accepted for publication, Mine Closure 2015 is expected to make a
significant contribution to the mining industry by the dissemination of valuable lessons,
networking amongst mine closure practitioners and “raising the bar” with a broad adoption
of improved leading practices related to mine closure.
Like earlier mine closure conferences, Mine Closure 2015 has a broad reach based on the
many countries from which the accepted papers originate. Over 40 of the accepted papers
were submitted by international contributors (60% from Canada). This balance fulfils the
vision of the conference organisers that Mine Closure 2015 would offer Canadian mine
closure practitioners a forum for learning important research results from international
guests and for sharing important findings and successes at mines in Canada. The overall goal
is to build on available information and thereby improve mine closure and reclamation
practices, reduce the environmental footprint and provide for a sustainable economy
wherever mines exist.
xv
The papers presented in these Mine Closure 2015 proceedings cover a broad array of topics
pertaining to mine closure and reclamation. They were carefully selected to make an effective
contribution to the sustainability of the mining industry, with many of the papers presenting
the results of applied research into various aspects of mine closure and others describing
successes and failures in the form of case histories. Some papers pertain to regulatory systems
designed to achieve closure landscapes that will meet minimum requirements related to the
environment, preservation of resources and development of stable landforms. Several papers
address the need for stakeholder engagement to optimise return of the mine disturbed
landscape for beneficial uses. Other papers address the difficult issue of integrating mine
operations with mine planning.
Presenting the results of applied research into sustainable mine closure is a very important
element of these proceedings. Each mine is obligated to assess the unique local physical
setting and expected mine disturbances in developing a closure plan that meets the
established goals.
Case histories presented in these proceedings provide valuable information to enable mining
companies to replicate and even improve on successes at other mines. Case histories of
failures will help readers avoid repeating the mistakes of others while successes will help
readers emulate the beneficial practices of others.
Without effective regulatory systems the lessons of the past and the availability of state-of-
the-art technologies might not by themselves result in excellent mine closure and
reclamation. Several papers concern the development of appropriate regulatory systems to
encourage and, if necessary, enforce mine closure planning and execution. These proceedings
should encourage the desirable trend towards regulatory consistency across international
boundaries.
Technical procedures and guidelines for effective mine reclamation are covered in many of
the papers in these proceedings. These technical “guidance” papers will contribute to the
development of a global “standard” that may define an improved state-of-practice in the
future.
Perhaps the most effective factor in driving mine closure excellence is the motivation of senior
mining company executives, a motivation that is rooted in corporate values related to
community engagement, respect for aboriginal stakeholders, minimisation of the
environmental footprint, long term sustainability and local economic benefits. Accordingly,
several plenary sessions pertain to corporate social responsibility.
Stakeholder engagement in the mine closure planning process and the development of
sustaining economic benefits to local land owners and aboriginal land owners are highly
relevant topics of the Mine Closure 2015 conference, both in terms of papers presented,
plenary sessions and panel discussions. Chief Clarence Louie of the Osoyoos Indian Band in
British Columbia addresses this important topic. He encourages mining companies to build
relationships and take advantage of valued input from local stakeholders and aboriginal
landowners as a prerequisite to meaningful mine closure planning.
Finally, there is the difficult issue of integrating mine operations into closure planning. Mining
staff who are responsible for developing the mine closure landforms and for reclaiming to
mine disturbed land, commonly witness the fact that the work of mine closure planners is
often disregarded in favour of more “practical” treatments. Mine operators complain that the
xvi
plan cannot be easily constructed, that it is not practical. The challenge is to involve the mine
operators in the closure planning process. Several papers herein give some guidance on this
issue.
We are pleased with the high quality of papers that were submitted and are printed in this
volume. We congratulate the authors for their valued efforts and we encourage readers to
take advantage of the results of these efforts as presented herein. Equally, we trust that all
conference attendees and everyone who reads these papers, understand that conference
presentations and associated papers are important steps towards a goal that will require
many more. We must continue to challenge the state of global mine closure as we strive for
community acceptance, environmental excellence and economic well-being. We need to
understand that excellent mine closure is a prerequisite to maintaining social license for
continued mine operations.
Les Sawatsky, P.Eng., M.Sc. Michael Davies, P.Eng., Ph.D.

Conference Chair Conference Co-Chair
Senior Consultant, Golder Associates Ltd. Vice President – Environment,
Teck Resources Limited
xvii
The Organising Committee acknowledges with gratitude the authors’ contributions of high quality,
detailed and innovative papers. Their papers reflect ingenious and proactive solutions to challenging
mine closures around the world.
We would also like to thank the technical reviewers, the employees of InfoMine and all those involved
in the organisation of this conference and the preparation and production of these proceedings. The
support of the committees is greatly appreciated. We especially wish to thank our chairman, Les
Sawatsky, and his co-chair, Michael Davies.
The Organising Committee also wishes to thank all of our exhibitors, sponsors, institutional partners
and media partners. These are all listed on the following pages.
Finally, we would like to thank all the delegates who attended the conference to exchange their
valuable knowledge and expertise, thus contributing to the great success of the 10th International
Conference on Mine Closure.
ORGANISING COMMITTEE
10th International Conference on Mine Closure
xviii
This list was correct as of May 22, 2015
xix
This list was correct as of May 22, 2015
xx
Mine Closure 2015 – A.B. Fourie, M. Tibbett, L. Sawatsky and D. van Zyl (eds)
© 2015 InfoMine Inc., Canada, ISBN 978-0-9917905-9-3
D.E. Hockley SRK Consulting (Canada) Inc., Canada

L.C. Hockley AJBL Consultants Ltd., Canada
Mine closure, the idea. Where did it begin and how did we develop our current understanding of the
associated technical, environmental and social challenges? This paper will present three histories of mine
closure, developed using different methods and leading to somewhat different conclusions.
The first history examines societal factors that contributed to the early ideas about mine closure. It uses the
normal methods of historical research, specifically review of primary and secondary written sources, and
focuses on how both competing resource uses and the development of environmental movements led to early
mine closure legislation.
The second history addresses scientific and technical developments related to mine closure. It uses a method
known as bibliometric mapping. Over 4,000 technical references related to mine closure were collected and
analysed by a series of text-mining, statistical and graphical techniques. The results show the growth in the
field as a whole from the 1960s to the present day, and also the shifts in emphasis from reclamation to more
holistic closure, from the major element composition of affected waters to various minor and trace elements,
and from bio-physical and chemical effects to socio-economic concerns.
The third history reviews the development of mine closure practices over the last twenty years. It draws on
concept mapping exercises completed by mine closure experts in 1998 and again in 2006. It shows the
transition from isolated specialties to a mature discipline, complete with higher level management processes,
continuing growth in technical fields, and new interests in stakeholder and sustainability issues.
Even these three approaches together do not constitute a complete history. But perhaps they will enrich our
understanding of how the idea of mine closure has changed and continues to change over time and in different
parts of the mining industry.
The website for this conference lists approximately 110 accepted papers. Judging by titles only, over half of
them will provide at least a partial history of a mine closure project. But most of us do not think of our work
as history. Even the case studies typically include only a summary of facts and dates, just a bridge to get us
quickly to what we see as worthy of discussion with our peers. The main focus of each paper is typically a
technology, or more broadly a methodology, that we hope will be relevant to other mine closures.
As we will show later, the presentations over the next few days will add to a body of mine closure literature
that now includes over 4,000 papers. There are also over 500 delegates at this conference. If each of us can
name ten colleagues or associates who work in this field but are not here, the population of mine closure
practitioners numbers in the thousands. And as some of your papers will make clear, mine closure is a subject
of keen interest to many other stakeholders, including environmentalists, regulators, investors, mine
employees and of course affected communities.
Should we continue to believe that our work in mine closure is purely the story of ever improving
methodologies? Certainly many individuals with a purely technical or socio-economic focus continue to make
very significant contributions to mine closures. The remainder of this talk will seek to advance the question
of whether there really is anything more than that.
Mine Closure 2015, Vancouver, Canada 3

Keynotes and Plenaries D.E. Hockley et al.
If we accept that mine closure is a historical phenomenon, it is appropriate to apply the normal methods of
historical research, specifically review of primary and secondary written sources. But it is important to
recognise that our topic is an idea, the idea of mine closure, rather than a particular event or series of events.
It is in the nature of ideas that they are conceived, possibly independently by several individuals, and then
communicated, understood, misunderstood, debated and further developed by many other people. A
significant additional complication is that the idea of mine closure has developed differently by region and
by industry sector. We will not succeed in writing the comprehensive history of all that. Instead we will
attempt to present a history, and hope that future efforts will fill in any critical themes that we miss.
The history that we present focuses on the United States between roughly 1870 and 1980. That period is
bounded by the earliest recorded legal dispute about mining impacts on the surrounding environment, and
the rapid development of federal and state mine closure legislation.
Even the earliest documents about mining indicate an appreciation for what we would now call
“environmental impacts.” Agricola’s De Re Metallica, first published in 1556, included statements such as the
following:
When the woods and groves are felled, then are exterminated the beasts and bird, very many of
which furnish a pleasant and agreeable food for man. Further, when the ores are washed, the
water which has been used poisons the brooks and streams, and either destroys the fish or drives
them away. Therefore the inhabitants of these regions, on account of the devastation of their
fields, woods, groves, brooks and rivers, find great difficulty in procuring the necessaries of life.
But efforts to do something about these impacts appear to be a relatively recent phenomenon. In the United
States the late nineteenth and early twentieth centuries saw an increase in the use of larger-scale mining
methods like hydraulic mining, dredging, and strip-mining. At the same time the growth of communities and
other uses of the land meant that mines had direct impacts on landscapes and water uses valued by others.
The most important and vocal of these others in the early years were farmers whose lands were directly
disrupted by mining.
The legal dispute that arose from hydraulic mining of gold in California’s Sacramento Valley offers a well-
documented example. In the 1870s and 1880s, the damage caused to farmland became so great that it
prompted farmers to take legal action against the miners. It was the first legal case of its nature in the US,
and quickly escalated into acts of bribery, sabotage, and violence. Legislation intended to create a
compromise set up a tax funded program to reclaim the affected rivers and help in the construction of dams
and levees. However, it failed to resolve the conflict and, after several further court processes, the issue had
to be settled by a federal judge in 1884. The farmers, who would settle for nothing less than a ban on
hydraulic mining, won a clear victory when the federal court ruled that discharge of tailings into the river was
a “destructive public and private nuisance” and, ultimately, unlawful (Kelley, 1959).
The situation in Sacramento Valley is indicative of most early opposition to mining impacts. It was not
motivated by the environmentalism that we know today, but by competition over resources. There were
debates about the moral obligation of a miner to avoid damaging the property of others and whether or not
the prosperity of a few miners should outweigh the livelihood of several hundred others. But fundamentally
the farmers went to court to protect their property rights and their livelihood (Kelley, 1959).
Early efforts to reclaim closed mines were underway by the early twentieth century. By the 1920s articles
were being published on the scientific reclamation of the strip-mined lands (Croxton, 1928; McDougall,
1925). But by the late 1930s, reclamation was still largely experimental and limited to the planting of trees
(Plass, 2000; Holmes, 1944). And where reclamation of closed mines was substantial, it was entirely focused
on making the land economically useful (Morrissey, 2010).
4 Mine Closure 2015, Vancouver, Canada

Keynotes and Plenaries
The Appalachian coal sector in the eastern United States provides a good example of how mine closure
transitioned from being the concern of a few directly affected individuals to an issue of national importance.
Surface coal mining in Appalachia began in the late 1800s, but had limited impacts. The introduction of steam
shovels, then electric shovels and draglines, led to a fivefold expansion in “strip mining” over the period 1900
to 1930 (Montrie, 2003).
As with prior disputes, the conflict over strip mining began with directly impacted farming communities. In
this case however, they were quickly joined by others whose livelihood was threatened, specifically
businessmen and deep-coal miners who together argued that strip mining undermined the regional tax base
and put deep miners out of work. The efforts of these groups did culminate in State legislation requiring
reclamation of mine-impacted lands. West Virginia was first in 1939, followed by Indiana in 1941,
Pennsylvania in 1945, Ohio in 1948, and Kentucky in 1954. Each state implemented its own set of regulations,
but common requirements were that the land be left smooth enough for agricultural tillage, that miners post
financial security to cover the cost of reclamation, and that there be penalties for violations. Unfortunately
enforcement of the State regulations was variable, to put it mildly, and actively compromised in some states
(Montrie, 2003; Plass, 2000; Brooks, 1966).
There were also those who defended strip mining and rejected calls for regulation. A 1940 article explained
how resourceful land owners had successfully transformed strip-mined land into productive uses. In one
example the land owner created a successful hunting ground, while in another a recreational park was built,
both without any substantial reclamation by the coal company (Hall, 1940). Concerns over the economic
impacts of excessive regulation were also a common theme (Brooks, 1966).
In the 1960s, grassroots opposition to Appalachian strip-mining strengthened. Farmers, local businessmen
and deep miners were joined by conservationists and sportsmen, or environmentalists and recreational
hunters and fishers in today’s terminology. Well organised cells of opposition emerged, many of them calling
for an outright ban on strip mining. Environmentalists wrote articles, both academic and popular, on the
need for further regulation (Mink, 1976; Udall, 1979). Hunting and fishing groups, often including influential
public figures, were supported by national entities like the Sierra Club. As the movement gained steam, calls
for outright abolition of strip mining became common (Montrie, 2003).
There had been occasional calls for federal regulation of the Appalachian coal sector since the 1940s, but the
pressure for federal action reached a peak in the early 1970s. State laws were no longer seen as adequate.
They were insufficiently enforced, caused unfair competition across states, and failed to distinguish between
private, state, and federally owned land (Schechter, 1980). The first significant discussion of federal
legislation took place before the House Interior Committee in 1971. A surface mining control bill was
introduced in 1973, but vetoed by President Ford. A similar bill was approved by Congress in 1975, but again
vetoed by President Ford. Ford justified his vetoes with arguments that the act would cause a loss of jobs,
raise consumer prices, make the US more reliant on foreign oil, and unnecessarily reduce coal production
(Mink, 1976).
In 1977, the new President Carter passed an updated version of what is now known as the Surface Mining
Control and Regulation Act or SMCRA. But the successful passage of SMCRA in 1977 was not simply a result
of presidential change. An additional factor was the willingness of its supporters to lessen their demands.
Leaders of the pro-regulation movement dropped calls for banning strip mining and settled for a more
realistic compromise. Therefore the bill passed by Carter was significantly less restrictive than those vetoed
by Ford (Montrie, 2003). Nonetheless it included many of the components of modern mine closure
legislation, including requirements to restore land to pre-mining conditions, protect water courses and water
quality, and post financial security.

Ending this historical review with SMCRA is not intended to imply that it represents the last stage of legislative
interaction with the idea of mine closure. On the contrary, the 1970s, 80s and 90s saw the development of
mine closure legislation in many other jurisdictions. By the early 2000s review articles could list nine western
US states, nine Canadian provinces and territories, and 45 other countries that had some form of mine closure
legislation or regulations (Berger, 2002; Clark and Clark, 2005). Requirements and enforcement varied
significantly, but there was enough common ground that it was possible to talk in terms of clear trends and
generally agreed principles (MMSD, 2002).
Perhaps more importantly, this brief history illustrates how the interplay between increasing mining impacts,
competition from other land and water users, and broader environmental concerns has shaped modern ideas
about mine closure. We believe this process has parallels in many other regions and mining sectors.
Histories such as the above allow insights about the early development of mine closure as a concept, but how
do we track development of an idea that has since come to involve hundreds or thousands of practitioners?
One option is to resort to meta-analysis using methods that are being developed in the fields of bibliometrics
and text mining. In general these methods seek efficient ways to identify patterns or major themes in large
bodies of written material.
To apply this approach to the idea of mine closure, the authors gathered over 4,000 articles published over
the period from 1900 to 2010. Roughly half of the articles were published in refereed journals and half in
conference proceedings. Numbers were not statistically significant in many of the early years, so analysis was
focused on the period after 1960. However, to link back to the previous section, it is worth noting that over
170 articles on mine reclamation and over 200 on acid mine drainage were published before 1960.
The first step in the text mining was to assemble the “corpus” or body of documents. The library databases
GEOBASE and GeoRef proved to be good sources for journal articles, but many conference proceedings
needed manual searching and input. The reference management programs RefWorks and Zotero were used
to organise the assembled results into author, title, publication information, abstract, and key word fields,
and remove any duplicates arising from the different searches. The software WordStat (Provalis Research)
was used to transform the text fields into data that could be analysed by statistical methods. The statistics
were then further analysed using MS-Excel and a clustering and visualisation tool known as VOSviewer (Van
Eck and Waltman, 2007; Waaijer et al., 2010).
Figure 1 shows the most basic statistic, namely the number of mine closure articles published per year. The
raw data is quite variable, so the smoothed curve gives a better perspective. The number of papers in the
scientific literature as a whole is estimated to double every 15 years. The number of articles related to mine
closure grew significantly faster than that, reflecting the very rapid growth of the field, especially through the
1990s.

300
Total in database
250
5-Year Averages
Publications per Year Doubling every 5 years
200
Doubling every 15 years
150
100
50
0
1950 1960 1970 1980 1990 2000 2010
Year
Figure 2 shows another interesting trend, the shift over time from use of the term “reclamation” to “closure”
and more recently “remediation.” There are at least two underlying reasons. First the field has clearly shifted
away from a focus on land reclamation to a more balanced consideration of physical, chemical and socio-
economic factors that are incorporated in the term “closure.” Second, increasing emphasis on chemical
contamination has led to mine sites being thought of as problems needing “remediation.” External factors
also contribute to these trends, such as the development of more holistic mine closure guidelines in some
areas and contaminated sites legislation in others.
100%
90%
Percent of Publiications using Term
80%
70%
60%
50%
REMEDIATION
40%
CLOSURE
30% RECLAMATION
20%
10%
0%
1960
1964
1969
1973
1976
1979
1982
1985
1988
1991
1994
1997
2000
2003
2006
2009
Figure 3 shows the frequency of contaminant names in the mine closure literature. For clarity, only a few
contaminants are shown. But the full data show a growing interest in major element chemistry and acidity in
the 1980s, followed by metal contaminants in the mid-1990s, and continuing through progressively more
“exotic” minor metals and oxyanions in recent years.

25
IRON
20 COPPER
LEAD
ARSENIC
Publications per Year
15
SELENIUM
ANTIMONY
10
0
1970 1975 1980 1985 1990 1995 2000 2005 2010
Year
All of the above examples arise from analysis of individual factors. The true power of text mining is clearer
when two-factor and multi-factor interactions are examined. For example, the two-factor analyses of
location and commodity type show the growing regionalisation of mine closure issues. Papers about the
closure of precious metal mines are widely distributed geographically, but papers on the closure of coal mines
are primarily from the US east and midwest, papers on iron and aluminium mine closures are primarily from
Australia, and papers on uranium mine closures are primarily from Canada and Europe. In most cases these
patterns reflect the historical development of mining in these regions, but large projects that generate many
papers also skew the numbers.
Figure 4 presents the results of a multi-factor cluster analysis. It displays individual papers as numbered dots
positioned so that more similar papers are closer together, and with a coloration to indicate the main
groupings. This result and many others like it indicate dominant themes associated with the idea of mine
closure, and how they change over time.

One weakness of bibliometric methods is that they look at quantity of publications and do not easily account
for differences in quality. For mine closure, there has been exponential growth in the number of papers being
published, but has there been exponential growth in knowledge? Here we use the term knowledge to imply
something that is greater than data or information, and capable of being put to practical use.
A measure of mine closure knowledge can be gained by polling experts in the field. In 1998, SRK gathered
twenty senior professionals from around the globe to create a “mine closure knowledge map” that sought
to represent all of the areas of expertise related to the practice of mine closure. That exercise was repeated
in 2006 and the two resulting knowledge maps provide another set of insights into development of the idea
of mine closure.
Very briefly, the knowledge maps were constructed in facilitated workshops. In the heart of each workshop,
subject matter experts selected a topic related to mine closure, gave it a simple name and a one-sentence
definition, and then created lists of inputs, outputs, typical subtasks, and example projects. Table 1 shows a
typical topic description. The topic descriptions were then grouped and sub-grouped, and assembled into a
two-dimensional map.
The initiative started without the assistance of professional knowledge managers, but the approach meets
most formal knowledge management criteria. Specifically the mapping is a form of classification system or
taxonomy, the descriptions represent a vocabulary, together with the logical input-output relationships they
form an ontology, the project examples serve as a bibliography, and the project contact lists create an expert
directory. As SRK has been one of the most active consulting groups working in mine closure during this
period, the maps represent a reasonable perspective on the development of mine closure knowledge.
Content type Example content

Topic Pit backfilling
Backfilling of waste rock into a pit to reduce surface impacts, place
Summary description potentially acid generating material below the water table, and/or
prevent formation of a pit lake
Geochemical characterisation of waste rock
Key input links Pit volume capacity curves
Site water balance and site hydrogeology
Backfill volumes
Requirements for alkalinity addition or other treatment
Key output links
Control program, including long term planning, short term …
Water quality prediction …
Assess geochemical characterisation of waste …
Subtasks
Design program for short term control of backfilling ...
Flambeau Mine, Wisconsin – Project F107108
Example projects
Lichtenberg Pit, Germany – Project W104108
Daryl Hockley, Vancouver
SRK Contacts
John Chapman, Brisbane
Of most interest for this paper is how the mine closure knowledge map has changed over the period since
1998. The rapid development of the field required that the entire map be re-organised in 2006, but it is

convenient to use the “continents” of that map and ask what “regions” were added or heavily modified
between 1998 and 2006.
At the highest level of the current map, the categories are closure strategy and planning, closure
management, closure methods and technologies, and closure investigations. None of these are completely
new, but two of them were not initially considered high level categories. By 2006, closure strategy and
planning included many sub-categories, ranging from the development of corporate closure policy to the
assessment of methods to involve stakeholders in closure decisions. In broad terms these are higher level or
second order functions that typically accompany the maturation of a complex field. The closure management
category was also significantly expanded by 2006, and now includes a range of systems that are typical of
large project management in other fields. Many of those systems were put in place as a response to problems
encountered when ambitious closure plans first began to be fully implemented.
Changes at lower levels of the map are too numerous to list. A clear pattern is the increasing number and
level of detail related to investigation methods. Significant advances are also apparent in areas driven by
external pressures. A good example is the improvement in methods for estimating mine closure costs, driven
in part by recognition that closure costs are material, but also by changes to both financial security
requirements and international accounting practices. A second example is the significantly greater emphasis
on communities, both as participants in closure decisions and as a key factor in post-mining sustainability.
Perhaps surprisingly, the number of truly new closure technologies is quite limited, except perhaps under
the category of water treatment.
Another theme that is apparent in the map development is the extent of interaction with other disciplines.
Adjacent knowledge maps for fields like geochemistry and mine planning have always shared borders with
mine closure, but the extent of overlap has increased. Whether this is a reflection of a more holistic approach
to mine development or simply an artefact of growth in the various disciplines is one of the interesting
questions worthy of further debate.
These three analyses, a selective historical review, text mining of the technical literature, and review of
knowledge mapping by expert groups, paint an admittedly incomplete picture of the history of mine closure,
the idea. Despite these limitations, the methods provide a number of insights into how the concept has
changed, and continues to change today. A few broad patterns that emerge are:
 Requirements for mine closure became a subject of contention in the late nineteenth and early
twentieth centuries, in particular where the scale of mining impacts began to threaten other uses
of land and water.
 Continuing growth in the scale of mining, the weakness of initial regulatory efforts, and increasing
environmental awareness led to the development of comprehensive mine closure requirements
in the 1960s and 1970s.
 Mine closure legislation spread around the globe in the 1980s and 1990s, and the range of topics
considered part of mine closure expanded from surface reclamation and water use to include
much broader environmental protection and remediation.
 Technical literature related to mine closure experienced a boom in the 1990s and early 2000s, and
the field took on many of the characteristics of a mature discipline, including the development of
guidelines, procedures and higher level management processes.
 The range of considerations, implications and requirements related to mine closure continues to
expand, with notable recent examples being the increased attention given to stakeholder input
and the sustainability of mining-affected communities.

The authors gratefully acknowledge the assistance and advice provided by several colleagues at both SRK
and AJBL.
Agricola, Georgius (1556) De Re Metallica. Translated from the first Latin edition of 1556 by H.C. Hoover and L.H. Hoover, Project
Gutenberg, November 2011.
Berger, Alan (2002) Reclaiming the American West, New York: Princeton Architectural Press.
Brooks, David B. (1966) Strip mine reclamation and economic analysis, Natural Resources Journal, Vol. 6(1), pp. 13–44.
Clark, Allen L. and Clark, Jennifer Cook (2005) An international overview of legal frameworks for mine closure, available on line from
Environmental Law Alliance Worldwide, viewed www.elaw.org
Croxton, W.C. (1928) Revegetation of Illinois coal-stripped lands, Ecology, Vol. 9(2), pp. 155–175.
Hall, Harry H. (1940) The romance and reclamation of the coal lands of southeastern Kansas, Transactions of the Kansas Academy of
Science, Vol. 43, pp. 57–67.
Holmes, Leslie A. (1944) Reclaiming stripped land in Illinois, The Scientific Monthly, Vol. 59(6), pp. 414–420.
Kelley, Robert Lloyd. (1959) Gold vs. grain: the hydraulic mining controversy in California’s Sacramento Valley. Glendale, USA: Arthur
H. Clark Company.
McDougall, W.B. (1925) Forests and soils of Vermilion county, Illinois with special reference to “Striplands,” Ecology, Vol. 6(4),
pp. 372–379.
MMSD (2002) Research on mine closure policy. Mining minerals and sustainable development report No. 44 for Cochicl, Chilean
Copper Commission.
Mink, Patsy T. (1976) Reclamation and rollcalls: the political struggle over strip mining, Environmental Policy and Law, Vol. 2(4),
pp. 176–180.
Montrie, Chad (2003) To save the land and people: a history of opposition to surface coal mining in Appalachia, Chapel Hill, USA:
University of North Carolina Press.
Morrissey, Katherine G. (2010) Rich crevices of inquiry: mining and environmental history, in D.C. Sackman (ed), Companion to
American Environmental History, Blackwell Publishing Ltd., pp. 394–409.
Plass, William T. (2000) History of surface mining reclamation and associated legislation, in Richard I. Barnhisel, Robert G. Darmody
and W. Lee Daniels (eds), Reclamation of drastically disturbed lands, American Society of Agronomy, Madison, USA, pp. 1–
20.
Schechter, Mollie E. (1980) Surface mining control and reclamation act of 1977: its background and its effects, New York Law School
Law Review, Vol. 25(4), pp. 953–1000.
Udall, Morris K. (1979) Enactment of the Surface Mining Control and Reclamation Act of 1977 in retrospect, West Virginia Law Review,
Vol. 81(4), pp. 553–558.
Van Eck, N.J. and Waltman, L. (2007) Bibliometric mapping of the computer intelligence field, ERIM Report Series reference number
ERS-2007-027-LIS.
Van Eck, N.J. and Waltman, L. (2014) VOSviewer Manual, Manual for VOSviewer version 1.5.4, Universiteit Leiden, January 2014.
Waaijer, C.J.F., Van Bochove, C.A. and Van Eck, N.J. (2010) On the map: nature and science editorials, CWTS Working Paper Series
paper number CWTS-WP-2010-001.

M. Davies Teck, Canada
Teck has been discovering and operating mines for 100 years. Our people live in the areas where we operate,
and we care about doing the right thing for our community, the environment and future generations. To guide
all of our actions, Teck established a comprehensive strategy with goals that stretch through to 2030 in areas
that represent the most significant challenges and opportunities facing our company in the area of
sustainability. One such area is water, which is fundamental to our ability to operate and to which we ascribe
goals around use intensity, water quality and fair access for all potential users. Further to Teck’s commitment
to sustainability is our formal and robust management approach to our reclamation activities and our
dormant/legacy properties. Teck’s approach to water, reclamation and closure are briefly described as a
window to our sustainability practices and how Teck is establishing the basis for operating for the next 100
years.
1
Please note that this presentation is available only as an abstract; there is no full paper.

C. Digby Centre for Sustainability in Mining and Industry, University of the Witwatersrand, Johannesburg
D. Limpitlaw Centre for Sustainability in Mining and Industry, University of the Witwatersrand,
Johannesburg
The Proceedings of the Mine Closure conferences from 2006 to 2014 represent a valuable collection of
knowledge on mine closure and the evolution of closure practice over the past decade. There have been nearly
700 papers presented at the ten conferences held in Australia, Canada, South Africa, Chile, and the UK. Subject
matter ranges from detailed technical papers on cover design for tailings storage facilities to inclusive
approaches required to facilitate the participation of first nations people in closure planning. There are site-
specific case studies outlining closure plans and recording closure outcomes relating to revegetation, water
management, and land and infrastructure use. There are contributions from regulators relating to policy and
regulatory issues and papers on public sector interventions dealing with orphaned and abandoned mines that
can help to inform today’s closure practice.
However, to date these papers have been largely inaccessible, available only in the hard copy proceedings
produced for each conference. The intention of this presentation is to open up the collection of papers and to
promote the use and application of the knowledge contained in the series. We report on an assessment of the
collective output of the Mine Closure series and attempt to show what the series has contributed and why it
is important. The assessment identifies areas where knowledge has evolved and contributions that have stood
the test of time.
2

E.J. Gardiner NOAMI, Canada

W.R. Cowan Cowan Minerals Ltd., Canada
G.A. Tremblay Natural Resources Canada, Canada
C.M. Hogan Natural Resources Canada, Canada
The National Orphaned/Abandoned Mines Initiative (NOAMI) was launched in 2002 in Canada in response to
a request by federal, provincial and territorial mines ministers. It is a national multi-stakeholder initiative that
addresses issues related to orphaned and abandoned mines in Canada. One of the guiding principles of
NOAMI is that “work toward eliminating future abandonment must continue, including the tightening of
regulatory approaches.” Recently, NOAMI discovered that while processes of closure planning and the
provision of financial assurance are well-developed and consistently applied across Canada, policy and
regulatory approaches focusing on long-term management of sites beyond closure, including potential
methods of returning mining lands to the Crown, are almost non-existent. This paper will describe various
initiatives undertaken by NOAMI during the past four years that explore this issue in detail and offer
recommendations for change.
A 2010 NOAMI report entitled “The policy framework in Canada for mine closure and management of long-
term liabilities: a guidance document” presents a policy framework and recommendations for preventing
further accrual of abandoned mine liabilities. Subsequently, a 2011 multi-stakeholder workshop, Exploring
the Management of Long-term Liabilities and the Return of Mining Lands to the Crown, resulted in the
development of a roadmap for managing long-term liabilities. In 2013, Cowan Minerals Ltd. produced a report
for NOAMI based on a two-part study. The first part examined six case studies from different Canadian
jurisdictions that describe closed sites that either were returned to the Crown or were in the process of making
such an application. The second part, a decision tree or process, identified key issues and questions that need
to be addressed in order to determine whether a site should be brought under government jurisdiction or
remain the responsibility of the operator. The report, “Case studies and decision-making process for the
relinquishment of closed mine sites” (2013), lays out a five-step approach for regulators and industry to
consider when determining if a site could, or should, ultimately be returned to the Crown. Finally, in 2014,
NOAMI commissioned a study that describes key criteria for the effective long-term stewardship of closed
mine sites. The purpose of the key criteria is to provide a summary of site aspects that will aid the user in
identifying, analysing, and evaluating potential site hazards, including those that may pose a risk to public
health and safety, to the environment, to ecosystem services and to future land use.
The legacy of orphaned and abandoned mines, including environmental liability, human health concerns, and
the cost of clean-up and long-term monitoring and maintenance is a serious issue facing all Canadians.
Orphaned or abandoned mines are those for which the owner cannot be found, or is financially unable or
unwilling to remediate the site. These mines can pose environmental, health, safety and economic problems
for communities, the mining industry and governments in many countries, including Canada.

The National Orphaned/Abandoned Mines Initiative (NOAMI) was formally established in 2002 at the request
of the federal, provincial and territorial mines ministers in Canada. The groundwork for the initiative was laid
out in 1999 when a number of stakeholders put forward requests to the Canadian mines ministers to
establish a multi-stakeholder working group to review the issue of orphaned and abandoned mines. Ministers
supported the establishment of an industry-government working group that would examine the steps to be
taken to facilitate action in this area. They requested that a multi-stakeholder workshop be organised to
identify key issues and discuss priorities for action.
This workshop, which was held in Winnipeg in 2001, determined the key issues associated with
orphaned/abandoned mine sites, identified common ground among various communities of interest, and
identified processes and procedures for moving forward. Operating principles and a series of guidelines were
laid down at the workshop, which ultimately would apply to NOAMI as it exists today.
NOAMI is guided by a multi-stakeholder advisory committee that brings together representatives from the
Canadian mining industry, federal, provincial and territorial governments, non-governmental organisations
and Aboriginal Canadians. Together, they assess issues and make recommendations for collaborative
implementation of remediation programs and policies for orphaned and abandoned mines across Canada.
The NOAMI Advisory Committee takes direction from the federal, provincial and territorial mines ministers
and in turn, reports progress annually to the Energy and Mines Ministers Conference. NOAMI’s activities are
jointly funded by the federal, provincial and territorial governments, the Mining Association of Canada and
the Prospectors and Developers Association of Canada and are administered by a secretariat at Natural
Resources Canada. NOAMI does not directly clean up orphaned and abandoned mine sites. Rather, it
examines the legislative, policy and program framework in Canada for addressing issues associated with
orphaned and abandoned mines, and makes recommendations for improvement. A pan-Canadian effort,
NOAMI has made tremendous progress in more than a decade, in fulfilling this mandate.
One of the key guiding principles of NOAMI is: “Work toward eliminating future abandonment must continue,
including the tightening of regulatory approaches.” In 2009, the NOAMI Advisory Committee, recognising
that there is a policy void in the area of long-term closure, began to examine the legislative tools and policy
approaches across Canada to ensure that current operating mines can be closed properly so that they do not
become abandoned mines in the future. Related to this, members of the committee have long believed that
there is a need for a clear policy framework for mine closure, long-term liabilities and return of mining lands
to the Crown. It is believed that such a framework would address ultimate closure of mine sites and long-
term management and related liabilities, in a manner that clearly sets out the responsibilities of the mine
owner/operator and government regulatory agencies. The policy framework would also examine the
questions surrounding the transfer of mining lands back to the Crown through the issuance of a release,
including questions of when and under what conditions such a return would be appropriate (Holmes and
Stewart, 2011).
In 2010, Cowan Minerals Ltd. of Sudbury was commissioned to conduct a survey that found that while
processes of closure planning and provision of financial assurance are well-developed and consistently
applied across Canada, policy around long-term management of sites beyond closure, including methods of
returning mining lands to the Crown, is almost non-existent. Cowan Minerals proceeded to produce a report,
“The policy framework in Canada for mine closure and management of long-term liabilities: a guidance
document” (Cowan et al., 2010). The document examined major components related to mine closure and
post-closure site management, which can include long-term maintenance and monitoring, financial
assurance, relinquishment and institutional care. A valuable reference tool, the report presents a policy
framework, together with recommendations for preventing further accrual of abandoned mine hazards. A
key message of the report is that jurisdictions should have a managed relinquishment process, which is clear
and unfettered and is specific about what will not be accepted. The report notes that closure plans are

normally prepared on a “design for closure” basis, and it suggests that a more forward-looking approach be
embraced in the form of a “design for relinquishment” (Cowan et al., 2010).
Building on the “Cowan Report” of 2010, a multi-stakeholder workshop, Exploring the Management of Long-
term Liabilities and the Return of Mining Lands to the Crown, was held in 2011 (Tunis and Associates, 2011).
Advice and guidance was obtained from representatives of various communities of interest that assisted
NOAMI in developing a roadmap for managing long-term liabilities and issues relating to the return of lands
to the Crown. The themes of risk management, funding and legislation, and policy and regulation are
discussed fully in the workshop proceedings, which are available on the NOAMI website,
http://www.abandoned-mines.org. While several recommendations were formulated by the workshop
participants, a key recommendation was that NOAMI create a decision tree for the return of mining lands to
the Crown.
Following the workshop, the NOAMI Advisory Committee held its Annual General Meeting, which included a
strategy planning session to analyse the results of the workshop and to develop the next steps for the return
of mining lands to the Crown project.
The committee believed that the creation of a decision tree for considering the return of mine lands to the
Crown would be a useful tool, and would provide information and options for Canadian jurisdictions to
consider in planning to develop legislation or to revise current legislation. Case studies have been used in the
NOAMI program to examine factors involved in successful approaches for dealing with orphaned and
abandoned mines, such as funding, regulatory and policy frameworks, and engagement with communities.
Examination of several mine sites in Canada that were successfully returned to the Crown would provide
information on processes involved and some lessons learned. The information collected from the case
studies, along with other relevant documents, could be used to establish the decision tree.
Based on these discussions, a new Return of Mining Lands to the Crown Task Group was formed in 2011, and
a research plan was developed for the next steps in the return of mining lands project.
Cowan Minerals Ltd. conducted a two-part study, Case Studies and Decision-Making Process for the
Relinquishment of Closed Mine Sites. The first part examines case studies from different Canadian
jurisdictions that contribute information towards relinquishment. The second part, a decision tree or process,
identifies key issues and questions that need to be addressed to determine if a site should be brought under
government jurisdiction, or remain the responsibility of the operator. A five-step approach was laid out for
regulators and industry to consider when determining if a site could, or should, ultimately be returned to the
Crown.
Initially, case studies were to be examined from different Canadian jurisdictions that describe closed mine
sites that are being returned, or have made application to be returned to the Crown. Upon further
investigation, few sites in Canada were found that were both available for study and would contribute
information towards the development of a decision tree (Cowan et al., 2013). To complete the study, the
criteria for case studies were amended to include closed mine sites that illustrated relevant information and
lessons learned towards their potential return.
Case studies were undertaken for sites in multiple jurisdictions: Quebec Lithium Mine, Quebec; Renabie Gold
Mine, north-central Ontario; Gregg River Coal Mine, Alberta; Contact Lake Gold Mine, northern
Saskatchewan; and Farley East Tailings Management Area, northern Manitoba. The sites displayed a diverse
range of information due to age of projects; changes in ownership, management and operational personnel;
regulatory regimes and reporting requirements; and variable accessibility to information (Cowan et al., 2013).

The six case studies gave rise to a series of lessons learned. The most salient are noted below (Cowan et al.,
2013):
1. Jurisdictions must have clear and well-written legislation and policies in place to facilitate the
return of lands. It is necessary to have a system to receive and manage funds to ensure the
viability needed to address long-term issues.
2. Where planning and procedures for relinquishment are not in place, defaulting will eventually
occur at taxpayers’ expense. This can occur in several ways including forfeiture of corporate
charters and seizure by municipalities for unpaid taxes.
3. Few jurisdictions have a satisfactory funding regimen in place to deal with surrendered, or about
to be surrendered lands.
4. Most jurisdictions providing for rehabilitation releases and/or relinquishment of lands do not have
complete releases for environmental responsibility or liability.
5. It appears that only some jurisdictions have institutional control plans for relinquished lands, i.e.
there is no administrative unit with direct responsibility for institutional issues following
surrender, e.g. records maintenance, land use planning inputs.
6. Public consultation requirements/efforts are generally considered inadequate relative to current
expectations, especially with regard to aboriginal communities. The mines reviewed for this
project came into production prior to comprehensive consultation becoming either an expected
practice or mandated by government. Consultation with community and First Nation Stakeholders
in the Lynn Lake area of Manitoba was well planned and carried out with regard to the Farley East
Tailings Management Area. Consultation requires good information on the issues and needs of
impacted communities.
7. Third-party involvement by concerned special interest groups and others is becoming common;
there is a lack of perceived government credibility.
8. The length of post-closure monitoring periods needs greater evaluation especially where
documentation/identification of features is lacking.
9. Well-defined risk assessment procedures for property returns are not available for many
situations. More information on quantitative risk assessment is required to support chemical and
physical stability determinations.
10. It appears that many technical assessments are accepted at face value and that peer review by
qualified persons is not commonly used. Peer review was used in the review of the Human Health
and Environmental Risk Assessment at Lynn Lake Manitoba.
11. Where cutting-edge technology is involved, longer-term monitoring and scientific assessment may
be expected, e.g. new schemes related to permafrost or climate change issues.
12. Effective cost-estimation procedures for long-term care and maintenance need development.
13. Storage and maintenance of mine plans and records are essential for technical assessment.
Several of our case studies were hindered by the loss of or unavailability of documentation.
14. Provision of a final closure report detailing all completed decommissioning and reclamation work
is an excellent best practice.
15. There is no protocol for the public to inform jurisdictions of observed reclamation issues.
As stated earlier, a principal recommendation of Cowan et al. (2010) was that:

… jurisdictions should have a managed relinquishment process, which is clear and unfettered and
is specific about what will not be accepted. Hitherto closure plans have been prepared on a
“design for closure” basis. It is suggested that a more forward-looking approach be embraced and
that a “design for relinquishment” approach be adopted.
The authors go on to state that it is not intended that relinquishment be the only option, but rather that it
be an important option or objective because of its role in protecting the public from the inadvertent
accumulation of abandoned mine environmental and financial risks (Cowan et al., 2013).
The simplest form of relinquishment would be for fully reclaimed sites that are physically and chemically
stable, and require little or no further monitoring, care or maintenance. More complex forms of
relinquishment would be needed for reclaimed sites that will require ongoing monitoring, care and
maintenance. Future care of these sites would need to be fully funded by the proponent and managed by
the jurisdictions through some form of institutional care. In general, relinquishment is unlikely for sites where
risks are too high, or if there are specific no-go conditions, such as water treatment (Cowan et al., 2013). One
important aspect of relinquishment is whether liability would be transferred or not. The proponent maintains
the liability in the Saskatchewan “institutional control program”, while under the Ontario “surrender by
agreement” legislation, if applied, the liability would be transferred to the jurisdiction.
To determine if a site would be acceptable for relinquishment, a decision tree or process could be put in
place that outlines key factors that need to be assessed or developed. As discussed at the NOAMI 2011
workshop, there are many elements to consider when developing a process for relinquishment. In addition,
each jurisdiction would have its own set of elements. However, many of the elements could be considered
at different stages of the process, and would be amenable to a step-wise approach for planning.
These steps would ultimately determine whether responsibility for the site should remain with the operator
or be transferred to the Crown. The five-step process developed by Cowan Minerals (2013) provides a starting
point for jurisdictions developing or revising a program for relinquishment. Although national consistency
would be beneficial and a desired goal, each jurisdiction would need to establish a decision- making process
that met its own regulatory regime and policies.
A five-step decision-making process has been created following the natural progression from “submitting the
application” through to “implementation” (Figure 1). For each step, consensus must be reached by the review
committee prior to moving to the succeeding step, i.e., before proceeding from one step to the next all “yes”
components must be met. It is suggested that an appeal process be available through every step.
The five-step decision-making process provides guidance on whether a mining property has been closed out
as per requirements and closure plan, and on whether technical and financial long-term monitoring and
maintenance needs have been addressed sufficiently (Cowan et al., 2013). The steps are briefly summarised:
1. Submitting the application: ensures that the proponent qualifies for relinquishment, specifically,
the closure plan has been implemented, closure and post-closure land use objectives are completed
or near completion, the site is physically and chemically stable, and existing permits are fulfilled or
transferable to the jurisdiction. All necessary parties, and impacted stakeholder and Aboriginal
peoples must be notified of the application.
2. Site assessment: locational and technical issues are evaluated to determine if the site is
appropriately reclaimed to meet legal requirements under the closure plan, and will meet proposed
future land use requirements. Any interim monitoring requirements must be identified for
evaluation.
3. The long term: requirements for long-term monitoring, maintenance or capital replacement of
rehabilitation works and associated cost estimating are evaluated. A peer-reviewed risk assessment
by qualified persons is required for unforeseen events. This is a critical point, as the proponent for
mines with prohibitively high funding requirements for relinquishment may elect to retain the
properties.

4. Funding: considers what future activities will need to be funded, their costs, and the method of
fund management. All funds are provided by the proponent.
5. Implementation: once a site is approved for relinquishment, the regulator must have a
management system in place to manage funds, ensure the site is monitored and maintained, and
manage data securely. Emergency protocols should be in place.
The report concludes that mining projects should be designed with the objective of reclaiming the site for
relinquishment and future beneficial use. It is clear that relinquishment will not be possible where the
environmental, social, political or financial risks/costs are too great. However, Cowan Minerals believes that
a well-designed and well-managed relinquishment policy and program can lead to a win/win situation in
many instances. The five-step decision-making process proposed herein should assist jurisdictions in
developing policies and procedures for relinquishment that reflect the regulatory environment of the
jurisdiction within the context of mining as an economic development instrument for the jurisdiction.
An effective long-term monitoring and maintenance program at closed/post-closure, orphaned/ abandoned

mine sites and mineral exploration sites presents a number of challenges. A strategy needs to be in place to

address the aspects involved in management of these sites. This would provide an additional tool that could
be utilised in the ongoing effort to eliminate future mine abandonments in Canada.
Kingsmere Resource Services Inc. has undertaken a study for NOAMI entitled “Key criteria for the effective
long-term stewardship of closed, orphaned/abandoned mine and mineral exploration sites” (2015) to
develop criteria to assess these sites in order to evaluate their condition. A list of site attributes was
developed, along with a review of national and international best practices of the management of post-
closure mine sites. The report notes that effective stakeholder engagement at every stage of the assessment
is important and can result in significant benefits.
The initial step toward long-term stewardship is to assess the current state of the site. The report provides a
summary of features that could exist at the site, with a focus on identifying the hazards. Before the
inspection, research needs to be carried out to collect information on the site, and a site visit safety plan
should be developed, containing potential hazards and mitigation controls, logistical details, and safety rules.
The site inspection aspects are numerous, and include site access and condition, mine type, facilities and
infrastructure, and physical and chemical hazards. To assist in this, a field report was created that can be
used to record observations during the site visit.
The next step is an assessment of the level of risks posed by the identified hazards. Various types of risk are
considered in the report: public safety risk and ecological and human health risk. Subsequently, a decision
can be made as to whether remediation is required, and to what level. Reaching consensus on the level of
risk posed by a hazard is a difficult process, and must take into account the concerns of all the stakeholders.
An effective stakeholder engagement plan must be in place to address site risks and possible remediation of
that risk. The next step is site remediation; a difficult stage, as a balance must be achieved between the level
of risk reduction desired by the stakeholders and the cost to achieve that level. The last step for long-term
stewardship involves data keeping of the activities on the site and regular inspections by qualified personnel.
To ensure the long-term stewardship of closed, orphaned and abandoned mine sites, an institutional control
program is recommended.
The return of mine lands project has produced several important tools and guidance documents that
contribute to the prevention of future abandoned mines. The report “The policy framework in Canada for
management of long term liabilities: a guidance document” provides a robust policy framework for mine
closure, with recommendations towards prevention of future abandonments. A key recommendation was
that projects should be designed for relinquishment – a higher standard. The report “Case studies and
decision-making process for the relinquishment of closed mine sites” lays out a five-step process for
jurisdictions to follow in moving through the steps to possible relinquishment. Consensus must be reached
at each step along the path. It is recognised that relinquishment may not always be possible due to
environmental, social, political or financial risks or costs. Regardless of whether a site remains with a
company or not, the company must provide necessary funding to carry out any required long-term
monitoring and maintenance.
An effective long-term monitoring and maintenance program at orphaned and abandoned mines sites and
post-closure/decommissioned mine sites is an important part of achieving the goal of effective long-term
stewardship. A list of criteria has been developed to assess these sites, in order to evaluate their condition
and provide direction for the planning and delivery of long-term stewardship. This will be an additional tool
to be utilised in the ongoing effort to eliminate future mine abandonments in Canada.
Taken together, these projects constitute important tools that will make a major contribution towards the
development of a pan-Canadian policy framework that would address all aspects of managing orphaned mine
liabilities in the long term, and preventing future abandonments. Recommendations from this work will be
reported in full to the federal, provincial and territorial mines ministers during the 2015 Energy and Mines
Ministers Conference.

One of NOAMI’s most important projects is the development of a national inventory of orphaned and
abandoned mines based on compatible inventories from each province and territory. The level of detail and
completeness of these inventories varies from jurisdiction to jurisdiction. A national inventory has been
designed that would introduce standardisation of information, provide a single-window, web-based access
to information, and facilitate the addition of more detailed information in the future. The feature-based
inventory has been completed for all jurisdictions across Canada, except for Quebec. The NOAMI Secretariat
is working with Quebec provincial staff to determine if the Quebec database can be incorporated into the
web portal. A beta test of the site is currently underway in preparation for its launch this year. Due to the
immense importance of this work, NOAMI determined the best venue to launch the inventory website would
be at the Energy and Mines Ministers Conference 2015, to be held in Halifax, NS in July.
NOAMI recognises the need for tools to help communities understand the legacy issues associated with
orphaned and abandoned mines, and to build their capacity to deal with these sites in an effective, practical
and meaningful way. NOAMI sponsored two training workshops organised by the First Nations of Quebec
and Labrador Sustainable Development Institute. The workshops will ensure that environmental
representatives of First Nation (FN) communities are familiar with the regulations and tools for rehabilitation
in Quebec, and identify best practices in the management of abandoned mines. The first workshop was held
in Val-d’Or, Quebec, on 17–18 December, 2014; the second is planned for 17–18 March, 2015 in Sept-Îles,
Quebec. The first workshop was a success, with twenty participants from different organisations, including
seven FN communities. Participation of FN communities in partnership-based approaches for the
rehabilitation of abandoned mine sites was discussed, with a presentation on the FondsRestor-Action-
Nunavik (FRAN). The FRAN project concept was of much interest; the First Nations participants eventually
would like to develop similar projects.
This pilot study will provide an opportunity to evaluate NOAMI findings at the First Nations community level,
and in turn, for NOAMI to share the results of these workshops with other communities across Canada.
NOAMI’s first Performance Report, covering the years 2002–2008, was a major undertaking and described
the activities and infrastructure of the initiative in considerable detail. It was widely distributed in Canada
and abroad, in English and French, and effectively promoted NOAMI itself, while showcasing the activities of
Canadian jurisdictions in the remediation of orphaned and abandoned mines sites.
NOAMI has made further progress in the last few years and plans to release a second performance report in
2015 that will cover the period 2009–2014. While the format is more focussed than that of the first report,
it will once again highlight both NOAMI’s major achievements, and the efforts of Canadian jurisdictions to
address the potential legacy issues associated with orphaned and abandoned mines across the country. The
report will be distributed at the Mines Ministers Conference 2015, as part of NOAMI’s continual commitment
to update Canadian mines ministers on activities across Canada on the status of orphaned and abandoned
mines.
Cowan, W.R., Mackasey, W.O. and Robertson, J.G.A. (2013) Case studies and decision making process for the relinquishment of closed
mine sites. Prepared for the National Orphaned/Abandoned Mines Initiative, viewed 26 February 2015, http://abandoned-
mines.org/publications-e.htm
Cowan, W.R., Mackasey, W.O. and Robertson, J.G.A. (2010) The policy framework in Canada for mine closure and management of
long-term liabilities: a guidance document. Prepared for the National Orphaned/Abandoned Mines Initiative, viewed 26
February 2015, http://abandoned-mines.org/publications-e.htm

Holmes, R. and Stewart, G. (2011) A guidance document for mine closure and management of long-term liabilities: examining a policy
framework in Canada, in Proceedings Sixth International Conference on Mine Closure, 18–21 September 2011, Alberta,
Canada, pp. 21–28.
Kingsmere Resource Services Inc. (2015) Key criteria for the effective long-term stewardship of closed, orphaned/abandoned mine
and mineral exploration sites, viewed 27 March 2015, http://abandoned-mines.org/publications-e.htm
NOAMI (2009) National orphaned/abandoned mines initiative 2002–2008 Performance Report, viewed 26 February 2015,
http://abandoned-mines.org/publications-e.htm
Tunis, E.S. and Associates. (2011) NOAMI Workshop Proceedings – Exploring the management of long-term liabilities and the return
of mining lands to the Crown in Canada. 31 May – 1 June, 2011. Ottawa, ON, viewed 26 February 2015, http://abandoned-
mines.org/workshops-e.htm

D.G. Hovdebo Kingsmere Resource Services Inc., Canada

K.E. Cunningham Saskatchewan Ministry of the Economy, Canada
D.M. Kristoff Saskatchewan Ministry of the Environment, Canada
M.S. Webster Cameco Corporation, Canada
Environmental practices and regulations have advanced significantly since Canadian mine operations were
developed in the early 1900s and now cover all aspects of construction and operation. Similarly, most
jurisdictions around the world now require mining operations to prepare closure plans and to post a bond or
other financial assurance of sufficient value to cover the cost of closure from the time a mine is first approved.
However, few jurisdictions have developed a formal institutional control management framework that
provides for custodial transfer and effective long-term stewardship of sites once the operator has fulfilled its
closure obligations and is eligible for release from further financial bonding (closed sites). Post-closure
management of such sites is an issue that has been identified by the public, industry, and government
stakeholders.
The Province of Saskatchewan successfully undertook the processes necessary to develop a formal, effective
institutional control program (ICP) that defines the conditions under which it will accept custodial
responsibility for closed mine sites and provide for the long-term stewardship of each site. The process was
undertaken by a working group led by a consultant and composed of representatives from the Executive
Council, Ministries of Environment, Energy and Resources, Northern Affairs, Justice, and Finance. It was
initiated in 2005 with a comprehensive assessment of policy and legislative requirements and the risks and
liabilities associated with undertaking such a program. This process also involved extensive consultations with
industry, Aboriginal traditional users, and other stakeholders in order to secure their input, inclusion, and
support. The entire process was completed with the promulgation of the Reclaimed Industrial Sites Act (RISA)
and associated regulations in 2007 and the formal acceptance of the first mine sites in 2009.
The Saskatchewan ICP addresses all aspects of conventional closed mines as well as the uranium-specific
issues of radioactive waste management, including those defined in the articles of the International Atomic
Energy Agency’s (IAEA) Joint Convention on the Safety of Spent Fuel Management and on the Safety of
Radioactive Waste Management, all applicable provincial acts and regulations, and the federal Nuclear Safety
and Control Act (NSCA). The program includes a formal, publically accessible registry and document
repository. It has been designed to be revenue neutral and sustainable and to ensure that future generations
are not burdened with the costs of the long-term monitoring and maintenance of former mine sites in the
province. In addition, the Saskatchewan Ministry of the Environment now includes the cost for transferring a
closed site into the ICP as a component of the financial assurance required for operating sites.
Since its inception, the Saskatchewan ICP has accepted one gold mine/mill and five uranium mine sites into
the institutional control registry and has completed the first of the requisite scheduled inspections of each
registered site. This case history describes the actions and process used to develop the program, discusses the
challenges encountered, and summarises the outcomes of the creation of a sustainable formal institutional
control management framework in Saskatchewan.

For more than two decades, the Government of Saskatchewan had consistently stated that, once the
operator of a mining facility had fulfilled its decommissioning and reclamation obligations and demonstrated,
by transition phase monitoring, that its site was chemically and physically stable, the government would
accept custodial responsibility of the property. The province’s development of an institutional control
program (ICP) fulfilled that commitment and provides the regulatory framework to manage that
responsibility.
In the context of the Saskatchewan Institutional Control Program (ICP), institutional control consists of those
actions, mechanisms, and arrangements implemented in order to maintain control or knowledge of a
remediated site after project closure and to transfer custody to some form of responsible authority or
custodian.
The primary objectives of the Saskatchewan ICP are as follows:
1. Protect human health and safety;
2. Protect the environment;
3. Permanently preserve knowledge of the site including all relevant documentation;
4. Ensure, to the extent possible, that future generations are not burdened with the costs of long-
term monitoring and maintenance following decommissioning and reclamation activities;
5. Be sustainable; and
6. Recognise and respect federal jurisdiction, regulatory roles, and responsibilities for national and
international obligations.
In Saskatchewan, the responsible custodian under the ICP is the ministry or ministries assigned responsibility
for implementing and managing the program. The legislative authority to implement and enforce the ICP is
the Reclaimed Industrial Sites Act (RISA) and the Reclaimed Industrial Sites Regulations (RISR). These
authorities have been designed to allow one or more ministers and their respective ministries to be granted
that responsibility. To date, Saskatchewan Energy and Resources (currently within the Ministry of the
Economy) is the provincial ministry that has been assigned the responsibility for managing the ICP.
Activities undertaken by the custodian under the ICP can range from permanently recording the location of
a remediated site to conducting regular inspections, sampling, and maintaining the property. The custodian
also has the authority to address unforeseen events that could potentially arise at a particular site.
It must be noted that the ICP does not manage the responsibility for the decommissioning and reclamation
regulatory process; it manages the steps that follow.
A site cannot be accepted into the ICP until remediation activities have taken place and regulatory authorities
have issued a release (discussed below). The Saskatchewan ICP is specific to, and restricted to, remediated
mine and/or mill sites on provincial Crown land. However, the institutional control framework has been
designed to manage a broader scope of sites, such as private land or other types of industrial sites.
Throughout the life of a mine in Saskatchewan — from construction through operations to final remediation
— mine/mill operations are carefully governed under environmental regulations, beginning with an
environmental assessment. The environmental assessment process for a proposed mine and/or mill,
administered by the Saskatchewan Ministry of Environment, requires the proponent to include a conceptual
decommissioning and reclamation plan in its environmental impact statement. A proposed mine and/or mill
may also be required to undergo a review under federal regulation through the Canadian Environmental

Assessment Agency (CEAA). Uranium mines and/or mills may also be required to undergo a review through
the Canadian Nuclear Safety Commission (CNSC). Further information on the environmental approval process
is available from Saskatchewan Ministry of Environment, the CEAA, and the CNSC.
Once ministry approval for the proposed mine and/or mill is received under the Environmental Assessment
Act, the Mineral Industry Environmental Protection Regulations, 1996, issued pursuant to the Environmental
Management and Protection Act, (EMPA), require any person seeking an approval to operate a pollutant
control facility, mine, or mill to submit a detailed decommissioning and reclamation plan; this must include
a post-closure transitional phase monitoring plan for review and approval by the minister. Once the plan has
been approved, the operator is required to post a financial assurance instrument of sufficient value to cover
the cost of the decommissioning and reclamation. The same regulations require the operator of the facility
to conduct a detailed review of the decommissioning plan and the financial assurance at least once every five
years, whenever requested to do so by the minister, or within the 12 months preceding the permanent
closure of a facility. A flowchart of the regulatory processes as they apply to final closure is presented in
Figure 1.
After mining has ceased and the operator has completed the approved decommissioning and reclamation
activities, the site enters a period called “transition phase monitoring.” During this period, the operator is
required to continue monitoring and maintaining the site as per the requirements in the plan at its own
expense. The operator is required to maintain financial assurances sufficient to cover the cost of the
remaining obligations outlined in the plan for the balance of the transitional period as well as a contingency
for any unexpected occurrences. During this transition phase, the province and, in the case of uranium
facilities, the CNSC continues to conduct periodic regulatory inspections and reviews of monitoring results.
Also during this period, the operator continues to remain fully liable for any impacts the site may have on
the environment, surrounding communities, and public safety, both provincially as per the requirements of
EMPA and federally as per the requirements of the Nuclear Safety and Control Act (NSCA).
If the site performs in accordance with the decommissioning and reclamation plan and achieves the predicted
stability during the transition phase monitoring period, the operator may then make an Application for a
Release from Decommissioning and Reclamation (Release) to the Saskatchewan Ministry of Environment (SE)
in order to obtain a release from further monitoring and maintenance responsibilities and from the obligation
to maintain financial assurances. Having received the Release, the operator may then proceed to apply for
and receive a release from its surface lease (i.e., the surface tenure obligation). The release from the surface
lease will allow the operator to make an application to transfer the property to the ICP for long-term
stewardship. In Saskatchewan, mineral tenure and surface tenure are separate; this is not true for all
jurisdictions.


The national and international regulatory framework surrounding the institutional control of radioactive
wastes (and therefore uranium mining and milling facilities) has changed significantly since uranium was first
mined. The Government of Canada is a member of the International Atomic Energy Agency (IAEA) of the
United Nations, and is party to that body’s safeguards and protocols. In 1998, Canada became a contracting
party to the Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste
Management (Convention). As a contracting party to the Convention, Canada is required to take steps to
ensure that an appropriate institutional control framework is in place to address the long-term management
of decommissioned uranium mine/mill wastes, including those in Saskatchewan. The responsible federal
authority for the Government of Canada’s obligations under the Convention is the CNSC.
The Government of Canada’s regulatory framework, as it applies to mining and milling facilities, is exercised
through a number of departments and agencies and the legislation that empowers them. A site is required
to meet the environmental objectives required by those authorities, in addition to provincial authorities, in
completing its decommissioning and reclamation plans prior to a site being considered for entry into the ICP.
Once a site has entered the ICP, it will be required to maintain those objectives. Federal authorities that
provide objectives and regulatory oversight over the development of a mine or mill facility can include the
following:
1. Canadian Environmental Assessment Agency (environmental assessment);
2. Environment Canada (environmental monitoring);
3. Department of Fisheries and Oceans (mining effluent regulation); and
4. Human Resources Development Canada (health and safety).
Canada has also promulgated the NSCA and associated regulatory framework as it applies to uranium
mining/milling facilities. This grants constitutional jurisdictional authority over such facilities to the federal
government. The responsible federal authority for the NSCA and Regulations is the CNSC. The Institutional
Control Registry (Registry), in conjunction with the monitoring prescribed by the Registry, has been designed
to be comparable to an active licence issued by the CNSC. The Registry has also been designed to meet
Canada’s obligations under the IAEA’s Convention. The CNSC will decide on the exemption from licensing on
a site-by-site basis. If a site holder does not receive consent for exemption from the CNSC, the site will not
be accepted into the ICP and Registry.
The development of the Saskatchewan ICP was initiated in 2005 with the establishment of an
interdepartmental Institutional Control Working Group (ICWG) led by a consultant and composed of a
consistent group of senior management-level representatives from the Executive Council, Departments of
Environment, Energy and Resources, Northern Affairs, Justice, and Finance. The ICWG was granted approval
in principle by the Provincial Executive Council to develop an institutional control management framework,
undertake stakeholder consultation, and bring forward draft legislation and regulations for consideration by
the Legislative Assembly. The representative from the Executive Council demonstrated the government’s
commitment to the process and was charged with overseeing ICWG activity and providing regular updates
to a select group of ministers of the Crown.
The choice of the ICWG member departments was critical to the success of the process, as was the dedicated
participation of a consistent, senior management-level member from each department in the working
group’s activities. In making that choice, consideration was given to the mandate and responsibilities of each
government department in order to ensure that all appropriate departments were at the ICWG table and
privy to its deliberation from the onset. Not unexpectedly, a number of the ICWG’s participating departments

had little or no knowledge of or interaction with mineral development, mine operation and decommissioning,
or environmental or engineering standards. Alternately, those departments familiar with these aspects often
had a limited knowledge of how a government manages surface and mineral tenure on provincial Crown
lands, drafts legislation, accounts for liability, or manages funds outside of the General Revenue Fund. The
inclusion of a single, consistent representative from each department in the ICWG’s deliberation from the
outset allowed each individual to fully understand and effectively communicate with his or her department
on the issues arising from the development and implementation of the institutional control management
framework.
Another key to the ICWG’s success was the existence in 2005 of a number of former uranium mine sites that
had been successfully decommissioned and had undergone more than a decade of transition phase
monitoring to demonstrate that they were chemically and physically stable. These sites offered concrete
examples of sites for the working group to consider. The ability to use existing sites considered ready for
transfer into an institutional control management framework significantly enhanced the ability of all of the
members of the ICWG to identify, understand, and address potential issues during their deliberations.
The first task undertaken by the ICWG was to conduct a comprehensive assessment of policy and legislative
requirements, risks, and liabilities associated with undertaking such a program. This was followed by
extensive consultations within government and a detailed review and consideration of Canada’s international
obligations (particularly as they apply to the management of nuclear wastes) and all relevant federal and
provincial legislations, regulations, guidelines, and objectives.
The ICWG then developed a detailed intergovernmental background paper, which fully outlined a proposed
institutional control management framework and anticipated benefits, risks, and potential liabilities. Once
the framework was reviewed and approved in principle by Cabinet, the ICWG proceeded to draft and publish
a public discussion paper and entered consultations with a variety of stakeholders, both internal and external
to government.
The public consultations were conducted during 2006. Key stakeholders included mining companies and
related industries; northern communities; the North Saskatchewan Environmental Quality Committee
(NSEQC), composed of representatives nominated by approximately 32 northern municipal and First Nations
communities that are potentially impacted by northern mining operations; northern First Nations; Métis
leaders and communities; federal regulators; and various environmental nongovernmental organisations
(NGOs). Consultations included multiple public and northern meetings; meetings with federal regulators, in
particular the CNSC; and the establishment of an industry/government working committee.
The results of the public consultations and the industry/government committee discussions were
incorporated into the final development and implementation of the ICP. One important issue was that the
ICP be governed by standalone legislation. Meetings and discussions between the ICWG and representatives
of the mineral industry continued through 2006, culminating in the promulgation of the RISA and the RISR
during the first quarter of 2007 to implement the ICP.
The Reclaimed Industrial Sites Act (RISA) grants the Ministry of Economy, Energy and Resources (SME) the
legislative power to establish the Institutional Control Program (ICP). The stated purposes of the ICP are as
follows:
1. to set out the conditions by which the Government of Saskatchewan will accept responsibility for
land that, in consequence of development and use, requires long-term monitoring and, in certain
circumstances, maintenance;
2. to ensure that the required monitoring and maintenance are carried out on that land;
3. to provide a funding mechanism to cover costs associated with the monitoring and maintenance
on that land; and

4. to ensure that certain records and information are preserved with respect to that land.
(Saskatchewan Government, 2007)
The ICP has two primary components, the Registry and the Institutional Control Funds, namely the
Monitoring and Maintenance Fund (ICMMF) and the Unforeseen Events Fund (ICUEF). The RISR prescribe the
conditions under which SME will accept a closed site into the ICP, the requirements of the ICP to monitor and
maintain a closed site, the funding method, and the method for enforcing the preservation of records and
information.
A site holder wishing to initiate the transfer of a closed site into the ICP is required to submit an application,
which describes the site, the results of the transition phase monitoring, and a detailed assessment of any
remaining liabilities on the site, to SME. The documentation must also include site holder (corporate)
information; the Release or proof of eligibility to be released from all licenses, permits, and leases; an IC
monitoring and maintenance plan; and, when available, site-specific historical information. The application
is subject to a review by SME with input from the Ministry of Environment in order to ensure that it is accurate
and that all required documentation has been included. Assuming the application meets the ministry’s
approval, the ministry will enter into discussion with the site holder to determine the amount to be deposited
into the ICMMF and the ICUEF. Once the application is approved, the site holder is required to submit a
registration fee and the prescribed fund deposits.
The monitoring and maintenance plan and the present value of the future costs associated with the
monitoring and maintenance obligations will undergo a detailed review to ensure they are sufficient and
appropriate to meet the long-term environmental, health, and safety objectives required of the closed site.
Companies may find it beneficial to develop such plans as part of their detailed decommissioning and
reclamation plans and submit them at the time of application for a Release. During the initial years of the
ICP, the Registry does not anticipate employing full-time technical expertise to complete this review. SME
will consult and act on the recommendations of Saskatchewan Ministry of Environment and Saskatchewan
Ministry of Finance in making the decision to accept a closed site into the Registry and in determining what
the appropriate fund deposits will be for the site to be accepted.
One of the prescribed conditions for acceptance into the Registry is that a site holder must have submitted
a monitoring and maintenance plan that identifies both the monitoring and maintenance obligations that
need to be undertaken once the site is accepted into the ICP and the present value of future costs associated
with those obligations. SME has established the Institutional Control Monitoring and Maintenance Fund
(ICMMF) and manages the funds as monies separate from the province’s General Revenue Fund.
The site holder’s contribution to the ICMMF must be of a value to generate sufficient revenue to pay those
future costs in perpetuity. The contribution is calculated based on the net present value of the obligation at
a forecast inflation rate and on forecast investment return rates. The total ICMMF contribution is the sum of
the individual contributions for each monitoring and maintenance obligation at the site.
For a future cost that has been submitted in current dollars, the future value of an individual obligation in
the year it occurs is calculated by applying the rate of inflation to the current dollar value. The rate of inflation
is determined by SME in consultation with Saskatchewan Finance and assigned following an application by a
site holder. The present method is to calculate the rate of inflation as a 10-year average based on annual
values reported by the Bank of Canada.
The future dollar value of the obligation is then discounted back to current dollars based on a rate determined
by the rate of return on the invested fund contribution and the number of years in the future that an
obligation occurs. The rate of return is determined by SME and assigned following an application by a site
holder. For the sites currently accepted, the rate was determined by SME in consultation with Saskatchewan

Finance and was established in reference to the inflation rate (i.e., Rate of Inflation + X%). The new
Investment Advisory Committee may fulfil the Saskatchewan Finance role for future sites.
SME manages the ICMMF in its entirety, but each site-specific deposit on account is tracked individually.
Responsibilities include maintenance costs anticipated at the time the closed site is accepted into the ICP,
and for any other general costs that should have reasonably been anticipated at the time the closed site was
accepted into the ICP, and costs incurred for the purpose of determining the required monitoring and
maintenance of the closed site.
SME can only access site-specific monies for site-specific monitoring and maintenance and cannot access
monies on deposit for one site to fund expenditures at a separate (different) site.
A comprehensive monitoring and maintenance plan with sound cost estimates significantly reduces the risk
of a cost overrun on monitoring and maintenance activities. Should such an event occur, a root-cause analysis
should be performed to determine cause and responsibility. Funding sources for such events may include the
Institutional Control Unforeseen Events Fund (ICUEF), a financial assurance, the former site holder, or the
province.
Understandably, concerns were raised by stakeholders, in particular the mineral industry, that the fund
should be prudently managed and that their site-specific monies not expended elsewhere, subjecting them
to cost liabilities at a future date after having made initial deposits in good faith.
The RISA grants SME the authority to invest monies in an account of the ICMMF, which are not presently
required for the purposes of that account, in any security or class of securities authorised for investment
pursuant to the Pension Benefits Act, 1992. SME has established an Institutional Control Funds Investment
Advisory Committee comprised of stakeholders including former site holders and industry representatives to
provide investment advice, review, and recommendations to manage the monies. Once the fund accrues to
a sufficient level to bear administrative costs of third-party management, the longer-term investment
strategy will likely be to include a diversified portfolio of bonds, equities, and assets, prudently managed to
meet investment objectives.
In addition to the contribution to the ICMMF, a site holder must include a contribution to the ICUEF. The
contribution to the ICUEF must be of sufficient value to generate revenue to pay the costs of future
unforeseen events and eventually release a site holder from a financial assurance requirement. The
contribution is calculated as follows:
 for a closed site without tailings or engineered structures, 10% of the Total ICMMF Contribution;
and
 for a closed site with tailings or engineered structures, 20% of the Total ICMMF Contribution.
The calculation and assignment of percentages was determined in negotiation with industry. In short, it is
based on the assumption that a site with an engineered structure presents twice the risk of a site without such
a structure and that the contribution should be at a sufficient monetary level to provide growth potential.
It is difficult to accurately forecast or estimate the extent of unanticipated future costs at any individual
decommissioned property. However, modern mine decommissioning and reclamation strategies in
Saskatchewan are based on the implementation of passive control methods wherever possible. These
methods significantly reduce the potential for such costs to arise. Unforeseen events could include such things
as the failure of a containment dyke, the collapse of a pit wall, the premature degradation of a shaft cover, or
a change in regulatory requirements. The ICUEF will fund these contingent events.
However, a site holder cannot be granted complete absolution from site responsibility. The EMPA provides
for absolute liability of a person responsible for a discharge to continue indefinitely. The authority to waive
this liability does not rest with the Minister of Environment as no such authority is provided in the EMPA. It

is for this reason that neither the Minister of Environment nor the Minister of Economy can issue a deed of
custodial transfer that states that the operator is completely absolved from responsibility for environmental
contamination at a particular site.
The ICP will limit the province’s liability to be held responsible for future clean-up costs arising from
unforeseen circumstances or company dissolution that are not provided for in EMPA or identified in the
Application for Release from Decommissioning and Reclamation approved by Saskatchewan Ministry of
Environment and upon which the basis of the custodial transfer was undertaken. Only in the case of the
original operator no longer being in existence would such costs have to be addressed by public means.
As with the ICMMF, concerns were raised by stakeholders and in particular, industry site holders, that the
money be prudently managed. The ICUEF is, in effect, a “rainy day fund” to manage the cost contingencies
of unforeseen events. Notably, this fund, with no forecast withdrawal, can reach significant levels in the
future. With continued deposits and interest earnings from sound financial management, the province and
stakeholders agree that the fund could reach significant levels in the future such that further financial
assurance or deposits will not be required. While SME will manage the ICUEF in its entirety, unlike the ICMMF,
it will not be tracked by individual site-specific monetary deposit.
A financial assurance requirement has been implemented to minimise the ICP’s financial risk during the initial
years, while the ICUEF is building in value. In negotiation with industry, while implementing a condition to
reduce its risk, the province also took steps to minimise the financial impact on good corporate citizens
through the acceptance of corporate guarantees. It is understood that, once the ICUEF has reached a
sufficient level to manage the total cost for unforeseen contingent events, the financial assurance
requirement will be removed. For SME to accept a closed site into the ICP and thus, the Registry, a site holder
must post a financial assurance in an amount equal to the cost of a maximum failure event that could occur
at the closed site or any such reduced amount agreed on by the minister. The maximum failure event is to
be identified in the monitoring and maintenance plan submitted to the minister in the application for entry
into the Registry. The maximum failure event is site specific and typically references a failure of the largest
engineered structure that exists on the site.
The form of an acceptable assurance is prescribed in the RISR and must be of one of the following types:
1. cash;
2. cheques and negotiable instruments;
3. government bonds, government guaranteed bonds, debentures, term deposits, certificates of
deposit, trust certificates or investment certificates;
4. corporate guarantees, irrevocable letters of credit, performance bonds or surety bonds; and
5. any other financial instrument or security that is acceptable to the minister.
For a financial assurance that is posted as a corporate guarantee, the minister requires that the site holder
providing that corporate guarantee have a sufficient financial credit rating to ensure the ability to cover the
amount of the assurance. Acceptable investment-grade credit ratings are defined as the rating assigned by
Standard & Poor’s, Moody’s, or Dominion Bond Rating Service of, respectively, BBB/Baa3/BBB(low) or higher.
A financial assurance for a specific site is to be reviewed approximately once every five years to determine
whether it continues to be required and at what level. In conducting the review, the Registry shall consider:
the condition of the closed site, the amount estimated as standing to the credit of the closed site in the
ICMMF, and the financial solvency of the former site holder.
The Registry can only access the financial assurance under two conditions: 1) if the minister is obligated to
undertake significant maintenance obligations at the site and withdrawing the funds for the required
maintenance will unduly deplete the amount standing to the credit of the closed site in the ICMMF or the

ICUEF or 2) if the minister determines that the security provided as part of the assurance fund is about to
expire and that no replacement security has been provided.
The final cost component of the ICP is the prescribed registration fee of $500. This is a fixed fee prescribed
in the RISR and is designed to provide for processing of applications, document transfer, and program
administration.
The total cost to transfer a site to the Registry will be site specific and largely based on the type and condition
of the various aspects and residual risks present on the site; however, the cost for most sites is anticipated
to be under C$ 100,000.
The ICMMF and the ICUEF must be managed openly and transparently. As they are provincial government-
managed funds, the Provincial Auditor is required to audit the accounts and transactions and table an annual
report on the business of the funds and a financial statement in each fiscal year. In this manner, the
information is available for public and stakeholder review.
The legislation also requires for SME to do a review of the legislation after five years and for the Registry to
prepare a report, to be known as the Institutional Control Report, every five years identifying the condition
of all closed sites accepted into the ICP. As with the Registry records, this report is made available for public
and other jurisdictional access.
The Institutional Control Registry (Registry) is a primary component of the ICP. The Registry has been
implemented by SME and is a record and information archive for accepted sites. It is responsible for ensuring
that monitoring and maintenance is performed as prescribed and for enforcing compliance with established
land use restrictions. It is also a portal for public access to information and is responsible for reporting on all
its activities to national and international regulatory authorities.
The RISR require the Registry, as a record and information archive, to secure and maintain certain prescribed
records and information. At a minimum, each site holder must submit the following records and information:
1. location of the closed site;
2. identification of the final holder of the closed site;
3. a description of the closed site and the activities that were conducted on the closed site;
4. the release from decommissioning and reclamation issued pursuant to the Mineral Industry
Environmental Protection Regulations, 1996;
5. a reference to and the location of the documents provided by the site holder pursuant to the
Mineral Industry Environmental Protection Regulations, 1996 for the purposes of applying for the
release mentioned in clause (d), including a reference to and the location of a full and complete
set of ‘as-built’ reports;
6. a description of the monitoring and maintenance obligations;
7. a reference to and the location of the documentation provided to the site holder when the site
holder is released from any surface lease agreement that governed the closed site;

8. in the case of a closed site that was a uranium facility, a reference to and the location of Canadian
Nuclear Safety Commission licensing documentation and all Canadian Nuclear Safety Commission
decisions related to the closed site;
9. a notation of the location of all documentation that the minister considers applicable to the
closed site and that is in the control of the site holder; and
10. surface and underground plans submitted pursuant to the Mines Regulations, 2003 or any
predecessor to those regulations.
Currently, a site holder’s submission is not specifically required to include all historical records and data
previously submitted to a government department, agency, or commission and retained on file, as some
historic records may have been submitted by a previous site owner or are department specific. That
information is to be submitted by the responsible government department, agency, or commission to the
Registry.
In addition, as defined in the RISR, the Government of Saskatchewan, the Government of Canada or any
agency or commission of those governments is required to submit the following items to the Registry:
1. notation of the location of all documentation that the minister considers applicable to the closed
site and is in control of the relevant Government, agency or commission;
2. a description from the department of the Government of Saskatchewan responsible for the
management of the surface lands that are part of the closed site and that are owned by the
Government of Saskatchewan, identifying and specifying any surface land use restrictions for the
closed site;
3. a description from the department of the Government of Saskatchewan responsible for the
management of mineral lands that are part of the closed site and that are owned by the
Government of Saskatchewan, identifying and specifying any mineral disposition restrictions for
the closed site;
4. in the case of closed site that was a uranium facility, a reference to and location of Canadian
Nuclear Safety Commission licensing documentation and Canadian Nuclear Safety Commission
decisions related to the closed site;
5. a note indicating whether the property is registered as a land disposition administered by the
minister responsible for the Environmental Management and Protection Act;
6. reference to and location of any final surface lease agreement respecting the closed site provided
by the department of the Government of Saskatchewan responsible for the management of the
surface land; and
7. a copy of any surface, underground and final closure plans respecting the closed site that are
provided to any department of the Government of Saskatchewan responsible for management of
those plans.
The submission of these government-retained historical records may take the form of a letter identifying the
applicable records for which that agency is responsible (i.e., a list and location); however, that agency is
expected to maintain those records in perpetuity. Alternately, the submission may take the form of a physical
transfer of documentation should the agency not wish to maintain the requisite records. Responsibility for
records already archived must be transferred to the Registry.
The Registry is required to monitor and maintain a closed site in accordance with the monitoring and
maintenance plan as submitted by the site holder and approved as a condition of acceptance into the ICP. As

the site matures, those monitoring requirements may be amended based on the site having proven the ability
to further stabilise at or below environmental and safety regulatory objectives and standards.
The Registry is also required to carry out all maintenance on sites within the Registry to ensure the sites
continue to meet environment and safety regulatory objectives. For example, this could include the
replacement of a concrete cap on a mine shaft once every 75 years (funded by the ICMMF), the repair of a
containment dyke following a once-in-a-thousand-year rainfall (funded by the ICUEF), or the maintenance of
long-term revegetation and reclamation objectives.
In carrying out the monitoring and maintenance responsibilities when there is a limited number of sites, the
Registry has chosen not to employ full-time qualified personnel but to retain the services of private technical,
professional, or other adviser, specialist, or consultant personnel. The Registry will also provide a copy of
monitoring reports to the original site holder that completed the original decommissioning to further ensure
the site is performing as planned and that the Registry has undertaken any corrective maintenance prior to
failure of engineered structures. The Registry will access the appropriate account of the ICMMF or, as the
case requires, the ICUEF for payment of the performance of those services.
At the time of application into the ICP, a site holder is required to relinquish the associated surface lease. For
sites in northern Saskatchewan, the lease will have been negotiated by the Ministry of First Nation and Métis
Relations (Northern Affairs). Based on a site’s operational use and the remediation measures incorporated
in the decommissioning and reclamation plan, it may be necessary to restrict future surface use to ensure
the stability and function of engineered structures. As an example, it would not be prudent to allow an access
road to be built across a decommissioned engineered tailings facility. The Registry may restrict or prohibit
access to a closed site that has been accepted into the ICP if it is considered to be in the public interest. The
land use restrictions are to be determined at the time an application is made to enter a site into the ICP, and
the Registry will retain the authority to enforce those restrictions. The surface responsibility will revert back
to the management of Saskatchewan Ministry of Environment or a delegated authority on surrender of the
lease. The ministry or delegated authority will have the responsibility of permitting permissible land use
subject to the land use restrictions on record within the Registry.
Similarly at the time of application into the ICP, a site holder is required to relinquish the associated mineral
dispositions, currently managed by SME. Based on the engineered structures incorporated in the remediation
of the site, it may prove necessary to the public interest to restrict future mineral activities such as
exploration on a site. When required, SME intends to establish a mineral reserve at a closed site that in effect
prevents future dispositions being issued. A reserve can be established at the time of application into the
ICP, when a site holder is required to relinquish the associated mineral rights.
In May 2009, the former Contact Lake gold mine/mill site was the first to be accepted into the ICP. Five other
former uranium mine sites were subsequently accepted into the ICP later the same year. All six sites were
released or exempted from further decommissioning and reclamation obligations by the applicable
regulatory authorities; the applications into the ICP were reviewed and accepted, and the sites were entered
into the Registry.
The necessary components for these sites to be accepted into the ICP were provided, and the coordination
between industry and the appropriate regulatory bodies, including several federal regulatory organisations
and numerous different provincial ministries and branches, was unprecedented. The success and relatively
short timeline for the coordination of release, exemption, and acceptance into the ICP was the result of the
commitment to the process by industry, its representatives, and the regulatory authorities. The commitment
and determination of a relatively small group of people to successfully initiate and drive this process to
completion should be recognised.

A description of the requirements for application, such as details of the site, the site holder information,
operational history, copies of all regulatory approvals, the monitoring and maintenance plan, the
identification of all engineered structures, and all financial obligations, have been discussed. The first six
properties’ applications were extensively reviewed and amended so that they did meet the requirements as
outlined. However, now that the six initial sites have been accepted and the first scheduled
inspection/monitoring program has been completed, the application requirements and review process
should ensure that some additional details are included.
Specifically, future applications must include a detailed list and description of all aspects and features of the
site to be monitored and inspected. Since the inspections will generally be several years apart and will
potentially be conducted by personnel unfamiliar with the application, the need for a detailed identification
and description of all elements to be inspected (and specifically what components of each element must be
examined) is imperative. For example, while an adit or horizontal underground mine working may be
identified and summarised in an original application, details of the margins of the closure method may not
have been provided in the application because they appeared to be relatively obvious at the time of closure.
However, as a result of natural revegetation, normal rock face slumping, and the inspectors’ lack of
experience with the site, the preliminary indicators of the failure of the closure method could be missed. The
inclusion of physical markers, detailed maps, global positioning system information, photographic evidence,
and additional details of the closure methods will greatly reduce subjectivity and improve the efficiency and
effectiveness of the inspection. Other examples of site features that may require additional details in the
original application include waste rock slumping, location of seeps or other drainage paths, waste disposal
grounds, and features that may slowly deteriorate and potentially cause concern in the long term.
In addition to adequate descriptions of the physical features of the site, access to all site-specific
documentation is essential. Currently, the Registry requires only the notation of the location of all
government-retained documentation considered applicable to the closed site. Similarly, as a result of the
extended timeframe between inspections or acceptance into the registry, personnel familiar with the site
and even the ICP may be limited. As a result, a comprehensive review of the site documentation will be
required. If access to this information is difficult for the reviewer, the cost and efficiency of the review, the
inspection, and the overall results will be restricted. For example, if historic operation information or
environmental sampling results are difficult to locate, this may cause difficulty in conducting an effective
inspection and comparing current analytical results to historical values. Consideration is being given to
alternate methods to improve the accessibility and availability of this documentation to the Registry in order
to facilitate review prior to future scheduled inspections.
Another lesson learned is that, for the long-term sustainability of the program and funds, investment returns
and costs of monitoring and maintenance may vary significantly in future years. For example, the ICP did not
anticipate the stock market crash of 2008. Such incidents highlight the value of investment advisory
committees and the importance of a review of the RISA and of funding, particularly during the initial years.
The Province of Saskatchewan successfully undertook the processes necessary to develop a formal, effective
institutional control program (ICP) that defines the conditions under which it will accept custodial
responsibility and long-term stewardship of closed mine sites once the operator has fulfilled its
decommissioning and reclamation obligations and demonstrated that the site is chemically and physically
stable. The process was undertaken by a working group composed of representatives from the five different
ministries and an executive council, which completed a comprehensive assessment of policy and legislative
requirements and the risks and liabilities associated with undertaking such a program and undertook
extensive consultations with industry, Aboriginal traditional users, and other stakeholders in order to secure
their input, inclusion, and support.
The primary objectives of the ICP are to do the following:
1. Protect human health and safety;

2. Protect the environment;

3. Permanently preserve knowledge of the site including all relevant documentation;
4. Ensure, to the extent possible, that future generations are not burdened with the costs of long-
term monitoring and maintenance following decommissioning and reclamation activities;
5. Be sustainable; and
6. Recognise and respect federal jurisdiction, regulatory roles and responsibilities for national and
international obligations.
The ICP has two primary components: the Registry and the Institutional Control Funds, namely the
Monitoring and Maintenance Fund (ICMMF) and the Unforeseen Events Fund (ICUEF). The Reclaimed
Industrial Sites Regulations (RISR) prescribe the conditions under which SME will accept a closed site into the
ICP, the requirements of the ICP to monitor and maintain a closed site, the funding method, and the method
for enforcing the preservation of records and information. It addresses all aspects of conventional closed
mines, as well as uranium-specific issues of radioactive waste management, including those defined in the
articles of the IAEA’s Joint Convention on the Safety of Spent Fuel Management and on the Safety of
Radioactive Waste Management, all applicable provincial acts and regulations, and the federal Nuclear Safety
and Control Act (NSCA). The program includes a formal, publically accessible registry and document
repository, has been designed to be revenue neutral and sustainable, and ensures that future generations
are not burdened with the costs of the long-term monitoring and maintenance of former mine sites in the
province.
Since its inception, the Saskatchewan ICP has accepted one gold mine/mill and five uranium mine sites into
the IC Registry and has completed the first of the requisite scheduled inspections of each registered site. As
with any program, as SME gains experience, it is expected to regularly assess the program with input from
other ministries and to explore ways to further define and improve the approval process and review funding
levels, investment management, and emergent issues such as a proposal to redevelop a site that has been
accepted into the Registry. It is anticipated that the review of the program and any changes will require
consultation with all relevant stakeholders and will include a review and potentially revisions to the
Reclaimed Industrial Sites Act (RISA) and the RISR.
Saskatchewan Government (2007) The reclaimed industrial sites act, being Chapter R-4.21 of the Statues of Saskatchewan, viewed
25 February 2014, www.qp.gov.sk.ca/documents/english/Regulations/Regulations/R4-21r1.pdf.

R. Knapp SENES Consultants, Canada
Effluent treatment standards are evolving with lower limits on conventional pollutants and the introduction
of new parameters for control. Of specific note are requirements for sulphate and possibly TDS control. Not
only do these represent material costs to existing mining operations, but the ramifications to mine closure
and long-term water treatment are even more significant. This presentation will review the implications of
these new regulatory requirements with specific emphasis on how this could impact mine closure and long
term costs.
3

P.A. Varris Golden Star Resources Ltd., Ghana

M.B. Thorpe Torex Gold Resources Inc., Canada
The Bogoso operations of Golden Star in Ghana are like a time capsule of closure planning: dig around and
you can find all sorts of treasures. Mined by a number of operators over the past 80 years, including Marlu
Gold Mining Areas Ltd., Canadian Bogoso Resources Ltd., and Billiton Bogosu Gold Ltd., before its acquisition
by Golden Star in 1999, the concession hosts more than 40 pits and 25 waste dumps under various stages of
rehabilitation and closure. With such a vibrant and extensive mining history, the site is host to legacies that
date back to the early formal mining of the 1930s, sites mined and closed in the challenging economic
conditions of the 1990s and examples of modern day integrated closure planning and implementation.
In stark contrast to the 1930s belief that “nature will take care of herself,” ecosystem reconstruction by Golden
Star now incorporates the careful identification of next land use alternatives based on engineering
requirements for the management of reactive materials and the capability of the available resources to
sustain the nominated use, overlain by community desires and expectations. Assisted by the bi-modal rainfall
pattern of the region, reclamation programs include the propagation of species of conservation significance
to complement demonstration farms (forest timber production) and re-forested areas.
The kaleidoscope of past and present closure planning efforts highlights the importance of well-executed
baseline assessments and proactive closure planning that adapts to reflect changes in mine plans, the
operating environment and community needs. These aspects, combined with stringent application of
techniques to understand the suitability of the materials available for reclamation, as well as techniques to
validate execution, are critical to successful outcomes, and the converse avoidance of the need for
interventions and the associated liability.
Showcasing ecosystem reconstruction that has been applauded by executives of the Forestry Commission,
and sites that have been deemed to have achieved final completion against requirements of the
Environmental Protection Agency, the Golden Star Bogoso operations provide examples of the evolution of
mine closure and associated expectations. The examples range from sites exhibiting the long-term outcomes
of failure to remedy poor substrate quality to rehabilitated sites that provide livelihoods for local farmers. The
developed dump rehabilitation strategies and opportunities for mined-out pits are presented in this paper.
The ability to follow the evolution of mine closure in tropical West Africa provides a unique insight into the
opportunities and pitfalls of closure planning, the integration of community needs, constraints to next land
uses and how these remain unchanged across the decades.
Golden Star operates two gold mines and three processing plants in Ghana, West Africa. Golden Star Bogoso
operates a series of open pits and two processing plants near the towns of Bogoso and Prestea, while Golden
Star Wassa operates a series of open pits and a processing plant in the Mpohor and Akyempim area. All the
operations are in the western region of Ghana (Figures 1 and 2) and are between 230 km and 270 km west
of the capital, Accra. Golden Star owns 90% of the two companies, and the government of Ghana owns a
10% free-carried interest in the companies. Under a similar arrangement, Golden Star also owns 90% of the

Prestea underground mine, which is presently under care and maintenance while project evaluations are
completed following a positive preliminary economic assessment.
GSR Operations
GSBPL
GSWL, Wassa
GSWL, Benso
GSWL, Hwini Butre
Golden Star purchased the Bogoso Concession in 1999 and has operated the Bogoso non-refractory plant
since that time, processing gold-bearing ores when non-refractory feed was available. In 2001, Golden Star
acquired the Prestea Concession, located adjacent to and south of the Bogoso Concession, and began mining
the open pit deposits in late 2001. In late 2002, Golden Star acquired the Wassa operation, and constructed
the Wassa gold recovery plant, which began commercial operation in April 2005. In July 2007, Golden Star
completed construction and development of the Bogoso refractory (sulphide) plant to process the refractory
ores using bio-oxidation (BIOX®) technology.
In late 2005, Golden Star acquired the Hwini-Butre and Benso (HBB) properties. Benso mine development
activities started in late 2007, and in 2008 the first ore from the Benso mine was transported to the Golden
Star Wassa plant for processing. Development of the Hwini-Butre mine started in the fourth quarter of 2008,
and in May 2009, the Hwini-Butre mine began transporting ore to the Golden Star Wassa plant for processing.
The HBB mines are in phases of suspended operations and rehabilitation, respectively. Operations are
continuing at the Wassa mine with a planned underground development concurrent with open pit

operations, and oxide and sulphide mining is ongoing at the Bogoso operations. The mining operations in
Ghana continue to be the primary revenue-generating stream for GSR.
Mined by a number of operators prior to the purchase by Golden Star, the Bogoso and Prestea concessions
host more than 40 pits and 25 waste dumps, that is, a virtual “treasure-trove” of insights into the successes
and failures of closure planning, execution and management over a period of eight decades. To further add
to the complexity of obtaining a “release” for reclaimed areas, the Ghanaian regulatory environment has
evolved considerably, particularly in the last three decades.
At the time of mining/construction of the majority of the more modern day pits and dumps, the reclamation
was subject to what was then Ghanaian law, as well as the prevailing World Bank Guidelines as required by
a financing contract with the International Finance Corporation (IFC). The subsequent development of mining
environmental law, including the enactment of the Environmental Protection Agency Act, modified the
regulatory landscape with respect to closure.
The IFC involvement in the operations ceased with Golden Star’s purchase of Bogoso Gold Ltd. (BGL);
however, the World Bank guidelines remained the primary international standards guiding reclamation and
closure until the establishment, in 2005, of a reclamation security agreement (RSA) between Golden Star and
the Ghanaian Environmental Protection Agency.
Mining in the Prestea area dates back several centuries. The first direct involvement by Europeans in the area
occurred in the 1880s with the establishment of the Gio Apanto Gold Mining Company and the Essaman Gold
Mining Company. These companies became Apanto Mines and Prestea Mines Ltd. in 1900. Both companies
merged to become Ariston Gold Mines in 1927, and companies associated with Ariston Gold Mines carried
out exploration and some mining to the northeast of Prestea at Quaw Badoo and Brumasi. The company also
prospected concessions immediately to the southwest of Prestea at Anfargah.
Prospecting and some mining had been carried out independently on the adjacent Ekotokroo, Bondaye and
Tuapim concessions located to the south of Anfargah. These concessions were acquired by Ghana Main Reef
Ltd. in 1933 and operated continuously until 1961.
The Ghana State Gold Mining Corporation (SGMC) was set up with effect from March 1961 under an
instrument of incorporation signed by the president of Ghana. The various Ghanaian gold operations were
regrouped from April 1963. The Government of Ghana purchased both the Ariston Mines and Ghana Main
Reef companies and merged them to form Prestea Gold Ltd. The Buesichem concession to the northeast and
along strike from Brumasi was subsequently added to the Prestea concessions. The Buesichem concession
contained a small historical open pit, which was one of several operated by Marlu Gold Mining Areas until
1955.
Tailings and mine waste dumps that date back over 80 years can be found in the area. Some of the current
material being processed by Sankofa Mines is tailings produced back in the first half of the twentieth century.
Mining in the Bogoso area has a relatively long history, although there is no official record of any workings
prior to 1882. Small scale, commercial mining began near Bogoso in 1906 and continued to the early 1930s.
In the early 1930s, several, large, low-grade gold deposits were recognised and developed as open-pit
operations by the Marlu Gold Mining Areas Ltd. When the Marlu and Bogoso North deposits were mined out,
other open pits were developed further south, that is, Chujah, Dumasi, Ablifa and Buesichem. Gold ore
production for Marlu Gold Mining Areas Ltd. peaked in the period from 1936 to 1942 at nearly
800,000 tonnes per year. About 924,000 ounces of gold were recovered from 7.5 million short tons of ore

during the period 1936 to 1955. The operations then closed as a result of declining grades and reserves of
the oxide ores, and recovery problems associated with the remaining sulphide ores.
In the Bogoso area, the SGMC carried out examinations and sampling programmes in underground adits
during the early 1970s, and the United Nations Development Programme financed a two-year drilling
programme in 1976–78, which concentrated on the Chujah-New Dumasi deposit. This work indicated
substantial low-grade reserves. In 1981, a study sponsored by the United Nations Development Programme
evaluated the available information on the old workings at Bogoso.
The Bogoso prospecting concession was awarded to Denison Mines Ltd. in May 1986. Denison entered into
a joint venture agreement with Sikaman Gold Resources Ltd. to form a Ghanaian company, Canadian Bogoso
Resources Ltd. (CBRL). Within the same year, a Canadian consulting group conducted preliminary exploration
of the licence. Reconnaissance sampling of various exploration adits on the property and drilling of the Marlu
tailings were carried out, which, in 1987, resulted in a prefeasibility report on some of the technical aspects
of a potential mining operation at Bogoso.
In April 1988, Billiton International Metals BV (a subsidiary of the Royal Dutch Shell Group) acquired the
Denison Mines’ share in CBRL; in 1992, the company’s name was changed to Billiton Bogosu Gold Ltd. Further
study indicated that, owing to the apparent inadequate supply of oxide ores to sustain a viable project,
emphasis should be directed towards providing a sustainable project based on the refractory (sulphide) gold
resources. It was concluded that an operation based on the mining and processing of sulphide ores, using
open-cut mining methods and the flotation and roasting of sulphide concentrates, was viable, at a processing
rate of 1 million tonnes per year. Design and construction of the treatment plant and associated facilities
were completed in 1991, when mining and processing activities commenced.
In early 1994, following a lengthy period of technical and ore grade problems, it was recognised that the
roasting process was not viable, and in February 1994, the sulphide flotation and roasting sections were
suspended. Mining and processing of oxide ores commenced. In November 1994, Gencor Ltd. purchased the
majority of Shell’s exploration, mining and processing interests.
In a 1997 review of the BGL reclamation and rehabilitation plan, the IFC recognised the concession as hosting
32 pits, of which a third had flooded; 12 waste rock dumps, of which four were generating acidic drainage;
and six ore stockpiles, of which two were transitional ore and generating acidic drainage. At the time, BGL
had active open pits at Bogoso North, Marlu, Nankafa, Dumasi and Chujah. Gencor disposed of its interests
in the project in March 1998 to the senior lenders and other banks.
A closure plan for the Bogoso Mine developed in 1999 highlighted the impending sale of the business by the
project lenders, and in September 1999, Golden Star purchased its interest in the Bogoso Mine from the
consortium of banks led by the IFC. The IFC continued to assist BGL for some time after their funding
arrangements ceased, and were still advising on environmental aspects into 2001.
When Golden Star acquired BGL, a feasibility study was completed into the processing possibilities for the
sulphide ore. The feasibility work resulted in the design and construction of a processing plant expansion to
process the sulphide ore using BIOX.
Reflecting the long-term nature of mine closure, the legislation and standards applicable to reclamation and
closure at the Golden Star Bogoso operations have evolved with time.
While mining and construction of many of the pits and waste dumps occurred under the Ghanaian Minerals
and Mining Law (PNDCL 153) 1986, these projects were simultaneously developed under a financing
arrangement with the IFC that started in 1989 and continued until the financing agreement was paid out in
1999. As a member of the World Bank Group, the IFC and its funded projects are subject to the relevant
World Bank Guidelines. Initially, the projects were required to comply with the environmental and ecological
guidelines of the International Bank for Reconstruction and Development, as well as the IFC-approved

environmental management plan; the guidelines were subsequently revised in the 1998 guidelines contained
in the World Bank Group Pollution Prevention and Abatement Handbook.
In May 1994, Ghana’s Mining and Environmental Guidelines were developed as the result of a three-year
collaboration between the Minerals Commission and the Ministry of Environment. In the same year, the
Environmental Protection Act was promulgated, and subsequently a series of procedures, guidelines and
ultimately regulations pertaining to environmental permitting, assessment, reporting and bonding were
established.
In December 2005, with the signing of an RSA, reclamation standards became applicable for pits and waste
dumps developed from that time onwards within the Golden Star Bogoso mining concessions.
In 2006, the Minerals and Mining Act (Act 703) was enacted to establish further law on the process for
obtaining mineral rights, administration and management of these rights and for the protection of the
environment. This was followed by the 2012 enactment of the Minerals and Mining Regulations, which
repealed the 1970 Mining Regulations (LI 665).
The Bogoso mining leases have a strike length of more than 65 km, with the location of the Birimian and
Tarkwaian contacts varying along the strike. This underlying geological structure affects the ore, waste and
pit walls, and variations along the trend result in more sulphides in the Birimian than in the Tarkwaian.
Thereby, the performance of the various operational units (e.g. pits, dumps) is determined, in large part, by
their location within the concessions and the management of the various operational units.
The waste rock dumps generated in this period consisted mainly of waste from oxide sources, with occasional
transition material. These waste rock dumps, while relatively easily rehabilitated, were not reclaimed with
any detailed view of the next land use. The primary focus was to cover the dumps with vegetation to stabilise
the area and prevent erosion. To that end, much of the rehabilitation of these early dumps consisted of
Acacia monocultures.
The advantages of this approach result from the nitrogen fixing of these species, as well as parallel biomass
generation, which is a key driver of ecosystem development in the tropics. As the soil conditions improve,
species that require higher soil nutrient levels can be established. On many of the waste dumps, the biomass
integration into the upper layers of the soils does not occur quickly, as the Acacia leaves do not break down
quickly, even in a high rainfall environment. The main method of transition on these older dumps occurs
when local stakeholders cut the trees and burn them to make farmland. This improves the soil conditions
and provides a sustainable next land use — albeit unintentional.
In a number of these areas, Golden Star Bogoso has progressively thinned the Acacia stands and completed
supplementary planting with indigenous species. Managing the gradual reuse of these areas by local farmers
while protecting the establishment of the indigenous species creates challenges to the closure process
(Figure 3).

With the commencement of the refractory operations and mining of transitional and fresh ores came the
challenges of management of reactive materials. Improved handling and management of growth materials
(e.g. oxide and topsoils), reactive rock segregation and implementation of successive improvements in
capping of reactive materials for encapsulation have resulted in improved reclamation outcomes.
The actions of earlier operators in reactive rock management have provided considerable guidance to Golden
Star in the development of rehabilitation and closure standards. While we have benefited from the lessons
of others, our own actions have also held their share of lessons. Expanding pit shells affected the achievement
of progressive rehabilitation objectives for adjacent, space-constrained waste dumps; less-than-diligent
supervision of waste dump construction has seen dump toes placed too close to permanent infrastructure
(e.g. high-voltage power lines); and inexperience and inappropriate task allocation have seen unqualified
persons supervising large earthworks projects. These lessons are outlined more fully in the following sections.
Open pit mining in the Bogoso area was limited before the 1930s. However, extensive open pit development
occurred during the era of the Marlu mine. The primary ore source during that time was the non-refractory,
which essentially limited the open pits to a depth of about 30 m below the existing surface, with more
extensive excavations where hills were involved (e.g. Bogoso North and Marlu pits). When the Marlu mine
closed in about 1955, most of the pits were simply abandoned and filled with water. There was limited work
on the pit walls for stabilisation and almost no rehabilitation work at the time. Additionally, the concept of
acid/mineralised drainage was not considered.
These abandoned pits provide some guidance for the rehabilitation and closure of the current pits, as they
provide indicative behaviour of various components:
 pit wall stability;
 pit lake development and evolution;
 potential closure concerns within a sector of the concession.
Mining in some of the open pits on the concession commenced as early as the 1930s and carried on through
to the closure of the Marlu mine in 1955. The original pits and their associated pit walls were characterised
by the following:

 long-term (50+ years) stable pit walls;

 poor regeneration on the benches;
 varying water quality — from neutral clean water to acidic;
 varying connections with ground water/outflows;
 No active rehabilitation completed (e.g. re-sloping, planting, integration of water bodies into
surface drainage).
The challenges to the rehabilitation of the pits from this era remain today, with poor growth of vegetation
on the benches, resulting in high walls only able to achieve aesthetic land use criteria. Figure 4 shows the
Ablifa pit, where re-vegetation has only occurred on the upper benches from soil and seed movement from
the benches above. The lower benches show very little re-growth; in a tropical environment, this suggests
limitations of the substrate (e.g. lack of nutrients) rather than a lack of seed.
The quality of water in established pit lakes shows variation across the concessions with noted variation from
north to south, reflecting changes in the geochemistry and hydrochemistry. However, most of the pit lakes
were established in areas hosting oxide-only ore deposits, and as such have not been influenced by acidic
drainage. The primary issues for pit lake water quality have resulted where pits have mined into adjacent
hills and exposed transition materials.
The pits mined in the modern era (post-1990) show specific trends for rehabilitation:
 Pits were generally deeper, as the refractory ore could be processed consistently.
 Backfilling was completed as often as possible (e.g. Plant North pit).
 Pit wall stability concerns increased as pit walls were steepened to reduce mining costs.
 With deeper pits, predicting pit water quality at closure became more problematic.
 Ongoing pit expansions and cutbacks were not reflected in the closure plans, thus minimising
opportunities for operational synergies and increasing re-handle and cost.
 Limited thought was given to pit mine designs to support closure (e.g. wall/bench requirements,
littoral zone).
Where backfilling was completed, either as part of the mining sequence (Marlu pit) or as a permit condition
(Plant North pit), the rehabilitation of the pits incorporates the placement of a compacted oxide cap, followed

by a layer of growth medium before vegetation is established. When completed to required standards,
backfilled pits can sustain a variety of next land uses, including community uses (Figure 5) and agro-forestry
plantations (Figure 6).
The long-term stability of the pit walls was a concern, as greater pit depths and pressure to steepen slopes
to reduce waste mining came with the impact of wall failures and non-conformance to evolving regulatory
requirements. Given the extended period of mining of some pits, lay-backs in the upper oxide layers were
completed to improve long-term stability and protect the safety of the local stakeholders who will ultimately
use the area.
The general pattern for next land use for the larger open pits has been to nominate the use as water resource,
although early commitments to pit lakes were often made in the absence of detailed information regarding
groundwater conditions and potential future interactions of transitional ore materials and seasonal water
levels.
In the modern era, Golden Star has undertaken extensive reviews of baseline studies, historical feasibility
studies and reviews by development organisations to obtain a clear understanding of the hydrogeological
regime of the larger pits and ensure that pit lake closures will result in complete, year-round inundation of
reactive transitional and fresh rock materials (Figure 7).

Tailings management (or lack thereof) has varied over the past decades. In the early years (1800s and early
1900s), tailings were released into the landscape with little or no containment. With increasing
environmental concerns, tailings were stored in impoundments. On the Bogoso concession, there are tailings
from the 1930s to 1950s that, even after more than 50 years, have only weed grasses growing on them. This
is in contrast to the more recent tailings management of Golden Star’s TSF 1, where tailings exhibit good
growth of Typha sp. after only a few years (Figure 8). The vegetation establishment has been so successful
that the vegetation has impacted the tailings re-treatment operations underway at the facility.
The limitations to growth on the old Marlu tailings are both density and toxicity related. The tailings remain
waterlogged, limiting oxygen penetration into the substrate and, hence, to the roots. Other factors likely
include chemical contamination from early processing technology. In contrast to the historic tailings areas,
in most locations in a tropical environment, excellent growth is expected within a 50-year timeframe. Golden
Star has recently obtained a permit to re-treat some of these tailings, with an associated plan that enables
the encapsulation of remaining materials within a potential waste dump expansion.
The historic mining companies provided infrastructure for both the operations and the workers. While largely
still owned by the SGMC, many of the old structures have been taken over by people living in the area. Early
houses (from the 1920s) are visible in the Prestea area within the Golden Star Bogoso operations. These

buildings vary in the level of maintenance, and historic ownership legal contests have limited investment
(Figure 9).
In the modern era, infrastructure developed by Golden Star has been purpose-built to address community
needs. The current model for community infrastructure development and integration within the Golden Star
Development Foundation (GSDF) requires that the initial request be from the communities. Once a need is
identified, the GSDF seeks a long-term sponsor for the project (e.g. Ghana Health Service for a clinic) (Figure
10). Once a sponsor for the ongoing operation and maintenance of the facility is obtained, the GSDF builds
the infrastructure and equips it with everything required for operation (e.g. desks in schools). The eventual
closure of the mining operations in the area will, therefore, not affect the ongoing operation of GSDF-
sponsored infrastructure.
At the start of 2014, Golden Star (Bogoso/Prestea) Ltd. (GSBPL) commenced an exhaustive process of
reviewing the history, status and condition of a large number of reclaimed areas on its Bogoso and Prestea
concessions. Interestingly, it was not the desire to relinquish these areas, nor even for the recovery of
financial guarantees held by the Government, that motivated this review, but rather the combined impact of
recognition that the liability need not burden the balance sheet any longer, and pressure associated with
community encroachment in the rehabilitated areas.
This process of review involved unearthing long-buried mining records. By combing available due diligence
documents associated with the BGL purchase of the concessions and laboriously chronicling interventions
reported in monthly reclamation and annual regulatory reports, we were able to understand the
rehabilitation history of the various sites.

The literature review was complemented by field inspections involving a team of mining engineers, survey
and GIS specialists, as well as environmental and reclamation professionals. With time and successive
owners, pit and dump names had changed, tailings facilities became ore sources, pits were re-mined and
backfilled and waste dumps were amalgamated and reworked. Even until the week prior to the official
submission to the EPA for final completion, historic data were being reconciled to locations and activities in
the field.
Using the relevant and applicable closure standards for the various facilities, and benchmarking against the
requirements of the currently applicable RSA, field inspections and post-reclamation monitoring indicated
that a large number of formerly mined areas met not only the historic closure criteria but also the landscape
criteria, land-use completion and criteria for final completion specified by the RSA (Table 1).
Criteria category Criteria Specification

Landscape completion: Berm width (m) 6 m minimum
waste rock dumps
Slopes Stable, free from significant rock
fall/soil movement;
erosion levels similar to non-
mined areas;
stable, reshaped and
rehabilitated
Overall slope (°) 30° maximum;
3H:1V;
bench terracing for benches >
10 m in height
Water management Run off managed to control
erosion
Acid rock drainage ARD prevention measures
Aesthetics Gentle rolling landforms similar
to existing aesthetics
Land use completion: water Cycling Fresh water inflow assured
resource
Water discharge Meet EPA effluent guidelines
Aquatic fauna Self-sustaining population
established
Littoral zone Depth 20 m;
area equal to 10% of pit lake
surface
Final completion: areas with Retain criteria for land use with
acid or mineral drainage or no additional maintenance (after
potentially acid-generating reclamation) for a period of no
materials less than 7 years
On this basis, in August 2014, GSBPL officially applied to obtain certificates of final completion for 19
backfilled pits, five pit lakes and 12 reclaimed waste rock dumps (Table 2), covering 112 hectares of land,

over which a closure guarantee of some US$ 0.9 million was held, and for which the 2013 balance sheet
recognised US$ 1.3 million in liability.
Following the process of detailed review, and application under the RSA, on 21 January 2015, the 19 backfilled
pits, five pit lakes and 12 reclaimed waste rock dumps rehabilitated by ourselves and in part by our
predecessors were deemed as achieving final completion.
Landscape type Number ARD potential Land use category

Backfilled pit (PAG) 5 Refractory pits with PAG Re-forested areas;
encapsulated in backfilled agriculture — agro-forestry
Backfilled pit (NAG) 14 Oxide-only pits Agriculture;
reforested areas,
some with aesthetic high walls
Pit lake 5 Oxide-only pits Water resource
Waste dump 12 PAG dumps Accommodation areas
NAG dumps Agriculture;
agro-forestry;
reforested areas;
accommodation areas
NAG: not acid generating, PAG: potentially acid generating
The Golden Star concessions showcase a long history of mining oxide and refractory ore deposits, the exploits
of various operators (small and larger scale) and the impacts of rapidly evolving expectations. The lessons
learned in closure, under the influence of these factors, provide evidence that clearly dispels a number of
“old wives’ tales,” busts myths and reveals some truths that may be hard to acknowledge.
Table 3 provides examples of the successes and challenges of closure planning, execution and management
in this environment using a root cause analysis approach.
Just as closure planning sometimes feels a little like looking into a crystal ball, the age and history of the
Golden Star concessions affords us with the benefit of hindsight. Some of the lessons of our predecessors
and our own experiences are shared in the following:
 Major earthworks programmes are best managed and supervised by bulk earth movement
specialists. Your environmental team should overview the work for quality control.
 Consider mining pit wall slopes that are suitable for closure. When thinking practically, re-
accessing benches higher in the pit profile at closure to modify slope angles may not be possible.
 High wall rehabilitation should be completed in the early stages of mining, if feasible. Obtaining
access at closure will be difficult, as most pits are designed such that the pit walls are stable for
the “mining” life, not the longer “mine life.”
 Having sufficient and appropriate resources for reactive waste encapsulation, and for growth
media, is fundamental to achieving the next land use. If your resources are insufficient, plan early
to locate borrow locations.
 Designs for reactive waste management and implementation of controls are critical to closure
feasibility. Failure to implement appropriate controls has the potential to materially impact your
closure costs and schedule.

What happened Basic cause Contributing Outcome Lesson

ASM breached Sabotage Oversight; ARD and TSS; replace Security; condition
oxide cap, resulting quality control compacted cover / growth reviews; capping
in ARD media; delay to standards
relinquishment
Large farm areas Problem not Preventative Large farms costly to Demo farms trialled
hard to maintain anticipated maintenance maintain; crops transplanted and found to be a
by farmers feasible alternative
Toe not Procedure not Management Infrastructure removal Anticipate failures; use
marked/too close used /not system; work constrains progressive multiple controls to
to infrastructure followed direction; quality rehabilitation achieve specifications
control
Tailings placement No procedure Training Virtually no revegetation Vegetation
not planned for after almost 80 years establishment needs
closure suitable materials
Assumed waste Problem not Oversight; Waste dump drainage likely Understanding detail
dump should drain anticipated worker selection to result in acidic pit lake of geo and
to pit hydrochemistry
Oxide cover not Problem not No training ARD; replacement of Assess capping
installed to anticipated compacted cover required; materials suitable to
standard delays to relinquishment specifications
Pit designs Management of Training needs Impact to materials Update closure plan
modified, closure change improvement quantities, mine sequence, with mine plan;
plan not updated controls and commitments validate key
assumptions
PAG materials not SPAC not Oversight; Larger volume to manage as Clarity in
segregated followed supervision PAG; increased cost; delayed accountability for key
closure mitigations
High wall Problem not Independent Bench access no longer Conduct high wall
rehabilitation left anticipated review needed possible, resulting in an rehabilitation early in
until end of pit life aesthetic next land use pit life
Farm development Oversight No communication Vegetation burnt; backfilled Reach agreements
impacts site PAG landform not suitable early; monitor to avoid
for farming interventions
New regulation Problem not Corrective action Delay to sterilisation; delay Delaying pre-feasibility
requires approval anticipated not implemented in land access; added activities poses risk to
for backfilling complexity in permitting the business
Inexperienced Selection of Supervision during Inappropriate equipment Environment team
supervisor worker work selection; inefficient may not be
operations; increased cost experienced in bulk
earthworks
ASM breached Sabotage Oversight; ARD and TSS; replace Security; condition
oxide cap, resulting quality control compacted cover / growth reviews; capping
in ARD media; delay to standards
relinquishment

 Update the closure plan as mine plans are update. Each pit cutback or waste dump expansion
should involve critical review of the validity of key assumptions in the closure plan and explore
opportunities for operational synergies.
 Accept that stakeholder communities and regulators may not have complementary objectives for
closure, and commence negotiation to resolve these impasses early. Once you have already
revegetated to farmland is not the time to find out that the regulator will only accept forest.
 When your reclaimed landforms are successful, the community may wish to re-access them. This
does not mean that they accept liability for the area, share your objectives for next land use or
protect the landscape until you obtain relinquishment.
 Understanding the baseline hydrogeological conditions is fundamental to proving the feasibility of
pit lake closures. For a pit lake to be sustainable, the hydrology must at least seasonally support
recharge and flow conditions.
 To ensure that pit lakes have a good littoral zone, you must design for that outcome. Pit ramps
might be okay for the purpose, but chances are they will not be, and additional work may be
required on the upper pit slopes to provide an appropriate littoral zone.
 There is likely to be a conflict between having enough real estate at closure for settlement ponds,
wetlands and littoral zones, versus the objective of keeping the waste dump as close to the pits as
possible. If you have a closure design, it is easier to demonstrate this need.
 Always validate a closure plan in the field. Does the plan show all infrastructure (existing and
required)?
 There is risk to the feasibility of the closure plan until you are confident of the properties of the
mined materials. The geochemical and hydrochemical aspects pose considerable risk.
 Being conservative and assuming an entire lithology is potentially acid generating because you did
not do field testing is often not feasible in the longer term. Do the work to understand the
materials management requirements.
 An underground portal in your pit will save on underground mine development costs but has the
potential to significantly change your closure plan and associated closure costs.
Billiton (1999) Bogoso gold mine — mine reclamation, decommissioning and closure plan (pre-draft document), internal report, May
1999.
Bogoso Gold Ltd. (BGL) (2001) Annual environmental report 2001.
Burgéap (1998) IFC, Bogoso Gold Ltd. (Ghana), review of Bogosu mine reclamation and rehabilitation plan, responsibilities and costs,
final report, April 1998.
Golden Star (Bogoso/Prestea) Ltd. (GSBPL) (2013) Annual environmental report 2013.
Knight Piésold Ltd. (2001) Bogoso Gold Ltd., Bogoso gold mine, audit report 2, rehabilitation works, 1 August 2000 – 28 February
2001.

C.A. Pelletier Environmental Resources Management (ERM), Canada

E. Prelypchan Environmental Resources Management (ERM), Canada
K. Etches Environmental Resources Management (ERM), Canada
With more than half a century of involvement in providing environmental services to the mining sector in
Canada and abroad, Mr. Clem Pelletier is well-placed to reflect on the past, present, and future of mine
closure. Mr. Pelletier’s career has shared a timeline with the approach to mine closure’s shift as ad hoc and
opportunistic to a priority in the mine cycle. Much has changed in Canadian mining over the past five decades
to shape this emphasis on closure. A vastly increased number of mines spread across the country now include
project sites that occupy areas of significant size and complexity owned by juniors, mid-tiers, and majors, all
focusing on profitability in a highly dynamic financial market. Technologies for mining have advanced and the
scientific understanding of geochemistry and remediation technologies have progressed, along with the
expectation that they will be applied during closure. The environmental movement of the 1960s and 70s,
which started a public conversation about the legacies of mining, gave a voice to affected communities, and
helped shape domestic policy on environmental and mining regulations, which in turn has changed how mine
closure and reclamation is done in Canada.
The increasing understanding of long-terms risks related to mine closure has heightened industry awareness
around financial assurance for closure, with the view that closure cannot be an exercise in perpetual care and
maintenance; it needs to be a means of returning the land to other useful purposes. An eye towards closure
must therefore influence every stage of the mine cycle, from exploration onwards. In Canada, this emphasis
is supported by federal and provincial/territorial legislation, financial security guidelines, and strict
enforcement. While advancements in mine closure and reclamation have been made, closing mines remains
one of the key challenges for this industry. Water and tailings management in particular are still key technical
risks for successful mine closure, and lasting benefits to local communities remains an essential consideration
in planning mine developments and thus closure.
This paper presents an overview of the shifting approach to mine closure from 50 years ago to today, as
witnessed by Mr. Pelletier, and anticipates how increased public awareness, changes in risk tolerance,
technical innovation, and an emphasis on environmental accountability will continue to drive advancements
in an increasingly complex and vital practice.
4

C.C. Ortiz Department of Biology, University of Santiago, Chile

M.A. Wilkens Department of Biology, University of Santiago, Chile
D.V. Barros Department of Biology, University of Santiago, Chile
J.K. Pizarro Department of Geography, University of Santiago, Chile
This article presents results of studies of the performance of biobased technologies for treating copper mine
tailings under field conditions. The technologies were developed to be applied during operation and/or the
closure and abandonment stage of mines to mitigate the environmental impacts of the residues. Both solid
and liquid wastes have a complex physicochemical composition characterised by a basic pH, heavy metal
content (e.g., copper), and high sulphate content. We first developed a strategy at the laboratory stage in
order to establish the critical parameters to consider in the pilot scaling stage. The systems are currently
operating to treat a hectare of tailings substrate and an approximate volume of 8L/s of water derived from a
tailings dam in the Second Region of Antofagasta, in Chile.
For biobased technologies, tolerant native plants and microorganisms were identified, characterised, and
acclimated for use in the field. After 13 months in the field, 72% of the individual plants were established,
dispersion of fine particulate matter (PM2.5) had decreased up to 46%, and about 60% of the copper was
removed from the site near the plant roots. For the treatment of liquid residues, we used a native consortium
of sulphate-reducing bacteria, achieving up to 23% of sulphate abatement after two weeks of treatment.
Copper removal occurred at a lower rate.
The main contribution of this work is the evaluation of biobased technologies in field experiences under
natural conditions on a scale that allows a realistic cost and feasibility assessment. The results obtained to
date indicate that biobased treatments can be recommended as an effective strategy with low investment
and operating costs. Also, they provide an excellent option for rehabilitation and reclamation of mining sites.
The current challenge is to improve the performance of the biological systems to render them more effective,
competitive, and accepted by environmental regulators.
Metals extraction generates economic benefits for society and creates wealth and jobs. However, mining
also generates environmental impacts such as water mine drainage and particle emissions from mine tailings
facilities. These impacts are felt in the deterioration of the quality of life and health of people, animals,
microorganisms, and plants. Therefore, the challenge is to develop technologies that are efficient, low cost,
and environmentally friendly in order to improve environmental standards both nationally and
internationally.
Chile is currently in a phase of greater concern regarding issues associated with environmental risks caused
by mine tailings deposits. One example is new regulations that favour the use of environmentally friendly
technologies; these include the law governing mine closure and mining facilities number 20551 (Ministry of
Mining, 2011) which incorporates issues of chemical stabilisation, and the incorporation of classical physics
stabilisation measures into documents of previously existing regulatory bodies, such as the Supreme decree
N° 248 (Ministry of Mining, 2007). Among the technologies favoured in these documents, those aimed at
rehabilitating degraded areas, such as phytotechnologies (technologies based on the use of plant species)

have proven to be efficient, environmentally sustainable, and affordable (Brix, 1999; Berti and Cunningham,
2000; Johnson and Hallberg, 2005).
The use of plants and the microbiota associated to their rhizosphere give rise to what we call biobased
technologies for the treatment of liquid and solid mining wastes. The use of metallophyte plants coupled
with microbiota isolated from tailings facilities offers an alternative for mining wastes treatment that can act
as a first step for rehabilitating a site (Terry and Bañuelos, 1999). The effect of the biological systems helps
to stabilise and improve the physical, chemical, and biological characteristics of the substrates, diminishing
the toxicity of the wastes. The use of biobased technologies prevents the dispersion of metals into the
environment, reduces the bioavailability of metals present in the substrates lowering their potential risk for
the biota and human populations nearby, and supports various subsequent uses of the area (Rousseau et al.,
2006).
The objectives of a mine closure project and rehabilitation plan may be to obtain environmental solutions
involving physical and chemical stability of soil and water. Therefore, biobased technologies aim not only to
control pollution but also to integrate the sites into the surrounding landscape. While the characteristics and
efficiency of biobased technologies are not yet state of the art in Chile, these technologies are ready to be
tried at the field scale. The Plant Biochemistry and Phytoremediation Group at the Universidad de Santiago
de Chile (USACH) has jointly performed a series of pilot studies with the mining industry and other universities
in order to introduce biobased technologies into tailings impoundments and landfills. For this purpose, a
methodology based on plants inoculated with microorganisms was developed to treat solid tailings residues
at the Talabre tailings facility in the Second Region of Chile. At the same time, a biological treatment
consisting of three treatment pools was installed to treat clear water from the Talabre tailings. Both systems
are currently in operation.
In 2009, the Plant Biochemistry and Phytoremediation research group at USACH started a research and
development project that aimed to develop technologies for stabilising residues derived from mining
activities. The methodology used in an initial field trial of 3,000 m2 was based on a selection of plant species
according to their tolerance to the substrate (Figure 1). The species were identified after being collected in
the area surrounding the Talabre tailings facility (Ortiz-Calderon et al., 2008).
The Talabre tailings facility belongs to Codelco and is located about 2,680 meters above sea level in the
Atacama Desert, 13 km southeast of Chuquicamata and 10 km northeast of Calama in the natural depression

formerly known as Salar de Talabre (Figure 2). The Talabre tailings facility began operations in 1952 and
currently extends over 46 km².
Field research included identification of vegetation present in the area surrounding the Talabre tailings
facility (50 km). Plants of the genus Baccharis, Phragmites and Atriplex with population densities of about
eight plants per square meter were identified. Plants of the genus Lycium were also identified; these showed
a lower population density. Plants were collected and propagated in hydroponics and then transferred to soil
for acclimatisation for five days under shade. The plants were subsequently transferred to nursery beds and
transported to the experimental plots. Then, one by one, the plants were planted according to a previously
designed matrix and inoculated with bacterial cultures previously isolated from a tailings dump. The
installation of the plant species in the field trial was monitored over 13 months, and the results obtained
were promising for the parameters evaluated (Table 1 and Figure 3). The project allowed researchers to
validate the methodology for treat tailings substrates. This involved an initial diagnosis and assessment to
preselect plant species from the nearby zone and characterise the substrate to be stabilised. The second
stage of the methodology involved elaboration of a general layout for implementation, including technical
feasibility, costs, and time. The main activities at this stage were selecting plant species through tolerance
assays, allowing for reproduction of the biological material, and acclimating and installing plants in the site
after the substrate was prepared. The final stage of implementation includes a plan for monitoring and
following the performance of the technology from day one. The development of the field trial of this
phytotechnology made possible to quantify the costs and time necessary to implement the technology in a
larger scale. This stage in the implementation is currently being undertaken under the brand name Fitoestab,
and the responsible for the technology transfer is a spinoff company named Ambiotek and created by the
researchers. This a successful case of a commercialisation of a biobased technology resulting from research
and development projects funded by the Chilean Government and with the support of a mining company.

Parameter
Parameter evaluated Before treatment After treatment difference
(%)
pH 4.0 5.8 45
Electrical conductivity 33 15 55
(mS/cm)
Cd (ppm) 1.3 0.9 31
Pb (ppm) 505 343 32
Zn (ppm) 882 474 46
Cu (ppm) 2,784 1,120 60
Organic matter (%) 0.7 1.1 57
Particulate material (20– 43 36 16

50 µm)
Particulate material 41 22 46
(≤2 µm)
In a tailing dam, the water stored normally is collected and recovered back for the process. This clear water
contains high levels of sulphate and mineral elements. Because the water pH is neutral to basic (7.5 to 8.0)
the metals are not highly solubilised, but the sulphate present can reach up to 3 times the level of the
regulation permitted contents. With the aim to develop a biobased technology to mainly remove the
sulphate present in the clear water of Talabre tailings, a biological treatment consisting of three pools was
installed in the site (Figure 4). For that purpose, a native bacterial consortium was isolated from tailings of
Salar de la India, Chuquicamata, at the Second Region of Atacama. The bacterial sample was grown and
stored in a modified Postgate C medium at 35°C in dark conditions. In the field, the bacteria grown as
previously described were mixed in a container with wheat straw as the carbon source, and the mix was
covered with plenty of wheat straw. The mixture was allowed to stabilise for three to five hours, and the
mixture was placed on a bed of wheat straw and arranged on the treatment pool. For the secondary
treatment at the third pool, plants of the genus Baccharis and Phragmites were used. The plants were
previously characterised and selected according to their tolerance to the water to be treated. Tolerance, in
this case, refers to the ability to generate leaf and root biomass.

Before the system was installed, the workplace was defined, the topographic study was conducted, and
manuals were developed for the design and installation of the pools. The excavation site was delineated in a
section of an artificial drainage stream. To proceed, the landfill site was sealed with a geotextile surface, and
the basin was filled with gravel and covered with earth removed from a hill near the work site. After that, a
new topographic study was performed in order to proceed for the excavation. Once the pools were excavated
to the depth and dimensions established, each of the pools was sealed with high-density polyethylene
(HDPE), which was then heat-sealed and fixed to prevent wind uplift. Three pools were built: the first for
water collection, the second for sulphate reduction with the laboratory-selected consortium, and the third
for secondary treatment with endemic plants. Hydraulic systems to connect each of the treatment systems
were installed next, and hydraulic tests were performed. We proceeded to fill the pools under a flow rate of
4–6 L/s, checked the system for possible leaks, and then proceeded to install the biological material. The total
filling of the three pools took 48 hours. A turbidimetric method based on the reaction of barium chloride
(BaCl2) was used to analyse the sulphate-reducing activity of the consortium in the field experimental pool.
For all samples, the concentration of metals was determined using atomic emission spectroscopy with
inductively coupled plasma (ICP-AES). The ICP-AES equipment was calibrated using standard dilutions of the
elements to be measured and the corresponding calibration curve. Finally, we measured all elements
simultaneously for each sample. Determination of sulphate, metals, and some physicochemical parameters
was made at points 1 (input), 2, 3, and 4 (output) in the pilot plant (Figure 4). The results showed that the
sulphate concentration in the water released by the tailing dam is variable over time, probably due to the
circulation of the residues deposited (Table 2).
4
1
2 3
Sampling Sulphate Sulphate Sulphate

points concentration, week 1 concentration, week 3 concentration, week 6
(ppm) (ppm) (ppm)
Input 2,850 8,758 3,883

Output 2,470 6,723 3,260
Sulphate 13% 23% 16%

removal
(%)
* The pilot plant was installed in April 2013.

The revegetation undertaken in Chile to stabilise mine tailings deposits has not considered some of the most
important constraints for these cases, such as physical fertility, metal toxicity, and the absence of the
microbiota responsible for recycling dead plant biomass. The use of microorganisms as biofertilisers is useful
in accelerating the treatment and rehabilitation of mining soils impacted by industrial activities, using few
amendments. The methodology has to be customised after a diagnosis of the system to be treated in order
to prevent dust dispersion, stabilise the particulate material, immobilise metals, and enhance physical and
chemical properties of the substrate. It is important to validate the biobased technology proposal through
on-site pilot assays to minimise the technological risk associated with the biology of the system.
In this context, phytostabilisation technologies have proven to be a much more suitable tool for revegetation
and physical and chemical stabilisation of solid mining waste. Expertise is required to support efficient use of
these new technologies at the local level and to raise more information about relevant native/endemic
species that are exclusive metallophytes. Chile has a high rate of endemic species (species that are unique to
a defined geographic location); these could be used successfully in different biogeographic regions of the
country, along with local breeders, to ensure that the system can sustain itself over the long term. This field
trial allowed a technological approach that costs about fifty times less than other solutions such as chemical
covers or removal.
On the other hand, a biobased approach to improving the quality of industrial water derived from mining
extraction activities was shown to be partially effective due to the variation in input sulphate levels. The
rather conservative results might be improved after long-term monitoring of the system. However, it was
possible to estimate that the cost of implementing this technology is between ten and forty times less than
chemical precipitation and reverse osmosis. In both cases, using the biobased solutions developed by our
group produces cost savings at the operational stage because of the small amount of energy required for the
systems in the field.
Project CORFO-INNOVA 09CN14-5795

Project FONDEF D06I1097 and V09P0022
Project FONDEF ID14I10151
Berti, W.R. and Cunningham, S.D. (2000) Phytostabilization of metals, in I. Raskin and B.D. Ensley (eds), Phytoremediation of toxic
metals: using plants to clean-up the environment, New York: Wiley, pp. 71–88.
Brix, H. (1999) How “green” are aquaculture, constructed wetlands and conventional wastewater treatment systems? Water Science
and Technology, Vol. 40, pp. 45–50.
Johnson, D. and Hallberg, K. (2005) Acid mine drainage remediation options: a review, Science of the Total Environment, Vol. 338,
pp. 3-14.
Ministry of Mining (2007). Regulations for Approval of Design Construction Operation and Closure of Tailings Deposits Projects,
viewed http://www.sernageomin.cl/pdf/mineria/seguridad/reglamentos_seguridad_minera/DS248_Reglamento_ Deposit
os Relave.pdf
Ministry of Minining (2011). Law 20551, Regulates Mine Closure and Mining Facilities. Last modified on the 14 th March, 2015. Id
N°1032158. http://bcn.cl/1q27k
Ortiz-Calderon, C., Alcaide, O., and Li Kao, J. (2008) Copper distribution in leaves and roots of plants growing on a copper mine-tailing
storage facility in northern Chile, Revista Chilena de Historia Natural, Vol. 81(4), pp. 489–499.
Rousseau, D., Lesage, E., Story, A., Vanrolleghem, P., and Pauw, N. (2006) Constructed wetlands for water reclamation, Desalination,
Vol. 218, pp. 181–189.
Terry, N. and Bañuelos, G. (1999) Phytoremediation by constructed wetlands, in Phytoremediation of Contaminated Soil and Water,
Boca Raton, FL: CRC Press, pp. 19–29.

R.S. Siwik Siwik Consulting Inc., Canada

M.D. Clemens McMaster University, Canada
Mines and metallurgical processes may last 10 to 20 years. During this operating period, tailings facilities are
utilised as water and tailings storage facilities. However, once ore processing ceases, a tailings facility may
continue to operate in perpetuity. Designing a tailings facility only for storage presents an interesting
paradox; facilities need to meet design requirements for short-term storage but must also address waste
disposal in perpetuity.
This paper explores the Canadian mining industry’s waste storage practices and their impact on local
communities, First Nations groups and the natural environment. Historically, most waste-related decisions
were made between mining companies and provincial government representatives responsible for the
promotion of mineral extraction. This paper reviews the history of waste storage in Canada, including the
evolution of design and decision-making criteria for site selection, containment methodology and closure
standards.
The paper then investigates current processes for mine closure design from a variety of angles, including
regulatory, engineering, social and environmental impact, and financial and management perspectives. It
explores the ways in which mining companies approach waste disposal facilities in perpetuity. Design for
closure in perpetuity requires a strategy and management process that goes beyond the traditional approach
used to design storage facilities.
This paper thus introduces a new paradigm that challenges the mining industry to move from mere conceptual
design of tailings and waste facilities to execution-level closure plans. Rather than designing these sites for
tailings storage, mining companies should design facilities that accommodate final disposal and permanent
closure. This paper examines the management approval and closure evaluation processes. Establishing a
more rigorous decision-making and project delivery process for closure design that includes community
engagement will improve the forecasting of planned decommissioning and restoration provisions and reduce
perpetual closure risks.
This paper provides a brief history of the mining industry in Canada from a mineral development and
regulatory perspective. It argues that Canada is not just a place of mineral wealth but a complex socio-
physical environment in which nature, local communities and First Nations groups intersect. The subject of
corporate social responsibility (CSR) in the context Canadian lands is explored. The approach entities use to
obtain a social license to operate is evaluated in terms of mineral wealth and mineral wastes, the extent of
their management systems and the decision-making process. This paper develops aspects of community
social responsibility in light of Canadians’ ownership of mineral wealth as citizens and explores an opportunity
for Canadians to become better advocates for Canadian lands. Existing gaps with respect to full engagement
in closure planning are uncovered to reduce the caretaker responsibility placed on communities. The final
section of this paper proposes the development of a closure agreement between a mining entity and the
community and/or First Nations communities. The principles of closure policy required to achieve an
acceptable closure agreement are elaborated. Part of this new engagement process for closure includes a

closure design process that would parallel the traditional mineral recovery design process but occur in
advance of the impact assessment process.
Early First Nations communities used copper for a variety of purposes prior to the arrival of the French
explorer John Cabot in 1497 (Allen, 1933). In 1535, Jacques Cartier observed indigenous peoples using gold
and silver in their tools and jewellery. Similarly, Jesuit missionaries in search of an overland route to the Far
East found First Nations mining sites near Lake Superior, Ontario, in the late eighteenth century. Europe’s
interest in the vast resources of the New World precipitated increased exploration, settlement and mineral
extraction. Under the auspices of France’s Louis XIV, furnaces were constructed in 1736 and operated until
1883 to smelt iron ore in present-day Quebec and Ontario. Underground coal mining began in 1720 to fuel
the Louisburg fortress in Nova Scotia. However, mining was sporadic and uncoordinated. As Allen (1933)
indicates, the lack of mining development between 1800 and 1842 was due to a lack of “systematic
investigation of geological conditions under some form of Government control.” Recognising the economic
potential of Canada’s resources, the federal government directed a comprehensive assessment of the
landscape. In 1842, the state established the Geological Survey of Canada (GSC) with a mandate to “read
rocks” and aid the fledgling mining industry in several ways (Zeller, 2009). The GSC provided detailed maps
of Canada’s diverse geography and various mineral deposits. Moreover, the completion of the first
transcontinental railway line in 1885 was an “outstanding milestone in the mining industry” (Allen, 1933).
The metal mining industry was established in British Columbia by the end of the nineteenth century.
In the second half of the nineteenth century, mining companies secured and exploited vast tracts of land
under the principle of free entry to mining. They acquired ownership rights to provincial mineral resources
and gained the right to extract these minerals by paying royalties and fees (Laforce et al., 2009). This “best
way to ensure that people exploit our mining resources” was enshrined in the Mining Law in Quebec in 1880.
This became the central theme in most Canadian mining jurisdictions: free entry to resources, right to obtain
a mining lease and security of tenure.
In the late nineteenth century, Canadian policymakers preached a non-interventionist approach to mineral
extraction. Raw minerals like coal, gold and nickel were critical to the country’s growth, and mining
companies were allowed to dig deep beneath the earth seemingly at will. As Canadian historian Morris
Zaslow (1971 notes, the government “treated the resources under their stewardship as if they were
inexhaustible and completely valueless in their natural state.” Even the conservation movement, which
vociferously contested the brazen individualism and wastefulness of Canada’s resource industry, was notably
silent on mining’s indiscriminate practices.
The state further aided the mining industry through political legislation: the establishment of the federal
Department of Mines in 1907; the first federal Mining Act in 1917, which specified that ores mined in Ontario
be refined in that province instead of in the United States; provincial mining legislation; the Roads to
Resources Program of 1958, which provided funding for the development of transportation corridors in
regions where mineral development was constrained by difficult access; and the reorganisation of the
Department of Mines and Technical Surveys in 1966 to the Department of Energy, Mines and Resources, with
a mandate expanded to include economic research of resource markets and the development of policies and
plans for resource management.
Significant revisions to mining regulations occurred between the 1980s and the 1990s, with federal and
provincial legislation addressing the requirements for impact assessment studies. Provincial governments
updated their environmental and mining legislations to require impact assessments as well as closure plans
and financial assurance to address the physical and chemical stability of mining activities in Canada.

By the late 1960s and early 1970s, Canadian provinces began introducing legislation that would regulate strip
mining and reclaim lands impacted by defunct coal mines (Errington, 2001). Alberta was one of the first
provinces to introduce guidelines for closing mines. In 1976, the Coal Development Policy of Alberta stated,
“The primary objective in land reclamation is to ensure that the land will be returned to a state which will
support plant and animal life or be otherwise productive”(Smyth and Dearden, 1996). In 1979, the Science
Council of Canada, a government advisory board established in 1966, recommended that the costs of
reclamation “be determined by testing small integrated programs that combined environmental protection,
land-use planning, and mining reclamation” (MacDowell, 2012). However the recommendation did not gain
much traction in parliament or within the mining industry, and many orphaned coal mines were left
unattended.
The history of reclamation activities in the mining industry dates back to at least 1916, when Inco (now Vale)
established an agricultural department to deal with erosion and landscape aesthetics (Peters, 1978). The
thinking during the 1970s was that revegetation would provide an oxygen-consuming barrier and reduce acid
mine drainage. Subsequent studies at the revegetated sulphide tailings dam showed no improvement in
water quality with revegetation (Veldhuizen et al., 1978). The issue of acid mine drainage came into focus
during the mid-1980s, when the Canada Centre for Mineral and Energy Technology (CANMET) and the
Canadian mining industry developed the Reactive Acid Tailings Stabilisation (RATS) program. The program
looked for cost-effective closeout technologies to allow rehabilitation of tailings and waste rock, allowing
companies to “walk away” from mining sites with the knowledge that the environment would be protected
in the long term. Subsequent collaborative efforts were seen in the Mine Environment Neutral Drainage
(MEND) and International Network for Acid Prevention (INAP) programs. Closure planning is a requirement
in all jurisdictions in Canada.
Provincial governments have the responsibility, on behalf of the public, to provide and enforce efficient and
effective legislation with respect to the development of mineral resources and management of wastes on
Canadian lands. In addition, First Nations have authority over activities on their land claims and have a say
with respect to impacts and benefits.
The notion of place goes beyond the physical; it also accounts for the social, cultural and ecological
relationships attached to a specific space. First Nations groups and other communities have a multi-faceted
view of a place defined by traditional knowledge, cultural values and material interactions with the
surrounding environment. Haluza-DeLay et al. (2013) describe place in terms of the physical and social world
and the realm of meaning, which comprises ideas, values and beliefs: place requires a “conscious attention
to the social and ecological relations of Canadian environments.” Haluza-DeLay et al. (2013) further state
that the “pluralism and the attendant cultural politics of the environment form a basis for Canadian-styled
just sustainability” that requires “equitable recognition of different cultural ways of understanding places —
the ways these may affect quality of life, environmental health, future generations, and equity.”
The relationships between land, government, mining companies and minerals and waste are illustrated in
Figure 1.

In the early 1900s, mining companies would develop resources on Canadian or First Nation lands in
conjunction with the provincial governments, as shown in Figure 1(a). There was little to no engagement with
local communities to occupy the surface rights for disposal wastes from mines and mills. Once a mine was
closed, mining companies abandoned their mining leases, and the provincial and/or federal government was
responsible for mining wastes on Crown land, as shown in Figure 1(b). There are more than 10,000
abandoned mine sites in Canada (NOAMI, 2010).
As shown in Figure 1(c), mining companies are now required to submit closure plans based on guidance
provided from provincial regulatory authorities and to engage communities in closure plans and social and
environmental impact assessments (SEIA). Under existing legislation, mining companies are held accountable
for their wastes until such time that closure is attained as a result of a walk-away closure scenario.
Impact and benefit agreements (IBAs) are instruments First Nations can use to address their social, cultural
and ecological relationship to their land. In an analysis of environmental issues associated with IBAs, Schafrik
and Kazakidis (2011) indicate that environmental guidelines, assessments and closure plans are the key
components in obtaining permitting rights. Today, mining companies must go through a more rigorous
permitting process to get community acceptance and regulatory approvals for new mines. Although
government regulations outline a regulatory process for engagement, there is complexity in dealing with
expectations from communities and First Nations, especially if there is uncertainty in the design for closure.
The Brundtland Commission defined sustainable development (SD) as “development that meets the needs
of the present without compromising the ability of future generations to meet their own needs” (WCED,
1987). Mining companies can and must remain sustainable to pay for the legacy of their previous mining.
However, once a mine operation is closed, the liability associated with mined rock, tailings and water
management facilities and the associated infrastructure is only sustainable if there is a corporate entity to
cover these legal obligations.
The mining industry faced a major milestone in the mid-1980s with the emergence of the concept of SD. The
industry looked for both self-regulation and the development of a normative consensus (Dashwood, 2014)
around economic growth, environmental protection and social responsibility. Mining companies have since
initiated CSR programs unilaterally, collaboratively with the rest of the mining industry or both. Major mining
companies have attempted to change the culture of their organisations by securing commitments from their

employees on environmental, social and economic issues facing their operations. A CSR policy is executable
within an organisation if there is competence and knowledge within a company’s management team.
The term “right to mine” has been replaced by the term “social licenses to operate” to address the need for
some form of social engagement. According to Boutilier (2014), the “danger is that the meaning [of this
concept] could become so blurred that its value as a tool for promoting collaboration becomes obscured or
lost.” The mining industry may continue to engage communities and, indeed, gain acceptance for a social
license to operate, but, in the end, communities may still not believe there is an actual agreement in place.
Acceptance comes only with some form of agreement, and communities may hold out on acceptance in order
to be clear about their disagreement. Hodge (2014) suggests the key success factor for moving forward in
any mining operation is “the creation of relationships with host communities and countries that are
characterised by authenticity, respect, integrity, inclusiveness and transparency.” What used to be a mining
company’s right to extract resources now requires a strategy and roadmap to manoeuvre through the
regulatory, community and First Nation landscape.
Mining companies use environmental management systems (EMS), such as ISO 14001, for policy
development, planning and execution, checking and corrective action and management review. EMS are
used to implement key management processes related to CSR. Levy (1997) argues that EMS seek “to reduce
environmental hazards to a politically tolerable level.” They function “on an ideological level to legitimise the
role of corporate management as stewards of the environment, deflecting the threat of more radical change
to the economic and social system.” Levy concludes that EMS limit a “broader debate over economic goals,
organisational governance mechanisms, and alternative approaches to addressing environmental problems.”
Taylor et al. (2001) found that in enterprises with EMS in place, the highest rated objective of an EMS audit
function was system conformance. Due process was the lowest rated concern with respect to environmental
performance. Therefore, conformance to EMS for certification purposes may lead an organisation to believe
it is legitimate in the eyes of its stakeholders, but in reality the organisation may not have gained validity. As
Nash and Ehrenfel (1997) note, “Without cultural change, practices will be of little consequence” because
the underlying cultural structures have not changed. However, mining companies that seek continual
improvement of their EMS and their environmental performance will likely see a dramatic improvement in
their legitimacy with stakeholders.
Effective change in the way an organisation addresses closure requires sound technical processes that include
an integrated project and EMS, as well as a buy-in strategy for the organisation to meet its CSR commitments.
Transforming a company’s existing management systems to meet closure-planning objectives requires a
change in management approach. Developing a closure strategy and execution plan will allow the
incorporation of a design for closure approach.
Closure is the longest part of a site’s life cycle, but usually receives the least attention during project
development and may remain conceptual until the mine nears closure. Figure 2 illustrates the timeframes of
a typical CEO tenure, mine life and functioning of waste facilities. The timeframe for decision making goes
beyond the typical five-year tenure of a CEO. Decision making with respect to closure is diligently executed
over a long period of time. Knowing that some mining companies come and go, governments and
communities need to be diligent in ensuring that good decisions are made with respect to closure, and that
closure provisions are available.

The core business of companies, Collins (2001) argues, needs to operate within three dimensions: what the
company can be the best at in the world, what drives the company’s economic engine, and what the company
is deeply passionate about. Since one core of the mining business is waste rock and tailings management
(90–99% of the production tonnage) mining companies need to plan to be the best at tailings and waste
management, and the closure of these facilities. This is achieved through “a series of good decisions diligently
executed and accumulated one on top of another.” As with SEIAs, closure plans must address environmental
impacts, community exposure and technical feasibility in order to meet communities’ expectations. Design
for closure has not always been part of the decision-making process. Traditionally, decisions relied on capital
and operating costs, and mining and processing selections. Whereas technical and financial aspects are
developed to the feasibility level, decisions with respect to closure planning usually remains conceptual.
The mining industry in Canada is required to comply with a regulatory process on par with other industrialised
first-world countries. The process of seeking government approval includes extensive preparations to plan
out a SEIA and engineering studies as part of the permit applications for a new project. This includes the
submission of a conceptual closure plan. This regulatory process is shown in Figure 3.
In addition to navigating the regulatory process, mining companies need to engage with communities early
on in the process with respect to a SEIA and closure plans. If mineral resources are near communities or on
First Nations land claims areas, additional engagement and consultation are required. The mining industry
may have to negotiate IBAs and community development plans to secure land access and infrastructure
development. Simply following the regulatory process in Canada is no guarantee that a mine will be allowed
to open.
Companies have developed specific policies to address major risks throughout their organisation. In the mid-
1980s, mining companies wrote environmental, health and safety policies. Eventually, SD policies were
written and CSR became a theme within policy writing. With the failure of the Omai tailings, some companies
developed internal tailings management policies. With the recent failure of the Mount Polley tailings dam, a
tailings policy is now the minimum requirement for corporate governance.
Another major issue facing industry is closure planning. Integrating closure planning into business practices
requires a policy statement to operationalise closure into the organisation’s management system. We
suggest the missing link in gaining acceptance of wastes on Crown or private land is a negotiated closure
policy, as shown in Figure 3.
Provincial departments regulate mine development in Canada and promote mine expansion in the public
interest. Boothe (2013) contends that a well-conceived regulation is one “made in the public interest” that
does not “serve the private interest of individuals or firms.” Based on regulations and policy, provincial
governments negotiate with mining companies for economic development, royalty payments, tax revenue

and environmental protection. The reality is that after mining development, lands will never be returned to
their original state. An opportunity exists for communities and First Nations to influence decisions related to
what closure looks like and how it is prepared for. Communities should understand that if their role is to act
as custodians of the public interest then they should be part of the decision making regarding the lasting
effects of mining operations. An inclusive agreement that accounts for the concerns of a community will help
the community to accept mineral wastes on its public lands and places.
Communities need to understand their role in the planning for the life cycle of a mine. First Nations can
influence decisions about mining wastes by incorporating conditions in IBAs for activities on their lands. In
the same fashion, communities need to be able to establish practices that allow decision making associated
with best practices in environmental protection and minimise risks from operating and closed mines in their
areas of influence. The missing link for communities negotiating closure with mining companies is a closure
policy. Developing a site-specific closure policy with local communities would improve the closure-planning
process.

A closure plan submitted through the regulatory process will never be able to tackle the requirements for
social responsibility fully unless a negotiated closure policy is in place that addresses the expectations of First
Nations and communities. If the industry wants to implement an effective closure plan, it should endeavour
to replace the “social license to operate” with a “license to close” based on a closure policy signed by the
local communities and First Nations.
This section proposes an alternative approach to securing an agreement for mines to properly close their
facilities. From a social responsibility perspective, rather than manage expectations of value, industry needs
to adapt the opposite message — that there will be no value remaining after closure to settle expectations
of perpetual social and economic benefits. A closure policy is the first step in negotiating closure. The industry
seeks social acceptance to operate. Gaining this acceptance requires agreement on what closure looks like
for the community.
The leader of a ship crossing the ocean is not the captain, but rather the designer of the ship. All operations
on a ship are a “consequence of design, which is a result of human intention” (McDonough, 1998). The final
layout of all waste facilities situated in watersheds is a result of the design and designer of those facilities.
Rather than closure plans being the outcome of a design process, closure should lead the design process. It
will require strong leadership to properly execute a closure plan within the governance, sustainability and
regulatory requirements, let alone within the scope, schedule, cost and quality requirements to meet
planned decommissioning and restoration provisions.
The challenge will be to engage communities in a design process that has traditionally been led by subject
matter experts and is based on technical jargon such as maximum credible earthquake and probable
maximum flood. The development of closure criteria is usually based on closure technology, regulations and
policies, cost, environmental and safety considerations and communities of interest. Incorporating
community concerns for safety and the environment could be facilitated with a process of engagement in
closure planning. Closure policy negotiation will be successful if there is belief that certain fundamental
conditions can be addressed. The ownership of lands and places must not be premised on the right to mine
but on a clear recognition that mining is taking place on Canadian and First Nations lands and places. A closure
policy will provide a clear process for developing socially acceptable closure outcomes.
A closure policy needs to have a set of both principles and instructions that include a commitment to partner
in the closure design process with industry, First Nations and community landholders. Principles for a closure
policy and their operationalisation are provided in Table 1. The design process for closure should lead rather
than conclude the mineral design process in order to ensure that mineral design processes address both
mineral wealth and waste.
The basis of design for waste facilities is well described within the industry based on internal standards,
regulations, industry guidelines, hazard classifications and operating requirements for those facilities. The
site selection criteria and constraints for placement of waste facilities should be developed in concert with
local communities to determine the ranking of selection criteria from a community perspective. The basis of
design for closure usually deals with cost, feasible closure technology, environmental considerations,
communities, regulations and safety. Communities should have a voice in the selection process for criteria
for closure, as environmental considerations, safety and closure technology impact local communities. To
properly engage communities in closure planning, they need to be part of the design process so that they
can understand the consequences of the design.
Independent reviews are an important part of any design process. A review of the design and closure
execution is important to identify fatal flaws and unacceptable risks to the community. A strategy to include

the community and First Nations in closure planning requires an agreed-upon mechanism for change. A
closure policy should state that it does not replace all the legal obligations of an entity as required by
provincial and federal legislation.
Principle Operationalisation
Community is part of the design process Community engaged in closure decision making
Commitment to design for closure Closure planning drives design process
Closure leads the design process rather Institute a closure design process in advance of
than lagging after the mineral design mineral design process
process
This process does not exclude Compliance with regulatory requirements
regulatory requirements
Independent review is required Independent review of closure plans
Change management is addressed Changes to closure design subject to review
Closure policy is operationalised Policy implemented within life cycle of company
and project management systems
Closure policy is a binding agreement Signed by community leader and company officer
Mining companies use some form of project management to execute advanced exploration projects, and use
an executive management decision-making process to accept or reject recommendations for the various
project gates. Figure 4 illustrates the traditional mineral recovery design process and the associated mining
life cycle. The challenge of developing an executable closure plan is getting the technical side not to
overshadow the social and environmental requirements of design. Too often, the environmental and social
aspects are considered at the back end of the production process rather than leading or being level with
production process design decisions. Front-end loading with economic and production metrics may leave no
room for innovative environmental and socially responsible design. Many SEIAs are amended because the
project definition changed during the design process. Rather than rush into developing conceptual closure
plans and SEIAs based on preliminary design concepts, it is more efficient to select technologies that will
meet both closure and operating requirements.
The three major aspects in the life cycle of a mine are bringing a mine into production, closure of a mine and
dealing with potential for new resources. There is an exhaustive body of literature on project management
techniques in developing a mining project. However, there is limited literature for a closure design process.
The process is compared to the mineral processing design process in Figure 4.

For this closure design process to work, a closure policy based on closure principles and operable goals needs
to be negotiated early on, during the project-stage gate. Front-end loading of a closure policy, budget and
schedule can be integrated into existing design and project approval processes. The accountability for
negotiating closure with communities should not be delegated to consultants, but rest with an officer of the
corporation.

The selection of the closure technology for waste should not be the output of process design, but rather an
input to process design. Being able to articulate design for closure parameters during the conceptual and
design phase may, for example, eliminate the need for cyanide in the process or a saturated tailings area in
a seismic area, or promote the requirement for sulphide separation. Zero maintenance requires that a tailings
dam is no longer energised, that is, the water table is so low within the tailings basin that liquefaction as a
failure mode is minimised.
“Getting the future right often seems more difficult that getting it wrong,” but “with foresight and
determination, pondered safeguards may spell out beforehand the success or failure of an undertaking”
(Richardson, 2009). Traditional process design looks at operating requirements to determine the amount of
material to be disposed of. A search for the most suitable sites subsequently prompts a selection based on a
life-of-mine model. The selection of alternatives should be based first on basin capacity and environmental
characteristics. In essence, the selected basin is the base case from which process design options will be
evaluated.
As shown in Figure 4, the life of mine will likely change. The selection a preferred basin should be based not
only on the proposed life of mine but the additional life of mine quantities. The selected basin helps set the
design basis and mineral processing technology. The proven closure technology selected during the project
phase may be improved upon during the operational life.
With the closure policy in place, negotiations about acceptable locations for mineral wastes have already
been addressed, leaving one less uncertainty with respect to project approvals. From the pre-feasibility level
through to execution, deliverables for mineral process design and closure design are the same, that is, true
costs are developed to +/− 15% in project financials, and engineering is completed at the same level of effort.
As the mine starts operating, the closure design process continues with progressive reclamation and auditing
for compliance with the closure policy and agreed-upon closure plan. Any changes in the life of mine and/or
closure technology can be accommodated.
In the past, Canada operated under laissez-faire principles of free mining, where there was unrestricted
access to resources and few barriers for obtaining a mining lease. Much of the focus of the mining industry
has been the extraction of resources on Canadian and First Nations land, which has left a legacy of post-mine
wastes. As regulations required more transparency and engagement, mining companies transitioned to
management system to address environmental concerns, and eventually to CSR. The fundamental
uncertainty communities have about a project is due to a lack of understanding or confidence in a mine’s
ability to mitigate adverse impacts. The unbalanced high level of effort in engineering studies versus the
conceptual nature of studies associated with wastes and impacts forces communities to question the risk
associated with mining, or even reject a mining project outright.
This paper proposes an approach to reduce the uncertainty about how mining wastes will be handled to
negotiate a closure agreement that communities and First Nations can accept. A closure policy process with
principles and operationalisation techniques is proposed to ensure that closure planning leads the design
process. A closure process design is proposed to guide the mineral recovery design process, which will lead
to certainty in design. Negotiating a closure policy with First Nations and community representatives will help
solidify agreements amiable to both parties concerning mine wastes placed on Canadian and First Nations
lands and places.
Allen, J. (1933) Milestones in the mining industry in Canada, presidential address, the Canadian Institute of Mining and Metallurgy,
pp. 155–168.
Boothe, P. (2013) Making good regulations, Canadian Public Policy, Vol. 39(3), pp. 359–370.
Boutilier, R.B. (2014) Concepts and Emerging Ideas, Frequently asked questions about the social licence to operate, Impact
Assessment and Project Appraisal, Vol. 32, No. 4, pp. 263–272.
Collins, J. (2001) Good to great, Harper Collins Publishers Inc.

Dashwood, H.S. (2014) Sustainable development and industry self-regulation: developments in the global mining sector, Business
and Society, Vol. 53(4), pp. 551–582.
Errington, J.C. (2001) Mine reclamation in British Columbia, twenty-five years of progress, in Proceedings 25th British Columbia Mine
Reclamation Symposium, Campbell River, Canada, pp. 20–31.
Haluza-DeLay, R., DeMoor, M.J. and Peet, C. (2013) That we may live well together in the land: place pluralism and just sustainability
in Canadian and environmental studies, Journal of Canadian Studies, Vol. 47(3), pp. 226–256.
Hodge, R.A. (2014) Mining company performance and community conflict: moving beyond a seeming paradox, Journal of Cleaner
Production, Vol. 84, pp. 27–33.
Laforce, M., Lapointe, U. and Lebuis, V. (2009) Mining sector regulation in Quebec and Canada: is a redefinition of asymmetrical
relations possible? Studies in Political Economy, Vol. 84, pp. 47–78.
Levy, D.L. (1997) Environmental management as political sustainability, Organization and Environment, Vol. 10(2), pp. 126–147.
MacDowell, L.S. (2012) An environmental history of Canada, University of British Columbia Press, Vancouver, Canada.
McDonough, W. (1998) A centennial sermon: design, ecology, ethics, and the making of things, Perspecta, Vol. 29 pp. 78–85.
Nash, J. and Ehrenfel, J. (1997) Codes of environmental management practice: assessing their potential as a tool for change, Annual
Review of Environment and Resources, Vol. 22, pp. 487–535.
National Orphaned/Abandoned Mines Initiative, Canada (NOAMI) (2010) The policy framework in Canada for mine closure and
management of long-term liabilities: a guidance document, pp. 1–140.
Peters, T.H. (1978) Inco Metals reclamation program, CIM Bulletin, pp. 104–106.
Richardson, J. (2009) Weighing foresight with due diligence and the precautionary principle, Foresight, Vol. 11 No. 1, pp. 9–20.
Schafrik, S. and Kazakidis, V. (2011) Due diligence in mine feasibility studies for the assessment of social risk, International Journal of
Mining, Reclamation and Environment, Vol. 25(1), pp. 86–101.
Smyth, C.R. and Dearden, P. (1996) An ecological audit of the regulatory requirements for surface coal mine reclamation in nine
western North American jurisdictions, in Proceedings 20th Annual British Columbia Mine Reclamation Symposium, Kamloops,
Canada, pp. 122–141.
Taylor, D.W., Sulaiman, M. and Sheahan, M. (2001) Auditing of environmental management systems: a legitimacy theory perspective,
Managerial Auditing Journal, Vol. 16(7), pp. 411–422.
Thornton, D.B. (2013) Green accounting and green eyeshades twenty years later, Critical Perspectives on Accounting, Vol. 24,
pp. 438–442.
United Nations World Commission on Environment and Development (WCED) (1987) Our common future, Oxford University Press.
Veldhuizen, H., Blowes, D.W. and Siwik, R.S. (1987) The effectiveness of vegetation in controlling acid drainage from base metal
tailings, in Proceedings Acid Mine Drainage Workshop/Seminar, 23–26 March 1987, Halifax, Canada.
Zaslow, M. (1971) The opening of the Canadian North, 1870–1914, McClelland and Stewart, Michigan, USA.
Zeller, S. (2009) Inventing Canada: early Victorian science and the idea of a transcontinental nation, McGill-Queens University Press,
Montreal, Canada.

H.W.B. Lacy MWH Australia

K.E. Bennett MWH Australia
Throughout 2014/15, with contributions from 13 highly competent sub-authors, the authors completed a
major revision of the 2006 Mine Closure and Completion Handbook for the Australian Government. The
original document went through a thorough gap analysis, with the intention being to close these gaps and
bring advancing paradigms and ideas from a rapidly evolving mining industry into the handbook so that it
truly represents current leading practice.
The 2015 Mine Closure Handbook (the handbook) introduces a number of aspects intrinsically linked to
closure, including legal and regulatory requirements; cumulative impacts; impacts on local and regional
biodiversity; climate change; post-mine land use opportunities; physical, chemical and geochemical
characterisation of soils and mine wastes; and engineered landform design. Interaction and consultation with
the community regarding closure during mining is considered integral throughout the handbook. The
interrelationships of these aspects throughout the seven phases of the mining cycle are described in the
handbook and now include post-closure management as a site moves toward post-mining land use and
relinquishment.
This handbook introduces the mineral resource legacy framework for general discussion around the issue of
legacy associated with the mineral industry, and the cyclical nature of mining and subsequent responsibilities
in managing that legacy. The interrelationships in the discovery and utilisation of minerals involve mining
companies, communities and government, and are represented simply in the framework.
Planned mine closure and relinquishment is still at the early stage of implementation in Australia. There are
limited examples in Australia’s mining history in which mine closure planning has been applied from
conception through to relinquishment. This is largely attributable to the timeframe of most mining operations
and the relatively recent development of integrated mine closure planning throughout mining life cycles.
However, the handbook showcases some of the work undertaken by the mining industry and minerals sector
in applying the principles of leading practice mine closure, and the authors touch on those examples
particularly relevant to legacy and community as case studies.
With the guidance of a steering committee of mining professionals, Australia’s federal government produced
its first mine decommissioning guideline in 2002 (Lacy and Koontz, 2002) as part of the best practice
environmental management in mining series. In preparing the original guideline, the authors were strongly
influenced by the release of the Strategic Framework for Mine Closure (ANZMEC/MCA, 2000), a robust and
simple framework established for the mining industry’s consideration in the closure aspect of its business,
and prepared by a working committee of the Australian and New Zealand Minerals and Energy Council and
the Minerals Council.

During 2006, an updated version of the original 2002 decommissioning guideline, the Mine Closure and
Completion Handbook, was written by a committee of 14, led by a chairperson, and prepared as part of the
publication of 14 themes developed by working groups of government, industry, research, academic and the
community. In the foreword, federal Minister Ian Macfarlane provided a foreword summarising the purpose
of the guidelines: “To assist all sectors of the mining industry to reduce the negative impacts of minerals
production on the community, and the environment, by following the principles of leading practice
sustainable development” (Australian Government, 2006).
The steering committee defined the purpose of the review as to “ensure the guide series to remain credible,
current and relevant in sharing LPSDP practices globally.” Among the general goals of the revision were to be
able to assist and benefit developing economies by sharing the experience of Australia’s mining industry and
to further demonstrate how industry expertise could improve and strengthen developing economies’
capacity through cooperative efforts between governments and industry both domestically and
internationally, and by sharing leading practices.
This current 2015 version, the Mine Closure Handbook (the handbook), is a general return to the original lead
author model of 2002, but differs in that it involves 13 subject experts and guidance from a qualified steering
committee. As the lead authors, we conducted a gap analysis, and were involved in coordination, completing
sections and editing to advance a considerable expansion in the document. The handbook also needed to
complement the other leading practice guides while retaining its quality as a standalone handbook. The
document has been enhanced with an update and increase in the number of case studies from six to 10.
The first action to improve the handbook and ensure the updated handbook reflected the rapidly improving
and advancing mine closure management process in Australia was to complete a gap analysis. We were asked
to conduct that process and provide a report to the steering committee that covered the following:
 an indication of any changes required in the existing Mine Closure and Completion Handbook to
maintain its credibility with the LPSDP audience;
 any changes to structure or content;
 a list of other experts we believed should be involved in reviewing or contributing to the updating
of the handbook;
 an estimate of how long such a review might take and the amount of work required.
After we provided the gap analysis, the steering committee provided feedback on our key findings. They
commented on the sub-author experts we wished to engage (currently listed in the handbook itself) and the
process we would follow, and recommended additional sub-authors. We proceeded to contact, guide,
manage and edit the work of the co-contributors to the gap analysis. We believe this approach ensured a
comprehensive document in which all contributors had the opportunity to incorporate their expertise and
thoughts within the handbook.
The key areas identified in the initial gap analysis as requiring enhancement included, but were not limited
to:
 the information and flowcharts representing closure as being an integral part of the life of the
mine;
 collection of environmental baseline data and material characterisation and incorporation of the
material characterisation and environmental baseline data into the landform design;
 consideration of local and regional biodiversity;
 development of closure objectives and completion criteria;
 the community and mine closure;

 identification, mitigation and management of risks associated with closure;

 financial provision and aspects of closure and risk management;
 the stakeholder and legal components;
 mine completion and relinquishment;
 the conclusion;
 the case studies; and
 the guidance on closure issues, closure options and techniques in Appendix B (which has been a
feature since the original handbook), including a reference to closure within the eEnduring vValue
Framework (MCA, 2006).
We concluded that to enhance the handbook, we needed to reinforce that a closure plan for an operation
must consider up-to-date industry practice, opportunities for developing technologies and the best available
and proven technologies on the market for sustainable and safe closure.
We wished to include more from the work of the International Council of Minerals and Metals, particularly
the key principals and elements to expand the primary audience for the handbook (ICCM, 2011).
In addition, we wished to reinforce that it is important for closure plans to be regularly reviewed and updated,
that is, that closure plans developed during the definition phase of a project must be updated after
construction to reflect the as-built status and changes. We also identified that a number of the items are
important enough to be framed within independent sections under broader subject sections; these included:
 Baseline data and biodiversity: More information on the type of baseline data that can be
collected and how these can be incorporated into the landform design at an earlier stage, an
aspect related to local and regional biodiversity.
 Characterisation and selective placement of materials, and mine waste management: Place
greater value on the importance of the characterisation and placement of waste materials and
discuss best practice, that is, incorporation of waste units in block models, scheduling and specific
handling when problematic waste will be mined. We wished to convey the importance of
designing a landform based on the parameters of the material, and constructing it in accordance
with the design from the onset of mining.
 Financial assurance and provisioning: The requirement to accurately identify the scope of current
liabilities and the cost of engineered solutions to close the operation to standards agreed upon
with stakeholders, and within all applicable regulations, was broadly discussed in the past but
could do with improvement and expansion.
 Mature operations stage: The information incorporated under the mature operations stage in the
2006 handbook was to be considered at a much earlier stage of planning in the revised handbook,
as this was leaving a critical planning component too late in the mine cycle.
Other gaps identified brought a greater focus to ongoing stakeholder consultation, agreement on closure
objectives, completion criteria, improving monitoring standards and performance by revising with up-to-date
methodologies.
Risk management is a key process today and was considered a gap in the 2006 handbook. In today’s mining
business, we believe it is essential to identify, mitigate and manage both current and future risks associated
with closure.

The lack of a legal section was a gap that had to be addressed along with a review and update of the mine
completion and relinquishment section within the 2006 handbook to capture additional information on
monitoring, maintenance and relinquishment.
The first mine decommissioning manual stated that “mining operations needed to approach mine
decommissioning and closure on a systematic basis from the very beginning of the operation” and added
that “mine closure planning must be a dynamic process including regular review and updating” (Lacy and
Koontz, 2002). We believe nothing has changed; this concept is as relevant as it was 15 years ago, and
effective implementation of a mine closure system requires:
 support from the company board or mine owners;
 commitment from the operation management, particularly the senior manager;
 an accepted closure systems framework;
 involvement of stakeholders;
 adequate resources (financial and human) to implement;
 managers and champions in charge of the system at the site level;
 regular system audit and actioning of outstanding items;
 regular reporting to the board from mine managers;
 ongoing commitment to funding for closure options research;
 acceptance by the regulators;
 monitoring to ensure long term viability (Lacy and Koontz, 2002).
These concepts have been reinforced in the 2015 handbook, and, rather than just discuss the creation
through many staged closure plans, we believe the handbook encourages companies and mine operators to
make closure a life of mine (LoM) process through the seven cycles contained within an integrated business
planning framework, as illustrated in Figure 1.

The 2015 handbook encourages planning for mine closure to be undertaken progressively throughout the
LoM. In order for mine closure planning to be successful, the management team needs to ensure closure
planning is systemised early in the mine life and integrated deeply into the normal mine business planning
and practice of the company, adopting an ethos of closure planning as part of normal business.
This approach ensures the practice flows into planning, consultation and implementation rather than being
attended to at the exhaustion of known ore resources. The initial groundwork, even at the exploration phase,
can impact the effectiveness and success of closure planning. The company can ensure optimal results if staff
actively engage in all the steps of closure implementation, and should ensure that stakeholder engagement
occurs strategically throughout the process of planning for mine closure and into the post-mining land use
relinquishment phase (Australian Government, 2015).
In conducting the upgrade and review of the important LoM cycle section, we specifically worked to improve
planning and design, which has doubled in content, followed by operations, which is slightly expanded, and
decommissioning and closure, which has also doubled. This is followed by a new section on post-closure
management and a section on closure planning that reflects a major change in addressing this gap. This was
such an extensive rewrite of the guideline that it is impossible to fully represent in this paper, and so we only
touch upon key aspects that we believe will be of most interest to the reader.
In many Australian states or territories, the regulatory authorities require a closure plan as part of the
approval process. This plan is used to assess the project, the environmental controls required and the long-
term potential liability posed by development of the mine. The revised 2015 handbook extended this section
markedly to stress that during planning, particularly for effective landform construction, some critical
information is required in relation to mined waste:
 the estimated volumes of waste material to form the landforms;
 the physical, chemical and geochemical characteristics of the soil and waste materials;
 preferred material positioning in the landforms to accommodate material characteristics and
volumes available of, for example, those materials that may be suited to placement near the
surface, and those that may be best contained deep within the body of the landform;
 the sequence and timing of different materials being made available via the mining schedule.
The nature of the site and the degree to which there is choice for the location of the waste landforms makes
every landform unique. If possible, it is important to consider the footprint location ahead of infrastructure
placement, as the location of the landform can have multi-million dollar closure implications. Critical factors
in location include:
 proximity to the open pit exit or exits;
 gradient of the footprint area, for both direction of drainage from the landform and the
implications for dumping costs;
 placement in relation to natural drainage, where possible avoiding blocking natural surface flow
or accommodating drainage beneath the landform;
 footprint constraints (tenement leases; future ore bodies; priority flora, fauna and ecological
communities; cultural heritage sites and infrastructure [current and future]);
 topography, consideration of visual impact and opportunity to complement the local landscape;
and stability of underlying material.

Mine and infrastructure construction has a major impact on a local region, and consultation has to be stepped
up so that local landowners and the local community are not unnecessarily inconvenienced, as this is often
when the foundations for long-term relationships are built. Sustainability requires that the complex
relationships between various risks be well understood, especially the potential for links between
environmental, social, political, economic and reputation risks, which can be seriously tested during mine
and infrastructure construction.
There are some critical phases in a mine life during which systemised closure activities may be overlooked,
particularly the mature operations stage and pre-closure stage. The 2015 handbook stresses that it is
important during the mature stage for experienced personnel to be involved in overseeing the construction
and placement of deleterious material in landforms. Failure to maintain quality control in this phase of the
mine operation can jeopardise environmental protection during operations with considerable consequence
post-closure. Consistency of workflow process, audit and well-resourced control are vital.
A mine’s pre-closure stage – when known ore resources are exhausted – is often an uncertain and
unpredictable period, dependent on final ore reserves, discovery and unforseen events can drive closure. It
is often far too late to develop a detailed decommissioning and closure plan at this stage, yet in many
Australian jurisdictions, it is only a requirement to take this step two years out from pre-closure. Research
has shown that, historically, 70% of mine closures in Australia are unforseen, (Laurence, 2005) and so this is
perhaps a risky position for regulators to maintain.
This section was expanded considerably as we identified many aspects requiring attention for
decommissioning prior to the post-closure phase that have to be addressed in order for a company to
relinquish a mine site with all obligations for future maintenance and funding discharged. Particular attention
has to be paid to the early establishment of verifiable completion criteria. Acceptance and approval by the
regulator with an agreed-upon verifiable process to monitor and demonstrate completion criteria and gain
relinquishment are very important. It can take considerable persistence by a company to achieve
relinquishment, particularly if early rehabilitation is inadequate for the task. Activities that can run for some
years at the decommissioning phase are numerous and are expanded in the text (Australian Government,
2015).
The 2006 handbook only briefly discussed completion, relinquishment and post-closure monitoring. We
believe that since that time, a maturing management cohort in industry, government and within the wider
stakeholder community has led to greater awareness of the need for post-closure management. The 2015
handbook includes a sustained and detailed process that addresses this matter.
The mining industry is embracing the concept that completion and relinquishment incorporate delivery of a
defined post-mining land use rather than just closure when the operational stage of a mine ceases and
decommissioning is complete. Thus, post-closure management (Chapters 3.7 and 8) (Australian Government,
2015) was added to the handbook to accommodate guidance for companies to manage post-
decommissioning conditions en route to and at relinquishment.
Even after the bulk of the mine infrastructure has been demolished and removed and the site has been fully
rehabilitated, there is a requirement for ongoing management and monitoring of the site. This phase
continues until final sign-off and relinquishment are achieved and the new land users take ownership and
responsibility.

Common closure issues, like acid mineral drainage (AMD), for instance, can have a long lag time before they
become evident. It is often necessary to monitor the success of revegetation, the effectiveness of cover
systems and any impacts on water resources, for many years until evidence of stability is on hand and
relinquishment can be obtained from the regulator. While leading practice involves early definition of, and
agreement on, completion criteria and progressive sign-off, some criteria may need to be monitored for an
extended period (possibly 10 to 20 years).
This phase of the closure process can require considerable resourcing, as there are many tasks, logistics,
personnel, safety issues and responses to change to consider. Retaining company staff or caretakers to attend
to post-closure management may require the retention of offices, amenities and equipment, but can be more
cost effective than looking to others external to the site, who may lack site knowledge and have large
mobilisation costs. There is more awareness in the Australian industry that some closed sites may have to
retain a medium- to long-term presence on site to meet the requirements, particularly those related to long-
term water treatment.
While the 2006 handbook briefly discussed some of the complexities in closure planning and the financial
aspects of mine closure, merit was seen in expanding these elements within the handbook (Australian
Government, 2015). Having the right information to make the best technical and social decisions in closure
planning requires the collection, assessment and management of environmental, social and economic data,
which have to be adequately managed to be easily retrievable and accessible.
Mining engineers, geologists, planners and, in some instances, consultants generally have the most influence
in mine planning and design. They need to take into account the mine closure issues and integrate economic,
environmental and social elements into the decision-making process. In order for mine closure planning to
be successful, the management team needs to ensure it is integrated into planning early rather than being
attended to at the end of the mine life, particularly in the construction and closure of mine waste landforms.
The handbook also makes the point that the management of water through all phases of the mining cycle is
as critical to closure as is the management of mine waste. The handbook (Australian Government, 2015) links
the characterisation of prospective waste materials and process residues (tailings), and the potential risk of
future contamination to the environment via waterborne pollutants with appropriate design of waste
landforms. Experience has shown that appropriate landform design needs to incorporate the management
of surface water and have the capacity to achieve target completion criteria.
Early recognition of financial aspects and overall closure costs promotes improved strategies for operations
to plan mitigation strategies and anticipate progressive closure and rehabilitation activities. Closure planning
and consideration of closure costs throughout the LoM can create shareholder value if this long-term closure
liability can be reduced or eliminated during operations. Chapter 6 of the handbook presents three processes
that can be utilised for closure cost estimation. These processes reduce the potential environmental liability,
as they ensure investment, development and operating decisions made today are made in full recognition of
the potential financial impacts for closure in the future.
The future of the mining industry is dependent on the legacy it leaves. Industry reputation is affected when
mines are abandoned or long-term detrimental environmental impacts emerge because closure was
notappropriately addressed during the LoM (Australian Government, 2006). . The industry today recognises
that to gain access to future resources, it needs to demonstrate that it can effectively manage and close
mines with the support of the communities in which it operates (Australian Government, 2006). Ongoing
engagement with the community throughout the LoM makes good business sense for companies as they
seek to contribute to the building of sustainable regional communities through long-term partnerships.
The word legacy is mentioned many times in the three closure handbooks and is a critical term. The word is
generally used with a negative connotation by the industry and its detractors; however, in reality, legacy is

defined as, and implies “a gift handed down.” (OED Online, 2014). We acknowledge that, depending on your
particular point of view, legacy can be either positive or negative.
Figure 2 illustrates the prime relationships in the discovery and utilisation of minerals: between mining
companies, local communities and government, and lists and describes some of the broad effects as verbs.
The common interface of these three primary stakeholders is the resource legacy itself.
By providing the legacy framework (Figure 2) within the handbook, we provide a visual context for
approaching the broader discussion of the nature of the mineral resource legacy in the interest of fostering
understanding in the continuing LoM cycles between company employees, local communities, regional
stakeholders, shareholders, company managers, non-government organisations and society as a whole.
Social equity and a new way of looking at the resource legacy are needed more than ever as conflict over
humankind’s search for and use of minerals, and the resource input necessary to extract and process
resources, have massive global effects. However, mining is a vital primary industry concerned with obtaining
or providing natural raw materials for conversion into commodities and products for the consumer. In
economic terms, it is these primary industries that determine our economic success
Within the handbook, we were able to draw upon examples from a number of the earlier handbooks that
are still considered viable and valuable for this subject, including leading examples of leading practice. In
addition, we added such recent examples as replacement of streamlines (Yogunup – lluka, Maryland Creek –
Glencore), power generation and multiple land use (Woodlawn – Veolia), progressive relinquishment of
underground coal operations (Kestrel – Rio Tinto) and many others.

The handbook we prepared with our joint contributors outlines that the following essential elements are
required to achieve mine closure and relinquishment:
 recognising and addressing the closure issues a mining operation needs to consider in its planning
for closure through to relinquishment;
 development of a risk management approach to mine closure planning that applies from mine
concept to post-closure and is integrated with whole LoM planning;
 closure activities associated with each step in the LoM cycle, and the need to integrate these into
the business practice via progressive implementation of a closure system;
 processes and tools that can assist a mining operation in achieving good practices in mine closure
and relinquishment;
 the need for engagement with communities and regulators in establishing and implementing
leading practice closure outcomes, as the community inherits the resource legacy;
 the requirement to collect quality baseline data, and development of a high-quality knowledge
base that is easily accessible;
 the requirement to develop closure objectives and completion criteria in the planning phase of a
mine in consultation with key stakeholders, and then regularly review them as research,
monitoring and progressive rehabilitation are undertaken;
 recognition that the physical, chemical and geochemical characterisation of soils and mine waste
are important components of engineered landform design and construction;
 recognition that mine tailings rehabilitation and closure require a unique focus;
 recognition that water management and its interaction with the mine landforms is a critical
closure element;
 the requirement to consider mine closure planning and associated financial provisioning across all
LoM phases, with estimates generally developed for provisioning purposes, regulator reporting
for environmental bonding and long-term LoM planning and budgeting;
 awareness that the pre-decommissioning and closure planning stage is critical and requires a
focus on aspects of liability, final decommissioning planning, assets and divestment, remediation,
legacy infrastructure, and post-closure monitoring and management; and
 advanced and careful planning to ensure the transition to the post-mining land use and
relinquishment is as smooth as possible (DIS, 2015).
Mine closure and rehabilitation ultimately determine the nature of the legacy left behind as a post-closure
land use for future generations. If mine closure and rehabilitation are not undertaken in a planned and
effective manner throughout the LoM, a site may continue to be hazardous and a source of pollution for
many years to come. The overall objective of mine closure and relinquishment is to prevent or minimise
adverse long-term environmental, physical, social and economic impacts, and to create a stable landform
suitable for some agreed-upon subsequent land use.
The 2015 handbook emphasises the need for mining to focus on developing closure objectives and
completion criteria based on the post-mining land use as the goal of mine closure. The handbook encourages
systemised closure planning, as unplanned closures are not cost effective and often result in sub-standard
rehabilitation outcomes. Early recognition of rehabilitation and closure costs promotes improved strategies
for operations to plan additional mitigation strategies and anticipate progressive closure and reclamation
activities.

Mine closure is a process; to be successful, it should commence with early planning, involve progressive
rehabilitation during operations and culminate with final decommissioning, rehabilitation and
relinquishment. Closure may be only temporary in some cases, or may lead into a program of care and
maintenance. In this sense, the term mine closure encompasses a wide range of drivers, processes and
outcomes.
The global minerals industry is and will continue to be a cornerstone of human civilisation and development.
The Australian leading practice series supports the Australian industry, and we hope the handbook will assist
in and complement the progressive mine closure process in any country that seeks assistance and guidance.
We acknowledge the contributions of 13 sub-authors, the steering committee, and the team at the
Department of Industry and Science, who approved the writing of this paper and its presentation.
Australian and New Zealand Minerals and Energy Council and the Minerals Council of Australia (ANZMEC/MCA) (2000) Strategic
framework for mine closure, ISR 2000/155, ISBN 0 642 721386, Canberra Australia.
Australian Government (2006) Department of Industry, Tourismand Resources (DITR), Leading practice sustainable development
guideline on mine closure and completion, October 2006, viewed http://www.industry.gov.au/resource/
Programs/LPSD/Pages/ LPSDhandbooks.aspx.
Australian Government (2015) Department of Industry and Science (DIS) Leading practice sustainable development programme for
the mining industry, mine closure handbook April 2015, viewed http://www.industry.gov.au/resource/Programs/LPSD/
Pages/LPSDhandbooks.aspx.
International Council on Mining and Metals (ICMM) (2008) Planning for integrated mine closure: Toolkit, L. Starke (ed), ISBN 978-0-
9553591-8-7.
Lacy, H., and Koontz D. (2002) Environment Australia, Best practice environmental management in mining – mine decommissioning,
Commonwealth of Australia, Canberra Australia. ISBN of series 0 642 19418 1.
ICMM (2008) Planning for Integrated Mine Closure: Toolkit International Council on Mining and Metals ISBN: 978-0-9553591-8-7
Edited by: Linda Starke
Laurence, D. (2005) Optimisation of the mine closure process, Journal of Cleaner Production, March 2005, Volume 19 pp. 278-28.
Minerals Council of Australia (MCA) (2004) Enduring value – the Australian minerals industry framework for sustainable development,
viewed 1 March 2015, http://www.minerals.org.au/enduringvalue.
OED Online (2014) http://www.oed.com/, last visited October 2014.

E. Sanzana Codelco, Chile

B. Weeks Golder, Chile
S. Fernandez Codelco, Chile
CODELCO, the National Copper Corporation of Chile, is the single largest producer of copper in the world.
Drawing on mineral reserves located entirely within Chile, Codelco’s annual production makes up roughly 10%
of the world total. In 2013, taking advantage of extensive changes in Chile’s mine closure laws, Codelco
undertook an ambitious program of updating and harmonising the closure plans for all eight of its operating
divisions. These divisions represent a challenging variety of conditions for closure, ranging from mines that
have recently begun to operate to mines that have been in operation for over 100 years; some sites face
possible partial closure in the near future and others have 60 years or more of anticipated life. This long mining
history has created some of the largest mining facilities on the planet, including open pits, foundries, and
waste deposits, to be closed. These sites include Chuquicamata, which has the world’s largest open pit, and
El Teniente, which is the largest underground mine. The largest single tailings facility to be closed will have
an estimated area of roughly 6,000 ha at closure.
Climatic conditions at the mines range from the driest desert in the world to a moderate Mediterranean
climate. Mine site elevations range from sea level to sites at nearly 4,000 m.a.s.l. in the Andes Mountains.
This paper presents a case study of how one corporation confronted a changing regulatory landscape to
establish a unified vision for closure planning, incorporating the results of past closure planning work and
harmonising closure criteria while respecting radically different operating conditions and costs and planning
ahead to sustainably meet future legal closure requirements.
The paper discusses special challenges, including those related to the closure of historic sites, very large
deposits, and facilities located very close to inhabited areas. Meeting the challenges of closure planning for
all eight divisions of Codelco within a limited time frame required planning that incorporated historic,
operative and new project knowledge in each division; environmental skills; benchmarking studies to evaluate
and optimise closure techniques; and identification of methods to unify cost estimating criteria while
respecting regional differences.
CODELCO, the National Copper Corporation of Chile, is the single largest producer of copper in the world and
owner of approximately 10% of the world’s known copper reserves. In 2013, taking advantage of extensive
changes in Chile’s mine closure laws, Codelco undertook an ambitious program of updating and harmonising
the closure plans for all eight of its operating divisions. This paper documents how the corporation undertook
to update and harmonise the closure plans for its properties within a limited time frame.
An independent, state-owned company, Codelco was formed in 1971 through nationalisation of copper
resources, and its modern existence was formalised in 1976. In 2015, Codelco consists of eight divisions, all
of which are located within Chile. These divisions include mines with extensive histories. One of the largest
and oldest mines is Chuquicamata. Prior to the arrival of Europeans in South America, this deposit was being

used by the Inca and Tiahuanaco peoples; its copper was highly valued for its use in bronze. Copper
production continued to grow in Chile through the colonial period and into independence in 1810. In the
twentieth century, international capital input led to the construction of more modern mines, including the
early development of El Teniente mine in 1904 and the modernised exploitation of Chuquicamata in 1910.
Deposits at Salvador, another division of Codelco, were first exploited nearly 1,000 years ago by the
Atacameño people, and industrial-scale exploitation was initiated there in the 1950s. Closer to the capital of
the country, resources were identified in the 1920s in Río Blanco, but industrial-scale exploitation of these
resources within the Andina division did not begin until the 1970s.
The mineral riches of the region near Chuquicamata district continue to be developed. Nearby mines have
been established and now form their own divisions, including Radimiro Tomic, which was discovered in the
1950s but did not enter into operation until 1998, Gabriela Mistral, which started operations in 2008, and
Ministro Hales, which began operation in 2010.
Codelco operates several foundries as part of select divisions. One foundry, known as Ventanas, is not
associated with any one mine but forms its own division. For the purposes of the mine closure law, it too is
considered a mine site, and it forms one of the eight divisions discussed in this paper.

Source: Codelco company reports and the World Bureau of Metal Statistics
The development of a mine closure law in Chile has taken place over a relatively short period of time. In 1994,
an environmental impact assessment (EIA) system was put in place, and while closure was not directly
addressed by the system, the EIA for new mining projects was expected to include the closure phase. The
next major step forward for mine closure regulation came in 2002, with the introduction of a new mining
security regulation. This regulation established, for the first time, a requirement that all mines present
closure plans to the National Mining and Geology Service (Servicio Nacional de Geología y Minería, or
Sernageomin). When the regulation was introduced, it established a deadline for existing mines to assure
that they had filed closure plans with the authorities by 2009. This deadline was complied with by all of the
divisions that made up Codelco at this time.
The focus of closure plans in 2009 under this law was on physical stability and health and safety issues.
Notably, environmental issues were not covered, as they were considered to be outside of the jurisdiction of
Sernageomin, and there was no requirement to assess or indicate the cost of implementing closure
measures.
In 2012, a new closure law and associated regulation was published. These new documents dramatically
increased the role of the state. Responding in part to public pressure and increased attention to the large
number of abandoned sites in the country, the new law explicitly laid out a mission to prevent the future
generation of more abandoned sites. This would be done by means of financial guarantees provided by the
mining companies for each operation, providing amounts sufficient to execute site closure should the owners
default on their closure obligations. Such guarantees are now a relatively common practice internationally,
but they represented ground-breaking change in Chile. The new law also touched on the social aspects of
closure, obligating companies to indicate in the new plans when and how closure would be communicated
to stakeholders.

The new law included a phasing-in process, known locally as the “transitory regime,” which set out simplified
requirements. Under this regime, mining companies were required to present the authorities with an
estimate of the value of the closure commitments they had acquired through statements made in the EIA
process, through the previous closure plan presented to Sernageomin, or through other permissions with
Sernageomin that made reference to closure activities. This estimation was to be presented in a document
that summarised a description of the site, its environmental setting, the closure commitments from the
various sources, the methods the site planned to use to comply with these commitments and the costs.
At the time this regulation came into force, Codelco had all of its divisions in operation, and their closure
plans had been approved under the previous law. This meant that all divisions qualified for the transitory
regime. The transitory period ended in 2014, and all future updates to closure plans must be completed in
compliance with the full extent of the new law, which requires companies to undertake a risk evaluation for
principal installations, intended to ensure that closure measures are sufficient to provide physical and
chemical stability, and to present a communication plan for the closure.
Under both the full and transitory regimes of the new law, each site is required to present an annual
guarantee based on the nominal value of the closure. The nominal value of the closure is calculated in
Unidades de Fomentos (UF), an inflation-indexed parameter commonly used in contracts in Chile.
Once the cost estimate for a closure is approved by the authorities, the guarantee is provided using one or
more of a series of approved financial instruments. The present value of the closure guarantee is calculated
using a formula that takes into account remaining mine life and a discount rate based on an independent
state-provided index. Companies must guarantee 20% of the present value in the first year, and the
guarantee amount gradually increases to the full present value over a period that is either two-thirds of the
remaining mine life or 15 years, whichever is shorter. There are various provisions for partial reductions of
the guarantee to promote progressive closure.
When the new closure law was announced in 2012, Codelco began a centralised planning program to ensure
that all eight of the divisions operating at that time would comply completely with the letter and the spirit of
the new law.
An overall closure planning schedule was established, and realistic key milestones were defined for activities
needed to meet the legal deadlines for the new law and to ensure that annual updates of closure costs and
provisions would be completed on time and in accordance with international financial reporting standards
(IFRS) requirements. The schedule incorporated information from the updates to the plans, as it became
available.
Figure 3 shows key activities in this schedule, which included updating the corporate guidance for closure
plans, running a competitive bid process to select the engineering company to develop the updates for all
divisions, preparing the plans, permitting sufficient time for revision of each plan by the corresponding
division, and presenting the plans to the authorities on or before the legal deadline. This schedule was
maintained almost without variation through the two-year period from the publication of the new law to the
submission of the plans prior to the deadline.

2012 2013 2014 2015

Programmed Activity
N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J
1 New Mine Closure Regulation (Sernageomin) Award of Study
2 Update Corporate Provisions (Audit)
4 Update corporte closure directives and standards Presentation of Closure Plans
5 Bid Process for Closure plan Updates under 20.551 within Legal Timeframe
6 Develop Corporate Closure Plan studies

7 Corporate Approval of Bond amounts
Definition of Financial
8 Present Closure Plans to Sernageomin provisions
9 Questions by the Authorities (legal period)
6.1 Support for responses, cost updates, etc
Update of Financial
10 Approval of Closure Plans by Sernageomin Provisions
11 Payment of Bond (1st Installment of 20%)
6,2 Update of Closure plan under DS132
12 Technical visits/seminars/training
13 Administration and Owner's Engineer
Notes Owner Activity or Milestone

Primarily Activity of Consultant
Authority Activity
Codelco’s principal closure planning criteria are defined in an internal document, Manual for Mine Closure
Projects, which lays out the main aspects of closure planning for the corporation. Following publication of
the new closure law in 2012, this manual was updated in early 2013 to reflect the new legal scenario. It
includes clear definitions of the corporate roles and responsibilities for closure planning.
Following the publication of the new closure law, the following criteria and strategic considerations were
established for updates to the required closure plans:
 Full compliance with legal requirements will be maintained.
 A single corporate team will guide updating of all closure plans, to ensure a consistent approach.
 Where possible and practical, closure measures will be applied such that similar designs will be
used at each division.
 Cost estimates will be developed using a single methodology. While unit costs may differ between
divisions, the form of calculation will be the same.
 All cost estimate inputs, including unit costs and quantity take-offs, will be full traceable to
facilitate future updates.
 Financial provisions and cost estimates for the guarantee will be aligned. Due to differences
between the mine closure regulation in Chile and the IFRS guidelines, it will not be possible to
guarantee that the amount provisioned and the amount to be guaranteed will be the same.
However, both amounts will come from the same base set of information.
The previously described transitory regime of the closure law also defined specific criteria that turned out to
be important in defining the content of the updated closure plan. In the transitory regime, the owner was
required to put a value on the existing closure commitments that had been acquired through the approvals
system. The authorities indicated that, during the transitory regime, there would be no possibility of
modifying or updating the existing closure commitments, even if strong technical or economic reasons for
modifying the commitments existed.

As previously measured, the legal context placed significant restrictions on improvements and optimisations
that could be applied to the closure measures while still complying with the stipulations of the transitory
regime. However, there was still a significant margin for optimising closure measures. As part of the
development of the closure plan for each division, a complete register was developed of all of the closure
commitments that had been acquired for each installation. Closure specialists then defined, for each
installation, the specific activities that would be considered to ensure compliance with all the existing
commitments and that would represent the optimum measure from a technical and economic standpoint,
within the constraints of the existing commitments. For example, if an operating permission indicated in only
general terms that an area would be covered at closure, in the absence of more specific indications, the
closure team would define the thickness and type of cover to be used to achieve the stated or interpreted
aim of the closure measure.
During the development of the plans, it was found that measures related to demolition, construction of
covers, and closure of underground mines all contributed significantly to the closure costs at some or all
divisions; within the constraints of the existing closure commitments, these provided considerable margin
for optimisation.
Demolition and removal of the industrial installations and buildings associated with a site is one of the most
common and least contentious closure measures. However, the associated costs are a critical component of
overall closure cost. As others have noted, this cost is frequently estimated poorly (see for example Fletcher
et al., 2011). The time needed for demolition is another important component of the estimates for
completing closure work. In this study, the time needed to execute earthworks was evaluated and compared
with the time needed for demolition and removal of industrial site installations. The longer of the two times
was used to define the overall duration of the closure period — a factor which was extremely important to
the indirect costs.
A demolition specialist was brought in to evaluate closure needs for the buildings and equipment at each
site. A first-level review of demolition needs used information such as site plans, photographs, satellite
images and reports. The specialist worked closely with people familiar with the site to understand not only
the size of the industrial installations but also their age, the types of residual materials that could be present
(such as asbestos or contaminated soils), and the presence of other factors that could impact demolition
logistics. These other factors included considerations such as high altitude, difficult winter conditions, access
restrictions within the site and nearby environmentally sensitive zones or populations. Other special
considerations were also taken into account. For instance, the foundry at the division Ventanas includes a
stack that is approximately 155 m high. While this emblematic stack could potentially remain after closure as
a monument, its removal was factored in for costing and demolition planning purposes.
The closure commitments of seven of the eight divisions include the need to construct some type of cover.
These covers are generally required over tailings facilities or waste rock dumps at closure, most often to
provide control of dust generation or to promote chemical stability. Taken together, they represent a large
financial commitment, accounting for roughly 30% of the total direct closure costs.
As an example, the Talabres tailings storage facility at Chuquicamata mine is anticipated to have a total area
of 5,600 ha at closure. Because the tailings facility is located in an extremely arid zone, where average annual
precipitation is around 10 mm per year, and there are concerns about the potential for the long-term
generation of dust from the dry tailings surface, a closure cover of granular material has been specified. With
such a large areal extent, the thickness of the cover considered will have a large impact on the closure cost
for both the facility and the entire division. A cover thickness of 15 cm was selected for costing purposes,
largely based on constructability. Information in the literature suggests that an even thinner cover could

achieve the required reduction in dust generation. For instance, Blight (2007) reported on various
configurations of covers for reducing wind erosion at tailings facilities and showed dramatic results with a
cover of coarse material as little as 10 cm thick. For these sites, definition of rock cover particle size and
evaluation of weathering rate would not be considered until the detailed design phase.
Due to the extreme aridity of the mines in the northern divisions, covers for reclamation in the traditional
sense are not considered. Regionally, vegetation densities are so sparse that covers with self-sustaining
vegetation are neither reasonable nor feasible as a closure goal.
Several of the Codelco divisions include significant underground mining works. Most notably, El Teniente
mine has the most extensive underground mine in the world. Operated since 1904, the underground works
currently total drifts of over 2,400 km, with over 1,500 km of underground road and a major expansion with
a new mine level now in development. Major underground mines also exist or are planned in other divisions,
including Salvador, Andina and, in the near future, Chuquicamata.
Applicable closure practices from around the world were evaluated in order to refine the planned closure
activities for the mines. Some of the measures adopted were as follows:
 For material underground, salvage is considered for all of the equipment that can be removed
economically and safely.
 The scrap value of the equipment is to be defined in contrast to salvage value. This definition is to
be done as close to possible to the date of closure due to the variability of scrap values (see
Fletcher et al., 2011)
 As salvage value cannot be considered either for financial provisions or for the calculation of the
guarantee with the state, the majority of the equipment will be considered (for financial
purposes) to be abandoned underground.
 Hazardous or potentially hazardous materials will be removed, and appropriate costs will be
considered for disposal in accordance with environmental regulations.
 Access control at surface openings is generally considered to be by means of permanent controls,
which minimise or eliminate the possibility of third parties accessing the underground works.
These permanent controls are typically massive rockfill plugs, although other controls are
considered where specific needs dictate.
 Ventilation needs and possible water management controls at the access points require further
study and definition; these will form part of the detailed closure design for each site where
applicable. Hydraulic control needs are limited due to the hydrogeological conditions and
configuration of the underground works for the majority of the mines.
Many of the divisions present special challenges or unusual conditions. Some of the divisions are more than
100 years old and have been important to the history of mining in Chile. The Teniente division is home to the
historic Sewell Camp, which was designated a World Heritage Site by UNESCO in 2006, having been previously
declared a national monument by the Chilean government in 1998. Closure activities in the area will need to
be sensitive to the preservation of this now-historic site. Figure 4 shows an overview of the camp site as it is
today.

Most of the divisions have the advantage of remote locations, which isolate nearby communities from
potential direct impacts of closure activities. However, the Ventanas division is located beside the community
of the same name, and the implementation of closure activities will need to take into account the proximity
of the division to the community. This has been evaluated as part of the demolition planning. (As this division
is a foundry, the majority of its closure activities are related to demolition.)
Scale can present a special challenge for mine closure. In the case of the Chuquicamata division (DCH), the
main open pit that is anticipated at closure will be in the order of 4.5 km long, 2.8 km wide, and over 1 km
deep, with a surface area of nearly 900 ha. As the largest open pit in the world, it will likely remain after
closure as a permanent monument to the mining that took place. Closure measures will focus on ensuring
that the pit does not pose a health and safety risk, with access to the pit and its look-out points carefully
controlled. Other installations of notable size at closure include the Talabre tailings facility (5,600 ha), the
Chuquicamata spent heap leach piles (8,200 ha), the Ovejeria tailings facility (1,900 ha), and the Caren tailings
facility (3,300 ha), to name a few. Optimising closure measures for facilities with these sizes is critical for
controlling the overall cost of closure.
Due to the need to provide a financial guarantee to the state for the first time, special emphasis was given
to the cost estimates for closure and post-closure works. Closure costs for the eight divisions had previously
been calculated as part of internal planning exercises and establishment of financial provisions in accordance
with IFRS. However, the current study represents the first time that all eight divisions have had their costs
evaluated in parallel using unified criteria to develop cost estimates.
The methodology used for costing was conventional. Quantities were estimated based on the closure works
previously defined. Site-specific unit costs were developed for closure activities and applied to the quantities
to obtain direct costs. Indirect costs were calculated based on the anticipated duration of the closure
activities and the specific needs for engineering, supervision, studies and other supplementary work needed
to execute the closure. A contingency amount was added to the sum of direct and indirect costs; for post-

closure work, the present value at the year of closure was also added. Following the stipulations of the local
regulators, closure costs also included a percentage allowance for taxes.
For most divisions, quantities had been defined to a high level of accuracy due to detailed studies that had
been conducted on site by others. As a result, a significant part of the work in the current study consisted of
compiling and revising the previously collected data, incorporating more recent modifications in the mine
plan, and checking that numbers and assumptions were reasonable.
In the case of one division, the detailed in-the-field quantity take-offs had not been carried out. In this case,
quantities were estimated where needed using site plans, satellite images, and information provided by site
staff. This difference in methodology was accounted for in the final costing by applying a higher contingency
to the final cost estimate for this site, as discussed below.
Unit prices for closure activities were generally developed based on the cost of workers and machinery
needed to execute the work. As per the guidelines of the state, both of these were defined as the costs for a
third party to execute the work.
Labour-hour costs took into account typical salary and overhead costs for the region in which each site was
located, considering that, in the context of the closure of any of these sites, there would likely be an
unoccupied labour pool available. Rates were also adjusted for the level of complexity of the anticipated
work (a lower level of training is needed to take down an electrical distribution system than to set one up,
for instance).
Machinery costs were developed considering partial depreciation of the machinery over the duration of the
closure work as well as costs for fuel, supplies and maintenance. Productivity estimates for the machinery
were based on experiences at other mine sites with similar equipment. These also took into consideration
the expectation that there would be less interference at a site that is in closure compared to the stand-by
times that are typical at an operating site.
Other significant factors in the development of unit costs for each site included the hauling distances and
tipping fees for sites authorised to receive hazardous wastes and the distance to (or construction cost of) a
site adequate for receiving non-hazardous solid waste generated in the closure.
As mentioned, the closure guarantee must include a provision for post-closure activities at the site. The
current post-closure commitments generally consist of monitoring requirements for air, water, and
groundwater, and/or operating some type of treatment facility for a defined period of time after closure.
Values were assigned to these activities considering typical costs for the specialists, laboratory analyses and
other expenses needed to execute the work.
This calculation typically resulted in an annual cost for the duration of the defined post-closure period. Using
the discount rate defined by the state, the present value of this annual series at the start of the post-closure
period was calculated for inclusion in the total amount to be guaranteed.
The new regulation for mine closure explicitly requires that a provision for contingencies be included in the
closure cost estimate, but neither the regulation nor the authorities provide any direct guidance regarding
the amount of contingency that should be applied. Under Codelco’s corporate guidance policies, the level of
detail in the studies for seven of the eight divisions aligned most closely with a pre-feasibility study, and the
level for the remaining division was considered closer to a conceptual study. Based on corporate policies, the
appropriate contingencies were applied for studies of these levels.

For the owner of a site, the contingency has a very specific and clearly defined role. For any given engineering
cost estimate, there is a risk that the cost will be higher than estimated. A contingency amount is a planning
tool to help reduce this risk that the true cost may be in excess of the estimated cost. The amount of
contingency reflects both the confidence in the original estimate (which should be related to the level of
engineering) and the owner’s tolerance for risk. One could easily imagine a case of two owners: the one with
a low tolerance for risk would apply a 35% contingency to a conceptual study, and the other, with a higher
tolerance for risk, would apply a 15% contingency to the same study.
Not surprisingly, it became apparent during the review of the closure plans that the state had a different
tolerance for risk that the mining company, and in the absence of clear guidance, the amount of contingency
needed to be negotiated on a case-by-case basis.
The transitory regime ended on November 12, 2014. This means that all future updates to closure plans in
Chile will need to take into account all of the stipulations of the closure law, including the five-year maximum
period between updates. This means that all of the closure plans must be updated by 2019 at the latest.
However, it is far more likely that plans will be updated sooner, as the law requires a closure plan to be
updated in response to any significant change in the mine plan. While there is some uncertainty regarding
the interpretation of a “significant change,” several divisions are considering or developing major capital
investment projects that would certainly trigger updates if approved (environmentally and corporately).
For these updates, all closure plans will include a risk evaluation for the principal installations, and closure
measures will be defined to mitigate the risks. Sernageomin has developed a guide for conducting this risk
evaluation. While it is not legally binding, this guide is likely to be treated as the rule. It indicates that, in the
absence of specific studies, the evaluation must consider worst case scenarios. For example, in the absence
of geochemical characterisation, all waste rock should be considered acid generating. To address this,
Codelco is developing and refining a program of future studies and updates to ensure that the next
generation of closure plans is developed with sufficient backing studies.
Updating the closure plans for all divisions of Codelco was a technical and logistical challenge. Detailed
planning was needed to ensure that the closure plans complied with the stipulations of a new closure law
and incorporated the commitments for closure that had been previously made to both the national
environmental authority and the national mining service. At the same time, the updated plans incorporated
the knowledge and experience of the site staff and information obtained from specialist studies while at all
times maintaining a transversal alignment with corporate guidelines. Optimising and unifying closure
measures where possible, the eight closure plans with their attendant cost estimates were presented to the
authorities ahead of time. At the time of the preparation of this paper, the plans are being reviewed and
evaluated by the authorities, and the nominal corporate closure cost is around US$ 5,300 million.
The authors of this paper would like to thank the staff of Codelco and Golder who supported the development
of the updated closure plans, in particular Patricio Oyander (Teniente), Max Kobeck (Andina), Ana Maria
Galvez (Chuquicamata), Yermolay Peralta (Salvador), Alejandra Acuña (Salvador), Patricio Cordova
(Ventanas), Marcelo Navarrete (Gabriela Mistral), Leonardo Herrera (Radimiro Tomic), Alejandro Dagnino
(Ministro Hales), Ernesto Ortiz, Andres Lopez, and Julio Olivos.
Blight, G. (2007) Wind erosion of tailings dams and mitigation of dust nuisance, The Journal of South African Institute of Mining and
Metallurgy, Vol. 107(2), pp. 99–108.

Fletcher D., Hutton A. and Dick C. (2011) The “smoking gun” of detailed mine closure cost over-run – a review using case studies of
the real costs associated with the demolition and removal of infrastructure, in Mine Closure 2011, Proceedings of the Sixth
International Conference on Mine Closure, 18–21 September, Lake Louise, Canada, A. Fourie, M. Tibbett and A. Beersing
(eds), Australian Centre for Geomechanics (ACG), The University of Western Australia, pp. 97–106.

C.D. Grant Anglo American, Australia

P.R. Botha Anglo American, South Africa
Integrating closure planning with life of mine (LoM) and other mine-planning processes during various time
frames is not a new idea. While most would agree on the benefits of such an approach, examples where the
potential is realised remain the exception. We propose that this is often because of a lack of ownership by the
existing mine leadership teams as closure is seen as something that is distant and potentially someone else’s
problem. There is often a compelling case for changing current operational strategies from both an
operational expenditure perspective and a closure liability viewpoint. A sound business case can be made on
either or both fronts with immediate benefits to the current mine management team. It is critical that the
proposed change is framed in a business case context. This paper presents three case studies from Anglo
American mining operations around the globe. The first case involves a change in mining fleet and strategy
in Australia’s largest open-cut coal mine. Terrace mining, involving progressively backfilling voids, was
implemented instead of traditional dragline strip mining. The second case involves the use of a revised waste
rock deposition (WRD) strategy to impound fine and coarse tailings at an open-pit diamond operation in South
Africa. The fundamental shift in philosophy here was undertaken at effectively no additional operating cost
and resulted in the LoM closure liability being reduced by 35%. It also improved the stability of the tailings
deposits and eliminated the requirement for an additional storage facility, saving significant capital. The third
case involves a change in waste rock dump construction methodology at a large iron ore open-pit operation
in South Africa. Changing WRD to the final landform allowed progressive rehabilitation to commence,
reducing liability, while an additional opportunity to consolidate numerous waste rock dumps into one
megadump fully utilised available airspace and decreased the disturbance footprint. Work continues at this
location to further integrate planning processes and realise the full potential value, which has been estimated
at more than a 20% reduction in closure liability. These case studies demonstrate that changes to operational
strategies that reduce closure liabilities can be made if site leaders accept ownership based on a compelling
business case. The key to closure planning is not to defer closure costs but to eliminate future closure activities
through integrated closure and LoM planning.
Anglo American is a global and diversified mining business that provides the raw materials essential for
economic development and modern life. Anglo American has mining operations, growth projects and
exploration and marketing activities across southern Africa, South America, Australia, North America, Asia
and Europe; these involve a diverse array of commodities, namely coal (thermal and metallurgical), copper,
diamonds, iron ore, manganese, nickel, niobium, phosphate and platinum. Anglo employs almost 160,000
people worldwide.
The concept of planning for mine closure has now existed for many decades (Australian Government, 2006).
The philosophy is that closure should be considered throughout the life-cycle of a mine, from cradle to cradle
(ICMM, 2008). Regardless, there remain very few examples worldwide of successful mine closure and
subsequent lease relinquishment, with notable exceptions (e.g., Grant, 2006). Part of the problem is that
closure planning is not integrated with other mine-planning activities more relevant to the day-to-day

operation of the mine (e.g., short/medium mine planning, LoM planning). This is generally due to the lack of
ownership of closure issues by the current mine management teams, as closure is either seen as being too
far into the future or someone else’s problem. This issue is exacerbated by the often highly variable
professional backgrounds of the diverse people involved in integrated mine closure planning (i.e.,
environmental, engineering, planning, financial, social, legal). However, many opportunities exist to operate
sites differently today with little impact on operating costs and major decreases in closure liabilities. These
opportunities are often not realised for the reasons outlined above and because of the lack of integration of
the core elements involved in mine closure (see Figure 1). The objective of this paper is to provide three case
studies in Anglo American where the business case for changing operational strategies to consider mine
closure issues has been successful and to discuss the common themes of these successful approaches.
•Internal or external •Closure
standards •Life of Mine
•Policy •Short/Medium Mine
•Regulation •Rehabilitation
•Stakeholder expectations •Business (MOP)
Requirements Planning
Systems Financials
•Closure tools (e.g. MCT, •LoM closure liability

PRAC) •Premature closure liability
•Geographic Information •Site Opex
Systems •Site Capex
•Environmental Management
Systems
•Mine Operating Systems
The three case studies below are provided to demonstrate the importance of outlining a compelling case to
obtain ownership by site staff of alternative operating strategies that have neutral or lower costs and
decreased closure liabilities. Each case study first provides background to the mine, then outlines the
business case, defines the operational change and finally quantifies the value that was realised.
The Dawson mine is one of three large open-cut coal mining operations owned by Anglo American in
Queensland, Australia. The Dawson mine is located near the township of Moura in central Queensland (Figure
2). The natural topography of the area has gently rolling hills at an elevation of 150 m.a.s.l., with cattle grazing
the predominant premining land use. The mine was opened in 1961 with both underground and open-cut
operations and had various owners over the subsequent half century. Anglo acquired the mine in 2002 from

Shell as a joint venture with Mitsui (51:49). Dawson was the first Australian operation to use draglines in
1963 to best exploit the multiseam deposit. The steeply dipping coal seams result in increasing overburden
volumes for every strip. The mine currently extends over 50 km, with a footprint close to 9,000 ha and mines
10 Mt of metallurgical and thermal coal per annum, employing more than 1,000 people.

Dump space in certain areas of the Dawson mine was becoming limited due to the presence of infrastructure
to the east of existing dumps and future reserves to the west. A range of options was therefore investigated,
including an eastern out-of-pit dump, a western out-of-pit dump or terrace mining with trucks/shovels
backfilling existing voids. The age of the existing draglines assisted in making the capital expenditure case to
change the fleet. A lack of historic progressive rehabilitation and continuing new disturbances were also
leading to increasing closure liabilities.
Dawson mine has historically been operated as a large strip mine with multiple draglines exposing coal in a
westerly direction and dumping spoil to the east (Figure 3). Given the business drivers outlined above (section
2.1.2), a change in mining strategy was implemented to realign the mining operations by 90 degrees and use
operational expenditure to backfill mining voids. This involved changing the fleet for this part of the mine to
truck and shovel (Figure 3). Snapshots of the pit at various stages of development over a 10 year period are
provided in Figure 4. This simulation used Deswik software, which was instrumental in facilitating the
operational change, to provide a compelling visualisation of the development of the pit area over time.


The change to terrace mining at Dawson was undertaken with a reduction in operational expenditure of more
than $10/t. This was due to the relatively short haul distance that planners achieved by increasing the height
of existing waste dumps and minimising the open-pit area. However, this did take 18 months to establish,
and costs were slightly increased from historical levels as prestripping operations to establish in-pit dumping
were completed. The pit backfill eliminated the requirement on closure to fill and reshape substantial
portions of the final mining void, realising a reduction in the mine closure liability. This decrease in liability
was significant (multiple tens of millions of dollars) and was realised immediately with the release of the net
present value (NPV) held on the provision that contributed to profit that year at the Dawson mine.
De Beers Consolidated Mines (DBCM) currently operates the Venetia diamond mine in the Limpopo Province
of South Africa, approximately 80 km west of Musina, 40 km northeast of Alldays and 500 km north of
Johannesburg (Figure 5). The Venetia mine has been operational since 1992 and is the largest diamond-
producing mine within De Beers Mining South Africa, now part of Anglo American. The Venetia mine currently
exists as an open-pit operation with a life of mine (LoM) until 2023, whereafter it will be an underground
mine and the LoM will be extended to 2040. The mine currently has around 3,248 employees and contractor
employees (hereafter referred to as employees), and an additional 1,146 employees work on the Venetia
underground project. The current production rate ranges from 4 MTPA to 6 MTPA. The current pit
dimensions are approximately 1.4 km (north to south) and 2 km (east to west). The final open pit will extend
to approximately 450 m below the surface, and future underground mining will extend to about 900 m below
the surface (Figure 6).

During 2012 (Q1), the executive council (EXCO) of De Beers Consolidated Mines (DBCM) issued a directive
across the company, involving all DBCM operations including the Venetia mine, clearly outlining the
company’s expectations with regard to concurrent rehabilitation as a non-negotiable component of business
expectations.
The directive confirmed that the following objectives formed part of the project brief:
 Development mechanisms to track progress in the following categories:
○ Deposition to plan (waste rock, coarse residue, fine residue)
○ Profiling/reshaping to plan (waste rock dump, coarse residue deposit, fine residue deposit)
○ Environmental restoration/rehabilitation to plan (growth medium and seeding)
○ Budget spent (target vs. actual)
○ Reduction in premature and LoM closure liability
 Involve a multidisciplinary team of leads to participate in developing reporting dashboards to
track progress against set targets.
 Undertake concurrent rehabilitation during LoM operations.
 Achieve a rehabilitation efficiency rate (reduction in liability vs. total spending) of > 1.

The business case for following this approach was a lesson that the company learned from the Namaqualand
mines, where millions of rands are spent on rehabilitation during the rehabilitation and closure phase during
which the company cannot capitalise on the benefit that concurrent rehabilitation presents during the
operation of the active mine.
The Anglo American Mine Closure Toolbox (MCT) process (Botha, 2012) provided invaluable guidance during
the process. This comprehensive process involved strategic planning and the development of robust and
applicable restoration design criteria to meet the closure vision.
Given the volumes of material moved at the mine and the high proportion that waste deposits contribute to
the closure liability, Venetia identified an opportunity to reduce LoM and premature closure liabilities by
changing its current WRD strategy. Changes to the WRD strategy focused on the following activities:
 Maximising concurrent rehabilitation activities and eliminating future reshaping and waste rock
handling by undertaking selective and well planned waste rock placement;
 Increasing the stability and operational flexibility of the fine residue deposits (FRDs) through
waste rock impoundment; and
 Capping of the coarse residue deposit (CRD) as an operational activity.
This resulted in a cost-efficient decrease in the mine’s closure liability by maximising the use of existing
equipment and resources. Planning for the WRD, load and haul activities, dumping strategy and the use of
existing mining equipment and infrastructure was optimised, while future value destruction was prevented
by changing the WRD. Through this process, the Venetia mine identified an integrated solution to reduce
liabilities, which involved smart WRD and increased progressive rehabilitation (Figure 7).

The development of a revised WRD strategy involved modelling and reassessing current equipment and
resources, focusing on minimising potential additional capital requirements. In this instance, the new WRD
strategy was modelled using existing equipment and only resulted in the requirement of two additional D10
dozers over the remaining LoM.
The key performance indicators (KPIs) related to WRD were updated, including the resources responsible for
implementing the strategy. This also resulted in changes in tracking progress and success through the
financial reporting system. Venetia actively tracks the operational expenditure on closure-related activities
and compares the money spent to the reduction gained in reduced closure liability; the performance
expectation is that the reduction in closure liability gained exceeds the cost of concurrent rehabilitation.
Timing also played a critical role, as this opportunity would only deliver real benefits while the mine was
undertaking open-pit mining and had to be implemented prior to cessation of operations in 2023. As part of
the process, the mine has actively implemented a closure and rehabilitation plan, which is linked to the
mine’s strategic business plan. The plan addresses both the existing open-pit requirements and the proposed
underground closure.
The new WRD strategy resulted in the prevention of significant future value destruction at basically no
additional operating cost. The fundamental shift in philosophy resulted in waste being hauled to the extent
of the LoM deposition footprint and deposited systematically towards the open pit; this differs from the
traditional method of waste deposition, where waste is deposited in close proximity to the pit in an outward
direction. It also involved in-tipping to reduce the extent of waste rock dump faces to be rehabilitated and
waste rock impoundment of the FRDs and CRD, which in many cases had a shorter hauling distance than the
original WRD plan. This approach presents the following major improvements to the historical designs:
 Waste material can be deposited in benches rather than by advancing end tipping, facilitating the
rehabilitation of bench slopes while normal deposition strategies continue.
 Because the benches are lower, bulldozers are required to move significantly less material to
achieve the same slope angles and operate within a distance that optimises their efficiency; this
maximises concurrent rehabilitation efforts and significantly reduces LoM closure liabilities.
 The waste rock dump shaping and the cladding of the CRD and FRD will be undertaken using
operational expenditure and will be completed prior to cessation of open-cut mining at basically
no additional operating cost (Figure 8).
 Due to waste rock impoundment, the stability of the FRD will be improved significantly during the
operational and postclosure phases, and there will be an improvement in operational flexibility
(an increased rate of rise means that more tailings can be placed on the current facility in a year).
 The CRD will be cladded and shaped as an operating cost, significantly reducing the LoM closure
liability.
 The rock impoundment of the FRD resulted in a significant reduction of the LoM closure liability,
and no additional FRD facilities will be required for the underground operation, resulting in a
significant upfront capital saving to the project.
 By driving concurrent rehabilitation, the approach allows the mine to demonstrate
tested/sustainable rehabilitation methods over the life of the mine, such as sustainable slope
angles, storm water requirements and vegetation management.
 Overall LoM closure liability is reduced by 35%, and premature closure liability is reduced by
23.5%.

2011 2015
2020 2021
LEGEND:
WASTE ROCK DUMP SECUENCE (2011 – 2021)
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021

The Sishen iron ore mine is located in South Africa in the Northern Cape Province near the town of Kathu.
The closest major town is Kuruman, located approximately 60 km northeast of Sishen; Upington is located
approximately 200 km west of the mine (Figure 9). The Sishen mine is a large open-cut iron ore mine with a
land holding of approximately 36,000 ha and land within the mining fence of approximately 11,000 ha. The
pit is about 2 km wide (east to west), 10 km long (north to south) and more than 100 m deep. The mine’s
production rate is between 32 and 37 MTPA. The mining process entails exploration drilling and modelling,
topsoil removal and stockpiling, drilling and blasting of overburden and ore. Overburden is hauled to waste
rock dumps. Iron ore is hauled to the plant, where it is crushed, sieved, sorted and beneficiated before
transport by rail to the local and international markets via Saldanha Bay harbour. Mining started in 1953,
Anglo American (Kumba Iron Ore) took ownership of the mine in December 2003 and the current LoM is
2031. The mine currently employs more than 10,000 people.
Given the waste rock volumes moved at the Sishen mine and their contribution to the closure liability (> 80%
of total liability), the deposition of waste rock was identified through the MCT risk assessment process (Botha,
2012) as a critical value driver. Even though the outer shell method is primarily driven at this stage by
geotechnical and not rehabilitation considerations and the outer shell must be constructed of calcrete to
ensure clay can be accommodated on the inside of the walls, changing the way the current waste rock dumps
are constructed (moving to the outer shell method) led to significant value being identified and realised

(Figure 10). In addition to reducing closure liability, the outer shell method also significantly increases the
areas available for progressive rehabilitation, facilitating LoM placement of topsoil and compliance with
regulatory targets associated with concurrent rehabilitation.
The megadump concept is also being investigated at the Sishen mine; this involves reducing waste rock dump
batter, which will be rehabilitated by maximising the use of available air space on and between existing
dumps by infilling with waste rock and raising the WRD height (increasing the current design of 120 m to
160 m instead). The possible future requirement from the regulator to place backfill waste rock material on
the western side will have a significant impact on closure liability. Using the available airspace at the existing
waste rock dumps and reducing the requirement for future double handling of waste rock on the western
side of the mine presents a very compelling business case. This has the potential to significantly reduce
liabilities while eliminating or at least reducing the footprint of a western out-of-pit dump (Figure 11).

There has been increased effort at Sishen to address the rehabilitation backlog, the optimisation of
progressive rehabilitation and the resultant decrease in the mine’s closure liability. As part of the process,
the mine has actively implemented a five-year rehabilitation plan, which is linked to the mine’s strategic
business plan and an overall long-term rehabilitation plan. Construction of current waste rock dumps follows
the outer shell method, which facilitates progressive rehabilitation and a reduction of premature and LoM
closure liabilities at no additional operating costs. The current WRD strategy at Sishen is being reviewed with
the aim to significantly prevent value destruction by using the available airspace at the existing waste rock
dumps and reducing the requirement for future double handling of waste rock on the western side of the
mine.
The following key benefits realised by changing the WRD method will be enhanced by implementing the
megadump concept:
 Even though further detailed studies are needed, initial calculations indicate that the outer shell
waste rock dump construction method is a virtually cost-neutral operational that will result in a
decrease of over 90% in LoM closure liabilities for the western waste rock dumps through
progressive rehabilitation.
 The outer shell waste rock dump construction method facilitates placement of fresh topsoil and
compliance with regulatory targets associated with concurrent rehabilitation.
 Progressive rehabilitation will prevent the closure liability from increasing significantly by the end
of the life of the operation.
 Even though the megadump WRD strategy is only one of various possible solutions under
investigation, it can reduce the overall LoM closure liability by more than 20% if it is implemented
to its full extent.
To shift the paradigm on closure planning requires a multifaceted business case that emphasises tangible
and intangible benefits, examines perceived and real barriers, and enjoys the ownership and influence of
senior leaders in an organisation. Importantly, this often requires environmental professionals who are asked
to make a compelling case to change their perspective and view the issue in a new light, informed by the
backgrounds of the senior leaders they are trying to influence. The three case studies outlined above have
some or all of these elements, which can potentially be applied to other mine sites to realise value:
 Structured approach to closure planning: All three case studies had closure plans aligned with the
Anglo American MCT. The associated process involves risk assessment, gap analysis and action
plans, which assisted in identifying the opportunities in the first place.
 Closure liability estimation: The Dawson mine in particular had access to a leading practice tool
for calculating closure liabilities under different scenarios. This is critical to making the financial
business case and to accurately comparing closure liabilities.
 Sound financial business case: All three case studies have very sound financial business cases.
This is not always the case, and often more nonfinancial benefits of integrated planning projects
need to be emphasised. These benefits include compliance with regulatory requirements, social
licence to operate from external stakeholders, reputational development and facilitation of
project approval through demonstrated performance.
 Compelling visual and spatial outputs: All three case studies had access to mine-planning
software with compelling visual and spatial outputs (e.g., Deswik). This is critical to making the

compelling case to senior leaders, who are more likely to be convinced if they can visualise the
proposed mine plan changes.
 Multidisciplinary teams: All three case studies had multidisciplinary teams involved in identifying,
planning and implementing the operational changes. This involved site and corporate staff in the
disciplines of mine planning, mining engineering, finance and environment. Critically, individuals
were prepared to cross departmental and perceived boundaries and attempted to communicate
in common languages.
 Site ownership: While all three case studies involved corporate staff, it is critically important that
the projects were owned by the sites and managed by the site leadership teams.
 Progressive rehabilitation: At Venetia in particular, rehabilitation activities have been integrated
with general mine site activities, with the associated cost and management discipline. The focus
needs to be on eliminating future closure activities (i.e., rehabilitating large areas of land) rather
than on deferral.
 Aligning KPIs and quantifying value realisation: Ensuring that KPIs are aligned across a site drives
value realisation associated with tracking actual versus forecast costs and fully quantifying the
financial benefit; these are critical for planning and changing operational behaviour.
The authors would like to acknowledge the numerous people who contributed to the success of this project
including Damien Perkins, Larry Hantler, the De Beers Venetia team and the KIO Sishen team.
Australian Government (2006) Mine Closure and Completion, Leading Practice Sustainable development program for the mining
industry, Department of Industry Tourism and Resources, Commonwealth of Australia, Canberra.
Botha, P. R. (2012) The business case for mine closure planning and an overview of Anglo American’s mine closure toolbox approach,
in Proceedings First Life of Mine Conference, 10–12 July 2012, Brisbane, Australia, A. Fourie and M. Tibbett (eds), The
Australian Institute of Mining and Metallurgy, Carlton, Victoria, Australia.
Botha, P.R. (2013) Anglo American mine closure toolbox version 2. Anglo Operations Pty Ltd, London.
Grant, C.D. (2006) Decommissioning Alcoa’s first bauxite mine in the Jarrah Forest of Western Australia: Cradle to grave, in
Proceedings First International Seminar on Mine Closure, 13–15 September 2006, Perth, Western Australia, A. Fourie and M.
Tibbett (eds), Australian Centre for Geomechanics, Perth, Australia, pp. 287–297.
International Council on Mining and Metals (ICMM) (2008) Planning for integrated mine closure: toolkit, ICMM, London, United
Kingdom.

L.H. Fraser Natural Resource Sciences, Thompson Rivers University, Canada

H.W. Garris Natural Resource Sciences, Thompson Rivers University, Canada
S.A. Baldwin Chemical and Biological Engineering, University of British Columbia, Canada
J.D. Van Hamme Biological Sciences, Thompson Rivers University, Canada
W.C. Gardner Natural Resource Sciences, Thompson Rivers University, Canada
For the majority of mines, closure succeeds when healthy, self-sustaining ecosystems develop on
previously mined lands. In British Columbia, Canada, the regulations require reclamation of
ecosystems; however, there are few specified targets, and those that are presented are vague.
Genomics technologies may provide the key to both understanding the elements necessary to
recreate functional ecosystems and provide sufficient benchmarks for success. In this review, we
highlight the use of genomics to meet mine closure goals, enhance ecosystem development and
optimise ecosystem services inherent in self-sustaining reclaimed ecosystems. We outline practical
steps for applying genomics technologies to characterise the composition and activity of microbial
communities in soils and treatment substrates. From this framework, we address the state of the
science and how recently developed techniques have transferable value to mine reclamation. We
then define three areas in which genomics technologies have already proven effective at informing
management and reclamation of mine sites in the form of bioreactors, passive treatment systems
and novel gene discovery. Finally, we speculate on the future applications of genomics technologies
and the necessary steps to integrate these data into comprehensive management of mined sites.
In an ideal system, mines operate with minimal environmental impact within and outside of the operational
footprint. After the period of profitable extraction is reached, a desirable condition is recreated, be it forest,
pasture or suburban neighbourhood, so that the biota (soil microbes trees, and concerned citizens alike) can
function within a natural, self-sustaining ecosystem. In reality, mine sites often leave a legacy, including
perpetually altered plant communities (Holl, 2002); elevated contaminants in surface and groundwater (Cidu
et al., 2001); thin, compact soils (Skousen et al., 2009); altered soil function (Mummey et al., 2002); and
magnification of contaminants within the food chain (Allan, 1995; Muscatello and Janz, 2009).
In Canada, government regulations on mining operations set the standards for reclamation success, and as a
result, direct the fate of mined lands. The British Columbia Mines Act and Health, Safety and Reclamation
Code for Mines (the Code) specify that reclamation must satisfy the requirements of the chief inspector
(Government of British Columbia, 2008). The legislation is vague and subject to interpretation (by both the
mining companies and the inspectors), so there has been variability in goals and measures of success.
Historically, goals were set for plant productivity, likely because productivity is relatively easy to measure and
can be associated with ecosystem function (Britton, 1998). The current Code (10.7.5) expresses a goal for
equivalent land capability, which is a challenge to achieve because there are no set predefined targets. There
has been a recent trend to consider mine closure in terms of the whole ecosystem, with all of its functions
and services.

Ecosystem services can be defined as “the aspects of ecosystems utilised (actively or passively) to produce
human well-being“ (Fisher et al., 2009). The ecosystem services for reclaimed mined lands may include forage
for cattle and wildlife, timber, clean air and water, carbon sequestration, biodiversity and culturally
significant natural products related to traditional practices and medicines. Canadian regulations exist to
protect these services, or ensure their continuation, during and after mining activities take place. Despite
pervasive efforts to meet and improve upon regulatory guidelines for ecosystem services at previously mined
sites, there is a large degree of uncertainty in mine land restoration. Consistent application of revegetation,
soil amendments and regrading treatments can lead to very different results, even on the same site
(Martínez-Ruiz and Marrs, 2007). Why such inconsistency? Many authors attribute it to variability in starting
conditions (the prevailing mineralogical substrate), slope, aspect and myriad unmeasured factors (Martínez-
Ruiz and Marrs, 2007), with one emerging frequently ― the soil microbiota (Harris, 2003).
Microorganisms are catalysts for soil formation. They are responsible for creating complex
microenvironments that lead to nutrient uptake in plants, semi-homeostatic water and chemical regulations,
and to overall resilience to erosion, contamination and invasion by exotic plant species. New soils that form
on exposed bedrock often follow a predictable successional sequence. Microbial communities form on
weathered inorganic substrates, fix atmospheric nitrogen and CO2 and contribute the key elements of organic
life. Multi-cellular plants and animals are relative latecomers to the developing community, arriving when
soil formation processes have provided sufficient organic resources to sustain them (see Frouz et al., 2008,
for a discussion of this process on previously mined sites). Chronologically, this natural approach can be
considered bottom-up, with soil organisms and processes establishing and preparing the substrate before
larger plants and animals arrive. The science of restoration has not developed a mechanism for recreating
(much less accelerating) this bottom-up process. Rather, practitioners often attempt to recreate the desired
ecosystem irrespective of the natural chronological sequence by superimposing soil amendments, and with
seeding or planting plugged vascular plants (Tordoff et al., 2000). This practice is not undertaken out of
ignorance of the role soil microbial communities play, but because tools to observe, quantify and manipulate
these communities are not within the typical organisation’s toolkit.
Until recently, techniques for observing and quantifying microbiota have been extremely restrictive (Ficetola
et al., 2008). Less than one per cent of all soil microbial organisms have been cultured in a laboratory setting
(Hugenholtz et al., 1998; Harris, 2009). As one alternative, microbiologists have developed techniques for
quantifying the mass-action of microbes in soils, including the respiration and production rates of metabolic
products. Genomics technologies offer us an opportunity to observe the complexity of microbial
communities as they form on mined sites, and to apply ecological theory to soil community formation and
structure in a way that has until very recently been impossible. In recent decades, microbial genomics has
been applied to mine sites, as will be discussed, but it has yet to be incorporated into a comprehensive
monitoring and troubleshooting paradigm for mine closure and mine site restoration. In this paper, we
provide a context for genomics in the mining community and highlight specific applications of genomics and
metagenomics to mine closure.
Microbial genomics begins with a source, either an individual organism or sample material comprising many
organisms. Soil contains a combination of living, dead and dormant organisms, all of which contribute to the
soil’s genetic signature. Representative sampling of soils on mine sites for their microbial community
structures is a challenge, as mine sites are large, often incorporating entire landscapes with varying
geophysical properties. Furthermore, soil communities are notoriously patchy in distribution (among habitat
types and both horizontally and vertically within each soil stratum) (Foster, 1988; Ranjard and Richaume,
2001). Therefore, it is not feasible to sample every habitat type and soil stratum within a mined landscape.
Rather, sampling should target habitats and soil strata that may provide signs of imminent degradation of
water quality (streams and associated wetland areas within and adjacent to the mined land catchment),

contaminant sequestration (bioreactor substrates, or organic-rich soils receiving runoff from the mine site)
or reclamation success (processed parent material, organic amendments and unimpacted reference sites).
There are a few major concerns to consider when designing a sampling protocol for genomic analysis. First,
the soil chemistry and water content of samples change rapidly upon collection, possibly leading to shifts in
the activity and abundance of microbes within the soil, especially when collected from anoxic substrates.
Biological activity must therefore be suspended as quickly as possible following collection, which generally
involves flash freezing of samples with liquid nitrogen (−195°C) or on blocks of dry ice (−79°C) when liquid
nitrogen is not available or is too dangerous to use (e.g. on a boat). Ethanol (95%) can be sprayed on blocks
of dry ice to accelerate heat transfer. This freezing process suspends metabolic activity in the soil and
preserves in situ bacterial abundance. Second, replication should be large enough to account for the high
level of variation in soil microbial community composition and structure, with each replicate representing a
homogeneous mixture of multiple samples (often soil cores) taken in the field. Biogeochemical gradients
across physical interfaces and within sediments can be very steep. Therefore, preservation and assessment
of layers is preferable (see de Gruijter et al., 2006, for a comprehensive discussion of sampling designs specific
to natural resource monitoring). As microbial communities depend very strongly on local conditions
(water/soil chemistry and biological interactions), sufficient metadata should be recorded along with each
sample to enable this variation to be described for a particular community. As a minimum, this should include
pH, temperature, dissolved oxygen, conductivity, nitrate, nitrite and phosphorus. Finally, because of their
ability to bring oxygen and carbon resources into otherwise anaerobic soils, plant rooting zones can play a
pivotal role in determining microbial community composition (Marschner et al., 2001; Marschner et al.,
2004), and samples taken from vegetated substrates should account for the composition and characteristics
of co-occurring plant communities.
DNA extraction involves two steps: (1) breaking apart cell walls and membranes (cell lysis) through some
combination of heating, sonication and/or chemical treatment, and (2) isolation/concentration of DNA via
filtration (Picard et al., 1992; Zhou et al., 1996). Although conceptually similar, different lysis methods can
produce conflicting results (Martin-Laurent et al., 2001; Carrigg et al., 2007).
There are three major challenges to consistency in the extraction process. First, many microorganisms form
colonies or crusts on substrate components (sand grains and small stones). These colonies can be difficult to
break apart, and fundamentally protect many cells from the lysis procedure. This can lead to differences in
perceived community composition when comparing samples with different substrate grain sizes. Pre-
washing procedures have been developed to suspend adhered communities in solution before breaking apart
cells, leading to more representative extractions (Fortin et al., 2004; He et al., 2005).
Second, the thickness and material properties of cell walls and membranes are not uniform for all
microorganisms. Developing lysis procedures therefore involves identifying the major groups of
microorganisms of interest to the study, and considering performing parallel extractions to isolate distinct
groups for comparison. Sulphate-reducing bacteria (SRB) and methanogenic archaea are of particular interest
to the mining industry for their prevalence and activity in mined substrates. Bacterial and archaeal cell walls
comprise different materials (see Kandler and König, 2014, for a chemical and functional description of these
differences). As a result, using a single lysis procedure will invariably underrepresent one of these groups,
and multiple lysis procedures should be evaluated if a comprehensive understanding of microbial community
composition and prevalence is desired.
Finally, highly degraded organic material in carbon-rich soils (humic acids) are difficult to separate from DNA,
often leading to contamination of DNA extracts and poor sequence reads (Yeates et al., 1998). Procedures
have been developed to accommodate samples rich in organic acids; these involve removal of humic acids
with additional filtration steps (Tsai and Olson, 1992) or ion exchange chromatography (Tebbe and Vahjen,
1993).

DNA extracted from the mine environment sample provides information about the microbial community and
metabolic processes at that site. One method to describe the microbial community uses specific components
of the DNA, such as genes known to vary among species, that are amplified via polymerase chain reaction
(PCR) (Valentini et al., 2009). These reference genes found in the mine sample are compared to curated
databases of known annotated sequences (BOLD, GenBank, MG-RAST) (Ratnasingham and Hebert, 2007;
Meyer et al., 2008; Benson et al., 2013), thereby identifying which species are present at the mine site. Early
uses of genomics technology for bacterial community characterisation focused on the 16S rRNA gene, a
segment of DNA critical for the production of proteins by prokaryotic cells, and that is generally unique to
each species (Weisburg et al., 1991). Sequencing of 16S rRNA genes has been used to quantify microbial
diversity with sufficient resolution to detect shifts along major ecological gradients (Schmidt et al., 2014).
However, this approach is limited to determining which microbial species are present and provides little
information about the actual metabolic processes taking place (Eisen, 2007). Stable isotope (with 13C)
probing has been used in recent years in combination with environmental DNA sequencing to isolate
microbial community fractions involved in specific metabolic processes (Dumont and Murrell, 2005;
Verastegui et al., 2014), which makes the important link between bacterial community composition and
specific metabolic functions.
With the advent of high throughput sequencing technologies, targeting only specific regions within DNA has
given way to “shotgun” and whole-genome/metagenome sequencing using all of the DNA in the sample
(Tringe et al., 2005). The term metagenome is used when the sample contains many organisms (and thus
genomes), as is the case in soils. Shotgun sequencing involves breaking the metagenome into many small
fragments via physical shearing or enzymatic processes (Sharpton, 2014). After sequencing, these fragments
can be used to reconstruct entire genomes for the organisms present at the site by aligning the sequences
where they overlap. The metagenome includes both functional information (gene markers linked to specific
metabolic products) and compositional information (e.g. 16S and other species-specific reference sequences)
(Xia et al., 2011). The computational techniques required to compile, align and interpret these millions of
bases of genetic code arranged into short fragments is termed bioinformatics (discussed in the next section).
An additional use of environmental DNA fragments is functional screening. Cloning of these fragments into
live hosts such as the bacterium E. coli or the yeast P. pastoris enables expression of the protein products
and their functional screening. For example, growing of E. coli cells containing environmental DNA on
medium containing cellulose allowed for direct identification of novel cellulase enzymes (Mewis et al., 2013)
from a mine remediation environment.
Bioinformatics refers to the computational procedures used to extract meaningful information from the very
large datasets produced in metagenomic studies. Bioinformatics pipelines require considerable
computational power and unique algorithms to carry out the steps from quality control of the sequences to
assembling functional components (i.e. sequences coding for proteins used in cellular functions called open
reading frames or ORFs) that yield information about the metabolic potential of the organisms present at the
site. Bioinformatics for metagenomics are complex, as compositional as well as functional reconstruction is
required. To identify compositional structure, the marker genes (i.e. 16S rRNA) are binned, based on their
similarity, into operational taxonomic units (OTUs) that are likely to be derived from the same species.
Assembled functional components (the ORFs) and OTUs are interrogated against protein and gene databases
in bioinformatics pipelines using the blast tools (Altschul et al., 1990) in order to assign putative functions
and identities to the microbial community found in the mine site sample (see Sharpton, 2014, for a
comprehensive discussion of processing and interpretation of shotgun-derived sequences).

Given the much larger, more complex genomes found in plant and animal species, the molecular tools for
identifying them have developed more slowly than those for microorganisms. Whereas the 16S rRNA gene is
found in all bacteria and archaea, the corresponding 18S rRNA gene in higher eukarya, and even fungi, is not
as reliable as a species identifier. As such, much effort is being put into finding signature genes, called “DNA
barcodes,” for different groups of plants, invertebrates and vertebrates (Hebert et al., 2003). Each species
collected can be rapidly and cost-effectively identified, and with the proper sampling protocols, quantified in
terms of relative abundance and diversity by site.
New Gold’s New Afton Mine near Kamloops, British Columbia, Canada, began operation in 2012. Although
the mine is in its early days of operation, it has taken a proactive approach to future mine closure plans by
implementing a partnership with the Biodiversity Institute of Ontario to implement DNA barcoding for
environmental impact assessments. The pilot program involved four sites: two grassland sites (disturbed and
undisturbed) and two wetland sites (disturbed and undisturbed). Invertebrate samples from these four sites
were collected in the summer of 2013, and DNA barcode analyses were completed in August 2014. Between
294 and 5,560 individual invertebrates were captured in Malaise traps each week, and 3,956 species were
identified (D. Wilson and S. Davidson, personal communication 8 May, 2015). Differences were observed
between habitat types such that wetlands contained more species than grasslands, and the natural grassland
had more species than the disturbed grassland. The intention is to continue monitoring on a four- to five-
year time scale. Such baseline data provide invaluable information for future mine closure and site
reclamation.
Thanks to the conserved nature of, and relatively long history of collecting sequence data for, the 16S rRNA
gene in bacteria and archaea and, to a lesser extent, the 18S rRNA gene in fungi, researchers interested in
microbial communities (e.g. environment, health, industry) have reached a stage where using these
reference genes to describe microbial community composition and diversity is now routine (Schmidt et al.,
2014). A short number of years ago, publications would be based solely on 16S rRNA gene surveys of
microbial communities, whereas today these surveys are considered one of many standard analytical tools
for scanning microbial landscapes. While 16S and 18S rRNA gene surveys will continue to be powerful tools,
microbiologists have realised that these reference genes do not universally reflect the metabolic potential
and biochemical activities of individual microorganisms, let alone complex microbial communities. This
realisation continues to motivate the development of “meta” tools to qualify and quantify all of the DNA,
mRNA and proteins in microbial communities (metagenomics, metatranscriptomics and metaproteomics)
and to relate these data to biochemical fluxes (metabolomics) and, ultimately, ecosystem functions (Krause
et al., 2014).
At the present time, there are good genomic databases for pure bacterial cultures grown under laboratory
conditions. Generating good drafts of microbial genomes can now be done in days rather than years, by
individuals rather than teams, for hundreds rather than millions of dollars. This is a major shift from less than
a decade ago (Kyrpides et al., 2014). High throughput transcriptomic and proteomic tools have come online
and are being increasingly used for pure culturable microorganisms as well. The tools are available to
characterise natural microbial communities, but analysing data from even a few samples in a meaningful way
still requires the use of supercomputers or powerful computer clusters, and is based on imperfect databases
established in the early days of genomics (Howe et al., 2014). A “meta-omics” study is not for the faint of
heart, but the field is shifting rapidly, and these types of studies are becoming more common despite the
need for large monetary and personnel investments. Major breakthroughs are currently being made to
advance our understanding of the dominant roles microorganisms play in the metabolism and lifestyles of all
macroorganisms. Further, biotechnological applications are being realised in an array of fields, including

forestry, agriculture, animal husbandry, human health, and food, beverage and fuel production (Ekkers et al.,
2012).
Given the rich literature on metal-microbe interactions (Gadd, 2010), there is tremendous scope for applying
meta-omics to mine reclamation, particularly given the small but solid foundation of genomic work being
done in phytoremediation (Bai et al., 2014), metal-plant interactions (Hanikenne and Nouet, 2011; Bhargava
et al., 2012) and soil ecology (Howe et al., 2014) on which to build. Looking to the future, once current meta-
omics tools have sufficiently matured for microbial communities, the gene, transcript, protein and metabolic
signatures from plants and other macroorganisms must be integrated in holistic models in order to better
appreciate the holobionts (macroorganisms and their associated microbial and viral communities) essential
to healthy ecosystems.
Seepage from mine tailings storage facilities, waste rock piles, open pits and underground workings, as well
as excess process water, contains metals such as selenium, copper, molybdenum, zinc and arsenic, often
along with sulphate and nitrate (McDonald and Strosher, 1998; Wang and Mulligan, 2006). High-density
sludge treatment is used at many mine sites to treat this mine-influenced water, but this chemical process
consumes large quantities of reagents and produces high volumes of toxic sludge requiring long-term safe
storage (Zinck and Griffith, 2013). Bioreactors offer a sustainable, cost-effective alternative to reagent-based
water treatment technologies.
Biological processes use natural microorganisms to treat mine-influenced water. Some bioreactors reduce
sulphate to produce sulphide, which chemically binds with metal ions, causing them to precipitate out of
solution as stable metal sulphides (Barnes et al., 1994). Additionally, bioreactors have been designed to
selectively remove specific valuable metals, such as copper and zinc, that can be recycled to the metal
extraction facility (Zinck and Griffith, 2013). Bioreactors are used to remove nitrate, which is high in some
mine-influenced water owing to the use of explosives on mine sites (Koren et al., 2000). Selenium and arsenic
are metals that occur as anions, and there are natural microorganisms that transform these compounds in
order to gain energy for growth. Biological reactors using these organisms successfully remove selenium and
arsenic from mine-influenced water (Morita et al., 2007).
Given the benefits in economics and effectiveness promised by biological treatment, it is surprising that use
of bioreactors is not widespread on mine sites. A few reasons for this have been revealed through the
application of metagenomics. Bioreactors contain consortia of microorganisms, rather than one single type
(Baldwin et al., 2012), and the performance of these bioreactors depends on the types of microorganisms
used to inoculate them (Pruden et al., 2007). Microbial communities are dynamic, and their members
fluctuate in abundance and activity in response to changes in the influent water or operating conditions (Dar
et al., 2008). Shifts in microbial community composition can lead to failure to meet water quality
specifications, ultimately putting the receiving environment at risk (Mirjafari and Baldwin, 2011). To
overcome this limitation of biological treatment, it is necessary to monitor microbial community composition
in bioreactors and correlate this to operational settings and performance metrics. The microbial community
in mine-based bioreactors can be monitored using the metagenomics techniques, such as those targeting the
16S rRNA, described in Section 2.3 (Schmidtova and Baldwin, 2011; Baldwin et al., 2015).
High-throughput sequencing can generate enough information to characterise the entire microbial
community including very rare members. Often the important functional groups for metal removal in
bioreactors are rare, even though their action achieves successful treatment (Rezadehbashi et al., 2012).
Some other microbes present facilitate the activity of these desired groups, or some might compete for
nutrients and hinder their activity. Nutrient consumption is one of the major costs in running a bioreactor,
and when undesired microbes compete successfully for most of the nutrients, the desired organisms decline
in number and bioreactor performance fails (Silva et al., 2012).

Much of the work done to date on microbial communities in bioreactors, especially those related to mine
remediation, has focused on microbial community composition based on surveys of the 16S rRNA gene.
These surveys have revealed that microbial groups associated with metal-rich environments are unknown
and uncharacterised, such that new taxonomic groups have been invented to classify them (Khoshnoodi et
al., 2013). Thus, metagenomic studies are needed to discover the potential functions of these novel
organisms (Ellis et al., 2012).
Sequencing of DNA provides information on metabolic potential only, and does not reveal which of the genes
are being actively expressed. Active metabolic processes can be detected by sequencing the transcribed
genes using a technique referred to as transcriptomics (Luo et al., 2014). Although very recently developed
for microbial communities in bioreactors (Luo et al., 2014), microbial transcriptomics is possible but requires
careful sampling to preserve the active state of the microbes.
Given the affordability of high-throughput sequencing, use of metagenomics and transcriptomics for
bioreactor monitoring is bound to increase. Since these data sets are new, correlation of microbial
community composition, metagenomics and transcriptomics with bioreactor operation and performance is
still being investigated. As the bioinformatics improves, these metagenomics tools show promise for future
process control of bioreactors.
The benefit of using bioreactors for mine remediation is that they can be controlled using tanks, pumps,
valves, instrumentation and defined nutrients to achieve just the right conditions and rapid kinetics for
treating large flow rates (Zinck and Griffith, 2013). However, their capital and operating costs add to the
expense of mine operation and closure. In addition, although they are highly automated, operators are
needed.
The vast array of natural processes for biogeochemical cycling of metals and nutrients can be harnessed in
so-called passive or semi-passive treatment systems (Ziemkiewiczet al., 2003). These typically take the form
of constructed wetlands, either sub-surface flow anaerobic, surface flow aerobic or, most commonly,
combinations thereof. Instead of defined nutrients, mixtures of waste organic materials such as wood debris,
hay, compost, pulp and paper mill biosolids are used. These complex organic materials are decomposed into
the smaller carbon compounds needed for the sulphate- and metal-reducing bacteria.
If successful, passive treatment can remove metals at seeps distributed across the mine site for a fraction of
the cost of active bioreactor treatment (Zinck and Griffith, 2013, p. 14). The metal residuals are captured and
secured inside the organic matrix, most often as sparingly soluble metal sulphides (Khoshnoodi et al., 2013).
The ecosystem of a passive remediation system is as complicated as that of soils (Baldwin et al., 2015). Like
bioreactors, they are consortia of interacting species that shift in composition with geochemical gradients,
seasons and the age of the system (van der Lelie et al., 2012). Their performance may decline as the organic
material decomposes, evolving towards microbial communities with completely different metabolic
potential than at the start (Mirjafari, 2014). Preliminary studies of the microbial communities in these
systems reveal that the metabolic potential for metal removal in them is much wider than previously thought
(Baldwin et al., 2015). They contain species that tolerate high metal concentrations, many novel unclassified
candidate division groups found in other metal-contaminated environments and species capable of using
usually recalcitrant aromatic compounds for growth.
Successful metal removal occurs even in places where SRB are extremely rare, meaning either that many
other groups of organisms we do not know about are capable of sulphate reduction and/or metal
precipitation, or that only a few sulphate reducers are needed for successful treatment (Khoshnoodi et al.,
2013; Baldwin et al., 2015). Charting of microbial communities in passive remediation bioreactors has
revealed that they are not static, but fluctuate cyclically (Baldwin, unpublished data). Using metagenomics
and metatranscriptomics, we can learn more about the dynamics of these ecosystems as they respond to
changing conditions and use this knowledge to design better systems or diagnose performance issues.

Metal-rich ecosystems are considered extreme environments. They harbour highly specialised
microorganisms that have evolved unique metabolisms to transform, sequester or detoxify metals in order
to survive. Examples include sulphate reducers’ overproduction of extracellular polymeric material to bind
up copper ions, thereby creating nucleation sites for precipitation (Jalali and Baldwin, 2000), intracellular
mineralisation of tellurium to sequester this highly toxic metal (Amoozegar and Khoshnoodi, 2012), and
methylation and volatilisation of arsenic by Methanocorpusculum labreanum, suggesting that it may be a
significant contributor to metal cycling in anaerobic environments (Khoshnoodi et al., 2012).
Enzymes involved in metal cycling, or biochemical compounds with the ability to sequester specific metal
ions, can be used in future biotechnologies to improve bioremediation, and even develop methods for in situ
mining. The field of functional metagenomics, first mentioned in Section 2.3, is being used to screen large
DNA fragments from mine sites (Mewis et al., 2011). These large fragments of environmental DNA may
contain novel genes for metal cycling, and, using selective media in the laboratory, we can screen the E. coli
clones for metal resistance. It may be possible in the future to construct biochemical pathways for metal
removal using simple and easy-to-grow organisms and synthetic biology.
The mining industry has recognised for some time that planning for mine closure begins even before
overburden is removed from the site (Thirgood, 1986 in Polster 1989). Until recently, this has not included a
significant consideration of the pre-impact community of plants, animals and microorganisms residing in the
unaltered substrate, but see Morrison et al. (2005) and Jasper (2007). In the future, such assessments should
include evaluation of both unimpacted overburden soil communities and stockpiles of such material for
eventual recovering of the site. Microbial genomics may play a key role in expediting mine land restoration
by providing information on soil community dynamics in overburden stockpiles, allowing managers to
maintain these communities in a way that expedites recovery once these substrates are used to recover the
mined site.
Microbial genomics may also serve as an indicator for success in these systems, providing additional evidence
of soil community formation and ecosystem trajectory. To be able to set objectives for effective mine
reclamation and evaluate what is successful, the meaning of “equivalent land capacity” needs to be defined
for every mine site.
Barcoding and metagenomics can be incorporated into methods for evaluating ecosystem services. Microbial
community analysis provides valuable information about nutrient (carbon, nitrogen and sulphur) cycling,
green house gas emissions and metal transformations that can be fed into determining pre-mining land
capacity assessment. This will allow targets to be specified for post-mine reclamation, and these same tools
can be used to evaluate whether remediation strategies are working.
Microbial genomics fills a critical gap in efforts to simulate the formation of metabolic networks in
ecosystems. The presence or absence of key microbial communities may mean the difference between a
successfully remediated site and persistent degradation of water quality (e.g. acidification and metal
leaching). Efforts are underway to develop comprehensive predictive networks for mine sites, in which
environmental genomics are tied to environmental monitoring data to generate a comprehensive
understanding of ecosystem function. Probabilistic modelling frameworks may very well provide early
warning signs of acid generation in mined substrates, and genomics data may yield the necessary evidence
to determine what remediation options will be most likely to succeed in both the short and long term.

Defining benchmarks may very well be the key component necessary to leverage genomics tools for mine-
site management. Bioinformatics data processing workflows (pipelines) exist (sometimes incorporated
directly into sequencing hardware) that provide for rapid processing and interpretation of sequencing runs.
The great questions for the mining industry can be put quite simply: (1) What do we sequence?, and (2) How
do we use the resulting data to improve management, reclamation and containment?
Section 2.1 provides guidelines for sampling of these sites, but the industry requires a wealth of context-
dependent validation of these approaches to become widespread. Therefore, research and development
resources (both academic and industrial) should be leveraged in such a way that sampling procedures can be
standardised to the sorts of questions of interest at mine sites (e.g. How do soil microbial communities form
on mined substrate? What soil conditions promote recovery of native species?).
Ultimately, with standard sampling procedures, bioinformatics pipelines can be tailored to industry-specific
questions. For example, a sampling protocol might be developed to evaluate diffuse leachate exposure in
riparian substrates, with the question: Does exposure to leachate affect microbial community composition
and dynamics in a way that could limit containment in the future? Field sampling and sequencing following
established protocols would be processed through a tailored bioinformatics pipeline that yields the following
outputs:
1. Community composition matrices and planned comparisons (diversity, similarity, etc.)
2. Compositional differences and known associations (i.e. variations in groups of organisms known
to be involved in certain geochemical pathways)
3. Management recommendations and references to similar scenarios/responses.
We have outlined a range of genomics applications to mine closure, from characterising natural substrates
before overburden removal, to water treatment, to bioremediation and monitoring of healthy reclaimed
ecosystems. In performing this review, we identified two key constraints on the widespread application of
genomics for mine closure. First, industry-wide standard operating protocols need to be developed for mine
closure, including sampling procedures designed for representativeness and comparability (spatial extent,
replication, temporal frequency). Second, sequencing and data interpretation pipelines must be established
in parallel with the development of these standards, allowing mine managers to more easily discover what
works in a given system and to establish benchmarks for reclamation success. Mining is a key global industry
for development and for quality of life. The integration of genomics technologies into mine closure planning
and implementation may drastically improve the stability and reliability of ecosystem reclamation.
We would like to thank Jon Taylor (University of British Columbia) for helpful troubleshooting and advisement
on sampling design for mine-site applications. We would like to thank Genome BC, Genome Canada, NSERC
Engage and Mitacs Accelerate for funding to support preliminary R&D on metagenomics technologies. We
would also like to thank Dennis Wilson and Scott Davidson of New Gold for providing details about their
project. Finally, we would like to thank Mount Polley Mine Corporation and particularly Katie McMahon,
Colleen Hughes, Luke Moger and Art Frye for their partnership in projects aimed at addressing many of the
questions addressed in this review.

Allan, R. (1995) Impact of mining activities on the terrestrial and aquatic environment with emphasis on mitigation and remedial
measures, in Heavy metals: problems and solutions, U. Förstner, W. Salomons and P. Mader (eds), Springer Berlin Heidelberg,
Berlin/Heidelberg, Germany, pp. 119–140.
Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. (1990) Basic local alignment search tool, Journal of Molecular Biology,
Vol. 215, pp. 403–410.
Amoozegar, M. and Khoshnoodi, M. (2012) Tellurite removal by a tellurium-tolerant halophilic bacterial strain, Thermoactinomyces
sp. QS-2006, Annales de l’Institut Pasteur Microbiologie, Vol. 62, pp. 1031–1037.
Bai, Y., Liang, J., Liu, R., Hu, C. and Qu, J. (2014) Metagenomic analysis reveals microbial diversity and function in the rhizosphere soil
of a constructed wetland, Environmental Technology, Vol. 35, pp. 2521–2527.
Baldwin, S.A., Resadehbashi, M., Taupp, M., Mattes, A., and Hallam. S.J. (2012) The genetic diversity of microbes in an anaerobic
biological reactor treating metal-rich landfill leachate: what this tells us about treatment mechanisms, Proceedings of the 9th
International Conference on Acid Rock Drainage. 20-26 May, Ottawa, Canada. Vol. 1. pp. 128–139.
Baldwin, S.A., Khoshnoodi, M., Rezadehbashi, M., Taupp, M., Hallam, S., Mattes, A. and Sanei, H. (2015) The microbial community of
a passive biochemical reactor treating arsenic, zinc and sulfate-rich seepage, Frontiers in Bioengineering and Biotechnology,
Vol. 3(27), pp.1–13.
Barnes, L.J., Scheeren, P.J.M. and Buisman, C.J.N. (1994) Microbial removal of heavy metals and sulfate from contaminated
groundwaters, in Emerging technology for bioremediation of metals, Volume 2, J.L. Means and R.E. Hinchee (eds), CRC Press,
Boca Raton, USA, pp. 38–49.
Benson, D.A., Cavanaugh, M., Clark, K., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J. and Sayers, E.W. (2013) GenBank, Nucleic Acids
Research, Vol. 41, pp. D36–42.
Bhargava, A., Carmona, F.F., Bhargava, M. and Srivastava, S. (2012) Approaches for enhanced phytoextraction of heavy metals,
Journal of Environmental Management, Vol. 105, pp. 103–120.
Britton, J.M. (1998) An evaluation of public involvement in reclamation decision making at three metal mines in British Columbia, MA
Thesis, University of British Columbia, Canada.
Carrigg, C., Rice, O., Kavanagh, S., Collins, G. and O’Flaherty, V. (2007) DNA extraction method affects microbial community profiles
from soils and sediment, Applied Microbiology and Biotechnology, Vol. 77, pp. 955–964.
Cidu, R., Biagini, C., Fanfani, L., La Ruffa, G. and Marras, I. (2001) Mine closure at Monteponi (Italy): effect of the cessation of
dewatering on the quality of shallow groundwater, Applied Geochemistry, Vol. 16, pp. 489–502.
Dar, S.A., Kleerebezem, R., Stams, A.J.M., Kuenen, J.G. and Muyzer, G. (2008) Competition and coexistence of sulfate-reducing
bacteria, acetogens and methanogens in a lab-scale anaerobic bioreactor as affected by changing substrate to sulfate ratio,
Applied Microbiology and Biotechnology, Vol. 78, pp. 1045–1055.
de Gruijter, J., Brus, D.J., Bierkens, M.F.P. and Knotters, M. (2006) Sampling for natural resource monitoring, Springer-Verlag Berlin
Heidelberg, New York, USA.
Dumont, M.G. and Murrell, J.C. (2005) Stable isotope probing – linking microbial identity to function, Nature Reviews: Microbiology,
Vol. 3, pp. 499–504.
Eisen, J.A. (2007) Environmental shotgun sequencing: Its potential and challenges for studying the hidden world of microbes, PLoS
Biology, Vol. 5, pp. 384–388.
Ekkers, D.M., Cretoiu, M.S., Kielak, A.M. and van Elsas, J.D. (2012) The great screen anomaly – a new frontier in product discovery
through functional metagenomics, Applied Microbiology and Biotechnology, Vol. 93, pp. 1005–1020.
Ellis, J.T., Sims, R.C. and Miller, C.D. (2012) Monitoring microbial diversity of bioreactors using metagenomic approaches, Sub-Cellular
Biochemistry, Vol. 64, pp. 73–94.
Ficetola, G.F., Miaud, C., Pompanon, F. and Taberlet, P. (2008) Species detection using environmental DNA from water samples,
Biology Letters, Vol. 4, pp. 423–425.
Fisher, B., Turner, R.K. and Morling, P. (2009) Defining and classifying ecosystem services for decision making, Ecological Economics,
Vol. 68, pp. 643–653.
Fortin, N., Beaumier, D., Lee, K. and Greer, C.W. (2004) Soil washing improves the recovery of total community DNA from polluted
and high organic content sediments, Journal of Microbiological Methods, Vol. 56, pp. 181–191.
Foster, R.C. (1988) Microenvironments of soil microorganisms, Biology and Fertility of Soils, Vol. 6, pp. 189–203.
Frouz, J., Prach, K., Pižl, V., Háněl, L., Starý, J., Tajovský, K., Materna, J., Balík, V., Kalčík, J., and Řehounková, K. (2008) Interactions
between soil development, vegetation and soil fauna during spontaneous succession in post mining sites, European Journal
of Soil Biology, Vol. 44, pp. 109–121.
Gadd, G.M. (2010) Metals, minerals and microbes: geomicrobiology and bioremediation, Microbiology, Vol. 156, pp. 609–643.
Government of British Columbia (2008) Health, safety and reclamation code for mines in British Columbia, viewed 10 February 2015,
http://www.empr.gov.bc.ca/Mining/HealthandSafety/Documents/HSRC2008.pdf.
Hanikenne, M. and Nouet, C. (2011) Metal hyperaccumulation and hypertolerance: a model for plant evolutionary genomics, Current
Opinion in Plant Biology, Vol. 14, pp. 252–259.
Harris, J.A. (2003) Measurements of the soil microbial community for estimating the success of restoration, European Journal of Soil
Science, Vol. 54, pp. 801–808.
Harris, J. (2009) Soil microbial communities and restoration ecology: facilitators or followers? Science, Vol. 325, pp. 573–574.
He, J., Xu, Z. and Hughes, J. (2005) Pre-lysis washing improves DNA extraction from a forest soil, Soil Biology and Biochemistry, Vol.
37, pp. 2337–2341.

Hebert, P.D.N., Cywinska, A., Ball, S.L. and deWaard, J.R. (2003) Biological identifications through DNA barcodes, Proceedings of The
Royal Society B: Biological sciences, Vol. 270, pp. 313–321.
Holl, K.D. (2002) Long-term vegetation recovery on reclaimed coal surface mines in the eastern USA, Journal of Applied Ecology,
Vol. 39, pp. 960–970.
Howe, A.C., Jansson, J.K., Malfatti, S.A., Tringe, S.G., Tiedje, J.M. and Brown, C.T. (2014) Tackling soil diversity with the assembly of
large, complex metagenomes, Proceedings of the National Academy of Sciences of the United States of America, Vol. 111,
pp. 4904–4909.
Hugenholtz, P., Goebel, B.M. and Pace, N.R. (1998) Impact of culture-independent studies on the emerging phylogenetic view of
bacterial diversity, Journal of Bacteriology, Vol. 180, pp. 4765–4774.
Jalali, K. and Baldwin, S.A. (2000) The role of sulphate reducing bacteria in copper removal from aqueous sulphate solutions, Water
Research, Vol. 34, pp. 797–806.
Jasper, D. (2007) Beneficial soil microorganisms of the Jarrah Forest and their recovery in bauxite mine restoration in southwestern
Australia, Restoration Ecology, Vol. 15, pp. S74-S84.
Kandler, O. and König, H. (2014) Cell wall polymers in Archaea (Archaebacteria), Cellular and Molecular Life Sciences CMLS, Vol. 54,
pp. 305–308.
Khoshnoodi, M., Rezadehbashi, M., Taupp, M., Hallam, S. and Baldwin, S.A. (2012) Correlation of the microbial diversity In a
biochemical reactor treating metal-rich water with environmental factors, Canadian Society of Microbiology, Vancouver,
Canada.
Khoshnoodi, M., Dipple, G., and Baldwin, S.A. (2013) Mineralogical study of a biologically-based treatment system that removes
arsenic, zinc and copper from landfill leachate, Minerals, Vol. 3, pp. 427-449.
Koren, D.W., Gould, W.D. and Bédard, P. (2000) Biological removal of ammonia and nitrate from simulated mine and mill effluents,
Hydrometallurgy, Vol. 56, pp. 127–144.
Krause, S., Le Roux, X., Niklaus, P.A., Van Bodegom, P.M., Lennon, J.T., Bertilsson, S., Grossart, H.-P. ... Bodelier, P.L.E. (2014) Trait-
based approaches for understanding microbial biodiversity and ecosystem functioning, Frontiers in Microbiology, Vol. 5, pp.
1–10.
Kyrpides, N., Woyke, T. and Eisen, J. (2014) Genomic encyclopedia of type strains, phase I: The one thousand microbial genomes
(KMG-I) project, Standards in Genomic Sciences, Vol. 9, pp. 628–634.
Luo, F., Gitiafroz, R., Devine, C.E., Gong, Y., Hug, L.A., Raskin, L. and Edwards, E.A. (2014) Metatranscriptome of an anaerobic benzene-
degrading, nitrate-reducing enrichment culture reveals involvement of carboxylation in benzene ring activation, Applied and
Environmental Microbiology, Vol. 80, pp. 4095–4107.
Marschner, P., Crowley, D. and Yang, C.H. (2004) Development of specific rhizosphere bacterial communities in relation to plant
species, nutrition and soil type, Plant and Soil, Vol. 261, pp. 199–208.
Marschner, P., Yang, C.-H., Lieberei, R. and Crowley, D. (2001) Soil and plant specific effects on bacterial community composition in
the rhizosphere, Soil Biology and Biochemistry, Vol. 33, pp. 1437–1445.
Martínez-Ruiz, C. and Marrs, R.H. (2007) Some factors affecting successional change on uranium mine wastes: Insights for ecological
restoration, Applied Vegetation Science, Vol. 10, pp. 333–342.
Martin-Laurent, F., Philippot, L., Hallet, S., Chaussod, R., Germon, J.C., Soulas, G. and Catroux, G. (2001) DNA extraction from soils:
Old bias for new microbial diversity analysis methods, Applied and Environmental Microbiology, Vol. 67, pp. 2354–2359.
McDonald, L. and Strosher, M. (1998) Selenium mobilization from surface coal mining in the Elk River Basin, British Columbia: A survey
of water, sediment and iota, Ministry of Environment, Lands and Parks: Kootenay Region, Cranbrook, Canada.
Mewis, K., Armstrong, Z., Song, Y.C., Baldwin, S.A., Withers, S.G. and Hallam, S.J. (2013) Biomining active cellulases from a mining
bioremediation system, Journal of Biotechnology, Vol. 167, pp. 462–471.
Mewis, K., Taupp, M. and Hallam, S.J. (2011) A high throughput screen for biomining cellulase activity from metagenomic libraries,
Journal of Visualized Experiments: JoVE, Vol. 48(2461), pp. 1-4.
Meyer F, Paarmann D, D’Souza M, Olson R, Glass EM, Kubal M, Paczian T, Rodriguez A, Stevens R, Wilke A, Wilkening J, and Edwards
RA. (2008) The metagenomics RAST server – a public resource for the automatic phylogenetic and functional analysis of
metagenomes, BMC Bioinformatics, Vol. 9(386), pp. 1-8.
Mirjafari, P. and Baldwin, S.A. (2011) Factors affecting the start-up, operation and decline of a laboratory-based passive treatment
system for selenium and sulfate removal, in Proceedings Conference of Metallurgy, 2–5 October 2011, Montreal, Canada.
Mirjafari, P. (2014) Complex biochemical reactors for selenium and sulphate reduction: organic material biodegradation and
microbial community shifts. UBC Dissertation, pp. 1-181. http://hdl.handle.net/2429/46013.
Morita, M., Uemoto, H. and Watanabe, A. (2007) Reduction of selenium oxyanions in wastewater using two bacterial strains,
Engineering in Life Sciences, Vol. 7, pp. 235–240.
Morrison, B., Lamb, D. and Hundloe, T. (2005) Assessing the likelihood of mine site revegetation success: A Queensland case study,
Australian Journal of Environmental Management, Vol. 12, pp. 165–182.
Mummey, D.L., Stahl, P.D. and Buyer, J.S. (2002) Soil microbiological properties 20 years after surface mine reclamation: Spatial
analysis of reclaimed and undisturbed sites, Soil Biology and Biochemistry, Vol. 34, pp. 1717–1725.
Muscatello, J.R. and Janz, D.M. (2009) Selenium accumulation in aquatic biota downstream of a uranium mining and milling operation,
The Science of the Total Environment, Vol. 407, pp. 1318–1325.
Picard, C., Ponsonnet, C., Paget, E., Nesme, X. and Simonet, P. (1992) Detection and enumeration of bacteria in soil by direct DNA
extraction and polymerase chain reaction, Applied Environmental Microbiology, Vol. 58, pp. 2717–2722.

Polster, D. (1989) Successional reclamation in western Canada: new light on an old subject. Canadian Land Reclamation Association
and American Society for Surface Mining and Reclamation Conference, 27-31 August, Alberta, Canada. pp. 333-337.
Pruden A, Messner N, Pereyra L, Hanson RE, Hiibel SR, Reardon KF (2007) The effect of inoculum on the performance of sulfate-
reducing columns treating heavy metal contaminated water. Water research, Vol. 41, pp. 904–914.
Ranjard, L. and Richaume, A. (2001) Quantitative and qualitative microscale distribution of bacteria in soil, Research in Microbiology,
Vol. 152, pp. 707–716.
Ratnasingham, S. and Hebert, P.D.N. (2007) Barcoding BOLD: The barcode of life data system (www.barcodinglife.org), Molecular
Ecology Notes, Vol. 7, pp. 355–364.
Rezadehbashi, M., Khoshnoodi, M., Taupp, M., Hallam, S. and Baldwin, S.A. (2012) Microbial diversity of a biochemical reactor
removing metals from smelter waste leachate, Canadian Society of Microbiology, Vancouver, Canada.
Schmidt, T.S.B., Matias Rodrigues, J.F. and von Mering, C. (2014) Ecological consistency of SSU rRNA-based operational taxonomic
units at a global scale, PLoS Computational Biology, Vol. 10, pp. 1–10.
Schmidtova, J. and Baldwin, S. (2011) Correlation of bacterial communities supported by different organic materials with sulfate
reduction in metal-rich landfill leachate, Water Research, Vol. 45, pp. 1115–1128.
Sharpton, T.J. (2014) An introduction to the analysis of shotgun metagenomic data, Frontiers in Plant Science, Vol. 5(209), pp. 1-14.
Silva, A.F., Carvalho, G., Oehmen, A., Lousada-Ferreira, M., van Nieuwenhuijzen, A., Reis, M.A.M., Crespo, M.T.B (2011) Microbial
population analysis of nutrient removal-related organisms in membrane bioreactors. Applied Microbiology and
Biotechnology, Vol. 93(5), pp. 2171-2180.
Skousen, J., Ziemkiewicz, P. and Venable, C. (2009) Tree recruitment and growth on 20-year-old, unreclaimed surface mined lands in
West Virginia, International Journal of Mining, Reclamation and Environment, Vol. 20, pp. 142–154.
Tebbe, C.C. and Vahjen, W. (1993) Interference of humic acids and DNA extracted directly from soil in detection and transformation
of recombinant DNA from bacteria and a yeast, Applied Environmental Microbiology, Vol. 59, pp. 2657–2665.
Thirgood, J. (1986) Reclamation past, present and future, in Proceedings 11th Annual Meeting Canadian Land Reclamation
Association, 3-6 June, Vancouver, Canada.
Tordoff, G., Baker, A.J. and Willis, A. (2000) Current approaches to the revegetation and reclamation of metalliferous mine wastes,
Chemosphere, Vol. 41, pp. 219–228.
Tringe S.G., von Mering C, Kobayashi A, Salamov AA, Chen K, Chang HW, Podar M, Short JM, Mathur EJ, Detter JC, Bork P, Hugenholtz
P, and Rubin EM. (2005) Comparative metagenomics of microbial communities, Science, Vol. 308, pp. 554–557.
Tsai, Y.L. and Olson, B.H. (1992) Rapid method for separation of bacterial DNA from humic substances in sediments for polymerase
chain reaction, Applied Environmental Microbiology, Vol. 58, pp. 2292–2295.
Valentini, A., Pompanon, F. and Taberlet, P. (2009) DNA barcoding for ecologists, Trends in Ecology and Evolution, Vol. 24, pp. 110–
117.
van der Lelie, D., Taghavi, S. and McCorkle, S. (2012) The metagenome of an anaerobic microbial community decomposing poplar
wood chips, PloS One, Vol. 7(5), No. e36740, pp. 1-–6.
Verastegui, Y., Cheng, J., Engel, K., Kolczynski, D., Mortimer, S., Lavigne, J., Montalibet. J. … and Neufeld, J.D. (2014) Multisubstrate
isotope labeling and metagenomic analysis of active soil bacterial communities, mBio, Vol. 5(4), No. e01157-14, pp. 1-12.
Wang, S. and Mulligan, C.N. (2006) Occurrence of arsenic contamination in Canada: sources, behavior and distribution, Science of
The Total Environment, Vol. 366, pp. 701–721.
Weisburg, W.G., Barns, S.M., Pelletier, D.A. and Lane, D.J. (1991) 16S ribosomal DNA amplification for phylogenetic study, Journal of
Bacteriology, Vol. 173, pp. 697–703.
Xia, L.C., Cram, J.A., Chen, T., Fuhrman, J.A. and Sun, F. (2011) Accurate genome relative abundance estimation based on shotgun
metagenomic reads, E. Dias-Neto (ed), PloS One, Vol. 6, No. e27992, pp. 1–13.
Yeates, C., Gillings, M.R., Davison, A.D., Altavilla, N. and Veal, D.A. (1998) Methods for microbial DNA extraction from soil for PCR
amplification, Biological Procedures Online, Vol. 1, pp. 40–47.
Zhou, J., Bruns, M. and Tiedje, J. (1996) DNA recovery from soils of diverse composition, Applied Environmental Microbiology, Vol.
62, pp. 316–322.
Ziemkiewicz, P.F., Skousen, J.G. and Simmons, J. (2003) Long-term performance of passive acid mine drainage treatment systems,
Mine Water and the Environment, Vol. 22, pp. 118–129.
Zinck, J. and Griffith, W. (2013) Review of mine drainage treatment and sludge management operations, MEND report 3.43.1, viewed
10 February 2015, http://mend-nedem.org/wp-content/uploads/MEND3.43.1 ReviewofMineDrainageTreatmentSludge
Management Operations.pdf.


© 2015 InfoMine Inc., Canada, 978-0-9917905-9-3
R. Holmes Yukon Government, Canada

M. Flynn Canada Mining Innovation Council, Canada
M.B. Thorpe Torex Gold, Canada
Many significant advancements have been made in the mine closure process over the past several decades,
including “design for closure,” the development of comprehensive mine closure planning and the regulatory
requirement for financial assurances for mine site rehabilitation. However, mine closure remains a significant
challenge for all major mining jurisdictions around the world. This is largely due to the amorphous nature of
the process, inconsistencies in regulations across jurisdictions and inadequate management approaches. To
address these challenges in mine closure, the Canada Mining Innovation Council’s Environmental Stewardship
Initiative has prioritized the development of standardised, performance-based completion criteria for mine
closure and the subsequent relinquishment. The criteria will allow a clear and consistent path to mine closure
by defining the conditions acceptable to stakeholders, which, if met, will enable the government and
regulatory agencies to accept mine sites into their custody for long-term stewardship. The criteria will serve
as a key strategic planning and risk management tool for industry, regulatory agencies and mining
stakeholders at large. Mining stakeholders will have certainty regarding how a site will be rehabilitated,
decommissioned and closed, and it is anticipated that the criteria will drive closure management from mine
design to the ultimate closure of operations.
This paper provides an overview of the initiative by outlining the challenges in mine closure that the project
aims to solve. The standardised closure criteria concept is defined, with examples of criteria that will likely be
included in the framework. An overview of how industry and regulatory agencies will use the criteria is
provided. The benefits that will likely accrue to mining stakeholders are summarised. The main benefits for
industry include improved efficiency and consistency in the mine closure process as well as the promotion of
a more stable and competitive investment climate. For governments and regulatory agencies, the mine
closure process will become more managed and structured, thus enabling a more simplified and efficient
regulatory regime. Furthermore, with a managed relinquishment process, the accrual of improperly closed
(i.e. abandoned and orphaned sites) will be reduced. This will also provide a key public benefit, as public funds
will not be required for associated clean-up costs. Lastly, local stakeholders will be assured that environmental
performance will be central to the mine closure process with a view toward establishing positive post-mining
legacies.
The Canada Mining Innovation Council (CMIC) is a multi-stakeholder, open-innovation business ecosystem.
This approach brings together the key mining stakeholders to define the major research, development and
innovation challenges facing the mining business. The process is conducted in an open way, such that ideas,
technology, risk and innovation are leveraged and shared among the participating groups with a focus on
producing real solutions to current challenges.
In the fourth quarter of 2012, CMIC formed the Environmental Stewardship Initiative (ESI) with a mandate to
develop — through step-change technological innovation, project development and multi-stakeholder
collaboration — solutions to some of the myriad environmental and sustainability challenges facing the
Canadian mining industry.

Standardised, performance-based completion criteria for mine closure and mine site relinquishment R. Holmes, M. Flynn and M.B. Thorpe
A wide range of activities encompass the spectrum of environmental management related to mining; thus,
in 2012, the ESI commissioned the engineering consultancy Hatch Ltd. to undertake a scoping study outlining
areas where further research, development and innovation could be focussed. Several multi-stakeholder
surveys were also conducted to ascertain stakeholder priorities and needs in terms of research, development
and innovation. Based on the results of these activities, the ESI formed three working groups to prioritise and
develop specific projects: the closure working group (CWG), the tailings working group (TWG) and the water
working group (WWG).
Following a series of workshops and monthly meetings, the ESI prioritised the following projects for project
development activities:
 Standardised, performance-based completion criteria for mine closure and mine site
relinquishment;
 Passive systems for managing acid rock drainage (ARD) from waste rock piles;
 National database for water quality monitoring data;
 Remote, real-time sensors for water quality monitoring.
Figure 1 provides an overview/timeline of the project development activities undertaken for the projects.
This paper provides an overview of the first project: standardised, performance-based completion criteria
for mine closure and mine site relinquishment.
Multi-stakeholder Formation of Working

Scoping Study
Surveys Groups
Q1 2013
Q2 2013 Q2 2013
Project Implementation
Pre-feasibility Study Feasibility Study
Commences
Q2 2014 Q1 2015
Q2 2015
As Clark and Clark (2005) note, prior to the 1960s, most mining jurisdictions gave relatively scant
consideration to the post-operations phase of mining activities, including site decommissioning, landscape
reclamation and ecosystem rehabilitation, or what is now termed “mine closure.” Limited mine closure
activities took place, and mine sites were often abandoned with myriad negative consequences for the
surrounding environment. Over the past decades, 10,000+ mine sites in Canada alone have been abandoned
and/or orphaned, resulting in associated clean-up costs — since 2002 — of more than C$1 million (Tremblay
and Hogan, 2014).
However, with the increasing awareness of the environmental risks associated with inadequate mine closure,
along with the advent of “sustainable mining,” mining stakeholders are increasingly expecting a positive
environmental and social legacy from mining operations. Thus, mine closure has become a central
component of mining projects from the business, regulatory and stakeholder engagement perspectives.
Specifically, in most major mining jurisdictions, mine closure plans are now required before mining
commences and are updated at regular intervals throughout the life of mine cycle. In turn, mining companies

Regulatory
now plan for closure at the outset of mine development, typically through what is known as ‘design for
closure’ (Caldwell and McPhail, 2012). Financial assurances for closure activities are typically required to help
ensure that adequate funds are available for the closure costs. Companies are also expected to actively
engage stakeholders throughout the process to help ensure that community needs are met and positive
environmental and social legacies are maintained.
Despite these important advancements, comprehensive mine closure remains a challenge for virtually every
mining jurisdiction in the world, as it is an amorphous concept with no defined end in many instances. Closure
completion criteria are often ambiguous, ill defined and/or inconsistent across jurisdictions, resulting in
uncertainty in approaches to mine closure for mine operators, unclear standards for regulatory agencies and
confused expectations among mine communities and stakeholders at large. Furthermore, the lack of a
defined end often leads to the management of sites in perpetuity; should a company dissolve, the
responsibility of managing site rehabilitation and closure activities reverts to the government.
To address these challenges in mine closure and site relinquishment, the ESI is implementing a national
program to develop a framework for standardised, performance-based completion criteria for mine closure
and site relinquishment. These criteria aim to provide a clear path to mine closure and the subsequent
relinquishment by defining the conditions acceptable to mining stakeholders, which, if met, will enable the
governments to accept sites into their custody for long-term stewardship.
Many advanced mining jurisdictions have formal legal stipulations that allow the relinquishment of mine sites
to a “responsible authority,” defined as “any government body empowered to approve activities associated
with the mine closure process” (Australian and New Zealand Minerals Council and Minerals Council of
Australia, 2000). In Canada, the relevant responsible authority is typically the provincial government, or, in
the case of the territories and uranium sites, the provincial or territorial government in conjunction with the
federal government. In Canada, some of the major mining jurisdictions (i.e. British Columbia, Ontario, Quebec
and Saskatchewan) allow for relinquishment in mining legislation and/or regulations. For example, Section
149 (1) of the Ontario Mining Act states:
The Minister may accept a surrender of mining lands [i.e. relinquishment] from a proponent on the
conditions specified by the Minister if,
a) the project relating to the mining lands is closed out; or,
b) the project relating to the mining lands is not closed out only because it is subject to long-term
monitoring and maintenance by the proponent.
An example outside of Canada is Western Australia, where relinquishment of a mine site requires:
formal acceptance from the relevant regulatory agencies that all obligations under the Mine
Closure Plan associated with the tenement, including achievement of the closure completion
criteria have been met, and that arrangements for future management and maintenance of the
tenement have been agreed to by the subsequent owners or land managers. (Government of
Western Australia Department of Mines and Petroleum, 2014)
Despite some de jure stipulations across jurisdictions, few mine sites have been successfully relinquished to
the respective responsible authorities. For example, in Ontario, no mine site has been relinquished under the
current regulatory regime. In Australia, only three mine sites were found to have been successfully
relinquished to a responsible authority.
The aforementioned pre-feasibility study (see Figure 1) conducted by the ESI determined that a key gap
precluding more widespread relinquishment is the lack of comprehensive legal and regulatory regimes. In
the example of the Ontario Mining Act mentioned above, there are no formal stipulations as to what
comprises “the conditions specified by the Minister.”

An exception to this is found in Saskatchewan, which has developed the Institutional Control Program (ICP).
In 2007, the Saskatchewan government enacted the Reclaimed Industrial Sites Act and the Reclaimed
Industrial Sites Regulations to enforce the ICP. Under the program, a closed site enters a period of “transition
phase monitoring,” during which the mine operators are required to continue monitoring the site at their
own expense. If the site performs in accordance with the decommissioning and reclamation plan and
achieves the predicted stability (i.e. chemical, biological and physical stability) during the transition phase,
the mine operator may make an application for release from decommissioning and reclamation. Upon
approval of the release, the mine operator may apply for a release from its surface lease, which allows the
transfer of custodial responsibility for the property to the provincial government (Saskatchewan Ministry of
Energy and Resources (MoER, 2009).
Since the program was initiated, six sites have been transferred into provincial custody, including five
uranium sites and one gold mine. Between 2012 and 2017, the MoER anticipates that two more sites will
make applications for acceptance into the ICP.
The province of Alberta has established a comprehensive legal and regulatory framework for the
relinquishment of oil sands sites. According to the Alberta Environmental Protection and Enhancement Act
(EPEA), oil sands operators are required to work towards achieving reclamation certification for disturbed
sites. Once certification is granted, the operator is relieved of further reclamation responsibilities and the
site enters provincial custody for long-term stewardship.
As of 2014, one site has been returned to the Alberta provincial government: an overburden dump at
Syncrude’s Gateway Hill project. One of the challenges for operators and the regulatory agencies in having
more sites enter into provincial custody is the lack of streamlined criteria for determining whether
reclamation objectives have been met and what the regulatory agencies deem acceptable (Charette and
Poscente, 2012). This results in uncertainty for operators regarding how the regulatory agencies will evaluate
an application. Similarly, regulatory agencies experience difficulty applying reclamation criteria in a fair and
consistent manner (Charette and Poscente, 2012).
Consequently, oil sands stakeholders deemed the need to develop clear and consistent criteria for site
reclamation and release a priority through the Cumulative Environment Management Association (CEMA) —
a multi-stakeholder group comprising Aboriginal groups, environmental non-governmental organisations
(ENGOs) and industry that serves as a key advisor to the provincial and federal governments on
environmental issues related to the oil sands. In 2010, a three-year process was undertaken to establish the
criteria through an extensive, iterative stakeholder consultation process. In late 2014, the Alberta
Government preliminarily accepted the criteria framework subject to further review, development and field
demonstration/assessment (Alberta Environment and Sustainable Resource Development, 2014).
The standardised criteria will be structured as a suite of conditions operators will be required to meet when
closing a mine site in order for the site to be acceptable and ready for the ensuing transfer to a responsible
authority. Crucially, this includes a set of environmental performance endpoints that must be achieved — or
be on course to being achieved — to ensure that the overarching objectives of mine closure are met,
including minimising public health and safety risks, minimising ongoing negative environmental impacts, and
achieving an acceptable, productive and sustainable post-mining land use.
The performance-based completion criteria will be standardised and will be able to be applied broadly across
mine sites and jurisdictions using a similar approach as that used by Alberta for the oil sands sites. The criteria
will be structured using a framework approach, enabling flexibility for — and adaptability to — local, site-
specific conditions. The provincial and territorial governments will be able to adapt and tailor the framework
to align with their legislation and regulations, as well as the needs and priorities of local stakeholders.

Regulatory
Furthermore, the criteria will be designed with the aim of ensuring continual improvement so as to align the
criteria with new technologies, regulations and management practices.
It is important to note that the criteria will not serve as a one-off checklist of site conditions for mine closure
completion and relinquishment approval. Rather, the criteria will serve as a key planning and risk
management tool that will be used throughout the life of mine cycle, from mine conception design through
to operations and final closure. Thus, the criteria will serve as the basis for a comprehensive, managed and
structured process for mine closure planning, implementation and assessment. The mine closure process will
become more streamlined, efficient and consistent, thus improving environmental management, and,
ultimately, environmental outcomes, while providing investors with a clear understanding of the closure and
rehabilitation needs for site release.
The criteria will serve as a key catalyst/driver for the overall improvement of the mine closure process.
Specifically, with the development of the acceptable conditions for relinquishment, governments and
regulatory agencies will be able to devise regulations that enable the process to occur clearly and smoothly,
as has been done in Saskatchewan with the ICP. Similarly, the criteria will help to drive the development of
new technologies that will aid in achieving the defined endpoint conditions. For example, some performance
conditions that would be acceptable for stakeholders may not be currently achievable because of
technological limitations and/or the associated costs. Notwithstanding this, the criteria conditions will serve
as the goal that operators, service providers and researchers will need to work towards meeting through
technology development, adaptation of current methods/techniques and/or new management approaches.
The criteria will cover the major components of mine closure, including site/facility decommissioning, land
reclamation and ecosystem rehabilitation (among others). Where possible, the criteria will be quantifiable.
Some criteria and associated indicators may be achieved by meeting a milestone (e.g. geotechnical design).
Others may take several years to fully achieve; accordingly, some criteria may have an element based on time
(e.g. the criterion is being on trend to being achieved). Therefore, mechanisms to mitigate residual risk will
be incorporated into the framework.
Category Criteria example

Pit slopes; mine workings; underground
openings; tailings impoundments; waste
Physical stability
rock piles; dams, dykes, and other
containment structures
Water quality; water quantity; seepage
from waste rock piles; seepage from
Geochemical stability
tailings storage facilities; ARD; metal
leaching (ML)
Landscape integration with surrounding
Land reclamation landforms/topography; post-mining land
use; land capability
Ecosystem rehabilitation Biodiversity; wildlife habitat
Building demolition; asset transfers;
Site infrastructure and waste
hazardous waste disposal
Monitoring and maintenance funds;
Financial assurance unforeseen events funds; ongoing
liabilities

Table 1 outlines some examples of completion general criteria classes that may be included in the framework,
which will be further refined by the stakeholder consultation process. Table 2 provides a few specific
examples of completion criteria that could be included in the framework.
Criteria class Potential examples

Physical stability  Dams, dykes, and containment structures designed to
meet a minimum 1:100 year standard
 Pits are closed such that stability and erosion standards
are met
 Tailings impoundment structures meet design standards
and include detailed declarations of quantitative
performance objectives
 Non-critical structures are designed to meet 1:100 year
flood event; critical structures are designed to meet
1:1000 year flood event
 Mine rock piles have the required structural stability
 Mine openings are sealed to prevent access
 Underground workings are backfilled to blend in with
surrounding topography
Geochemical stability  Receiving water bodies meet water quality objectives
(e.g. regulatory, reference points, natural variation)
 Site does not require any active water treatment
 Site poses no further risk of ARD or ML
 Soil pH, salinity, nutrients, depth, water infiltration,
crusting and slope meet prescribed limits from reference
sites or regional studies (e.g. pH in range 5.0–8.5; salinity
< 0.2% chloride; EC (1:5 H2O) < Y dS/m-1)
Land reclamation  Land is reclaimed to allow for agreed upon post-mining
land use
 Landscape is integrated with surrounding landforms and
topography
Ecosystem rehabilitation  Reaching agreed species or ecosystem diversity targets,
such as areas have to have at least X of particular species
per m2
 Species richness is greater than or equal to Y% of the
mean value recorded in all 20 m by 20 m reference plots
in analogue sites in the target ecosystem
Site infrastructure and waste  All hazardous waste is disposed of safely
 All site infrastructure is dismantled or transferred to an
appropriate third party
Financial assurance  Monitoring and maintenance funds are posted
 Unforeseen events funds are posted

Regulatory
The criteria framework will serve as a key strategic planning and risk management tool for regulatory
agencies and governments. With the development of the criteria, the mine closure process will become more
structured, managed and streamlined in contrast to some of the current and past mine closure practices,
where mines are/were sometimes closed in an inconsistent and haphazard manner. As mentioned, over the
past decades, 10,000+ abandoned and/or orphaned mine sites with associated clean-up costs in excess of C$
1 billion have accrued to the Crown, and, ultimately, the public. The criteria approach will help to prevent
this in future, as the return of mine sites to the Crown will be done on managed terms and with far greater
environmental protection, thereby reducing the burden of long-term stewardship on the regulatory agencies
and governments.
The criteria will also help to promote consistency and efficiency in the approvals and permitting process. The
application of fair and consistent assessment standards will be promoted, as the regulators will be provided
with a tool for objectively determining the acceptability and readiness of a site for transfer into provincial
custody. In turn, this will help to create a more stable and competitive investment climate for current and
future mining developments, which will likely help to attract additional mining investment and general public
economic benefits over time. The general public will further benefit, as taxpayer funds will not be required
for the associated clean-up costs of abandoned/orphaned mines.
Lastly, as outlined above, one of the major gaps in relation to the long-term stewardship of closed mine sites
is the lack of comprehensive regulatory, policy and legislative frameworks for site relinquishment. With the
development of the criteria, governments and regulators will be provided with an understanding of
stakeholder expectations and requirements for achieving relinquishment. This will enable the criteria to drive
the development of the appropriate regulations, policies and legislation toward this end. Given that few
jurisdictions around the world have such programs in place, Canada will also further its position as a leader
in terms of sustainability and competitiveness in the mining business.
As with the regulatory agencies/governments, the criteria framework will serve as a key strategic planning
and risk management tool for operators throughout the life of mine cycle. With established completion
criteria/performance endpoints for relinquishment, mine operators will be able to proceed with mine closure
with greater certainty, and, therefore, less risk. The overall mine closure process will also become more
efficient. Taken together, these benefits will result in a more stable and attractive investment climate for
current mine operators as well as future mine proponents.
The criteria will be used for initial planning as part of the design for closure process. The initial planning and
design process will be strengthened, as operating sites and new mining projects will be able to use the criteria
to design for relinquishment (Cowan et al., 2010). Progressive rehabilitation will also be strengthened, as
there will be an increased incentive for operators to close sites in line with the completion criteria.
Specifically, once the performance conditions are met, operators will be released from all further monitoring
and maintenance requirements. This will help to decrease mine closure obligations over time, thus providing
a financial incentive for implementing and progressively achieving the completion criteria. To be sure,
operators will not be fully released from the liabilities associated with the site; in line with the “polluter pays”
principle, operators will still be liable for any deleterious environmental impacts (e.g. contamination).
Efficiency will be increased still further as mine proponents use the framework as the foundation for gaining
stakeholder acceptance for specific projects, thereby simplifying the permitting and consultation process.
Specifically, mine operators/proponents will have fewer requirements for gaining acceptance; instead, it is
anticipated that the main requirements for achieving acceptance will be tailoring the performance-based
conditions outlined as part of the criteria for the site-specific conditions.

As confusion surrounding the mine closure process will be reduced, it is anticipated that local stakeholder
confidence in mining will increase with the development of the criteria. Local stakeholders will have more
assurance that mining activities will cease without unacceptable consequences for the local communities and
the environment, which, in turn, will increase local buy-in for current and future mining development.
The criteria will provide local mining stakeholders with an increased understanding of the mine closure
process as a whole. These stakeholders will know what to expect from mining operators and the regulatory
agencies when mining ceases and the closure process is complete. Local stakeholders will also have more
assurance that mine closure will be conducted in a managed and structured way with robust protection for
the environment. Similarly, they will be assured that mines will not be abandoned, with the associated
environmental and economic costs externalised to local communities and the general public.
Given that the development of the criteria will include extensive stakeholder consultation, local stakeholders
will be further assured that the general needs of local mining communities (e.g. a productive, post-mining
land use) are accounted for. Local stakeholders will also be able to use the criteria as a tool/ framework for
tailoring site-specific criteria to help further ensure that local needs and priorities are met.
Stakeholder consultation is a key aspect of the mine closure process. Proactive and ongoing stakeholder
engagement throughout the life of mine — from initial planning to final closure — helps ensure that local
needs and priorities are met, thus helping to secure and maintain a project proponent’s social license to
operate. This includes the development of mine closure completion criteria aligned with the regulatory
requirements and stakeholder needs.
Accordingly, an extensive stakeholder consultation program will drive the development of the criteria. The
main objective of the consultation process is to enable stakeholders to arrive at a broad consensus regarding
the acceptable conditions for mine closure completion and the subsequent relinquishment.
The main stakeholders to be involved in the process are as follows:
 Governments and regulatory agencies
 Industry/industry associations
 Local/Aboriginal communities and community associations
 Service providers
 Technical experts
 ENGOs
 Public interest groups
 Mining research organizations
As the criteria will be standardised across jurisdictions, the stakeholder consultation program will be national,
with a focus on the major mining jurisdictions and regional groupings (i.e. British Columbia, Ontario, Quebec,
Saskatchewan, the Territories and the Maritimes). As mentioned, the criteria will be structured as a
framework approach, which will help to solve some of the challenges with standardisation. Jurisdictions can
tailor the criteria to their specific legal requirements and site-specific needs, based on varying risk appetites
and the requirements for local stakeholders. Furthermore, a key goal of the development of the criteria
framework is greater streamlining of the mine closure process across jurisdictions, which will allow investors
to understand their closure obligations at the onset of the project. Although these challenges may seem
daunting, the resulting framework should be beneficial to all, and, as such, should be pursued and achieved.

Regulatory
Following an initial period of partnership development (Phase 1), it is anticipated that the consultation
process will be carried out over 48 months beginning in the first quarter of 2016. The process will be broken
down into the following phases:
 Phase 2: Stakeholder sensitisation (12 months)
 Phase 3: Detailed, iterative stakeholder consultations (21 months)
 Phase 4: Government acceptance (15 months)
The stakeholder-sensitisation phase will provide the primary mining stakeholders with an understanding of
the purpose and benefits of establishing the standardised criteria. In turn, the project implementation team
will be provided an understanding of the stakeholders’ key initial needs, concerns and priorities related to
mine closure completion and the subsequent relinquishment. Parallel to these consultations, draft criteria
will be developed by amalgamating, streamlining and more clearly defining some of the existing mine closure
completion criteria across Canadian — and some international — mining jurisdictions. The initial stakeholder
consultations will help to further inform the required inputs, which will be incorporated into a second
iteration of the criteria.
Phase 3 will involve consulting with the various stakeholders on the various criteria components. As the
process is designed to be iterative, several rounds of consultations will take place to ensure that the feedback
from each stakeholder group is continually incorporated into the successive iterations of the draft criteria. A
key component of Phase 3 will be consultation with various technical experts to determine the feasibility and
practicability of the conditions acceptable to stakeholders. Consequently, expectation management will also
be a key component of this phase. Phase 3 will be completed once a broad consensus has been formed on
the criteria framework.
Government acceptance of the criteria will be the central focus of the final phase of the consultation process.
To be sure, governments and regulatory agencies will be engaged from the beginning of the entire process,
as early engagement is typically a key principle for achieving buy-in. Accordingly, this phase of the
consultations will focus on determining ways of formalising the criteria into regulations and legislation. This
will likely include modifying the criteria for individual provincial requirements and perhaps, as is the case in
Alberta, field demonstrations and assessment.
An overview of the stakeholder consultation process is presented in Figure 2.
In conclusion, the development of the standardised closure completion criteria will help to solve many of the
challenges most advanced mining jurisdictions currently face. A clear, consistent path to mine closure and
the subsequent site relinquishment will be established, aiming to provide certainty and efficiency for all

mining stakeholders. Industry, regulatory agencies and governments will be provided a key strategic planning
and risk management tool that will help ensure mine closure is conducted through a structured and managed
process. Crucially, environmental performance will be the central focus of the initiative, thus contributing to
the enhancement of environmental outcomes as well as helping to ensure that positive and sustainable post-
mining legacies are achieved. This will help to increase local stakeholders’ confidence and buy-in for mining
developments while creating a more competitive and stable investment climate. Taken together, these
benefits will help to make great strides in advancing the sustainability of the mining business as a whole.
The authors would like to acknowledge the contributions of the members of the Canada Mining Innovation
Council’s Environmental Stewardship Initiative. Special thanks are extended to the following closure working
group members (in alphabetical order): Mike Aziz (Goldcorp), Denise Chapman (O’Kane Consulting), Karen
Chovan (Enviro Integration Strategies), Scott Davidson (New Gold), Heather MacDonald (CH2M Hill), Robert
Reisinger (CH2M Hill), Bryan Schreiner (Saskatchewan Research Council) and Ian Wilson (Saskatchewan
Research Council).
Alberta Environment and Sustainable Resource Development (2014) Criteria and indicators framework for oil sands mine reclamation
certification, viewed 4 January 2015, http://esrd.alberta.ca/lands-forests/land-industrial/reclamation-initiatives-in-
alberta/documents/CriteriaIndicatorsFramework-Sep04-2014.pdf.
Australian and New Zealand Minerals and Energy Council and Minerals Council of Australia (2000) Strategic framework for mine
closure, Canberra, Australia.
Caldwell, J. and McPhail, G. (2012) Design for closure, TechnoMine, viewed 3 January 2015, http://technology.infomine.com/
reviews/DesignForClosure/welcome.asp?view=full.
Charette, T. and Poscente, M. (2012) Criteria and indicators framework for oil sands mine reclamation certification, Cumulative
Environment Management Association, Edmonton, Canada.
Clark, A. and Clark, J.C. (2005) An international overview of legal frameworks for mine closure, Environmental Law Alliance
Worldwide, Eugene, USA.
long-term liabilities: A guidance document, National Orphaned and Abandoned Mines Initiative (NOAMI), Sudbury, Canada.
Government of Western Australia Department of Mines and Petroleum (2014) Guidelines for preparing mine closure plans, viewed
3 January 2015, http://www.dmp.wa.gov.au/documents/Cuidelines_Closure_Plan.pdf.
Saskatchewan Ministry of Energy and Resources (2009) Institutional control program: Post-closure management of decommissioned
mine/mill properties located on Crown land in Saskatchewan, viewed 3 January 2015, http://www. economy.gov.sk.ca/ICP-
DiscussionPaper-Dec2009.
Tremblay, G. and Hogan, C. (2014). Canada’s National Orphaned and Abandoned Mines Initiative (NOAMI), Presentation to the
International Workshop on the Remediation of Uranium Legacy Sites (1-2 May 2014), Ottawa, Canada.

G. Stewart Province of British Columbia, Canada
The Crown Contaminated Sites Program (CCSP) leads the management of contaminated publically owned
lands in British Columbia, Canada. The program was initiated in 2003 and operates within the Ministry of
Forests, Lands and Natural Resource Operations (FLNR) under a Cabinet-approved policy that commits CCSP
to identify and prioritise contaminated sites that are a provincial responsibility. This policy uses a science-
based risk assessment approach to protect human health and the environment. Although CCSP manages a
wide range of contaminated sites, the vast majority of remediation effort has been directed at closing historic
mine sites. This paper focuses on these mine sites and reviews the program’s inventory, assessment,
prioritisation approach to site remediation.
Over the past decade, a tremendous amount of progress has been made towards the CCSP’s goals. As the
program matured, the focus shifted from developing various tools (e.g. site inventory, assessment guidance,
risk-ranking methodology), to their implementation, and then to remediation and risk reduction. Since
inception, the program has investigated 75 historic mine sites; eight have been fully remediated, and 15 are
characterised as priority sites where ongoing assessment and remediation is occurring. Examples from these
mine sites will be used to illustrate the program’s innovative and flexible approach. Of particular relevance to
mine closure is that CCSP has demonstrated leadership in using a risk-based approach to assess and prioritise
sites and protect human and ecological health. This approach improves the information available on
contaminated sites through site research and a database of sites and activities related to them. Regular
reports are made to government and the public on the progress of work and the financial burden of
contaminated sites. The program works collaboratively with First Nations, communities, environmental
organisations and other levels of government to tackle problems together. The end result after a decade of
effort is a comprehensive and coordinated government-wide framework for managing contaminated sites on
publically owned lands.
British Columbia, Canada, has an active history of mining, manufacturing, forestry, agriculture and many
related activities. While development activities have built a modern and vibrant province, there have been
some detrimental impacts to public lands from early in the province’s history, when the impacts of these
activities were not well managed.
Some historic industrial sites are contaminated with metals, acid rock drainage, hydrocarbons and other
substances harmful to human and environmental health. These sites can become the responsibility of the
Province of British Columbia. To protect the public interest in such sites, the province undertakes activities
such as removing contaminated soils, diverting watercourses away from toxic materials, capping hazardous
areas with engineered covers and replanting vegetation to create habitats for birds, animals, insects and fish.
Crown lands make up 94 per cent of the total land area of British Columbia. This includes the vast majority
of resource areas, wilderness, waterways, parks and ecological reserves in the province. Remediation of
contaminated sites located on this land becomes the province’s responsibility if the persons or companies
responsible for their clean-up cannot be found or no longer exist.

British Columbia, Canada’s Crown Contaminated Sites Program – a ten-year review G. Stewart
In 2002, a report by the Auditor General, Managing Contaminated Sites on Provincial Lands (Office of the
Auditor General of British Columbia, 2002), noted that contaminated sites that fall under the jurisdiction of
provincial ministries and agencies, including Crown lands, were being managed separately by diverse groups
including the ministries responsible for forests, mines, sustainable resources, transportation and the
environment, as well as the B.C. Buildings Corporation. The government therefore did not have the
information to provide oversight, or to manage contaminated sites in a consistent manner. The Auditor
General concluded that the province should establish a framework for managing work on contaminated sites,
gather information to develop management plans and support resource allocations and account for its
overall performance.
In response, the province established the Crown Contaminated Sites Program (CCSP) in 2003. The CCSP was
developed to take appropriate actions and provide public accountability on the management of
contaminated sites for which the province is responsible. The program manages high-risk contaminated sites
to ensure the protection of human health and the environment. It uses a scientific risk-ranking method to
determine which sites need remediation on a priority basis.
The CCSP is administered by the Liquid Natural Gas (LNG), Crown Land Opportunities and Restoration Branch
of the Ministry of Forests, Lands and Natural Resource Operations (FLNR). FLNR is responsible for developing
policies and programs to ensure that forests, lands and natural resources are managed in a sustainable
manner so they can continue to provide environmental, social and economic benefits to all British
Columbians now and into the future.
The CCSP operates under a policy approved by B.C.’s Cabinet: the Management of Provincial Contaminated
Sites Policy.
This government-wide policy defines the guiding principles that underlie all actions to manage contaminated
sites on Crown lands:
 reducing and eliminating risks to human health and the environment and minimising liabilities;
 applying standards that are cost effective, consistent and in keeping with the legislation;
 applying the “polluter pays principle” (The CCSP follows the polluter pays principle outlined in
British Columbia’s Environmental Management Act. Under this principle, the costs associated with
environmental remediation are borne by the parties responsible for the contamination when any
of those parties can be identified and are still in existence.);
 working cooperatively among ministries and consulting with stakeholders and First Nations;
 acting consistently and fairly;
 acting accountably and transparently;
 using innovative strategies for managing contaminated sites, such as public-private partnerships
(P3s);
 minimising the creation of future contaminated sites;
 applying sound science and technology to guide management.
CCSP remediates sites in compliance with the Environmental Management Act, contaminated sites regulation
and hazardous waste regulation. This legislation is a comprehensive and prescriptive process ensuring a

Regulatory
consistent approach and application of numerical and risk-based approaches to the remediation of
contaminated sites in the province.
For a contaminated site located on Crown land, the CCSP attempts to identify the organisations or individuals
responsible for causing the contamination so that they can be held accountable for the cost of clean-up.
Public funds are used only if private parties no longer exist and the site has defaulted to the province, or if
the site poses an immediate unacceptable high risk to human health or the environment.
The number and variety of contaminated sites on Crown land exceeds the resources available within the
CCSP to clean up all sites at once. As a result, the CCSP allocates its resources systematically to work on the
highest priority sites and identify other sites for future action. Sites that present the highest risk to human
health or to the environment receive priority. Sites that pose a lower risk will receive attention as higher risk
sites are remediated. In some cases, investigations determine that sites do not present an unacceptable risk
and do not require remediation.
Because of the potential for risks to human and environmental health, determining an appropriate
methodology for assessing and prioritising the risks is one of the CCSP’s key responsibilities. The methodology
determines what sites will receive action, what resources each candidate site will receive and when each site
will be scheduled for attention. Sites are identified through cross-government review of existing inventories
and the application of coarse filter criteria to known sites, including proximity to sensitive habitats, access by
humans and nature and extent of contamination.
Program sites fall into one of four categories:
 Remediated: Crown sites where clean-up activities are completed.
 Priority: Crown sites that have been identified for action based on site investigation and potential
risk to human health and the environment.
 Candidate: Crown sites on which initial investigation has begun to determine the degree of risk.
 Other: Sites that the province manages in order to fulfil obligations established under legal
agreements or other commitments.
In 2007, the CCSP developed a risk-ranking methodology (RRM) as its principle tool to establish program
priorities and allocate the funds available for remediation (Power et al., 2010 and various guidance
documents). The RRM is a scientifically based method for assessing the ecological and human health risks
presented by contaminated sites in British Columbia. The process uses two components to prioritise sites:
 a risk ranking support tool, which is a data entry and calculation spreadsheet that compares
contaminants in soil, water and sediment to regulatory standards;
 a risk-ranking workshop, which brings together experts on contaminated sites, including
geologists, engineers, biologists and toxicologists, to review and assess the information available
on candidate sites.
The process of assessing sites and the actions required follows these steps:
 identify potential candidate sites;
 determine land ownership and responsibility for clean-up;
 develop a list of candidate sites;

 conduct preliminary site investigations;

 enter field observations and analytical data into the risk-ranking support tool;
 hold a risk-ranking workshop to rank sites, identify priority sites and recommend next steps based
on a group consensus;
 develop management action plans appropriate for priority sites, subject to available financial
resources.
The site prioritisation process is shown in more detail in Figure 1.
Investigators gather available field-based data for candidate sites to populate the risk-ranking tool. The
information allows workshop participants to rank sites based on risk and to make initial recommendations
to CCSP on management actions for identified high-priority sites. The ranking process involves pair-wise
comparison of the multiple lines of evidence. Using the expert guidance of the workshop participants, a level
of action appropriate for each site is selected. The workshop participants can recommend any of the
following actions:
 Immediate action: The workshop recommends immediate action for sites where health or safety
risks require urgent attention.
 Detailed site investigation: Where the information indicates that a potential risk may exist, the
workshop recommends a more detailed site investigation.
 Supplemental investigation: When analysis shows a need for action for a ranked site, but requires
additional information to determine the most appropriate steps, the workshop recommends
specific, targeted investigations. These can include further study of various risks, sample taking or
other types of site analysis.
 Reduce uncertainty: Where the data in the tool are too uncertain for the site to be ranked, the
workshop may recommend gathering more information or performing analyses about specific
issues.
 Lower priority: Low priority sites are those with the lowest risks, where management action can
be deferred.
The recommendations from the workshop guide the CCSP in planning future actions.
A formal report on the state of the CCSP is published once every two years.
The two-year report period was chosen because typically there may not be significant reportable change over
an annual reporting period. In addition, remediation efforts tend to take longer than a single year, so the
period of reportable change in the indicators selected is more consistent with a biennial report. The report
includes an overview of principles, policy, process, governance and legislation driving the CCSP, case studies
from province-wide experience as well as comparison of B.C. progress to other jurisdictions worldwide.

Regulatory

The CCSP also reports out annually on financial information to meet the obligations of the Budget and
Transparency Act, as well as to operate in accordance with the provincial adoption of generally accepted
accounting principles. This information is prepared for inclusion in the annual public accounts report of the
Office of the Comptroller General (OCG) using approved financial reporting procedures as presented in the
core policy and procedures manual from the OCG, Ministry of Finance.
The information is subject to the requirements identified by OCG and may include sites identified as
contingent liabilities or financial liabilities, with estimated dollar amounts for financial liabilities as well as
expenditures in each fiscal year to date.
The CCSP continues important work to achieve its goal: protecting human health and the environment by
returning land to a clean and usable state:
 A total of 84 sites have been investigated since the program began in 2003.
 There are currently 17 priority contaminated sites warranting further study and investigation.
 Remediation has been completed at 15 mine sites. Many of these require ongoing monitoring to
ensure that the remediation is meeting the desired results and to ensure long-lasting integrity of
the remedial works.
The following sections highlight a number of high-priority mine sites managed by CCSP.
Britannia Mine, located 45 kilometres north of Vancouver next to Howe Sound on the Sea-to-Sky corridor,
was once the largest copper producer in the British Commonwealth. This site ranks as CCSP’s highest profile
remediation program owing to its size and complexity. It occupies a very large tract of land – the entire site
covers about 9,000 acres. Britannia Mine began production of copper ore more than 100 years ago and
stopped producing ore in 1974. Water entering the mine site continued to leach out an average of 300
kilograms per day of copper and zinc. Until 2005, this discharge flowed directly into Howe Sound (marine)
totalling approximately five million cubic metres each year, making it one of the largest metal pollution
sources in North America.
The province began its environmental remediation work at Britannia Mine in 2001, after a financial
settlement was reached with the former mine operators. The former operators contributed C$ 30 million
toward the clean-up cost and provincial funding provided an additional C$ 45.9 million. The CCSP has ongoing
responsibility for the investigation, remediation and environmental monitoring at this site.
A P3 was formed in 2005 with EPCOR Water Services Ltd. to design, build, finance and operate the new water
treatment plant. This is a conventional high-density lime water treatment plant processing on average 4.2
million cubic metres of acid water annually. Other detailed technical work completed included underground
mine rehabilitation, a power generation system and systems to divert surface water to reduce inflow of
uncontaminated surface water.
A long-term program of environmental effects monitoring was implemented and continues each year.
In 2010, a final mine closure plan was developed that includes a hydrogeological/hydrological investigation
at Jane Creek, the 2200 level of the mine and a risk assessment program for the Furry Creek watershed.
The success of the remediation work at Britannia is clearly visible, both in its creek and along the beach area.
In 2011, local citizens and the B.C. Rivers Institute reported observing fish in the lower reaches of Britannia
Creek. Fish have never been seen in this part of this creek. The B.C. Institute of Technology confirmed the
presence of fish through independent studies. Along the shoreline, the changes are equally dramatic – what
was bare rock covered with sporadic patches of copper-tolerant algae is now bristling with life. Much of the
shoreline is covered with fucus (rock-weed), barnacles, mussels and a host of other plants and animals. New

Regulatory
life is also visible under the water. Previously barren sediment is now sporting crabs, sea cucumbers and
anemones.
The Atlin Ruffner Mill and tailings site is located approximately 28 kilometres northeast of Atlin, B.C., on the
northwest slope of Mount Vaughan. The Atlin Ruffner Mine was developed following discovery of
mineralisation in an outcropping within the Crater Creek drainage in 1899. Mining operations commenced in
1900 and continued intermittently until 1981. A total of 3,535 tonnes of ore were milled, with recovery of
138,493 kg lead, 13,540 kg zinc, 2,079 kg silver, 920 kg copper, 15 kg cadmium and 3.4 kg gold.
The Atlin Ruffner Mill and tailings area was identified as posing risks to human health and the environment
and was prioritised for remediation. Other Atlin Ruffner mine workings, including numerous adits further up
the mountain, were determined to pose lower risks and were not included as part of the remediation plan.
The mine site included the following:
 An upper pad, which included the mine adit, a flat working area with metals-contaminated soils
and minor mining equipment debris, an ore chute, an ore cart track and several small surface
stockpiles and subsurface volumes of residual ore and waste rock.
 A lower pad area encompassing the mill building, a large flat working area around the north end
of the mill with a shack and contaminated soils with high metal concentrations (primarily arsenic,
lead and zinc) and two trailers (main trailer and single trailer). High levels of total and leachable
metals are present in the milling dust and debris that cover the inside surfaces of the mill building
and associated residual equipment such as the ball mill. Within and around buildings and debris
areas of the lower pad, there are small amounts of hazardous waste materials that include
hydrocarbon liquids and solids, asbestos-containing materials (ACM) and miscellaneous hazardous
materials.
 A tailings area, upper sedimentation pond and lower sedimentation pond comprising sand and
gravel dams with deposits of low permeability tailings behind them. The tailings pond was
partially covered with shallow water cover due to snowmelt and precipitation. The filled or
saturated sedimentation ponds are consistently refilled with overland and subsurface flows,
predominantly originating as groundwater exiting the mine adit. The adit water reportedly
originates as a subsurface flowing borehole that was formerly used as the mill water supply.
Water emanating from the sedimentation ponds infiltrated into the ground immediately down
slope of the ponds.


Regulatory
Test results on soil samples indicated that much of the soil beneath the upper and lower pads and within the
tailings and sedimentation ponds contained concentrations of metals that exceeded the numerical standards
of the B.C. contaminated sites regulation, including those classified as high risk. Leachability tests on some of
the higher concentration samples indicate some of the soil would also be classified as “leachable toxic waste”
and hence hazardous waste under B.C. hazardous waste regulations. Variability in the physical distribution
of metal concentrations meant that soils classified as hazardous waste or exceeding CSR numerical standards
could not be constrained to an area or depth suitable for ex situ remediation efforts. Site remedial efforts
were therefore directed at risk management, risk mitigation or (for some media) source removal.
Following detailed site investigations, a preliminary remedial plan was developed that detailed potential
suitable remedial actions for the site. These included demolition of all buildings and capping the site with
glacial till.
Access to the site for public recreation was discouraged by placing rocks and earthworks across entry routes.
Water will be monitored and the cover will be inspected once a year for the first five years, and every five
years after that.
Howard Mine in B.C.’s West Kootenays, about 30 kilometres south of Nelson, produced gold, silver and lead
during its operation from 1937 to 1938. The mine transported ore over a 7.5-kilometre aerial tramway to a
processing mill at the confluence of the Salmo River and Porcupine Creek. The mill operation left behind
many tonnes of mine tailings, much of which has since been washed away by the river.
Between 2009 and 2013, the CCSP commissioned a series of studies to assess the extent of mine tailings and
their potential risk to human health and the environment. The investigations showed that the mine wastes
contained lead and other metals at concentrations that pose a potential risk. Groundwater and surface water
in the Salmo River directly adjacent to the tailings deposit contained elevated concentrations of cadmium
and zinc. However, Porcupine Creek and the main channel of the Salmo River have not been affected by the
mine wastes.
The results showed that management of the mine wastes was necessary. In order to determine the best
approach, the CCSP analysed a number of different management options in late 2012 and early 2013 and
evaluated them against specific environmental, practical, social and cost criteria. Leachability tests on tailings
samples indicate that these mine wastes were leachable hazardous wastes and needed to be contained in an
approved hazardous waste facility. A conceptual plan to manage the wastes was presented at an open house
in August 2013, and local community members agreed with the recommended remedial option.
Remedial work commenced in the summer of 2014 and will be completed in spring 2015. The remediation
plan includes consolidation of mine tailings capped by a bituminous engineered cover to reduce water and
oxygen ingress into the mine waste.
Toquaht Bay recreation site includes a wilderness campground, marina and kayak launch site on the west
coast of Vancouver Island. The site became popular for public recreation use after the nearby Brynnor iron
ore mine closed in the late 1960s. Sheltered from the open ocean in Barkley Sound near Ucluelet, it was
transferred under the 2011 Maa-nulth Treaty to the Toquaht Nation.
In the treaty, British Columbia agreed to undertake environmental investigations and, if necessary, clean up
any contamination. Preliminary investigations in early 2013 found arsenic, selenium and cobalt in soil at
unacceptable levels, and also identified iron as a concern. The site was closed to assess the risks to humans.
The contaminants were found to be associated with mine tailings produced from the former Brynnor mine,
an unexpected result because iron mine tailings typically contain little arsenic.
The CCSP began detailed site investigations of all media at the site and adjacent beach to determine the
extent of contamination. The CCSP is also working in partnership with the federal Pacific Region

Interdepartmental Shellfish Committee to analyse clam, oyster and geoduck tissues from around Toquart Bay
(the body of water has a different spelling).
The CCSP is communicating closely with Toquaht Nation officials to share information and plans. In late
summer of 2013, the Toquaht Nation opened a campsite nearby to provide a replacement for the recreation
facilities, and in 2014, temporary measures were taken to allow the boat launch and marina to re-open.
First staked in 1915, the steep slopes and alpine meadows of the Emerald Glacier mine site produced over
three million pounds of lead and zinc, and smaller amounts of gold, silver, copper and cadmium. Active work
on the site ended in 1971, leaving a tailings pile and reservoir at a mill site in an adjacent valley about 2.5
kilometres from the mine.
The mining activity resulted in metals-contaminated soils and waste rock around the mill, an upstream
reservoir dam that had supplied water to the mill but had not been maintained since mining ended, and
metals-contaminated tailings beside a creek that flowed from the reservoir. The mill site’s remote location,
approximately 120 kilometres by road from the nearest town of Houston (430 kilometres from the larger city
of Prince George), made it particularly challenging.
Investigations between 2008 and 2011 identified a potential risk from derelict buildings, erosion from the
tailings pile and the possibility that the reservoir dam could breach and further erode the tailings pile. Using
a phased remediation strategy, buildings were first demolished in summer 2012, allowing safe access for
detailed assessments.
The second phase in fall 2013 involved capping all of the areas of the mill and tailings that had high
concentrations of metals. The capping material, consisting of a one-metre cover of compacted glacial till, will
protect the site from further erosion and prevent people and animals from being exposed to the metals-rich
tailings. Owing to the remote site location, an onsite source of the glacial till was developed for use during
remediation and then decommissioned. A berm was also constructed between the creek and the tailings to
prevent any future floods from reaching the tailings.
To remove the risk of the dam breaching, contractors pumped water out to lower the reservoir level and
then built a spillway in the middle of the dam to maintain the new lower level. The spillway mimics a natural
drainage channel and returns the reservoir to pre-mining levels.
The roads to the dam, ore pile and tailings were decommissioned by placing fallen trees on the road surface
and creating a ditch and berm at the road entrance. To begin revegetation, contractors planted willow stakes
on the berms and reseeded disturbed areas.
The CCSP will continue monitoring the former reservoir, the covers, the creek and groundwater. The new
vegetation will be monitored and enhanced as needed.
In 1942, prospectors discovered an outcropping of cinnabar, the ore from which mercury is produced, in
B.C.’s north-central region, about 180 kilometres northeast of Fort St. James. Bralorne Mines operated a
mercury mine on the site for about nine months in 1943 and 1944. One of only a few mercury mines in
Canada, it produced almost 60 tonnes of mercury (
Since the mine closed, its workings and wastes have produced a variety of contaminants, including mercury,
antimony, arsenic, cadmium and chromium. Mercury is a particular concern because it can change its form
and become concentrated through biological processes – called biomagnification – and its effects on wildlife
become more serious as the concentration increases. In addition, the region is rich in many minerals, and the
naturally high levels of several metals exceed the standards for permitted contaminants. This posed
challenges in assessing the mine contaminants and in designing an appropriate remediation strategy. The

Regulatory
remoteness of the site, almost 400 kilometres by road from Prince George, B.C., added to the logistical
difficulties.
The mine site is located within the traditional territory of the Takla Lake First Nation (TLFN). The Province of
British Columbia and the CCSP have been working with the TLFN to identify contaminants and assess the site.
TLFN technicians worked as members of the sampling team to assist with sample collection in 2012, 2013
and 2014. A technical working group made up of government and TLFN representatives, supported by expert
consultants, worked to identify and select a clean-up option that is effective, practical and acceptable to the
TLFN and the province. At community meetings in Takla Landing and Prince George from 2012 through 2014,
the technical working group reported its findings to the members of the First Nation and asked for their input.
By the end of 2013, objectives for the clean-up had been developed, and several options for remedial actions
were identified. The joint working group selected a preferred remedial option in the spring of 2014. Power
and Baker (2015) touch on the findings of an ecological risk assessment of the preferred remedial option.
Clean-up work is scheduled to begin in summer 2015.
When it was established in 2003, British Columbia’s CCSP inherited a legacy of contaminated sites – land,
water and air – from previous times when the impacts of industrial development were not well managed.
Compounding the problem, information on many of these sites was wanting, opening questions such as
which were the most highly contaminated, which required immediate attention and what kinds of
remediation would be needed.
This situation required (1) a system for gathering information, and (2) a way of ranking the urgency of the
sites and the related actions that needed to be taken.
Ten years later, a tremendous amount of progress has been made – but more remains to be done. To date,
we have identified and managed risks to human health and the environment at sites across the province. Our
innovative and flexible approach has created a comprehensive and coordinated government-wide
framework for managing contaminated sites on publicly owned lands.
Since the program was established in 2003, leadership has been demonstrated in:
 using a risk-based approach to assess and prioritise sites and protect ecological health;
 improving the information available on contaminated sites through site research and developing
inventories;
 reporting regularly to government and the public on the progress of work and the financial status
of contaminated sites;
 working collaboratively with First Nations, communities, environmental organisations and other
levels of government to tackle problems together.
The author would like to acknowledge CCSP program staff for ongoing dedication and professionalism in
delivering a world-class environmental program focussed on the remediation of Crown contaminated sites.
These include Katherine O’Leary, Joanna Runnells and Geoff Sinnett.
British Columbia Ministry of Forests, Lands and Natural Resource Operations (FNLR) (2014) LNG, Crown Land Opportunities and
Restoration Branch, Crown Contaminated Sites Program 2014 biennial report, Victoria, Canada.
McConkey, T., Stewart, G.G, Power, B.A., Gillett, B.A. and Runnells, J.L. (2015) Bralorne-Takla Mercury Mine: A case study for
collaborative remedial planning, submitted to Mine Closure 2015, A.B. Fourie, M. Tibbett, L. Sawatsky and D. van Zyl (eds),
1–3 June 2015, Vancouver, Canada, InfoMine, Vancouver.

Office of the Auditor General of British Columbia (2002) Managing contaminated sites on provincial lands (Report; 2002/2003: 5),
Victoria, Canada.
Power, B.A. and Baker, R.F. (2015) Using environmental risk assessment to meet changing expectations for mine closure, in
Proceedings Mine Closure 2015, A.B. Fourie, M. Tibbett, L. Sawatsky and D. van Zyl (eds), 1–3 June 2015, Vancouver, Canada,
InfoMine, Vancouver.
Power, B.A., Tinholt, M.J., Hill, R.A., Fikart, A., Wilson, R.M., Stewart, G.G., Sinnett, G.D. and Runnells, J.L. (2010) A risk-ranking
methodology for prioritizing historic potentially contaminated mine sites in British Columbia, Integrated Environmental
Assessment and Management, Vol. 6(1), pp. 1–10.

J. Noël Mine Reclamation, Agnico Eagle Mines Ltd., Canada

J. Cayouette Mine Reclamation, Agnico Eagle Mines Ltd., Canada
J.-F. Doyon Sustainable Development, Agnico Eagle Mines Ltd., Canada
M. R. Julien Environment, Agnico Eagle Mines Ltd., Canada
The Cobalt Mining Camp, in northeastern Ontario, Canada, was discovered in the early 1900s and has been
the host of more than a hundred silver mines and some 30 mills over the course of the last century. Mine sites
have changed ownership over the years and have been progressively closed to the standards of the time.
Agnico Eagle Mines Ltd. acquired properties in the Cobalt area starting in 1953, operated some mines from
1953 to 1989 and still holds the rights of a portion of the former mining sites in the area. A first series of
closure plans was submitted to the Ontario government in the 1990s and completed in the early 2000s. Agnico
Eagle is currently in the process of updating its closure plans. This process is a challenging one, since the plans
amendments include more than 230 properties for which more than 500 mine hazards (old shafts,
headframes, raises, trenches, waste rock piles, tailings accumulation areas, etc.) have been identified.
Moreover, a significant number of properties present complex ownership rights and are located near
inhabited areas. Many of them also present features of historical significance (the Town of Cobalt is
considered Ontario’s most historic town). The updated rehabilitation plans have to integrate all of these
features and take into account the dynamic reaction of the natural environment to this mining legacy.
The process to date has highlighted the importance of clearly defining the restoration objectives and of having
a detailed process to determine the appropriate rehabilitation measures, taking into account site specificity
and the unique nature of the Cobalt area. The selected approach is one based on risks assessment and
innovation. This exercise also proves to be a great opportunity to build further on the relationship with the
Cobalt community, proud pioneers of the Canadian mining industry.
Mining has been a driving force in the Ontario, Canada, northern economy since the beginning of the
twentieth century. Important mineral discoveries (silver, gold, copper, nickel) were made during the half-
century from 1890 to 1940 (Ontario Mining Association, 2010): Cobalt (1903), Porcupine Lake and Timmins
(1909), Kirkland Lake (1911) and, later, Red Lake (1926) and Pickle Lake (1929). These discoveries allowed
Canada to reach the first ranks of the world’s producers of gold and silver as some of the precious metal
deposits discovered in Ontario proved to be among the richest in the world (Baldwin, 1977).
During the same period, the Province of Ontario put in place a series of legislative acts as a framework for
exploration and mining activities. The Province of Ontario first established Mining Act in 1869. It underwent
a major revision process in 1906; this 1906 amended act governed prospecting and mining in Ontario during
most of the twentieth century (Government of Ontario, 2008). The act was modernised in 2009 (Government
of Ontario, 2009).

How to reconcile more than a century of mining history with today’s rehabilitation standards: a case study in Cobalt, Canada J. Noël et al.
Cobalt is located about 450 km north of Toronto near the Ontario-Quebec border (Figure 1). Prior to the
discovery of silver at Cobalt, metal mining played only a small part in the nation’s economy (Baldwin, 1977).
Silver was first discovered along the shores of Cobalt Lake in 1903 by workers of the Temiskaming and
Northern Ontario Railway, who were building the stretch from North Bay to New Liskeard. This triggered an
unprecedented period of profitability and growth in Canadian prospecting and mining, called by many the
“cradle of Canada’s mining industry.” In just a few years, Cobalt became one of the world’s most productive
sources of silver ore. In the years that followed, more than 100 mines came into production in the vicinity of
Cobalt. By 1905, there were 16 mines in the area, and in 1911, production exceeded 30,000,000 ounces
(Ontario Mining Association, 2010). Only a small number of these mines were long lived, but their joint
operations raised Canada to fourth place in world’s production of silver (Baldwin, 1977). Minerals mined at
Cobalt between 1903 and 1926 were worth almost a third more than all the gold extracted from the Klondike
(Baldwin and Duke, 2005). The discoveries at Cobalt led the way to further exploration, mining and
settlement in northern Ontario and Quebec. Cobalt was a training ground, the birthplace of hard rock mining
in Canada (Ontario Mining Association, 2010).
The ore of the Cobalt camp occurred predominantly in veins that include native silver, cobalt and nickel
arsenides, sulphides, sulpharsenides, sulpharsenites, antimonides, sulphantimonides, sulph-bismuthinites
and their secondary alteration products in a carbonate gangue (Reid et al., 1923). In general, 97% of the silver
occurred in native form, 85% to 96% pure with the most significant impurities being antimony, arsenic and
mercury (Anderson, 1993).
In response to this concentrated nature of the silver veins, Cobalt camp old mine workings are mostly narrow
and linear. The nearly vertical veins were at first mined from narrow cuts worked downwards, but as the
silver veins were followed to greater depths, underground mining methods (shafts, drifts and stopes) were
adopted.

Regulatory
Ore was at first hand sorted; the first concentrator mill was built in 1907 to recover silver, cobalt and copper
through gravity concentration processes, followed shortly after by cyanidation (Anderson, 1993). Low-grade
ore recovery rates spread from 80 to 95% (Anderson, 1993). High-grade mills were also established from
1911 to refine the concentrated ore using tube-mill amalgamation process followed by cyanidation and to
produce silver bullion locally (Anderson, 1993). A total of 28 mills and six refineries operated in the Cobalt
area from 1907 to 1989 (Anderson, 1993).
Cobalt experienced three main mining eras: a first mining boom from 1903 to 1935, a slow period from 1935
to 1950 and a second mining boom from 1950 to the end of the 1980s. When the price of silver fell, most of
the mines were shut down. A total of about 445 million ounces of silver were extracted from the Cobalt camp
(Marshall and Watkinson, 2000).
Cobalt today is a residential community with a population of 1,133 people (Canada 2011 census). At its peak,
more than 15,000 people lived in Cobalt (Barnes, 1986). The silver mines of Cobalt, and the prospectors and
miners that discovered them and worked the mines, have left an indelible mark on Canadian history. The
legacy of the mining past is still present.
In 1953, a handful of struggling mining companies in the area of Cobalt joined together to become the Cobalt
Consolidated Mining Company. Cobalt Consolidated was reorganised in 1957, and the name changed to
Agnico Mines Ltd. (Agnico standing for silver [Ag], nickel [Ni] and Cobalt [Co]]. Agnico Mines operated
23 mines in the area between 1953 and 1988 (Table 1), for a total ore production of about 1.7 million tonnes.
The Christopher/Cobalt Lode, Augnico, Beaver-Temiskaming, O’Brien and Coniagas mines were the most
productive ones, together accounting for three-quarters of the total production.
Ore was processed in four mills (Table 2). The Colonial Mill was used in the early years to process silver ore,
while the Mile 104 Mill was used to process cobalt ore. They both burned in the 1950s: the Colonial in 1954,
the Mile 104 in 1956. Milling was then directed to the Penn Foster Mill. More than 80% of the silver ore
extracted by Agnico Mines was processed at that facility between 1957 and 1989. This mill was sold in 1998
and is still operating. A fourth mill, the Hellens Mill, was mainly used to re-process tailings from Cobalt Lake
from 1966 to 1969. It was demolished in 1984–1985 (Anderson, 1993). In total, Agnico Mines extracted about
25 million ounces of silver from its Cobalt operation, about 5% of the total production of Cobalt’s camp.
Three areas were used for tailings discharge (Glen Lake, Cobalt Lake and Crosswise Lake). More than 70% of
the tailings produced by Agnico Mines, coming from Penn Foster Mill, were directed to the Glen Lake area.

Total production
Mine From To
(tonnes of ore hoisted)
Canadian Lorrain (Gilgreer) 1953 1953 2,174
Colonial 1953 1954 4,222
Crown Reserve and Kerr Lake 1953 1955 24,194
Right-of-Way 1953 1958 6,110
Cobalt Lode 1953 1966 73,535*
Augnico 1953 1960 266,080
Christopher 1954 1966 278,266
Foster 1955 1960 18,605
Beaver-Temiskaming 1957 1958 5,580
1977 1988 252,119
O’Brien 1958 1966 210,050
Brady Lake 1961 1963 1,797
Nipissing 407 1962 1971 132,520
Violet 1963 1964 1,772
Cart Lake (shaft 2) 1964 1967 6,711
Penn Canadian 1969 1970 17,705
Nipissing 96 1969 1972 41,887
Glen Lake 1970 1972 8,040
Trout Lake 1970 1975 45,837
Frontier Mine 1972 1973 751
Coniagas 1975 1982 159,395
Castle 1979 1988 119,911
Langis 1983 1988 22,716
Total 1953 1988 1,699,977
Cobalt Lake tailings re-process 1966 1971 279,153
*From 1962 to 1966, Cobalt Lode tonnage was included with Christopher Mine

Regulatory
Total production
Mill From To Tailings directed to
(tonnes of ore milled)
Colonial (silver ore) 1953 1954 50,816 Crosswise Lake
Mile 104 (cobalt ore) 1953 1956 258,250 Crosswise Lake
Hellens (tailings mill) 1953 1953 78,128 Cobalt Lake
*
1966 1969 275,102 Cobalt Lake
Penn Foster 1957 1989 1,599,190 Glen Lake
Total 1953 1989 2,261,486
*From 1966-1969, tailings from Cobalt Lake were re-processed at Hellens Mill (renamed “Tailings Mill”) and re-deposited in another
section of Cobalt Lake
Agnico Mines Ltd. merged in 1972 with Eagle Mines Ltd., operating gold mines in Joutel, Quebec, Canada, to
form Agnico Eagle Mines Ltd. (Agnico Eagle).
When Agnico Eagle ceased its operation in the Cobalt area in 1989, demolition work was done, followed by
rehabilitation work (mainly installation of concrete caps, backfilling and fencing) and then monitoring. These
works, done mainly between 1992 and 2005, were detailed in a series of closure plans submitted to the
Ontario Ministry of Mines and Northern Development (MNDM) between 1993 and 1998. These were filed
officially with the MNDM in 2000.
In 2012, the MNDM asked for an update and amendment of previous closure plans, which Agnico Eagle is
currently completing. More than 230 properties are to be covered by these updates, on which more than
500 mine hazards have been inventoried. Most mine hazards (about 65%) consist of former shafts, raises and
stopes to surface; 25% are trenches, pits and open cuts; and the remaining (10%) consist of lateral openings
like adits. Updating the information and prioritising further rehabilitation work to be done over the following
years has revealed itself to be quite a challenging exercise.
The evolution of the expectations and requirements in terms of defining the natural environment is one of
the main triggers of the work currently underway for the update of the 1990s closure plans. Requirements
with reference to sampling and measuring the environmental baseline conditions evolve quickly.
Elevated arsenic concentrations have been documented in surface waters in the area of Cobalt since the
1960s (Percival et al., 1996). Several environmental studies (Table 3) were further conducted in the 1990s
and 2000s.

Study Authors Study dates

Water quality and aquatic studies Dumaresq (1993) 1991–1992
Percival et al. (1996) 1994–1995
Beak (2002) 2001–2002
MOE (2009) 2006–2007
Watershed study MNDM (2010) 2008–2009
Private well study Harden Environmental (2007) 2006
Soil assessment MOE (2011) 2003–2008
Screening level health risk MOE (2005) 2003–2004
assessment
(self-guided mining tour)
Note: List is not intended to be exhaustive.
Arsenic is the major element of concern in the Cobalt-Coleman mining camp. Although concentrations of
other elements, such as antimony, cobalt, copper, lead, mercury, nickel, silver and zinc are also high, in most
cases, arsenic concentrations outweigh the possible effects of these other elements (MOE, 2011). Arsenic
occurs naturally in the rock, and part of the elevated concentrations is a result of the natural geology.
Moreover, some local groundwater shows high background arsenic concentrations that could naturally
contribute to surface water observed levels (Dumaresq, 1993; Percival et al., 1996). These high-level
groundwater concentrations were associated with mineralised zone, as the private drinking water study
found 98.5% of the samples below the Ontario drinking water standards (Government of Ontario, 2006).
Arsenic is also released from oxidation of exposed tailings and waste rock primary minerals, and from
dissolution of secondary minerals (Dumaresq, 1993). However, all tailings and waste rock are not the same,
which explains the high observed variability in the results. This variability calls for a thorough evaluation of
the materials present on the old mine sites to try to identify the major sources.
Water sampling programs conducted over the years have shown that arsenic concentrations are quite
variable in space, generally decrease downstream and appear to decrease with time (Percival et al., 1996).
Water reservoirs seem to play a key role in the attenuation of arsenic. Mean arsenic concentrations in surface
water–exposed areas is about 600 µg/L (MNDM, 2010); the provincial water quality objective for arsenic in
Ontario is 100 µg/L (MEE, 1994). The main arsenic species found in surface water is As (V), which represents
up to 99% of total arsenic, and is the less toxic form of arsenic (Percival et al., 1996). Arsenic is transported
in a weakly alkaline environment as the result of the carbonate weathering in the area. This probably explains
why, despite the numerous studies having shown potential impact to aquatic environments, this does not
appear in the field. Recreational use (i.e. swimming and fishing) is nevertheless suggested to be restricted by
the Health Unit, and signs have been posted to warn of the potential risks to human health (MOE, 2009).
The arsenic concentrations in the Cobalt area thus represent a complex issue, since the sources, on
watershed scale, are multiple, sometimes have migrated in space (and can still migrate when unconfined),
are located on different properties and originate from different substrates or mediums. The previous studies
have shown that it is important to characterise the materials we are dealing with thoroughly, and to have
good sets of water quality data to observe the long-term trends. This work is undergoing. The next step, once
data are available, will be to interpret them, and then to find innovative solutions in the context of a self-
sustaining environment.

Regulatory
The evolution of the practice and knowledge in this area since the release of the first closure plans, and
emerging new technologies (geo-referenced systems, metal leaching and acid-rock drainage testing, etc.)
also have to be integrated in the closure plan update. Moreover, new issues have emerged since the 1990s
expectations, and are now to be considered with reference to mine reclamation (e.g. groundwater, soils, and
human health risks).
An example of evolving considerations since the first closure plan submittal is the growing issue with bats.
When confronted with old lateral openings like adits and ramps, it is now required that their suitability and
use as bat habitat be documented. Some bat species are threatened by “white-nose syndrome,” a fungal
growth disease leading to individual death. Since January 2013, three species of bats have been added to the
Ontario species at risk list: the northern long-eared bat (Myotis septentrionalis), the eastern small-footed
myotis (Myotis leibii), and the little brown bat (Myotis lucifugus). Environment Canada has also recently
added the tri-coloured bat (Perimyotis subflavus) to its list of species at risk, which also includes the northern
long-eared bat and the little brown bat. Guidelines are in development with the governmental agencies to
try to reconcile the species habitat protection requirements with the public safety requirements.
One of the first steps that needed to be completed before updating the field information about the properties
was to conduct a land title search. The properties owned by Agnico Eagle in the Cobalt-Coleman area are
numerous (as many as 350 at one time), and many had undergone several owner exchanges over the years.
Land titles needed to be confirmed to clearly define the scope of work for the update of the closure plans.
The research took several months of work and required the review of more than 700 documents (patents,
parcel sheets, leases, transfer documents, reports, and maps). Investigation was also conducted through
discussions with former employees, consultants and government employees that have been active in the file
for a long period of time. The result: Agnico Eagle now owns 233 land titles in the Cobalt-Coleman area
(Figure 2).
Mapping information was georeferenced to build a geographic information system (GIS) to manage all the
mapping information. This led to the questioning of some boundaries (mapping information does not always
perfectly match between different sources). Discrepancies were highlighted and discussed with the
authorities. Specific surveys were also undertaken to confirm locations of some infrastructure (e.g. tailings
areas) with reference to property limits.
The history of the different properties is, in many cases, quite complex. The ownership of surface and mining
rights may vary from one parcel of land to another. Prior to 6 May 1913, any parcel of land granted as a
patent by the Crown included both surface and mineral rights. Land patents granted after this date may or
may not have included the surface rights. Some mineral rights are also leased by the authorities, with
expiration dates. The reclamation work planning is more complex for properties where private owners hold
the surface rights. Moreover, even where private owners do not own surface rights directly, we need to
carefully consider neighbours, as several of the former mine sites are located in the proximity of inhabited
areas.
As mentioned above, land title search can be an extremely complex exercise of reconciliation involving old
information in geodetic systems no longer used, or related to points on the ground that may have been lost
over the years, or simply resulting from human error during the transcription process. When crosschecking
the location information provided in different documents (property index block maps, mining land tenure
maps and property ownership documents such as patents and leases), some discrepancies were
encountered: for example, the boundaries were not the same, or portions of ownership were not clear or
characterised by overlap.
In addition to the discrepancies associated with mapping, another complexity arises with some property
limits shared by different owners. In some cases, it was found that different owners share some
infrastructure, making their reclamation even more complex. An example of this is the Beaver-Temiskaming

tailings (Figure 3) that show a complex ownership situation. For cases like this one, the planning of the
reclamation works necessitates discussions and agreements with the neighbours if we want to increase the
chances of success.
In 1991, the MNDM established the abandoned mines program. One of the triggers of the creation of this
program was the 1987 collapse of Highway 11B in Cobalt. An inventory of abandoned and inactive mine sites
and features was conducted over the following years, and an exhaustive database was created (the
abandoned mines information system; AMIS). The collapse of the highway also triggered a series of
geotechnical investigations done at the request of the town of Cobalt, which was concerned with public
safety issues, especially at old underground mine workings (e.g. Golder Associates 1987a,b,c; 1988a,b,c,d;
1989; 1990a,b; 1991; 1992; 1993).
For Agnico Eagle, mine hazards represent more than 500 former openings (shafts, raises, adits, open cuts,
etc.) that were backfilled, capped, fenced or, sometimes, left open. With the AMIS database as a starting
point, Agnico Eagle conducted an investigation of its properties in 1999 (Golder Associates, 1999), updated
in 2005 (Golder Associates, 2005). The work was expanded in 2013 and 2014, with a thorough investigation
and data search. All the data are gradually being merged into the GIS database, which will produce a solid
mapping reference.
One finding of the mine hazards investigation was that some of the former mine workings may not appear
on old drawings because they were exploited by leasers and not mapped. This adds to the challenge of
accurately evaluating crown pillars. Some hazards are hard to locate in the field, and the assessment activity
itself can be a real hazard: some veins mined to surface were “reclaimed” using layered timber and waste
rock, which can eventually create new hazards as timber rots.
Another aspect related to mine hazards that represents a challenge is that because of the numerous historical
mining activities that took place in the Cobalt-Coleman area, there is a huge amount of data in the AMIS
database for the sector. A significant effort was made to verify these data along with the historical archives.
This verification was made to remove some inaccuracies introduced over the years. For example, the hazard
locations were not always accurate because of the change in the coordinates system used (from NAD27 to
NAD83). Further, some mine hazards were entered twice but with different names. The review and re-
assessment of these different mine hazards has been a very complex task that required much effort both at
the office and in the field.

Regulatory


Regulatory
Cobalt is considered the silver capital of Canada. In 2001, TV Ontario named Cobalt “Ontario’s most historic
town.” The following year, Parks Canada declared the Cobalt Mining District a National Historic Site. The
designation includes buildings and remnants of past mining activities, as well as the unique physical fabric of
the townsite. In 2003, the Royal Canadian Mint even struck a special commemorative silver dollar to mark
the 100th anniversary of the discovery of silver in Cobalt.
The Cobalt Historical Society, with support from authorities and other contributors, established the Silver
Heritage Trail. This is a self-guided drive along the roads of Cobalt and through several former mine and mill
sites in the area. Sites were chosen to illustrate the features associated with mining in the early 1900s and to
make known to the public how the discovery of silver and development of Cobalt has contributed to Canada’s
mining history. The trail ends at the Northern Ontario Cobalt Mining Museum. This initiative has proven
successful from a heritage and communication point of view. However, the area covered by the trail and the
number of features to be preserved and maintained is quite large, and some buildings have started to show
signs of advanced degradation. These will need to be torn down for safety reasons, as the investment needed
to restore and maintain them is too significant. However, mine infrastructure is part of the history of the area
and represents a powerful component of people’s identity in mining towns as well as a memorial of a proud
past (Dumaresq, 2007). Thus, the community expresses concerns regarding the potential loss of built
heritage. This sensitive issue needs to be discussed in detail prior to taking any action regarding build
infrastructure.
This discussion leads to another growing reality with reference to reclamation planning: the need to consult
with stakeholders and local community, and to get social acceptance. People want to be informed of what is
going on in their community and want to be involved in processes, especially reclamation, aimed at restoring
the landscape, but these may generate some disturbance. The need to consult and get social acceptance also
works the other way around. For example, if local authorities are planning surface works, they will benefit
from consulting with the mining rights holders to make sure that mine hazards are considered in the planning
to minimise potential risk to people’s safety. The key is to talk and work together. For this reason, a multi-
stakeholder advisory committee was created for the Cobalt-Coleman area. This committee has been found
to be an efficient mechanism through which to discuss and address together some of the environmental and
social challenges associated with reclamation of the former mine sites.
In Cobalt, Agnico Eagle has to work with an already disturbed landscape that has naturally started to restore
itself, and with a multitude of former sites over a relatively large area. Mining, closure and reclamation
practices from the early days were obviously different from today’s standards. Mine waste was deposited
where it was most expedient to do so, without any knowledge of the potential contaminants and often
without any containment. At the end of operations, what did not have a salvage value was simply left in
place. There was little concern with mine openings. Practices of modern mine reclamation only started to
emerge in the 1970s. In Ontario, the 1991 changes to the Mining Act introduced the requirement to prepare
a closure plan and to post financial assurance. The Rehabilitation Code of 2000 came with the adoption of
Regulation 240/00 Mine development and closure under part VII of the Mining Act (Reg 240/00).
The challenge now is reconciling a century of mining activities, done to the standards of the time, with new
requirements and expectations concerning mine closure, health and safety, the environment and land use.
Current Ontario mine reclamation regulation was designed for single mine operations or new projects where
the baseline can be determined. Such an approach is difficult to apply to Cobalt’s legacy sites. When
contemplating rehabilitation work, there is a need to clearly state and reach agreement on targets and
objectives while being creative, innovative and realistic. Agnico Eagle in Cobalt is focussed on:
 measures to achieve protection of mine openings to surface;

 analysis of crown pillar stability and remedial measures to protect public safety as required;
 analysis of long-term physical stability of tailings and remedial measures as required;
 characterisation of waste (tailings, waste rock) for metal leaching and ARD;
 assessment of the existing surface water quality and aquatic ecosystems;
 assessment of the existing groundwater quality and beneficial use;
 monitoring surface water, groundwater, physical stability (crown pillar, surface openings, tailings,
rock pile);
 measures to establish sustainable vegetation.
The Mining Act defines the action of rehabilitation as:

measures, including protective measures, taken in accordance with the prescribed standards to
treat a site or mine hazard so that the use or condition of the site (a) is restored to its former use
or condition or (b) is made suitable for a use that the Director sees fit. (Government of Ontario,
2010)
For Cobalt, former use is hard to establish considering the heavy historical legacy of the area; instead, an
acceptable use and/or condition must be defined while respecting the intent of the Rehabilitation Code. The
approach selected to support the reclamation strategy for Agnico Eagle legacy sites in the Cobalt area is a
risk-based approach. All risks will be considered and weighted: health and safety, environmental, community
and social, final land use, legal and financial, and technical. Innovation will also be sought in developing
optimum solutions.
Taking a risk-based approach to legacy sites impacts assessment and reclamation strategy development, and
prioritisation allows for the following advantages (Laurence, 2001):
 Remediation issues are addressed in a structured and systematic manner.
 Decision-making processes are supported to prioritise the reclamation work to be done.
 Major issues and risks are highlighted, which allows management to concentrate resources
appropriately.
 Comparison between sites and features is facilitated.
The process in which Agnico Eagle is engaged for updating closure plans will see this risk-based approach
developed and completed.
Cobalt was rich in silver; its discovery allowed for the development of a region and the growth of the Ontario
and Canadian economies and expertise in hard rock exploration and mining. For this reason, Cobalt is now
rich with legacy. Its unique history poses a real challenge when planning to improve restoration of former
mine sites, given the physical hazards. However, it can also be seen as a great opportunity to think outside
of the box when developing and testing innovative restoration solutions targeting the essence of today’s
expectations. Just as silver was discovered while building a railway, the future will tell what will emerge from
rehabilitating legacies.
Anderson, P. (1993) Cobalt mining camp tailings inventory, Cobalt, Ontario, 122 pp., viewed 14 May 2015, http://www.Cobaltmining
legacy.ca/studies/Anderson_1993_Tailings_Inventory.pdf.
Baldwin, D. (1977) A study in social control: the life of the silver miner in Northern Ontario, Labour, Vol. 2, pp. 79–106.

Regulatory
Baldwin, D.O. and Duke, D.F. (2005) “A grey wee town”: an environmental history of early silver mining at Cobalt, Ontario, Urban
History Review, Vol. 34(1), pp. 71–87.
Barnes, M. (1986) Fortunes in the ground: cobalt, porcupine and Kirkland Lake. Boston Mills Press, Erin, Ontario, 263 pp.
Dumaresq, C.G. (1993) The occurrence of arsenic and heavy metal contamination from natural and anthropogenic sources in the
Cobalt area of Ontario, MSc Dissertation, Carleton University, Ottawa, Canada, 138 pp.
Dumaresq, C. (2007) Canada’s mining heritage – balancing the heritage preservation with the environment, health and safety, in
Proceedings Mining and the Environment IV Conference, 19–27 October 2007, Sudbury, Canada. Viewed 14 May 2015,
http://www.cobaltmininglegacy.ca/studies/Sudbury_2007_Dumaresq-paper.pdf.
Government of Canada (2011) 2011 Census. Statistics Canada. Viewed 14 May 2015, http://www12.statcan.gc.ca/census-
recensement/index-eng.cfm.
Government of Ontario (2006) Technical support document for Ontario Drinking Water Standards, objectives and guidelines. June
2003, revised June 2006, PIBS 4449e01. Viewed 14 May 2015, http://www.ontario.ca/document/technical-support-
document-ontario-drinking-water-standards-objectives-and-guidelines.
Government of Ontario (2008) Modernizing Ontario’s Mining Act – finding a balance, discussion paper, 20 pp. Viewed 14 May 2015,
http://www.pdac.ca/docs/default-source/public-affairs/lands-regulation---chronology---ontario-mining-act.pdf?sfvrsn=7.
Government of Ontario (2009) An Act to amend the Mining Act, Chapter 21, assented to October 28, 2009. Viewed 14 May 2015,
http://www.ontario.ca/laws/statute/s09021.
Government of Ontario (2010) Mining Act, R.S.O. 1990, Chapter M.14, last amendment 2010, c. 18, s. 23. Viewed 14 May. 2015,
http://www.ontario.ca/laws/statute/90m14.
Harden Environmental (2007) Occurrence of arsenic in private drinking water supplies, Township of Coleman, file 0615, prepared for
the Ontario Geological Survey, Sedimentary Geoscience Section, 16 pp.
Laurence, D. (2001) Classification of risk factors associated with mine closure, Mineral Resources Engineering, Vol. 10(3), pp. 315–
331.
Marshall, D. and Watkinson, D.H. (2000) The Cobalt mining district: silver sources, transport and deposition, Exploration and Mining
Geology, Vol. 9(2), pp. 81–90.
Ontario Ministry of the Environment (MOE) (2005) Technical report – screening level health risk assessment of the historical mining
tour of Cobalt, Ontario, prepared by the Human toxicology and air standards section, Standards development branch, 60 pp.
http://www.timiskaminghu.com/content.aspx?content_id=539, full text viewed 14 May 2015, http://www.cobaltmining
legacy.ca/ studies/SLHRA_Full_Report.pdf.
Ontario Ministry of the Environment (MOE) (2009) Water monitoring and reporting section, Environmental monitoring and reporting
branch, Biomonitoring assessment of Cobalt area lakes 2006–2007, 13 pp.
Ontario Ministry of the Environment (MOE) (2011) Terrestrial Assessment Unit, Air monitoring and transboundary air sciences
section, Environmental monitoring and reporting branch, Final report: Cobalt-Coleman mining camp soil assessment, 126 pp.
Ontario Ministry of Environment and Energy (MEE) (1994) Water management policies and guidelines, Provincial Water Quality
Objectives. Viewed 14 May 2015, https://dr6j45jk9xcmk.cloudfront.net/documents/3016/moeprovincialwaterquality
objectivesen.pdf.
Ontario Ministry of Northern Development and Mines (MNDM) (2010) A chemical and biological study to assess the receiving waters
in the Cobalt area, Ontario, prepared by LGL Ltd., 601 pp.
Ontario Mining Association (2010) Brief history of Ontario mining, Northern Ontario history, viewed 18 February 2015,
http://www.republicofmining.com/2010/07/23/brief-history-of-ontario-mining/, last updated 23 July 2010.
Percival, J.B., Dumaresq, C.G., Kwong, Y.T.J., Hendry, K.B. and Michel, F.A. (1996) Arsenic in surface waters, Cobalt, Ontario, Current
Research 1996-C, Geological Survey of Canada, pp. 137–146.
Reid, F.D., Denny, J.J. and Hutchinson, R.H. (1923) Milling and metallurgical practice in treatment of silver ores at Cobalt, Temiskaming
district, Ontario, Ontario Department of Mines annual report, 1922, Vol. 31(2), pp. 239–320, viewed 18 February 2015,
http://www.geologyontario.mndmf.gov.on.ca/mndmfiles/pub/data/imaging/ARV31/ARV31.pdf.
Beak International Inc. (2002) Water quality and in-stream loadings – Agnico Eagle Mines Ltd. Cobalt area properties, 122 pp.
Golder Associates (1987a) Preliminary engineering assessment and suggested remedial measures “P-S” stope intersection crown
pillar, HWY #11B, Cobalt, Ontario, 14 pp.
Golder Associates (1987b) Subsurface drilling investigation, factual results, phase I, “P-S” stope intersection and adjacent areas of
HWY #11B, south west of Galena Street, Cobalt, Ontario, 7 pp.
Golder Associates (1987c) Ground conditions, south lawn area, senior citizens home, Cobalt, Ontario, 8 pp.
Golder Associates (1988a) Subsurface investigations and preliminary engineering assessment phase II, Highway 11B corridor Galena
Street/Miller Avenue intersection area, Cobalt, Ontario, 23 pp.
Golder Associates (1988b) Subsurface investigation and preliminary engineering assessment, senior citizens’ building, Cobalt,
Ontario, 16 pp.
Golder Associates (1988c) Town of Cobalt mining hazard assessment phase IV A, Cobalt, Ontario, 20 pp.
Golder Associates (1988d) Hazard evaluation of #1 and #3 vein stopes, phases II-IV, senior citizens’ building, Cobalt, Ontario, 15 pp.
Golder Associates (1989) Mine hazard inventory for the municipality of the Town of Cobalt, Ontario, 17 pp.
Golder Associates (1990a) Precondition surveys in mine hazard areas in the municipality of the Town of Cobalt, Ontario, 11 pp.
Golder Associates (1990b) Subsurface investigation of zone #5, Cobalt, Ontario, 17 pp.

Golder Associates (1991) Subsurface investigation of zones 3, 4, 9 and 10, Cobalt, Ontario, Volume I, Main text, 27 pp.
Golder Associates (1992) Subsurface investigation of zones 7, 8, 11, 12 and 14 to 17, Cobalt, Ontario, Volume I, Background and
recommendations, 20 pp.
Golder Associates (1993) Geotechnical investigation of zones 13, 14, 17 and part of zone 2, Town of Cobalt, Ontario, 26 pp.
Golder Associates (1999) Report on near surface mine workings at Agnico Eagle’s Cobalt properties, report 991-1711, submitted to
Agnico Eagle Mines Ltd.
Golder Associates (2005) Abandoned mine shaft hazard investigation phase I, shafts listed by Abandoned Mines Information System
(AMIS) with inadequate protection, Cobalt, Ontario, submitted to Agnico Eagle Mines, 7 pp.

P.A. Steenhof CSA Group, Canada
A new international standards work program is underway through the International Organization for
Standardization (ISO) to develop a series of international standards that will provide requirements and
recommendations for mine reclamation management, with a focus on reducing the environmental risks,
associated financial liabilities and environmental impacts of mining. Nine countries are actively participating
in the standards development program, including all of the most significant mining countries in the world.
CSA Group has helped organise a committee of Canadian stakeholders and experts from the Canadian mining
industry, relevant government authorities, service providers, and academic researchers with the mandate to
provide Canadian leadership and input into this international standards program.
This article explains the international standards being developed for mine reclamation management with a
focus on an international work item and international standard that Canada has proposed on the topic of
mine reclamation management planning. The standard is intended as a crucial overarching document that
will provide requirements or recommendations on the topic of mine reclamation management planning. The
document will also help guide the development of related standards for the mine reclamation management
process.
The work item follows a lifecycle approach to mine reclamation management planning and recognises that
planning for mine reclamation must take place throughout the lifecycle of the mine and not just at closure.
This paper covers the main decision points and procedures that need to be considered in planning for mine
reclamation, organised around the following topic areas: approaches to reclamation, reclamation
technologies, reclamation activities, reclamation monitoring, financial provisions, and documentation and
reporting. The objective is that the ensuing standard will help ensure consistency and completeness in terms
of the mine reclamation planning process.
In 2013, a proposal was brought forward to establish an international standards development program
focused on mine reclamation. This was brought forward through an International Organization for
Standardization (ISO) technical committee focused on mining (ISO TC 82); the committee decided to form a
subcommittee (SC) focused specifically on mine reclamation management. Nine countries are actively
participating in the standards development program, including all of the most significant mining countries in
the world.
The purpose of this article is to provide an overview of this new international standards development
program, with a specific focus on a new work item being proposed on mine reclamation management
planning. This new work item is intended to provide a framework not only for mine reclamation management
and any standards developed for this, but also for the planning of mine reclamation across the lifecycle of a
mine, from inception to closure.
This article is organised as follows: To explain the ISO context, the article first reviews the general process for
developing international standards through ISO, including how the widest breadth of stakeholders is
consulted and how consensus is sought. Section 3 describes the need for standards focused specifically on
mine reclamation, and Section 4 provides an overview of the mine reclamation standards development

Development of international standards for mine reclamation management P.A. Steenhof
program, including Canada’s contribution to this. Section 5 focuses on the mine reclamation management
planning standard, including its need and purpose, scope, and international leadership. Section 6 provides a
summary.
The development of international standards through ISO follows a systematic process based on distinct
stages of development. There are six separate stages, which involve different activities at both the working
group (WG) level and the level of the committee to which the working group reports. These six stages are as
follows:
 Proposal
 Preparation
 Committee
 Enquiry
 Approval
 Publication
Several documents are generated through this process:
 Proposal (NP)
 Working draft (WD)
 Committee draft (CD)
 Draft international standard (DIS)
 Final international standard (FDIS)
 ISO standard (ISO)
This process is illustrated in Figure 1 and discussed in more detail in the paragraphs that follow.
The first stage is initiated by a proposal for a new work item (NP), which involves either a new piece of work
based on as little as a table of contents or the adoption of an existing document that can be used to seed the
international standard. New international standards are proposed through a specific parent committee with
a scope that includes the topic area of the standard. These parent technical committees are made up of
members from individual countries that can either have voting privileges (termed participating or P-
members) or act as observing countries to the process (O-members). Countries who have voting privileges
on a committee can propose new work items; in doing so, they are also responsible for appointing an
international convenor who will act to lead the working group responsible for developing the technical
content of the standard. New work items are voted upon in terms of their market demand and the relevance
of the standard on a global scale.

Regulatory
Another important element of this phase is the creation of a working group (WG), which will be composed
of international experts and will develop and vet the actual content of the standard. Depending on the size
of the parent committee, the WG will be made up of experts from a minimum of four or five different
countries.
The preparatory stage involves the preparation of a working draft (WD) of the international standard. This
work is done largely within the relevant ISO working group.
After the WG is satisfied on the technical merits of the working draft, this document is put forward to the
working group’s parent committee for comment and vote. There are three voting stages, and this is the first
one. This stage, which is termed the committee stage of the development process, is where each P-member
country registers its official comments and votes on the WD: a yes vote means that the working draft meets
the requirements to move on to the next phase. Successive committee drafts may be considered until
consensus, which is reached when at least two-thirds of the P-members register a yes vote. Once agreed
upon, the text is finalised for submission as a draft international standard (DIS).
The fourth stage, or the enquiry stage, involves the DIS being circulated by ISO to all of the member countries
of ISO for voting and comment over a three-month period. This is the second voting stage. The member
countries, in turn, are encouraged to make the DIS text available to the widest range of national stakeholders
possible in order to obtain the national vote of each member body. The text is approved if two-thirds of the
P-members of the committee under which the work item falls vote in favour and if not more than one-quarter
of the total votes cast are negative. The fifth stage of development is the approval stage. This involves the
final draft international standard being circulated to all ISO member bodies for a final two-month yes or no
vote. The text is approved for the final stage, the publication stage, using the same criteria as for the enquiry
stage. The final document is published by ISO as an official ISO international standard.
The ISO standards development process has been structured in order to allow for, and ultimately encourage,
the widest possible participation and consultation of experts and stakeholders involved with the subject at

hand. From the starting point, an international working group is comprised of individuals with expertise on
the subject, who each represent their own expertise. This is meant to encourage unfettered input of technical
content as a starting point for the standard.
The successive review and commenting stages are also structured to encourage a progressively widening
consultation process. The broadest of these stages is at the enquiry stage, when the ISO document under
development is made available to all member bodies of ISO. ISO is made up of 119 member countries that
have the ability to vote on ISO standards. This means that 119 countries will ultimately be given access to the
standard and encouraged to seek the input of stakeholders in these countries to inform their national
position. The map in Figure 2 shows ISO member groups in green and illustrates the wide scope of countries
involved.
Another mechanism that further encourages consultation during the development of ISO standards is the
possibility for liaisons between committees and between organisations and the committee at hand. Such
liaisons provide a number of privileges and opportunities to those in liaison, including the ability to observe,
to contribute technical content during meetings, and to make comments on drafts. Liaisons can also be
established at the working group level; these involve organisations that make a technical contribution to and
participate actively in the work of a working group.
International standards developed through ISO are developed following a consensus-based approach,
whereby consensus amongst member countries is sought at multiple stages. This process of consensus
building helps ensure unbiased content and technical integrity of the standard being developed.
Consensus building starts at the very beginning of the ISO standards development process, where new work
items being proposed will only pass if a majority of the voting countries of the particular committee approves
the work item. At the working group level, consensus building is focused on developing the technical content
of the work item and does not involve voting; during the committee stages of review, consensus is sought
through reviewing, commenting, and voting on the document. This occurs at up to three distinct stages,
namely, at the committee, enquiry, and approval stages, as discussed above.

Regulatory
ISO technical Scope of committee

committee
ISO/TC 82 (Mining) Standardisation of specifications relating to specialised mining machinery and
equipment used in open-pit mines (e.g., conveyors, high-wall equipment, rock drill
rigs, and continuous surface mining equipment) and all underground mining
machinery and equipment for the extraction of solid mineral substances, but
excluding the preparation and processing of the minerals; recommended practice in
the presentation of plans and drawings used in mine surveying; methods of
calculation of mineral reserves; mine reclamation management; and design of
structures for the mining industry
ISO/TC 127 (Earth- Standardisation of nomenclature, equipment use classification, ratings, technical
moving machinery) requirements, and test methods; safety requirements, operation, and maintenance
manual format for earth-moving and related machinery
ISO/TC 147 (Water Standardisation in the field of water quality, including definition of terms, sampling
quality) of waters, measurement, and reporting of water characteristics
ISO/TC 207 Standardisation in the field of environmental management systems and tools in
(Environmental support of sustainable development
management)
ISO/TC 282 (Water Standardisation of water re-use of any kind and for any purpose: this covers
re-use) centralised and decentralised or on-site water re-uses, direct and indirect water
uses, intentional and unintentional water uses, and technical, economic,
environmental, and societal aspects of water re-use. Water re-use comprises a
sequence of the stages and operations involved in uptake, conveyance, processing,
storage, distribution, consumption, drainage, and other handling of wastewater,
including water re-use in repeated, cascaded and recycled ways.
ISO/TC 262 (Risk Standardisation in the field of risk management
management)
ISO/PC 283 Development of a standard on occupational health and safety management
(Occupational systems: "Requirements with guidance for use"
health and safety
management
systems)
There are a number of broad uses of standards in the mining sector. Beyond the general objectives of helping
to mainstream best practices that help improve the safety and performance of operations, promote the safe
use of equipment, or improve the sustainability of mining itself, standards can be referenced in regulations
or can help mining companies demonstrate that they have met certain performance and operational criteria.
Although regulations usually stipulate what is required from a mine, they do not necessarily provide guidance
on how these requirements can be achieved. The development and implementation of international
standards therefore can assist mines to meet the requirements of such regulations as they provide key
guidance on the actions that should be undertaken. Regulations may also directly reference standards in
terms to help establish performance requirements (i.e., regulation by reference).
Another key use of international standards in the mining sector and by industry generally, is that these can
allow companies and practitioners to demonstrate conformance to the requirements set forth in
international standards. Prime examples include the use of the ISO 9000 family of standards, which sets forth

requirements for various aspects of quality management. This in turn can benefit companies through
marketing and improved public credibility.
There are two ISO technical committees (TC) focused specifically on mining — ISO/TC 82 (Mining) and ISO/TC
127 (Earth-moving machinery) — while there are many others developing complementary standards often
used within the mining sector. This latter group includes committees focused on topics such as water quality
(ISO/TC 147), environmental management (ISO/TC 207), and water re-use (ISO/TC 282); other TCs are
involved with developing standards for mining-related operations such as risk management (ISO/TC 262) or
occupational health and safety (ISO/PC 283). The scopes of these various committees are provided in Table
1.
Mining is a sector that has long faced heightened public scrutiny and attention from both local communities
and other industry observers and stakeholders in terms of its overall lifecycle environmental impact (Jenkins
and Yakovleva, 2006). This is especially true when considering the potential impacts of mining on immediate
and surrounding environments. Due to the requirements for extraction; the use of heavy equipment,
chemicals and other inputs during production and processing; and the large amount of disturbance to land,
mining can and often is viewed to have a potentially large environmental impact that can extend far beyond
the immediate surface disturbance and involve significant financial liability for the host country.
Over the last several decades, the topic of mine reclamation has therefore become increasingly intertwined
with the sustainability of mining itself. Here, it is important to emphasise that mine reclamation involves
planning and actions to reclaim the mining site and reduce its environmental impacts across the entire
lifecycle of the mine, from mine inception through its operational phase, as well as at mine closure and
decommissioning. As an activity, mine reclamation has emerged as a critical component of a project not only
for minimising the impacts of mining on the environment and protecting the environmental services local
communities often depend upon, but also for helping mining companies secure the social license they need
to operate (Slocombe, 2012).
Many different jurisdictions have now also made mine reclamation planning a key component in the approval
of new mines or the expansion of existing mines (Otto, 2010). This is especially the case across many of the
more developed economies of the world, although in lesser developed regions there are often lower
environmental requirements and safeguards in place (Cao, 2007; Dearden, 1998). Nonetheless, with the
advance of social networking, the rise of green consumerism, and ever increasing attention being placed on
the mining sector, companies operating throughout the world face a heightened impetus to manage the
environmental impacts of their operations even if they are not facing regulatory requirements (Jenkins and
Yakovleva, 2006).
There are significant jurisdictional differences in terms of both the requirements for mine reclamation and
what is included and practiced, and there is a need for consistency regarding how mine reclamation is
planned and carried out. This is especially the case given that the mining sector is global in nature and many
of the players involved with mineral and metal extraction and processing are vast corporate structures with
operations in many different resource extraction regions around the world. This is the impetus behind
developing mine reclamation standards.
There are a wide range of existing documents relevant to mine reclamation that will likely inform or, in some
cases, form the basis for the international mine reclamation standards being developed. While some of these
are international in scope, many are jurisdictionally specific, reflecting both the geographical realities of
mining and also the different environmental and social public policy priorities of various regions.

Regulatory
The Planning for Integrated Mine Closure: Toolkit published by the International Council on Mining and
Metals (International Council on Mining and Metals, 2008) is an example of a more generic document that is
internationally applicable. The document is intended to promote a more disciplined approach to integrated
closure and reclamation planning and to increase the uniformity of good practices across the mining sector.
Thirteen tools provide practitioners with work processes, examples and contexts within which to apply the
closure and reclamation planning discipline, ranging from stakeholder engagement to biodiversity
management. Each tool provides key references for the user so that they can refer to existing best practices
and provides general guidance on each topic area.
There are also many documents that are more specific to certain jurisdictional regions. In Canada, for
example, with mining regulated provincially or territorially, many Canadian provinces and territories have
developed guidance specific to their particular region. Examples include the Guide for Surface Coal Mine
Reclamation Plans (Nova Scotia Environment, 2009) produced by the Government of Nova Scotia. As inferred
by its title, this document is focused on surface coal mines, reflecting the historical importance of coal mining
in that region. However, the document also contains valuable and useful information for mine reclamation
more broadly, including, for example, principles of reclamation, landscape design, revegetation, etc. Another
example from a different jurisdiction is the British Columbia Reclamation and Closure Code (contained within
the Health, Safety and Reclamation Code for Mines in British Columbia) (British Columbia Ministry of Energy,
Mines and Petroleum Resources, 2008). Again, although this is specific to British Columbia, it does contain a
wide range of relevant information. Then there is the Mackenzie Valley Land and Water Board’s document
on mining and reclamation produced specifically for Canada’s Northwest Territories (Mackenzie Valley Land
and Water Board, 2013). Like the other documents highlighted, it is regional in nature but provides relevant
information that can be more widely applied.
Another important issue that affects successful mine closure is the management of mine wastes to prevent
metal leaching and acid mine drainage. There are several guidelines available, including the Global Acid Rock
Drainage (GARD) Guide (International Network for Acid Prevention, 2012), and the Guidelines for Metal
Leaching and Acid Rock Drainage at Mine Sites in British Columbia (Price and Errington, 1998).
The scope of the mine reclamation management standards development is set forth as follows:
Standardization of mine reclamation management to minimise mine hazards that occur during the
lifecycle of resource development, such as during exploration, exploitation, suspension of
operation, mine closure (restoration), and follow-up management.
However, it is expected that safety and health issues related to mining activity will be excluded
from this context.
Mine closure planning shall be re-established at every stage for sustainable resource development
and risk management. (International Organization for Standardization, 2013)
The standardisation program therefore focuses on the environmental aspects of mine reclamation while also
taking a lifecycle approach in terms of the framework being followed for the standards development
program.
There are currently nine participating (voting) member countries involved with ISO TC 82/SC 7 and six
countries that are observing the process. These are listed in Table 2.

Participating countries (9) Observing countries (6)

Australia Czech Republic
Canada Finland
Chile Russian Federation
China Sweden
France United Kingdom
Germany Zambia
Iran, Islamic Republic of
Korea, Republic of
South Africa
It should also be noted that communication with the U.S. Technical Advisory Group (the equivalent of the
Canadian Standards Mirror Committee) to ISO TC 82/SC 7 has indicated that the United States plans to submit
an application to become a participating member on the mine reclamation management subcommittee.
Potential title of Potential scope Proposing country and

preliminary work international leadership
item
Mine reclamation This work item and subsequent standard will provide Proposing country: Canada
planning requirements and recommendations for the purposes of Proposed work item leader:
mine reclamation planning. This will cover the main Professor Dirk van Zyl,
decision points and procedures that must be considered in University of British
planning for mine reclamation. The objective is to help Columbia
ensure consistency and completeness in terms of the mine
reclamation planning process.
Mine reclamation This document will help to maintain the required Proposing country: South
management − consistency of vocabularies within a group of standard Korea
terminology documents in SC 7 (mine reclamation) and other related Proposed work item leader:
technical committees. Dr. Sang-il Hwang, Korean
Agency for Technology and
Standards
Guidance on Target process: survey and analysis Proposing country: South
survey of mine Target area: abandoned mine area Korea
drainage in Proposed work item leader:
Sampling targets: adit water, drainage from open pit,
abandoned mines Mr. Sunbaek Bang, Mine
leachate from mine tailings and waste rock, background
water Reclamation Corp.
Additional item: collection of information required for the
plan of water treatment facilities
With the mine reclamation standards program still in its infancy at the time of writing, there are not yet any
official work items registered as part of the program. However, three work items are at the preliminary

Regulatory
proposal stage, where efforts are underway to put forward proposed new work items for review, comment,
and ultimately voting for their passage into the international standards development process. These
preliminary work items are presented in Table 3.
Additional work items can and will most certainly be proposed and added to the SC 7 standards program over
time.
Canada is not only putting forward the mine reclamation management planning standard as a new
international work item (discussed in detail in Section 5), but it is also putting forth some of Canada’s leading
thinkers and practitioners involved with mine reclamation planning and practices to provide input and
contributions to the entire mine reclamation management standards development program. Canada’s
involvement includes both contributions to the overall ISO mine reclamation management standards
development program and also the possible adoption or adaptation of any of the resulting standards into
Canada as National Standards of Canada.
The Canadian Standards Association has also established a Canadian standards mirror committee that
parallels the ISO activities on mine reclamation. This committee has two overarching purposes: First, by
bringing together Canadian experts and stakeholders involved with mine reclamation, this effort will help to
ensure both that Canadian experts have the ability and opportunity to provide input into the international
standards being developed through ISO/TC 82/SC 7 and also that Canada’s national position is informed with
respect to any decisions Canada puts forward. Second, the committee has the mandate and ability to adopt
or adapt any of these standards into Canada as National Standards of Canada. This specific ability to adopt
international standards into Canada comes about because the committee is officially recognised by the
Standards Council of Canada (SCC), Canada’s national regulator of the standards development landscape in
Canada and Canada’s national member body to ISO.
One of the basic requirements of any harmonised mirror committee is that it must be balanced in terms of
representation across interest categories, i.e., the different groups that have varying interests on the topic.
For the mine reclamation management committee, this includes five different interest categories: mine
operators, government/regulatory authorities, mine service providers, equipment suppliers, and a general
interest category that includes such groups as academic researchers and professionals. The rules of SCC are
that no one interest category can outweigh the two smallest in terms of actual voting members. The Canadian
harmonised standards mirror committee on mine reclamation is therefore populated by a cross-spectrum of
participants from varying interest groups and stakeholders.
It is important to emphasise the depth of expertise and experience in terms of the membership base of the
Canadian harmonised mirror committee. At the time of writing, the committee has about 45 members:
representatives from some of Canada’s leading mine companies, government, regulatory authorities from
both the federal and key provincial governments, and mining service providers, and some of Canada’s leading
academic researchers involved with different aspects of mine reclamation. This includes, for example,
members from industry who are directors of mine reclamation within their respective organisations;
government regulators (current and recently retired), often with decades of direct experience with mine
reclamation in Canada and internationally; many leading providers of mine reclamation-related consultancy
services; and leading academic researchers who are internationally recognised for their expertise on different
aspects of the topic. The committee also includes leaders from associations, such as the Mining Association
of Canada, that represent both the mining industry in Canada and also stakeholders involved with land
reclamation.
As a result of the breadth and depth of this membership, the mirror committee’s intention is to put Canada’s
leadership, experience, and high requirements forward as the basis of this international standards
development program.

The mine reclamation management planning project is important to the Canadian and global mining industry
as it will provide a crucial overarching document that will guide users and guide the use and/or development
of related standards for mine reclamation processes. The project will have direct benefits for Canadian
industry and governments, as it will set forth important requirements and recommendations that can be
referenced in regulations while also helping to save both money and time through the sharing of best
practices for operations, use of technologies and equipment, design and construction, etc.
This will also improve the competitiveness of Canadian industry in a highly globalised industry by helping
ensure that our high requirements and standards are being implemented in other jurisdictions. The mine
reclamation management planning standard will also help industry secure the social license needed to
undertake mining operations and activities.
It is important that Canada takes leadership of this specific standard, since this will help ensure that Canada
sets the direction of the subcommittee, in turn helping to prevent unintended difficulties that may arise from
unpractical members. Canada also needs to take a leadership role to ensure that mining is undertaken in a
sustainable manner in Canada and globally, recognising that Canada is a leader not only in mining, but also
specifically in mine reclamation management.
This work item and subsequent standard will provide requirements and recommendations for the purposes
of mine reclamation management planning. This will cover the main decision points and procedures that
must be considered in planning for mine reclamation. The objective is to help ensure consistency and
completeness in terms of the mine reclamation planning process. This includes the following issues and
activities:
 Approach to reclamation
 Reclamation objectives
 Reclamation technologies
 Reclamation activities
 Post-reclamation maintenance and monitoring
 Financial provisions
 Mine reclamation plan
Since Canada is proposing the work item on mine reclamation management planning, it has responsibility to
name an international convenor who will work to lead the associated international working group. For this,
Canada has put forward Dr. Dirk van Zyl from the Norman B. Keevil Institute of Mining Engineering at the
University of British Columbia. Professor van Zyl has more than 30 years of experience in research, teaching
and consulting in tailings and mined earth structures. During that period, he was a faculty member for 13
years at four universities in the United States and Canada. For the last 10 years, much of his attention has
been focused on mining and sustainable development. He has been involved internationally in many mining
projects. These projects covered the whole mining lifecycle, from exploration to closure and post-closure, in
a large range of climatic and geographic environments. His present research is in the area of the contributions
that mining makes to sustainable development and the application of lifecycle assessment to mined earth
structures.

Regulatory
This article has presented and discussed a new international standards development on mine reclamation
management, with a focus on a new work item around mine reclamation management planning.
In order to provide context for readers, this paper first describes the general process of developing
international standards. Two key aspects define this process: it follows a staged approach where, functionally
and operationally, the widest engagement of stakeholders and experts involved with a subject is sought; and
it is based upon a consensus-building process. The paper has also presented other ISO standards programs
directly or indirectly focused on the mining sector, showcasing the many different international standards
that already have been developed for use by the mining sector.
The need and purpose for an international standards program focused on mine reclamation has also been
discussed. Key aspects include the use of standards to help improve the environmental performance of mines
across their lives and to help mine operators gain a social license to operate. We then described the actual
work program now underway though ISO to develop standards on mine reclamation and the important role
that Canada is playing in this process.
One of the most significant ways Canada is contributing and taking leadership is by proposing a new work
item on mine reclamation planning. This new work item is intended not only to provide basic requirements
and recommendations for planning mine reclamation across the lifecycle of the mine, but also to provide a
crucial framework for standards focused on focal points of mine reclamation.
The author would like to acknowledge the review and helpful comments provided by Dr. Susan Ames, Dr.
John Errington and Bob Hart.
British Columbia Ministry of Energy, Mines and Petroleum Resources (2008) Health, safety and reclamation code for mines in British
Columbia, British Columbia Ministry of Energy, Mines and Petroleum Resources, Mining Division, Vancouver, British
Columbia.
Cao, X. (2007) Regulating mine land reclamation in developing countries: the case of China, Land Use Policy, 24(3), pp. 472–483.
Dearden, C.S. (1998) Performance standards and monitoring requirements of surface coal mine reclamation success in mountainous
jurisdictions of western North America: a review, Journal of Environmental Management, Vol. 53(3), pp. 209–229.
International Council on Mining and Metals (2008) Planning for integrated mine closure: toolkit, International Council on Mining and
Metals (ICMM), London, UK.
International Network for Acid Prevention (2012) Global acid rock drainage guide, International Network for Acid Prevention,
Melbourne, Australia.
International Organization for Standardization (2013) ISO/TC 82/SC 7 Mine reclamation management, Technical committees,
Standards Development, International Organization for Standardization, viewed 4 May 2015, http://www.iso.org/
iso/iso_technical_committee?commid=5052041.
Jenkins, H. and Yakovleva, N. (2006) Corporate social responsibility in the mining industry: exploring trends in social and
environmental disclosure, Journal of Cleaner Production, Vol. 14(3–4), pp. 271–284.
Mackenzie Valley Land and Water Board (2013) Guidelines for the closure and reclamation of advanced mineral exploration and mine
sites in the Northwest Territories, Aboriginal Affairs and Northern Development Canada, Ottawa, Canada.
Nova Scotia Environment (2009) Guide for surface coal mine reclamation, Government of Nova Scotia, Halifax, Canada.
Otto, J. (2010) Global trends in mine reclamation and closure regulation, Mining, society, and a sustainable world, J. Richards (ed),
Springer, London, pp. 251–288.
Price, W. and Errington, J. (1998) Guidelines for metal leaching and acid rock drainage at mine sites in British Columbia, Government
of British Columbia Ministry of Energy and Mines, Vancouver, Canada.
Slocombe, J.P. (2012) Exploring the origins of ‘social license to operate’ in the mining sector: perspectives from governance and
sustainability theories, Resources Policy, Vol. 37(3), pp. 346–357.


R.J.M. Wirtz WorleyParsons Canada Services Ltd., Canada

M.E. Urquhart WorleyParsons Canada Services Ltd., Canada
Design of closure drainage channels for reclamation of an oil sands mine site is unique, complex and
challenging. Reclaimed oil sands mines are expected to meet a predisturbance equivalency in terms of land-
use capability and to be sustainable over the long term with little to no maintenance. In terms of closure
drainage channels, this expectation could be interpreted to mean they should provide sustainable and
biologically productive aquatic habitats consisting of wetlands, lakes and watercourses. Research and
planning related to closure drainage design is being conducted within the oil sands industry. Historically, this
research and planning has primarily focused on understanding natural regional drainage systems and
applying the characteristics of the channels to the longitudinal slope, bed material, cross section and planform
design of the closure channels. This approach is commonly referred to as the geomorphic approach.
Oil sands mines are large scale operations and comprise various operational facilities and structures including
in-pit tailings ponds, out-of-pit tailings ponds, overburden dumps and licensed dams. At closure, these
facilities and structures will have been constructed using a variety of methods and materials (e.g.,
mechanically placed/compacted overburden, hydraulically placed tailings sand, etc.). The facilities and
structures would interface with in situ ground, upstream and downstream watersheds and often with
neighbouring mines. The design of closure drainage channels must consider the characteristics of the
materials that underlie the closure drainage channels from the perspective of fluvial geomorphology. The
consequences of channel erosion and migration on the sustainability of the closure landscape need to be
evaluated along with the expectations of the reclaimed mine landscape. Recently published guidelines for
delicensing oil sands dams suggest that applying the geomorphic approach might not be adequate for all
reaches of the closure drainage network within a reclaimed oil sands mine. It is expected that, at some
locations within the closure drainage network, channels will need to be more-robust against erosion and
channel migration than the geomorphic approach would allow in order to provide acceptable risk levels for
delicensing oil sands dams and successfully reclaiming an oil sands mine.
This paper provides an overview of various industry-specific guidance documents on closure drainage channel
design and discusses considerations and approaches for closure drainage channel design at oil sands mines
at various stages of mine development (planning, operation and closure). It also explores considerations for
future research into improving and demonstrating closure channel robustness.
Progressive reclamation of oil sands mines in northern Alberta has been ongoing for decades, and many
operators intend to ultimately return all or portions of the reclaimed mine sites back to public land.
Reclaimed oil sands mines are expected to meet a predisturbance equivalency in terms of land-use capability
and sustainability. Operators have been developing site-wide drainage plans for more than 20 years, with
increasing levels of detail, as more areas of the mines reach the end of their operating life and are ready to
be reclaimed.
The long-term sustainability of a reclaimed mine site and the surrounding watersheds, watercourses and
waterbodies must be carefully considered. A network of closure drainage channels must provide discharge
of surface water from the mine site to the natural stream systems that is long term (well beyond the typical

Closure drainage design for oil sands mining – robustness for dam delicensing R.J.M Wirtz and M.E. Urquhart
engineering timeframe) and requires little to no maintenance. Not only do the closure drainage channels
need to discharge surface water from the mine site to downstream receiving watercourses, the network must
often also receive stream flow from the existing natural watersheds upstream of the mine. The closure
drainage channels must provide discharge capacity for extreme floods potentially ranging from the mean
annual flood to extreme events such as the probable maximum flood (PMF). Consequently, the resistance of
channels against natural fluvial processes must also be considered to minimise the risk of exposing and
releasing mine waste or causing catastrophic failure of earthen structures. Design considerations such as
foundational materials, potential geomorphic changes, vegetation establishment and effects of wildlife on
the drainage system over long time frames make the design of closure channels complex and challenging.
Creating a sustainable drainage system is often one of the most challenging aspects of reclaiming a mine site.
A detailed evaluation of the robustness and discharge capacity of potential closure drainage systems is an
essential component of a mine closure plan. Evaluating the closure drainage system early in the life of the
mine could greatly reduce the level of effort required to achieve closure. The evaluation of a closure drainage
system should seek to identify potential high-risk and high-cost reaches of the closure drainage channel(s)
such that the overall closure drainage plan can be assessed, alternatives can be developed and mitigation
measures can be implemented if necessary.
Regulators, operators and the various stakeholders involved in oil sands mining operations expect mine
reclamation to meet provincial and federal criteria in order for the mine sites to receive reclamation
certificates and be returned back to public land. With respect to closure drainage design, expectations include
the following:
 the decommission and delicensing of all licensed dams;
 construction of an appropriately robust channel to protect against natural fluvial processes;
 adequate channel discharge capacity to limit potential high water levels during extreme
hydrologic events, such as the PMF, to maintain critical closure divides;
 restoration of hydrologic flow regimes to maintain premining watershed boundaries and
acceptable upstream/downstream stream conditions; and
 provision of natural and sustainable stream and riparian ecosystems for productive plant,
terrestrial and aquatic life.
In Alberta, dams are licensed through the Alberta Energy Regulator (AER) under the dam safety program.
Operations at older oil sands mines have reached the point where it is now necessary to begin the
decommissioning process for dams and move forward with closure and reclamation certification, as in some
cases, the active service life of the dam has ended. Mine operators are faced with an unprecedented task to
reclaim these structures in a way that adheres to all regulatory closure requirements. In 2009, the Oil Sands
Tailings Dam Committee (OSTDC), composed of professionals experienced in the field of oil sands mining and
tailings dams, was formed to further investigate the regulatory framework and outline a procedure through
which dams may be delicensed. The OSTDC produced a technical guidance document, De-Licensing of Oil
Sands Tailings Dams (OSTDC, 2014), which outlines a technical approach to responsibly delicense dams so
that they can be considered “solid earthen structures.” This guidance document incorporates a risk-based
approach for design, operation and regulation for delicensing; risk assessment is based upon failure mode,
probability of failure and the overall degree of consequence.
Delicensing a dam means the structure must be modified in such a way that it no longer meets the definition
of a dam as presented in the Canadian Dam Association’s (CDA) Dam Safety Guidelines (2007). The probability
of the facility ever reverting back to a configuration that meets the definition of a dam must also be extremely

Regulatory
low; this is where the risk-based assessment is incorporated. To qualify for a reclamation certification prior
to returning the land to the Crown, all environmental conditions must also be met. Although regulatory
agencies and industry committees have formed to develop and improve guidelines for final closure, the
framework for delicensing tailings dams in oil sands mines is still considered to be in its infancy (OSTDC,
2014).
In support of the dam delicensing process, it is necessary to provide a sustainable closure drainage channel
with an appropriate degree of channel robustness and discharge capacity. To meet dam delicensing
expectations, the closure drainage channel must demonstrate adequate performance and structural integrity
through postclosure monitoring, with little to no human intervention. If maintenance is found to be required
during the monitoring period, modifications can be made to the structure and potential failure modes can
be noted for future design. The period during which a dam is delicensed is defined by OSTDC (2014) to be
“sufficiently long to confirm that the performance target was achieved and the risk level is acceptable.”
In terms of channel robustness, it is anticipated that the highest standard for a closure drainage system would
be to provide long-term containment of mine waste against natural fluvial processes in the event of an
extreme hydrologic event, such as the PMF, over a geological timeframe. The appropriate degree of channel
robustness is determined through a risk-based assessment of the structure, depending on failure modes,
probability of failure and overall degree of consequence. The foundation material on which the closure
drainage channel will be constructed (i.e., in situ, mine waste, overburden, or tailings sand) is the main
contributing factor to the degree of channel robustness. For example, if the closure drainage channel is
situated on in situ material (typically outside of the mine footprint), a less robust closure drainage channel
design may be applied. On the contrary, if the closure drainage channel is situated above mine waste or
crosses through former dam structures containing potentially mobile material, a high degree of robustness
might be warranted. Overall, the appropriate degree of armouring must reflect the assessed level of risk to
ensure sustainable protection and to meet the expectations for dam delicensing and overall reclamation
certification. It would be impractical to apply a consistent level of protection throughout the closure drainage
network.
A key consideration related to channel robustness is the overall discharge capacity of the closure drainage
system. Upon closure and final reclamation, the closure drainage channel network is expected to convey
extreme floods without overtopping critical drainage divides. If a critical drainage divide were overtopped,
this could lead to rerouting of drainage channels along an unintended route. This could lead to catastrophic
changes to the reclaimed mine landscape and exposure of mine waste. For larger channels, continual low-
flow discharge would be required in order to establish aquatic and riparian ecosystems.
Broadly stated by the Government of Alberta, the oil sands mining industry “is legally obligated to reclaim all
disturbed land to an equivalent capability, and return it to the Government of Alberta” (Government of
Alberta, 2015). This statement can be interpreted to apply to many reclamation activities, including the
restoration of premining regional flow regimes.
In the Athabasca oil sands region, the restoration of regional watershed boundaries and flow regimes is highly
dependent on collaboration between operators. This collaboration is required due to the proximity of mine
sites, the hydrologic distribution of regional watersheds and the expectation of future mining expansions.
Watershed agreements have been established between operators to aid in the final restoration of regional
watershed boundaries and hydrologic flow regimes. AER requires that conceptual mine closure drainage
plans continually assess the interaction between neighbouring mines as part of the overall reclamation plans
for the mine sites.

The Government of Alberta’s statement regarding the obligation to return oil sands land to “an equivalent
capability” (see section 2.3) may also be interpreted to mean that oil sands mines must restore aquatic and
terrestrial ecosystems. Although not necessarily required for the delicensing of tailings dams, reclaiming an
oil sands mine to be environmentally productive and sustainable is required for reclamation certification. As
with the dam delicensing process, post-closure ecological monitoring is required to confirm the
establishment of natural vegetation and ecosystems that conform to the natural conditions found in Alberta’s
boreal forest. The duration of ecological monitoring may be lengthy, potentially years or decades, before a
reclamation certification would be issued.
With regard to the closure drainage channels, the expectation for “equivalent capability” is met through the
design and construction of closure drainage features. These features may include end-pit lakes, ponds,
wetlands and floodplains. Closure drainage channels are essential to providing flushing of fresh water
through ponded systems to improve water quality. These environmental criteria, conditions and
considerations are not discussed further in this paper. It should be noted that the floodplains along closure
drainage channels also provide flood attenuation, while the outlet channels from water bodies are critical to
minimising high water levels during flooding.
Research and planning relating to closure drainage design is ongoing within the oil sands industry. This
research and planning has primarily focused on understanding natural regional drainage systems and
applying the characteristics of these natural channels to the longitudinal slope, bed material, cross-section
and planform design of closure channels. This is commonly referred to as the geomorphic approach. The
geomorphic approach to channel design has many benefits over conventional engineering, which involves
designing a channel to withstand or convey a given flood event based on an accepted level of risk. However,
the geomorphic approach also has many limitations, and it might not adequately protect certain reaches
within a closure drainage system or meet the intent of the dam delicencing framework.
Conventional engineering practice for drainage channel design is to accept a risk of failure by selecting a
design flood recurrence interval and then sizing the channel and specifying armouring to withstand the design
flood. Under typical engineering circumstances, it is understood that if a hydrologic event occurs that exceeds
the design flood, failure or damage of the structure is expected, and subsequent repair and/or maintenance
will be required. When applying conventional methods of erosion control, such as riprap, to the entire closure
drainage system to guard against an extreme design flood, the required size and volume of riprap becomes
impractically large. Additionally, an extensive closure drainage network consisting only of riprap-lined
channels is unlikely to meet the land-use equivalency objectives related to creating sustainable aquatic
ecosystems. Given the limitations of conventional engineering channel design, the Canadian Oil Sands
Network for Research and Development commissioned the development of an alluvial channel design
manual for designing geomorphic channels in the Athabasca oil sands region (Golder, 2008).
The geomorphic approach is based on an environmental dynamic equilibrium in which a channel is designed
based on regional stream characteristics (e.g., cross-sectional dimensions, longitudinal slope, bed/bank
material) and is able to experience natural fluvial processes such as degradation, aggradation and
vertical/lateral migration. Channels designed using the principles of the geomorphic approach are
theoretically maintenance free, and the overall required size and volume of armouring becomes significantly
more practical. Proper design and construction of geomorphic channels also encourages the establishment
of natural vegetation and productive aquatic habitat.

Regulatory
Applying the geomorphic approach in all closure drainage channels within a mine site would generally
provide an equivalent level of protection against fluvial processes regardless of the substrate (mine waste or
in situ ground); however, the risk and consequence of natural fluvial processes is not equivalent in these two
substrates. Furthermore, the following key limitations to the geomorphic approach must be noted:
 Caution is recommended when applying the alluvial channel design guidelines to bank-full
discharges beyond 10 m3/s from 100 to 150 km2 watershed areas (Golder, 2008) and to
longitudinal channel slopes less than 0.75%.
 The geomorphic approach is based on relationships developed from regionally collected data on
natural channels, but areas of mine waste will not provide a foundation equivalent to that of a
natural watercourse. Reclaimed tailings impoundments will likely be filled with uniformly sized,
highly erodible material and will be prone to differential settlement; conversely, natural
armoured channels would generally have large extents of additional armouring material that
remains in the channel from the initial channel formation (i.e., this material could likely extend
across floodplains).
 Vegetation in the floodplains and channel banks will take time (potentially years or decades) to
mature into a robust erosion protection layer in the northern climate, and the channels would
likely need to be put in service shortly following construction.
The geomorphic approach is unproven, as large-scale channels of this nature have not yet been constructed
in a reclaimed oil sands mine. Additionally, the geomorphic approach tends to focus heavily on the bank-full
discharge, which is often considered to be the 2-year flood. Design consideration of managing more extreme
floods is often limited to assuming that the floodplain would limit channel erosion during extreme floods.
Natural fluvial processes such as lateral migration, down-cutting, head-cutting, localised scour and avulsion
could lead to failure of critical components of the mine landscape such as critical closure divides or cause
catastrophic failure, which could expose mine waste. Given that the geomorphic approach allows natural
fluvial processes to occur, the channels designed following this approach might not be sufficiently robust to
support long-term containment of mine waste in a manner that is acceptable for “walk-away” closure and
reclamation certification.
The risk-based approach outlined for delicensing dams is highly dependent on the design of closure drainage
channels because failure of drainage channels could lead to loss of containment of mine waste or
catastrophic failure of critical closure features. The process involves an iterative, risk-based assessment of
the probability and consequences of failure of the former dam structure whereby a potential closure scenario
is evaluated and then modified until an acceptable level of risk of failure is achieved. The risk assessment
framework comprises potential failure modes, the probability of failure and the consequence of failure. From
a hydrotechnical viewpoint, issues such as hydraulic stability, erosion, slope stability and overtopping are all
elements in which failure in a closure drainage channel, and therefore the delicensed dam, may occur. Under
extreme flooding events such as the PMF, these failure modes could lead to catastrophic consequences to
the human population, infrastructure and the environment (OSTDC, 2014).
Based on a review of these two approaches, it is evident that broad application of any one design approach
would not likely result in a closure drainage system that meets all design objectives. The geomorphic
approach may not be sufficiently robust to ensure containment of mine waste in the event of large-scale
floods or under certain closure drainage design configurations. Conventionally engineered channel design
might not provide desired habitat characteristics, might be expensive to construct and would typically require
maintenance. The situation for each channel at each oil sands mine will be unique, and it is likely that a “one-
size-fits-all” approach will not result in the robust closure landscape desired by stakeholders. Instead, a

rigorous approach is warranted: this involves identifying high-risk channels, assessing the consequences of
failure of those channels and implementing specific measures to guard against specific failure modes.
The proposed approach consists of the following steps:
 Screen the closure drainage system to identify risk high-risk channels.
 Adjust configuration to minimise high-risks channels.
 Assess the consequences of failure of high-risk channels.
 Design robustness and armouring to guard against specific failure modes.
The following sections provide further discussion on key considerations for implementing the above
approach.
The configuration of a closure drainage system dictates the risk associated with all aspects of the closure
drainage system, including its robustness against fluvial processes and its robustness for discharge of extreme
floods.
To minimise risks related to natural fluvial processes (such as lateral migration, degradation, head-cutting,
localised scour and avulsion), the configuration or alignment of the closure drainage system must primarily
address three key parameters:
 longitudinal channel slope;
 upstream drainage area (proportional to flow); and
 channel substrate (in situ ground, erodible tailings, overburden, etc.).
In a general sense, high-risk channels could be defined as the channels that have the steep bed slopes, large
upstream drainage areas and erodible substrates. If these conditions are present, channels should be
reconfigured to minimise slope and reroute drainage area away along a shallower slope or across a less
erodible substrate. The screening process provides a relative comparison of risk throughout the closure
drainage network.
Current closure drainage plans often include channel configurations that could be considered high-risk
configurations. This is often a result of retrofitting a drainage system through an existing mine landscape.
Challenging closure drainage channel configurations within an oil sands mine site might include the following:
 large watershed areas on top of overburden dumps, resulting in the need for steep drainage
channels to convey runoff from the dump;
 steep channels entering or exiting former tailings deposits, both in-pit and out-of-pit;
 drainage channels routed along the crest of a mine-pit wall;
 channels routed across highly erodible tailings deposits, where ground settlement could be
substantial, which could lead to channel relocation;
 discharge of flow from large upstream drainage areas across the mine area; and
 unknown and potentially variable interfaces with neighbouring mines/operators.
In order to minimise high-risk configurations, closure drainage system designs should incorporate the
following criteria:
 Minimise drainage area upstream of channels that are steep or located over soft erodible
deposits.

Regulatory
 Found landform outlets and critical elevations in areas of in situ material or competent
overburden.
 Minimise watershed area draining to constructed channels.
 Integrate the closure drainage channels with surrounding reclaimed mines and natural
watercourses.
To limit potential high water levels and maintain critical closure divides, it is essential to examine the
discharge capacity for a closure drainage channel closely. The topography of a reclaimed mine site is
fundamental to maintaining a sustainable long-term drainage plan and preventing extreme hydrologic events
from becoming catastrophic. A site-wide drainage plan includes not only closure drainage channels but also
end-of-pit lakes and wetlands. If, during an extreme flooding event, a closure divide were to become
compromised, the release of surface water may cause serious detrimental effects that could expose and
transport mine waste, flood neighbouring mines and/or put the human population at risk.
The discharge capacity should be assessed using a detailed hydrologic and hydraulic model that accurately
simulates flood storage and routing through the closure drainage network. The high water levels simulated
in the model should be compared to the elevation of critical closure divides to determine if the channel
capacity is adequate to convey design floods corresponding to an appropriate risk.
One challenge in evaluating the discharge capacity of the closure drainage channel is addressing the potential
for channel blockage and for high-channel roughness due to vegetation. Natural channel blockage could be
the result of debris accumulation, ice jams or beaver activity. Assuming the closure drainage channel will
become blocked at some point, contingency should be provided to limit potential high water levels during
extreme events.
Ideally, wide floodplains would be provided to convey extreme events and provide a high level of discharge
capacity. In many locations throughout a reclaimed mine site, the area to create wide floodplains might not
exist. In these cases, additional mitigation might be required. Appropriate sensitivity analysis must be
completed to allow consideration for changes to the channel discharge capacity.
Based on the screening analysis, the configuration of the closure drainage system would be adjusted to
minimise risks related to fluvial processes and high water levels during floods; however, given constraints
related to mine configurations and existing mine developments, reconfiguring the closure drainage system
is not likely to reduce risks to a point where the geomorphic approach would meet the requirements for dam
delicensing. Therefore, closure channel system design for a mine site can be expected to require a
combination of geomorphic and engineering approaches to increase robustness. These approaches might
include the following:
 Heavy riprap channel lining using conventional engineering methodology: this has high safety
factors to protect critical areas where erosion might lead to unacceptable failure (e.g., highly
erodible foundation material, valley walls and critical closure features).
 Additional robust riprap armouring in areas where degradation typically occurs: these include
channel constrictions, longitudinal slope changes, inlets/outlets to end-of-pit lakes or wetlands,
along valley walls oriented on a channel bend and transitions to/from natural stream systems.
 Construction of an alternative overflow channel to channel blockage: the location and orientation
of this potential mitigation strategy should be considered on a case-by-case basis due to the
highly variable circumstances of the mine planning, operation and closure.

Given the iterative nature of the proposed design process, closure drainage channel design considerations
are variable and depend on case-by-case assessment of conditions and mitigation measures deemed
necessary for the specific sites and closure scenarios.
This example outlines a design approach for site-wide closure planning based on reclamation of a
hypothetical oil sands mine. It assumes that the existing oil sands mine is being progressively reclaimed and
has an existing closure drainage plan and existing constraints. Many aspects of the approach are also
applicable to the planning of a new mine. Moreover, applying this approach and adjusting the mine plan to
meet the requirements of the closure plan could lead to a highly robust closure landscape at a lower overall
cost.
The operational state of the mine site is illustrated in Figures 1 and 2. Note that licensed dams would exist
within and around the mine pit and also to provide containment for the external tailings facility. The site is
located on the bank of the Athabasca River. The mine footprint extends across three watersheds and their
associated watercourses, as shown in Figure 3. The topographic map image for all figures was taken from
Natural Resources Canada (2015). The configuration of the mine, facilities and closure drainage plans are
entirely fictitious and were prepared only for illustrative purposes within this paper.

Regulatory
A hypothetical closure drainage configuration for the site is shown in Figures 4 and 5. Figure 4 provides detail
of the closure plan within the site. Figure 5 provides a watershed-level view of the closure drainage plan to
illustrate how the closure drainage plan integrates with the surrounding watersheds.

In Figures 4 and 5 the following high-risk configurations are evident:

 A very large upstream watershed (Watershed A) is conveyed through the external tailings facility,
across the reclaimed mine pit to the end-pit lake and then through the main outlet for the site.
The watershed area vastly exceeds the upper range noted for application of the geomorphic

Regulatory
approach. The large watershed area upstream of the channels that are cut through the berms (for
dam delicencing) would likely require heavy armouring to provide an acceptable level of
robustness for closure. Additionally, these large channels are routed across tailings deposits
where ground settlement could be substantial and substrate is highly erodible. These conditions
could lead to channel relocation. It is possible that the closure drainage channels across the in-pit
tailings deposits would not be sustainable in a closure time frame.
 There is a single large watershed on top of the overburden dump. The top of the dump is
approximately 100 m above the surface of In-pit A. These two factors make the outlet channel
from the dump high risk. The runoff from the overburden dump is also routed across the in-pit
tailings deposits, increasing the potential for flooding and damage to the closure channels.
 Runoff from Watershed B is routed between the toe of the overburden dump and the crest of the
mine-pit wall. This channel would be at risk of relocating directly over the edge of the pit wall,
leading to severe erosion of the closure landscape (In-pit A).
The hypothetical closure drainage plan was revised to minimise the high-risk configurations; the revised plan
is shown in Figures 6 and 7. Figure 6 provides a detail of the closure plan within the site. Figure 7 provides a
watershed-level view of the closure drainage plan to illustrate how the closure drainage plan integrates with
the surrounding watersheds.

The revisions to the closure drainage plan include the following:

 The overburden dump was relocated to upstream of the external tailings facility. The overburden
dump forms a divide and serves to divert upstream surface runoff around the mine area. Note
that this would only be practical if it were identified early in the mine development. The challenge
of rerouting channels adjacent to a mine site might not be feasible given the constraints of
neighbouring mines.
 The end-pit lake was relocated to minimise the watershed area upstream of the channels that are
cut through the berms. The upstream watershed areas for the external tailings facility berm is
reduced from 350 km2 to about 15 km2. Similar reductions would be realised for the channels cut
through the in-pit berms.
 Large plateau areas of the overburden dump are drained by several smaller channels rather than
one large channel.
The revised closure drainage plan includes construction of a long diversion channel, which is expected to be
preferable to the challenge of routing very large drainage areas through reclaimed tailings ponds.
Given that industry guidelines have been issued by the OSTDC, it should be expected that the dam delicensing
process in the oil sands mining industry is accelerating. With respect to closure drainage channel design,
further development that combines robust conventional channel design with the geomorphic approach will
allow for considerable improvements to existing conceptual closure drainage plans.
The following points summarise potential areas for further research and improvement regarding the design
of closure channels for the purpose of dam delicensing and overall final closure and reclamation of oil sands
mines:
 At this time, conceptual drainage plans are generally developed with the assumption that the
design will meet an acceptable level of risk of failure to the long-term containment of mine waste

Regulatory
within a reclaimed mine site. These drainage plans need to be further evaluated based on more
detailed analysis and interpretation of the risk assessment framework outlined by the OSTDC
(2014).
 Further consideration should be given to reducing the volume and size of riprap while providing
an appropriate amount of erosion protection for extreme flooding events. Various pit backfill
materials; other hydrotechnical control measures (e.g., buried spurs/groynes throughout the
floodplain or the installation of specified hard points to control bed elevations); and/or
bioengineering methods could be evaluated as possible alternatives.
 Large-scale channels should be constructed and monitored to quantify limitations of the
geomorphic approach and bioengineered channel protection. These channels should address
various potential subgrade materials (e.g., tailings sand, dispersive clay, coke) and replicate
extreme storm events in order to provide information that is applicable to the closure drainage
channels required for oil sands mines.
 The complexities and challenges related to creating closure drainage channels across large oil
sands mine sites must be recognised. It is anticipated that each significant drainage channel will
require substantial analysis, involving multidisciplinary teams, and adaptive implementation
strategies in order to meet the expectations of closure channels.
 Closure drainage plans should seek to reduce risks to channels in sensitive portions of the mine
(e.g., through tailings and/or dams) by minimising upstream watershed areas and the longitudinal
slope of channels. These reconfigurations should be identified as early as possible in the mine
development phase.
Canadian Dam Association (CDA) (2007) Dam safety guidelines 2007, Canadian Dam Association, Ottawa, Canada.
Government of Alberta (2014) Alberta’s oil sands, reclamation, viewed 23 February 2015, http://www.oilsands.alberta.ca/
reclamation.html.
Oil Sands Tailings Dam Committee (OSTDC) (2014) De-licensing of oil sands tailings dams, technical guidance document, Oil Sands
Tailings Dam Committee, Fort McMurray, Alberta.
Natural Resources Canada (2015) Atlas of Canada - Toporama, viewed 26 March 2015, http://atlas.gc.ca/site/english/toporama/.
Golder Associates (2008) Alluvial channel design manual, procedures for the Athabasca oil sands region, Canadian Oil Sands Network
for Research and Development and Department of Fisheries and Oceans.


Mine Closure 2015 – A.B. Fourie, M. Tibbett, K.G. Mercer and D. van Zyl (eds)
C.D. Grant Anglo American, Australia

L. Hantler Anglo Coal, Australia
In many mining operations around the world, the highest rehabilitation costs are associated with reshaping
mining voids. It is possible to use blasting techniques to move material from the highwall to the mining void
at much lower cost than by using heavy machinery such as bulldozers. While this is not a new idea, very few
trials have been conducted to determine how much material can be cast to the final landform position and
how much money can subsequently be saved from the closure liability. Trials were conducted at Anglo Coal
Australia’s Dawson Mine in 2013 to demonstrate that overall rehabilitation costs can be significantly reduced
by using sophisticated blasting techniques to move considerable amounts of highwall material into final
landform position and thus reduce the need for more expensive earthmoving options. Four blasts were
completed, and detailed preblast and postblast surveying allowed accurate estimates of how much material
was blasted to the final landform and how much reshaping was required. Detailed costs were collected for
every step of the process, risks were identified, and mitigation measures were trialled. Results showed that,
for a 25% postrehabilitation slope, at least two-thirds of the material could be cast blasted to the final
landform. Across a range of final landform slopes, cast blasting moved 25%–63% more material than
previously assumed in closure cost models. This led to immediate savings of tens of millions of dollars off the
premature and life of mine (LoM) closure liability. There are also significant nonfinancial benefits of the trials,
such as demonstrating to external stakeholders and regulators that open-cut mining can be a temporary land
use and that even the most difficult of areas (i.e., void highwalls) can be successfully rehabilitated to a
productive postmining land use. Further trials are planned, and an additional 10%–20% increase in cast blast
to final landform is seen as achievable.
Anglo American (AA) is a global and diversified mining business that provides the raw materials essential for
economic development and modern life. Anglo American has mining operations, growth projects, and
exploration and marketing activities extending across southern Africa, South America, Australia, North
America, Asia, and Europe; these involve a diverse array of commodities, namely coal (thermal and
metallurgical), copper, diamonds, iron ore, manganese, nickel, niobium, phosphate, and platinum. Anglo
employs almost 160,000 people worldwide.
The Dawson Mine is one of three large open-cut coal mining operations owned by Anglo American in
Queensland, Australia. The mine currently extends over 50 km in length, has a footprint close to 9,000 ha,
and mines 9 M t of metallurgical and thermal coal per annum. Of the area disturbed to date, only 20% has
been rehabilitated, largely due to uncertainty about future mine plans and sterilisation of coal. This means
that the mine has a substantial closure liability associated with progressive rehabilitation, void management,
and demolition costs. The largest proportion of this liability relates to rehabilitating mining voids and, in
particular, the regulatory requirement to reshape highwalls to defined slopes in order to achieve the final
land use values – in this case, returning rehabilitated areas to grazing land use. This represents multiple
hundreds of millions of dollars in liability and over 1,800 ha of disturbance. Demonstrating that highwall
blasting techniques could be used to cost-effectively move large amounts of material to the required
landform slope represents an opportunity to reduce this liability.

Highwall blasting: a cost effective way to reshape and reduce closure liabilities C.D. Grant and L. Hantler
The theory of highwall blasting involves drilling explosives into the highwall to cast material to the final
landform, rather than relying on expensive mobile equipment (i.e., bulldozers) to move the material (see
Figure 1). This is not a new idea, but very few trials have been undertaken to determine how much material
can be cast to the final landform position and how much money can subsequently be saved from the closure
liability. The objective of this paper is to demonstrate that closure liabilities can be significantly reduced by
using sophisticated blasting techniques to move considerable amounts of highwall material into final
landform position and reduce the need for more expensive earthmoving options.
The Dawson Mine is located near the township of Moura in central Queensland (Figure 2). The natural
topography of the area has gently rolling hills at an elevation of 150 m.a.s.l.; cattle grazing was the
predominant premining land use. The mine was opened in 1961, and had both underground and open-cut
operations and various owners over the subsequent half century. Anglo acquired the mine in 2002 from Shell
as a joint venture with Mitsui (51:49). In 1963, Dawson was the first Australian operation to use draglines to
best exploit the multiseam deposit. The steeply dipping coal seams result in increasing overburden volumes
for every strip. The mine currently extends over 50 km in length, has a footprint close to 9,000 ha, mines
9 M t of metallurgical and thermal coal per annum, and employs more than 1,200 people.

Landform
Four trial highwall blasts were undertaken at Dawson Mine in the second half of 2013. The four blasts are
referred to as 17B highwall, 18B northern endwall, 18A northern endwall, and 18A highwall, in that sequence.
The blasts varied in size and in the level of associated risks (e.g., size, highwall failures, endwalls, water in
pits). Smaller blasts were undertaken first, then progressively larger blasts; the fourth blast had water in the

void. Water in the void represents a risk of tidal wave and associated catastrophic failure of endwalls and
was identified as one of the major project risks. The drilling and loading was undertaken by contractors, with
Anglo staff supervising the blast procedure. Video footage was obtained for all blasts. Surveys were
undertaken prior to blasting, after blasting, and following reshaping. Survey data was analysed in Maptek in
Vulcan, using a digital terrain model (DTM). Moorvale earthmoving contractors were contracted to undertake
the bulldozer reshaping after blasting. All costs associated with each process were collected. The proportion
of material cast by the blast was calculated using the below equation:
𝑑𝑜𝑧𝑒𝑟 𝑝𝑢𝑠ℎ 𝑣𝑜𝑙𝑢𝑚𝑒 𝑖𝑛 𝑏𝑐𝑚
𝐶𝑎𝑠𝑡 𝑉𝑜𝑙𝑢𝑚𝑒 (%) = (𝑇𝑜𝑡𝑎𝑙 𝑑𝑒𝑠𝑖𝑔𝑛𝑒𝑑 𝑏𝑙𝑎𝑠𝑡𝑒𝑑 𝑣𝑜𝑙𝑢𝑚𝑒 𝑖𝑛 𝑏𝑐𝑚 − ( )) ∗ 100
1.1
All four blasts were successfully carried out, and no safety incidents were recorded. Knowledge obtained
from early blasts was incorporated into later blasts to increase casting (Table 1). The depth and angle of the
blast hole and the presence of highwall failures and endwalls were significant influencing factors. Increasing
angle and shallower depth increased casting while decreasing costs. The presence of highwall failures
significantly decreased cast blasting efficiency (Figure 3). The presence of endwalls and ramps also affected
the casting achieved by the blasts (Figure 4).
# Location Holes (#) ANFO (t) Mean depth (m) Volume (m3)
1 17B highwall 772 140 9.5 259,053
2 18B endwall (N) 447 105 10.4 160,318
3 18A endwall (N) 793 210 12.8 336,632
4 18A highwall 1,588 405 12.3 566,939
Total 3,600 860 11.2 1,322,942

Landform
The four blasts varied significantly in the percentage volume of material that was cast to the final landform.
Furthermore, certain sections within individual blasts exhibited significant variability. This was expected, as
small blasts with minimal risks were trialled first, and learning was incorporated into subsequent blasts. The
sites also varied in terms of the number of highwall failures and the existence of previous drill holes, which
influenced the effectiveness of cast blasting. The first trial blast achieved 54% of material cast to the final
landform design of a 25% slope (Figure 5). The second blast achieved a similar result, with 50% of material
cast to the final landform. The northern endwall of the third trial blast exhibited the highest volume cast to
final landform: 72% (Figures 6 and 7). There were a number of highwall failures in this blast area and, if these
are included, the average was 55%. However, for the fourth blast in pit 18A highwall, the average cast blast,
including some minor failures, was 71% (Figure 8). These results supported the conservative assumption that
casting 65% of material to a final landform slope of 25% was achievable.
Blast 1 Pit 17B Section 5

Rehab Floor Original Topo
Post-Blast
Section Line
Post Blast Floor
Original Topo
Rehab Floor @ 25%
12

Anglo Coal Australia has developed a rehabilitation liability tool (RLT) that incorporates cost models for a
range of domains and treatments. The cost information collected as part of the four trial blasts and associated
reshaping was run through the costing models, and a number of output options were presented (Table 2).

Landform
Sufficient evidence was collected during the trials to support increasing the assumed percentage of casting
from highwall blasting for all three slopes (a 25%–63% increase). This led to a decrease in cost of 20%–28%,
depending on the slope required for the postmining landform. Applied across the Dawson Mine, this led to
tens of millions of dollars in savings off the closure liability. The two final blasts achieved over 70% cast of
material in some areas, and it is believed that 75% casting for a 25% slope could be justified with further
trials. These trials were planned for 2014 but have been delayed to 2015 due to the existing unfavourable
economic climate in the global coal mining industry.
Cast efficiency (%)/ 40 50 65 75 80 90 100

Required slope (%)
75 100 85 72
33 100 90 84 80
25 100 93 82 75
*Bold numbers are those currently used, and numbers in italics are supported by results of this project.
A number of risks were identified throughout the project, and the following mitigation measures were
proposed or trialled:
 Previously drilled area: One highwall had been previously drilled for blasting an unknown number
of years ago. These holes are not suitable for loading, as many have become blocked, but they do
have the potential to reduce the effectiveness of the blast by allowing the escape of energy
through the holes. The strategy to combat this issue was to backfill as many holes as possible with
stemming. Attempts were made to open up the blocked holes either by manual means or by using
a small drill.
 Benched highwalls: A number of highwalls are not typical straight highwalls but contain a catch
bench in the middle. This geometry is not ideal for cast blasting, and extra ground preparation
work was undertaken to minimise the effect. Splitting the blast was also a potential mitigation
option.
 Highwalls with spoil: One highwall planned for blasting has had spoil dumped on it in the past.
This spoil has been shaped for reclamation purposes, but it will need to be removed to allow
drilling. Logically, it will be pushed into the void, taking up room that the cast blast would have
used. Detailed design will be undertaken to minimise this effect.
 Voids with water: A number of pits will not be available for blasting until the water in these pits is
reduced to a satisfactory level. This could delay the timing of blasts. Blasting without removing
the water could lead to undesirable consequences such as a large wave of water moving in an
uncontrolled fashion and/or a pit full of mud that dozers will not be able to work in. A pumping
plan was instigated to dewater these pits.
 Presence and risk of highwall failures: One large highwall failure existed in one of the trial areas,
which significantly decreased the efficiency of casting from the highwall blast. The geotechnical
stability of the highwall during drilling operations was monitored closely using ground sensing
radar to ensure the safety of personnel.

The four trial blasts undertaken at Dawson Mine in 2013 have demonstrated that significantly more material
can be cast to desired final landform slopes than previously estimated. All blasts were carried out safely, and
there was significant learning related to shot pattern and loading that could lead to further decreases in blast
costs. A number of project risks were identified, but these were successfully mitigated. The results indicate
that casting can be increased by 25%–63% over what was previously assumed, depending on the postmining
landform slope. Furthermore, the associated costs could be decreased by 20%–28%, leading to a savings of
tens of millions of dollars from closure liabilities. Further trials are planned in 2015 to demonstrate increased
efficiency at shallower postmining slopes, where a further 10% decrease in cost is seen as being achievable.
The authors would like to acknowledge the numerous people who contributed to the success of this project,
including Tom Hayes, Steve Simmonds, Mark Royes, Stephen Brady, and Karen Price.

K. Brand Klohn Crippen Berger Limited, Australia

M.J. Ind Klohn Crippen Berger Limited, Australia
M. Ridd Ok Tedi Mining Limited, Papua New Guinea
M. Thompson Ok Tedi Mining Limited, Papua New Guinea
L. Murray Klohn Crippen Berger Limited, Canada
The Lower Ok Tedi Dredge Project, which commenced operation in 1998, is located at Bige in the Western
Province of Papua New Guinea (PNG). The project is an important component of Ok Tedi Mining Limited's
(OTML) environmental management program to reduce the mine’s impact on downstream riverbed
aggradation, flooding, and vegetation dieback. Dredged river sediments (mostly sand and silt particles) are
stored in on-land stockpiles on the east and west banks of the Ok Tedi River (note the term “Ok” translates
to “River” in the local region language hence the term Ok Tedi used throughout this paper). Klohn Crippen
Berger Limited (KCB) (formerly Klohn Crippen Consultants Limited) undertook the original site selection,
geotechnical, and civil design for the project, including surface water management and landform design of
the stockpiles in the mid-1990s. The original design intent of the Bige project was to store sand in a “semi-
permanent” facility on the Ok Tedi flood plain, where erosion of the stockpiles by migration of the Ok Tedi
was seen as a possibility. The expectation at the time was that after mine closure and recovery of the river
system, gradual erosion of the stockpiled sand would not significantly impact the river system. In the early
2000s, following commencement of operations at Bige, further studies indicated that the dredged material
stored in the stockpiles could be potentially acid forming (PAF) and could lead to acid rock drainage (ARD).
Therefore, modifications to the landform design were undertaken to mitigate this risk.
The initial landform design of the stockpiles generally incorporated side slopes of 7.5H:1V. To facilitate surface
water management from this surface, a series of hard engineered structures including gabion baskets, Reno
mattresses, benches, graded bench drains and inter-bench spillways. This water management system has
operated successfully over the past 15 years on rehabilitated sections of the stockpile. Developments in
material availability, scheduling issues and construction practicalities associated with placing a proposed non-
acid forming (NAF) closure cover necessitated a change in design philosophy for the surface water
management system. In 2013–2014, OTML initiated studies to investigate the use of soft engineered water
management features including flatter side slopes, vegetated drainage swales, and ephemeral surface
wetlands. The updated closure design approach is intended to provide a landform that minimises long term
care and maintenance requirements, is sympathetic to the natural environment, and includes a stable and
resilient long-term closure cover.
The main objective of the Lower Ok Tedi Dredge Project, which commenced operations in March 1998, is to
alleviate floodplain die-back by reducing riverbed levels thereby reducing the frequency of overbank
flooding. River sediments are dredged and stored in on-land stockpiles on the East and West Banks of the Ok
Tedi River at Bige, Western Province, PNG (Figure 1). Riverine tailings disposal and erodible waste rock dumps
at the Mount Fubilan mine contribute approximately 65 million tonnes per year of sediments to the Ok Tedi,
of which approximately 18 million tonnes are dredged from the Ok Tedi and placed within the on land
stockpiles at Bige, approximately 100 km downstream (OTML, 2014).

Evolution of the Bige stockpile closure design for the Ok Tedi Mine K. Brand, M.J. Ind, M. Thompson, and L. Murray
The stockpile design has developed continuously over the life of the project and, after several expansions,
the East Bank stockpile is now nearing closure. The closure design has evolved significantly, from a structure
intended to be semipermanent to a traditional mine waste stockpile with bench, toe, and longitudinal drains.
A non-acid forming (NAF) closure cover is required over the existing stockpiles to reduce ARD processes and
enable revegetation. Because the stockpile is in a location with a high annual average rainfall and frequent,
intense storm events, the long-term integrity will be reliant on reducing erosion and controlling the surface
water.
The Bige site records 5.5 m of rainfall per year on average, while the upper reaches of the Ok Tedi and its
tributaries receive in excess of 10 m per year (OTML, 2013). The Ok Tedi catchment is approximately
1,300 km2 and ranges in elevation from approximately 3,000 m above sea level (m.a.s.l.) to 25 m.a.s.l. at Bige.
Studies were carried out in 2013 to validate the historic design approach to water management, update cost
estimates and schedules, and confirm the environmental requirements. As a result of these studies, an
alternative closure landform was proposed with reshaped slopes, removing benches and drain structures to
produce a design sympathetic to the surrounding environment.
The current mine plan sees the Ok Tedi mine operating until at least 2025 with the Bige stockpiles operational
until 2026. Planning for the progressive closure of both the East and West Bank stockpiles has shown that
the availability of suitable cover material, over the remaining mine life, is a critical factor. Dredge material
will be selectively placed depending on its ARD potential, with only the finer NAF material (generally with a
target acid neutralising capacity to maximum potential acidity (ANC/MPA) ratio greater than 2.0) used as the
final cover material.
This paper focuses on the East Bank stockpile at Bige; it briefly describes the historical stockpile design and
current surface water management intent, the drivers of change, the evolving closure design philosophy and
a review of cover trial performance to date. The East Bank stockpile closure will be used to inform the West
Bank sand stockpile closure and any future storage facilities.
Mining commenced at Ok Tedi in 1984 with the first production of gold. Production of copper commenced
in 1987 and OTML has produced approximately 4.5 million tonnes of copper, 400 tonnes gold, and 850 tonnes

Landform
of silver to date (OTML, 2014b). The mine is now a state-owned enterprise with a mine life currently approved
to 2025, and it is one of the largest contributors to the country’s economy.
Construction of a tailings storage facility (TSF) commenced in 1983; however, after several landslips and
foundation issues at the site, the TSF construction was abandoned (Murray et al., 2000). Extensive
investigations were undertaken to identify suitable alternative tailings storage sites, whereupon riverine
disposal methods were subsequently approved by the PNG Government.
In 1988, dredging of fine sands and silt downstream was identified as an option to reduce downstream
riverbed aggradation, flooding and dieback. At the time this option was not pursued, however after further
TSF studies in the mid-1990s the dredging project was reinvestigated by OTML.
The Bige stockpiles are located on an area of land partially affected by overbank flooding and die-back. The
OTML Mine Waste Management Project (MWMP) dredging trial commenced in 1998 on the East Bank. After
several expansions and design optimisations on the East Bank, placement of dredged sand commenced on
the West Bank in 2004 (OTML, 2006).
Around the year 2000, concern over potential acid formation (PAF) increased, whereupon the OTML Mine
Waste Tailings Project (MWTP) commenced in 2006 to remove sulphur (pyrite) from the tailings at the mill,
prior to disposal in the Ok Tedi (OTML, 2006). The pyrite concentrate (the removed sulphur) is piped from
the mill to Bige for long term storage in cells located beneath the water table under the West Bank stockpile.
The original intention of the MWMP was to dredge the mine waste from the river and construct semi-
permanent stockpiles on the flood plain. The stockpiles were assumed to erode in the medium term up to
10% of the annual stored volume, but with no unplanned release of material for approximately 100 years.
The expectation being that after mine closure and recovery of the river, gradual erosion of the stockpiles
over time would not significantly impact the system. In fact, minimal erosion occurred due to the adopted
surface water management practices including benching, runoff controls, and quick revegetation.
The design philosophy of the dredge program has evolved over the project’s life based on an increased
understanding of the impacts to the Ok Tedi River and the communities. The design criteria for the Bige
stockpiles subsequently changed to be stable long-term structures, and a closure plan to manage long-term
erosion and surface water has been developed. In this context, “long-term” design life of the post-closure
Bige stockpile landforms is defined as a nominal period of 200 years – greater than 100 years but less than
500 years (Logsdon et al., 2008).
Landform evolution modelling (LEM) carried out to assess how the structures may perform. The LEM
predicted that while some erosion should be expected, the stockpiles are stable in the long term under typical
conditions.
Because the Ok Tedi River runs between the East and West Bank stockpiles (Figure 2), the design criteria
examined whether river migration could impact the stockpiles. Studies undertaken predicted the river could
potentially meander eastwards, with a low chance of impacting the stockpiles within 50 to 100 years. An
average minimum offset of 500 m between the toe of the stockpile and the current river channel was
adopted, to reduce this risk.

Mine (~100km)
north
East Bank Stockpile
West Bank Stockpile
PCon Pit 1 North
PCon Pit 1 South
Dredge Slot
PCon Pit 3
It was recognised before mining commenced in 1984 that there was a risk of ARD and metal leaching as a
result of sulfidic deposits oxidising (OTML, 2008). However, ARD was not expected to be a significant issue
due to the high volumes of limestone waste rock combined with the naturally high alkaline concentrations in
the catchment runoff (OTML, 2008). By 2001, ARD studies undertaken by OTML indicated that the East Bank
stockpile could become potentially acid forming (PAF). The combination of increasing sulphur concentrations
and reducing acid neutralising capacity (ANC) of waste material caused by changing mine mineralogy (OTML,
2002) also indicated the PAF nature of future dredged sand was likely to increase over the remaining mine
life. As a result, OTML reviewed a number of options to mitigate ARD including: footprint and perimeter
length minimisation; external stream flow diversions to minimise contact water volumes; maximisation of
stockpile saturation levels; stockpile surface water management, including ability to collect contact waters
for treatment; installation of cut-off walls to intercept and collect foundation seepage flows; engineered
covers to minimise or maximise infiltration; improvement of fines retention; construction of an engineered
basal cut-off blanket; limestone addition across the desaturated stockpile regions to minimise ARD potential;
and options to increase the ANC/MPA ratio, including removal of sulphides from the dredge material.
In 2006, OTML initiated the Mine Waste and Tailings Project (MWTP) with one of the key objectives being to
reduce the ARD risk at the Bige stockpiles (OTML, 2006). Key outcomes from this project included a
commitment to install a new flotation plant at the mill to remove pyrite (sulphur) from tailings, excavation
of subaqueous storage pits on the West Bank at Bige for the storage of the pyrite concentrate (PCon),
construction of a PCon pipeline to Bige, and additional mining of limestone to provide more ANC to the
dredged sand. In 2008, the MWTP was commissioned and implemented, and the first PCon storage pit
(Pit 1 North) was filled with PCon between 2008 and 2010. Between 2010 and 2014, Pit 1 South has also been
commissioned and filled and is awaiting closure, and Pit 3 is now receiving PCon material (Figure 2).

Landform
The MWTP has been a success with sulphur content of river sediments, and hence ARD risk, gradually
decreased. From 2008 to 2013, sulphur concentration in dredged material has decreased from 3.0% to 1.6%.
As a result, the dredged material has, on average, changed from being PAF to NAF (OTML, 2014b). The
ANC/MPA target ratio for the dredge sediments under the MWTP’s was originally 1.5. This target has been
achieved since 2011, and for the year ending December 2013, the ANC/MPA ratio was 1.98 (OTML, 2014b).
A closure objective for the East and West Bank landforms is to limit the potential for acid generation within
the stockpile by reducing the availability of oxygen to PAF material within the dredge sand stockpiles (OTML,
2006). This is to be achieved by the following steps:
 Increase the elevation of the phreatic surface within the stockpile, as ARD generation is greatly
reduced below the phreatic surface.
 Decrease the rate of O2 available for ARD processes within PAF material above the phreatic
surface. The rate of atmospheric oxygen ingress into soil decreases with depth. Placing a NAF
cover over the PAF material means that the availability of atmospheric oxygen to ARD processes
within unsaturated PAF material is reduced. The simultaneous ARD and neutralising reactions
occurring within the NAF cover also consume oxygen and further reduce the rate of oxygen
diffusion into the underlying PAF materials.
In addition, the cover design also aims to reduce the formation of ARD hot spots on the surface area. This
requirement is intended to reduce the risk of revegetation failure through ARD hot spot formation. The
overall thickness of the required NAF cover for the East Bank stockpile varies– between 2 m and 4 m –based
on the results of close space geochemical sampling and characterisation. The final cover will be mulched,
seeded, and planted to aid in revegetation.
The intention of the original East Bank stockpile water management plan was to retain and attenuate rainfall
on the stockpile tops, before discharging through a series of spillways, bench, toe, and longitudinal drains;
and sediment control structures. The operations and proposed closure structures were typical of large waste
rock dumps and tailings dams, with the intention of controlling and directing runoff in lined structures to a
defined discharge location. This method was used on the perimeter of the operating stockpile, and a similar
approach was proposed for the closure stockpile.
The side slopes of the East Bank stockpile comprise near horizontal benches separated by 3H:1V, 4 m high
slopes that were formed for each lift of dredge sand. Overall slopes vary from 5.5H:1V to 20H:1V. A slope
management plan was implemented to manage seepage, control rainfall runoff erosion and reduce
development of bench slope washouts, (see Figure 3). The slope management plan included the following
elements:
 Benches: Benches were back-sloped towards the longitudinal drains to reduce uncontrolled
discharge over the bench and down the side slope face. This generally reduces slope washout and
erosion.
 Longitudinal drains: The longitudinal drain formed along each bench is designed to collect surface
runoff and seepage. Each longitudinal drain slopes towards a bench drain collection point.
 Bench drains: Bench drains collect seepage and surface water from the longitudinal drains and direct
the flow down the stockpile side slope to toe collection channels.
 Toe collection channels: The toe collection channels, at the downstream end of the bench drain, are
shaped for energy dissipation and to convey flow around the toe of the stockpile.

 Spillways and stilling basins: Runoff is directed by a series of spillways down the stockpile side slopes.
Chute channels were designed to carry peak discharges – up to the 1,000 year design rainfall event
– from the spillway inlet down the stockpile side slopes to the spillway stilling basin. The design
included stilling basins at the base of each spillway to reduce the hydraulic energy of the flow prior
to release into a sediment basin.
The design assumes that surface water flows will be detained on the stockpile crests to increase surface water
infiltration into the stockpile (and hence raise the phreatic surface) and help attenuate flows from extreme
rainfall events. This was achieved by designing a low gradient, labyrinthine flow path across the stockpile
surface to the spillway by constructing internal bunds across the hydraulically formed flow channel. The
design did not intend for significant ponding areas (greater than one-third of the stockpile surface area) to
occur on top of the stockpile under average rainfall conditions. During larger rainfall events, a pond may
form, which would attenuate peak runoff flows and reduce spillway capacity requirements. A perimeter bund
around the top of the stockpile provides freeboard to reduce the risk of uncontrolled release of water from
the top of the stockpile down the perimeter slopes.
The surface water management infrastructure including spillways and channels constructed with gabion
baskets and Reno mattresses was established during operations. The installation of drains, gabion baskets
and Reno mattresses was problematic due to the difficulty of sourcing suitably sized rockfill. The only locally
available rock is river gravel, and the available quantities were estimated during earlier studies to be
sufficient. Notwithstanding this, the local river gravel contains a relatively small percentage of suitable
oversize material, requiring some gabions to be filled with some undersized gravel.
Where gabions were installed in areas of low ANC/MPA dredged sand, some gabions became (due to impact
damage and corrosion) degraded and broke open. However, typically by the stage the gabions had degraded,
vegetation had established and the spillway channels had stabilised. Thus, the gabions acted as a sacrificial
measure to pass runoff flows while allowing the vegetation to take hold and provide a more permanent
solution. The use of alternative gabion types, such as stainless steel, was investigated for future installations
in low ANC/MPA zones; however, the estimated cost of installing these stainless steel units was prohibitive.

Landform
Although vegetation established naturally in the spillway channels, the spillway sizing remained capable of
passing the design flows.
OTML and KCB carried out regular monitoring, including formal twice-yearly (at a minimum) inspections to
observe the performance of the infrastructure. The stockpile was observed to perform suitably overall with
minimal erosion as observed by minimal accumulation of sediment in the sediment collection ponds.
Changes to the properties of dredged material and associated long term closure performance expectations
over almost 20 years of operations at Bige have driven changes to the design of the cover in terms of both
construction methodology and long-term closure requirements. The main drivers of the most recent design
changes are summarised below.
A key driver for an updated closure approach for the Bige landforms was the availability of a geochemically
and geotechnically stable cover material that was also able to be placed practically, economically and in a
timely manner.
Analysis of dredged material particle size and geochemistry, indicates the dredge slot in Ok Tedi tends to act
as a natural separator of this material. The coarse (and/or a higher SG) material settles out in the upstream
regions of the dredge slot (with a typical ANC/MPA ratio <1.5), this material is selectively placed in internal
locations within the West Bank stockpile to be covered later by the NAF cover material. Conversely the
material dredged from the downstream region is finer grained (generally with an ANC/MPA ratio of >2.0) and
selectively placed as final cover material.
A review of finer cover material properties was undertaken to assess potential constructability, water
management and geotechnical implications. This review identified a number of potential constructability
issues with using the finer material such as increased erodibility and loss of fines during the placement and
construction phase, reduction of infiltration into the stockpile and challenges to placement methods. The
loss of fines during the placement process (for example, during hydraulic placement the excess water is
decanted from the stockpile through a weir structure) was seen as a risk through decreasing the mean
ANC/MPA ratio and the dredge material placement efficiency thereby potentially increasing the time to place
cover material.
The results of this assessment concluded that placement of the finer cover material did not introduce any
significant additional geotechnical stability risk or materially change the overall runoff/infiltration
characteristics of the stockpile surfaces. However the selective dredging and hydraulic placement of these
finer cover materials on the final external stockpile slopes required a change to the existing stockpile
construction methodology and sequencing, given these finer material are somewhat more difficult to
hydraulically place, dewater and traffic.
OTML initiated studies into the stockpile closure landform in 2013 and a third-party review of the re-
vegetation program for the dredged sand stockpiles was undertaken. From this, a recommendation was
made to review the final landform design and assess the optimum batter and side-slope design.
This recommendation is supported by a number of studies which demonstrate that long term erosion is
increased and slope stability decreased when ‘traditional’ hard engineering design features such as berms,
benches and rock drains are employed on mine waste material landforms (Schor, 2007; McPhail and
Wilkinson, 2005).

The financial cost of closure planning is significant; in many cases, it is on a similar scale to the capital required
at the project outset (Howard et al., 2015). The often cited paper, “The Use and Abuse of Feasibility Studies”
(Mackenzie and Cusworth, 2007), highlights the need for comprehensive studies and planned financing
stages, but includes only a brief reference to the need to incorporate closure in these processes. Where
closure is considered from the outset, realising closure goals is more likely; early closure planning allows for
a progressive reduction of risks and unknowns (International Council on Mining and Metals, 2008) as well as
securing the required finances to allow for closure in an efficient manner.
The original stockpile closure plan envisaged primarily mechanical placement of cover on the slopes rather
than the more efficient and economical alternative of hydraulic placement of NAF dredged sand. In addition,
the cost of maintenance and replacement of the closure infrastructure, such as gabions and Reno mattress,
was predicted to be significant and the construction of hard infrastructure measures introduced scheduling
constraints.
Given OTML’s renewed focus on extending mine life, the development and implementation of a progressively
placed cover, using a lower cost construction methodology that meets the closure design criteria and
expectations, was essential to the continued success of the project.
As previously discussed, OTML initiated studies into the stockpile closure landform in 2013. From this, a
recommendation was made to review the final landform design and assess the optimum batter and side-
slope design. Subsequently, OTML reviewed and revised the closure design for the stockpiles, commencing
with an on-site workshop at the beginning of 2014. The design intent for surface management for the top
surface of the stockpile did not alter; however, the side slope design evolved greatly. As such, the remainder
of this paper focuses on the evolution of the cover design on the stockpile side slopes.
At the start of the cover options review, OTML and KCB undertook a high-level options assessment on
potential closure slope arrangements. The objective was to provide a qualitative understanding of the
relative merits of various overall slope angles and benches, prior to revising the closure design philosophy.
The varying parameters for each option are shown in Table 1. The ranking categories included
constructability, erosion loss, stability, and long-term maintenance amongst others.
Option 1 Option 2 Option 3 Option 4 Option 5

Overall stockpile slope 7.5H:1V 20H:1V 5.5H:1V
Individual batter slope 3H:1V 6H:1V 3H:1V 20H:1V 3H:1V
Nominal stockpile height (m) 24.0 24.0 24.0 24.0 24.0
No. of lifts 6 6 6 1 6
No. of benches 5 5 5 - 5
Bench width (m) 22 7 82 n/a 12
Total batter width (not including bench
72 144 72 480 72
widths) (m)
Total slope width (m) 180 180 480 480 132
Ranking (1 is best) 2 2 4 1 5

Landform
The move towards a softer engineering approach with fewer hard engineered structures such as gabions and
rock drains and potentially a flatter batter slope with no benches or berms, required further studies to
determine the most appropriate stockpile slope and batter configuration.
The general advantages of a benched slope during construction include reduced flow velocities, reduced
erosion, increased infiltration, and formation of a hydraulic break in the flow path. A continuous slope,
however, can produce a more sympathetic landform that mimics the natural surrounds and produces a more
consistent context for revegetation, but requires a longer overall slope length if erosion is to be managed
without lined runoff channels.
A key to selecting the preferred closure option was to assess whether the steeper slopes needed for Options
1 and 2 were preferable to the flatter but longer slopes for Option 4. Thee comparison of steeper or flatter
slopes requires evaluation of two competing slope effects: the decrease in overall slope length which will
reduce the area exposed for soil loss; and, the increase to the slope angle which will increase the soil loss per
unit area. To establish this relationship and to help develop the optimal landform design, the following four
scenarios, varying bench width, number of benches, and slope angle, were considered:
 Scenario A: continuous slope (no benches), varying slope angle;
 Scenario B: varying number of benches, fixed slope angle overall;
 Scenario C: fixed interbench slope, varying bench width, varying overall slope angle; and
 Scenario D: varying bench width, varying interbench slope angle, fixed number of benches, fixed
overall slope angle.
The Revised Universal Soil Loss Equation (RUSLE) (Hayes et al., 1994) was used to analyse various stockpile
slopes from 2.5H:1V through to 20H:1V and assess the potential erosion from each slope angle. An arbitrary
stockpile height of 20 m was used for all scenarios for comparison purposes, and it was assumed that hard
engineering runoff controls were not in place (this mimics a worst-case long-term closure condition).
Values used in this RUSLE assessment were for comparative purposes and may not represent actual field
parameters at the Bige site. However, as the purpose of this assessment was to compare various benches,
slope lengths, and slope angles, the parameters adopted for soil factors, conservation factors, and cover
management factors were kept the same for each scenario rather than trying to duplicate field monitoring
data. The RUSLE assessment is not a landform evolution model; LEM modelling would be expected to produce
a markedly different result, with the inclusion of gully, wind, and rainfall impact erosion.
Figure 4 shows that the soil loss per unit area is highest for this site at an overall slope of about 7.5H:1V,
while the total soil loss decreases monotonically as the slope flattens. For slopes flatter than 7.5H:1V, both
the erosion per unit area and the overall erosion reduce monotonically with reduced slopes.
Figure 5 shows that adding benches on a 7.5H:1V slope reduces the calculated potential erosion loss. As the
flow paths down the face of a slope are broken, the flow velocities and peak discharges reduce. Increasing
the number of benches (i.e., reducing the lift height between each bench) has a positive influence on
potential soil loss because of the reduced hydraulic flow length on each individual slope. The maximum
number of benches will be limited by practical considerations such as constructability and scheduling.
However, comparing Figures 4 and 5 suggests that adding four benches to a 7.5H:1V slope will result in a
reduction of total erosion from 18 t/ha per year to about 12 t/ha per year. Soil loss from a flatter overall
slope, such as 20H:1V as shown in Figure 4, is much lower – about 4 t/ha per year – even without benches.

Consideration of the options assessment, erosion analysis, and the practicalities of dredging and dredge
material placement led to a revised landform design for cover placement on the stockpile side slopes as
follows:
 Flattening closure side slopes to approximately 15H:1V;
 Providing a continuous slope with no benches; and
 Using swales formed by corrugations or undulations on the uniform slopes, which segment the
slopes into small catchments, avoid flow concentrations on the slope, and avoid the need for hard
engineering structures such as gabion-lined bench drains.
Note that runoff from the stockpile top surface still required the use of two engineered spillways, an internal
spillway between the two sections of the stockpile, and an external spillway discharging runoff from the top
of the stockpile into the local creek diversions. Engineered spillways are required due to the large peak flows
for closure; these spillways were largely unchanged from the original design and are not discussed further in
this paper.
Surface water management on the uniform 15H:1V slopes includes constructing swales lined with gravel and
spaced 50 m apart along the lower sections of the slope before revegetating the slope (refer Figure 6).

Landform
Trials of the cover placement using hydraulic placement of the finer dredge material were undertaken on the
West Bank in 2014, prior to embarking upon full scale cover placement on the East Bank stockpile.
The trial cover on the West Bank stockpile side slopes was placed in 2 m high lifts separated by berms to form
an overall 15H:1V slope. The berms were built using the placed dredge material, and sand was pumped
behind the confining berms similar to the hydraulic methods used for placement of the stockpiled sand.
Following placement of the overall bermed slope, the side slope was mechanically reshaped to achieve a
uniform slope and then further reshaped to create the drainage swales (corrugations).
The trials to date, demonstrate that placement of cover material using hydraulic methods combined with
subsequent mechanical reshaping is a more cost effective and faster technique than solely mechanical
placement of cover material.
One of the lessons learned from the West bank trial areas was the treatment of the emergent seepage zones.
These are areas located towards the toe on the side slopes, where seepage through the sand stockpile (either
from infiltration into the top of the stockpile or from higher up on the slope) emerges and creates a saturated,
looser area of material. These seepage zones, which had been experienced elsewhere on the East Bank
stockpile, are managed using vegetation covers and/or gravel filters to prevent loss of fines along with
seepage flows.
Another observation from the trial cover occurred after forming the swale drains (but prior to installing gravel
material in the invert),; a large storm event occurred, which resulted in erosion of the swale invert and caused
significant gullying of the slope (see Figure 7). Gully erosion originated in the saturated, looser areas of the
side slope where emergent seepage was present, and the erosion head worked its way up the slope assisted
by overland flow. The area was treated by excavation of loose material and placement of gravel mulch. A
further learning was to progressively reshape the swales and place the gravel/filter/vegetation in the
emergent seepage zones, rather than reshape large areas then have to repair and remediate.
Subsequent observations in early 2015 indicated that, once the downhill third of the swales were lined with
gravel as required by the design, they appeared to be operating well and were reducing restricting erosion
of the placed cover material (see Figure 8).

The current mine plan allows for mining at Ok Tedi until at least 2025, with dredging indented to continue at
Bige for a further year before five years of additional closure works are carried out. Although mine closure is
still at least 10 years away, OTML has already closed out sections of the East Bank, and it began scheduling
and progressive constructing the closure cover on the West and East Bank stockpiles in 2014. There are
approximately 1,000 ha of surface area which requires cover placement with an estimated cover material
volume of some 35 million m3 required. This equates to approximately 3.5 million m3 of cover to be placed
per year, over the next ten years. To place this volume of cover material using mechanical methods (for
example dump trucks, dozers, and compactors), would require a significant increase in operational
earthworks machinery requirements and associated significant increase in cost. The use of hydraulic
placement of cover material means that the cost of placement is not substantially different placing dredged
sand. Typically, the average unit cost of hydraulically placing dredged sand is around US$ 4 per cubic metre,
compared to an estimated unit cost of up to US$ 15 per cubic metre for mechanically placed cover material.
In addition to the placement costs, a further US$ 25 to US$ 45 million would have been required to construct
the hard engineered system of benches, interbench drains and spillways, and rock-filled gabions required for
the East and West Bank stockpile covers. The mechanical placement of some 35 million cubic metres of cover
material was therefore considered both uneconomical and impractical, whereupon the soft engineering
design approach using hydraulic placement methods was adopted by OTML.
With an estimated 35 million cubic metres of cover material requiring placement over the next ten years, an
updated closure approach was required. The key driver for a new landform design is the availability of cover
material that can be placed practically, economically and in a timely manner and meets the closure objectives
including being geochemically and geotechnically stable.

Landform
The previous landform design of the East Bank stockpile incorporated side slopes of 7.5H:1V to 20H:1V and
a series of hard engineered structures, including gabion baskets, Reno mattresses, benches, graded bench
drains and interbench spillways to facilitate surface water management. This water management system has
operated successfully over the past 15 years on rehabilitated sections of the stockpile.
Recent developments, however, led to a need to reassess the closure design for the remaining areas of both
the East Bank and West Bank stockpiles. Changes in closure design criteria expectations, material availability,
scheduling issues, and construction practicalities associated with placing a NAF closure cover, necessitated a
change in design philosophy for the surface water management system. The use of soft engineered landform
including uniformly sloped side slopes, vegetated drainage swales, and ephemeral surface wetlands has been
developed and trialled. The trials have been successful in showing that progressive placement of cover
material using slightly modified hydraulic methods, provides a practical and economic approach to meet the
overall stockpile closure design criteria. In addition, the trials have identified opportunities to improve the
landform construction methodology, prior to embarking upon full scale cover placement.
The intention of the revised closure design approach is to provide a landform legacy that is sympathetic to
the natural environment while enabling a stable long-term closure cover that is resilient to the effects of
erosion and damage from rainfall events. Regular review of the stockpile performance and mine plans have
provided OTML with the opportunity to progressively incorporate lessons learned over the life of the Bige
stockpile into the closure planning process.
Howard, B., Pershke, D. and Jones, M. (2015) Better mine closure outcomes for less money, The AUSIMM Bulletin, February 2015,
pp. 52–55.
International Council on Mining and Metals (ICMM) (2008) Planning for integrated mine closure: Toolkit, viewed 2 February 2015,
http://www.icmm.com/document/310.
Hayes, J., Haan, C. and Barfield, B. (1994) Design hydrology and sedimentology for small catchments, Academic Press.
Logsdon, M., Blowes, D., Wilson, W., Smith, L. and Portigal, M. (2008) Storage, management and monitoring of Ok Tedi pyrite
concentrates at Bige, Lower Ok Tedi, western province, Papua New Guinea: final report of the independent review team
concerning condition 10. Minister for Environment and Conservation Papua New Guinea, viewed 12 February 2015,
http://www.oktedi.com/media-items/reports/environmental/acid-rock-drainage/146-storage-management-monitoring-of-
ok-tedi-pyrite-concentrates-at-bige-condition-10-review-final-report/file.
Mackenzie, W. and Cusworth, N. (2007) The use and abuse of feasibility studies, Proceedings AusIMM Project Evaluation 2007, 19–
20 June 2007, Melbourne, Australia, pp. 65–77.
McPhail, G. and Wilkinson, D. (2005) Section 3: Planning – landform design concepts, creating cost effective rock dumps and
stockpiles, Perth.
Murray, L., Thompson, M., Voigt, K. and Jeffery, J. (2000) Mine waste management at Ok Tedi Mine Papua New Guinea: a case history,
Proceedings of Tailings and Mine Waste, 10, Colorado.
Ok Tedi Mining Limited (OTML) (2002) Annual Environment Report FY02. ENV020923 OTML Environment Department, viewed 12
February 2015, http://www.oktedi.com/media-items/reports/environmental/annual-environmental-reports.
Ok Tedi Mining Limited (OTML) (2006) Mine Waste Tailings Project Change Notice: Supporting Document.
Ok Tedi Mining Limited (OTML) (2008) Annual Environment Report FY08. ENV080921. Ok Tedi Mining Limited (OTML), Environment
Department, viewed 12 February 2015, http://www.oktedi.com/media-items/reports/environmental/annual-environ
mental-reports.
Ok Tedi Mining Limited (OTML) (2013) Annual Environment Report FY13. ENV130920 Ok Tedi Mining Limited (OTML), Environment
mental-reports.
Ok Tedi Mining Limited (OTML) (2014a) Annual Environment Report FY14. ENV140923. Ok Tedi Mining Limited (OTML), Environment
mental-reports.
Ok Tedi Mining Limited (OTML) (2014b) Annual Review 2013. Ok Tedi Mining Limited (OTML), viewed 12 February 2015,
http://www.oktedi.com/media-items/publications/annual-review.
Ok Tedi Mining Limited (OTML) (2015) Bige operations, viewed 12 February 2015, http://www.oktedi.com/our-operations/bige-
operations.
Schor, Horst J. 2007. Landforming: an environmental approach to hillside development, mine reclamation and watershed restoration.
Hoboken, NJ: John Wiley & Sons.


J.B.C Lowry Environmental Research Institute of the Supervising Scientist, Department of the Environment,
Australia
G.R. Hancock School of Environmental and Life Sciences, University of Newcastle, Australia
T.J. Coulthard Department of Geography, University of Hull, United Kingdom
An important goal of mine closure is the development of an erosionally stable landform that functionally
merges with the surrounding landscape. Poor landform design may result in severe erosion and gullying,
which may deliver excess sediment to the surrounding undisturbed landscape. In addition, erosion may result
in the exposure of hazardous material that was intended to be contained within the landform over geological
time, thereby posing an environmental risk. These risks mean that it is important that the erosional stability
of a landform be carefully assessed. In this study, a conceptual rehabilitated landform of a uranium mine in
northern Australia is assessed over a 1,000-year period using both the SIBERIA and CAESAR-Lisflood computer-
based landscape evolution models (LEMs). LEMS with the capability to capture a range of surface (erosion
and deposition), tectonic and near surface or critical zone processes (i.e. pedogenesis) have been developed.
Using two LEM models enables both independent evaluation of likely landscape evolution processes and
assessment and comparison of the relative performance and output of each model. Overall findings show that
SIBERIA and CAESAR-Lisflood produce erosion rates and patterns that are broadly similar. At millennial time
scales, short-term processes such as gullying appear to be the dominant erosion features on the proposed
landform, resulting in substantial erosion features in terms of size and amount of hillslope material eroded
and transported downslope. Overall, both models produce very similar results, highlighting the usefulness of
such modelling in terms of design assessment.
However, the use of LEMs to simulate extended periods of time raises a number of important issues. Current
knowledge of earth surface processes is based on long-term empirical understandings, short-term field
measurements and numerical models. In this paper, the ability to calibrate and parameterise models for mine
site rehabilitation over millennial time scales is investigated, specifically through assessing the effects of the
initial landscape dimensions and associated error as well as the parameterisation used to represent a
potential post-mining environment. The results demonstrate that subtle surface changes and
parameterisation can have a large impact on landscape behaviour, erosion depth and sediment discharge.
Rehabilitated mine sites, if not properly managed, have the potential to be the source of erosion from
disturbed areas such as waste rock dumps and tailings containment structures. Rehabilitation planning and
landform design therefore need to consider landform shape and surface treatments that reduce erosion and
encapsulate contaminants. This is particularly pertinent in the wet–dry tropics of northern Australia, which
experience an annual average rainfall of 1,584 mm (Bureau of Meteorology, 2014), most of which falls in the
wet season between October and April. The Ranger Uranium Mine operated by Energy Resources of Australia
Ltd. (ERA) falls within this geographic region. It currently accounts for about 10% of the world’s annual
production of uranium oxide. Mining of the open-cut orebody ceased in 2012, and milling and production
are scheduled to cease by 2021. The unique location of the mine site (Figure 1) – within the catchment of
Magela Creek, surrounded by the World Heritage–listed Kakadu National Park and upstream of floodplains
and wetlands listed as “Wetlands of International Significance” under the Ramsar Convention – has meant

Assessing the post-mining landscape using landform evolution models at millennial time scales J.B.C. Lowry, G.R. Hancock and T.J. Coulthard
that the development of closure criteria and rehabilitation plans for the mine has required special
consideration.
Given the environmental and cultural sensitivity of the area surrounding the mine, it is important to
determine the likely erosional stability of the proposed rehabilitated landform through time, to ensure that
post-closure environmental protection objectives are met. The environmental requirements for mine closure
specify that the final landform should possess “erosion characteristics which, as far as can reasonably be
achieved, do not vary significantly from those of comparable landforms in surrounding undisturbed areas”
(Supervising Scientist Division, 1999).
The rehabilitation of a uranium mine provides a further set of challenges beyond the climatic, cultural and
environmental characteristics of the region. Radionuclides represent a potential set of contaminants with
long half-lives and persistence in the environment (Schumm et al., 1984). Australian guidelines recommend
a design life for a tailings cap of a uranium mine of 200 years and a structural life of at least 10,000 years
(Commonwealth of Australia, 1987).
Landform evolution models (LEM) provide a means for assessing the long-term performance of constructed
mine landforms. Over the last 40 years, a variety of models have been used to evaluate erosion and simulate
post-mining landscape stability (Riley, 1994; Evans and Loch, 1996; Evans, 2000; Loch et al., 2000). These
models include the water erosion prediction programme (WEPP) (Laflen et al., 1991; Flanagan and Livingston,
1995), universal soil loss equation (USLE), modified universal soil loss equation (MUSLE), revised universal
soil loss equation (RUSLE) (Onstad and Foster, 1975; Wischmeier and Smith, 1978; Renard et al., 1994), and
SIBERIA (Willgoose et al., 1989). More recently, the CAESAR-Lisflood model (Coulthard et al., 2013) has been
modified and applied to study the evolution of proposed rehabilitated mine landforms in northern Australia
(Hancock et al., 2010; Lowry et al., 2011; Saynor et al., 2012).
Significant issues for LEMs are their (1) calibration, (2) parameterisation and (3) validation. Before use, each
model requires a set of data and parameter values for it to run reliably and, most importantly, produce results
with some measure of precision and accuracy. LEMs are now at the stage of development where, if
calibrated, they can provide meaningful and useful results. For example, Lowry et al. (2014) demonstrated
that model results are able to very closely match field-measured observations for periods of up to five years.
However, these models are largely calibrated with parameter values determined from current, or
comparatively recent, surface conditions that are themselves products of much longer-term geology-soil-
climate-vegetation interactions. How these parameters temporally vary based on both climate variability,
weathering and pedogenesis and the resultant soil-climate-vegetation interaction is largely speculative, with
insights based on qualitative assumptions. How these parameters may then be taken forward for hundreds
or thousands of years of future simulations is currently a source of great uncertainty.
In this paper, we use the CAESAR-Lisflood model to do an initial assessment of the geomorphic stability of a
catchment of a proposed rehabilitated landform of the Ranger mine for a simulated period of 1,000 years.
We concurrently use the SIBERIA model to independently assess the stability of the landform and the
reliability of the CAESAR-Lisflood predictions. We also identify and discuss some of the key issues for
modellers involved in long-term simulations, specifically the importance of parameter values and datasets
used as inputs to long-term simulations.
LEM were initially developed to examine landscape evolution and dynamics at geological time scales. In more
recent years they have been employed in more applied settings (i.e. mine sites) at much shorter time scales
(years, decades, centuries) than geological time scales.

Landform
The CAESAR-Lisflood landscape evolution model (Coulthard et al., 2013; Van de Wiel et al., 2007) simulates
landscape development by routing water over a regular grid of cells and altering elevations according to
erosion and deposition from fluvial and slope processes. The initial topography of the landscape drives fluvial
and hillslope processes that determine the spatial distribution of erosion (loss) and deposition (gain) that
occurs during a given time step. This altered topography becomes the starting point for the next time step.
Outputs of the model are elevation and sediment distributions through space and time, and discharges and
sediment fluxes at the landform outlet(s) through time.
Runoff over the landform is generated through the input of rainfall data. This is calculated using an adaptation
of TOPMODEL (Bevan and Kirkby, 1979). Surface flow is routed using Lisflood-FP (Bates et al., 2010), a two-
dimensional hydrodynamic flow model. Morphological changes then result from entrainment, transport and
deposition of sediments over nine size-based fractions, which are transported as either bedload or
suspended load, depending on the grain sizes. CAESAR-Lisflood provides two different methods of calculating
sediment transport, based on the Einstein (1950) and the Wilcock and Crowe (2003) equations. The Einstein
(1950) method was used for this study. In addition, soil creep and landslips are simulated.
A key attribute of the CAESAR-Lisflood model is the ability to use hourly recorded rainfall data from the study
area, enabling the modelling of the effects of specific rainfall events. Event modelling is critical, especially for
the early stages of landform evolution, since it is recognised that the majority of erosion typically occurs
during a limited number of high-intensity events (Moliere et al., 2002). As the climatic region in which the
Ranger mine occurs is dominated by seasonal, high-intensity rainfall events (McQuade et al., 1996), the ability
to model specific rainfall events meant that the CAESAR-Lisflood model was the model of choice for this
project.
Recently, CAESAR-Lisflood has been modified and applied to study the evolution of proposed rehabilitated
mine landforms in northern Australia (Hancock et al., 2010; Lowry et al., 2011,2013; Saynor et al., 2012).
SIBERIA is a mathematical model that simulates the geomorphic evolution of landforms subjected to fluvial
and diffusive erosion and mass transport processes (Willgoose et al., 1991). Importantly, SIBERIA describes
how the catchment is expected to look, on average, at any given time. The sophistication of SIBERIA lies in
its use of digital elevation models (DEMs) for the determination of drainage areas and geomorphology and

its ability to efficiently adjust the landform with time in response to the erosion that occurs on it. The model
itself and its application have been well described elsewhere (for a full description, see Willgoose et al., 1991;
Hancock et al., 2002, 2014a).
For all simulations, SIBERIA requires calibration of the sediment transport and area-discharge relationships,
and a DEM of the landform of interest (described in Section 3.3). The fluvial sediment transport equation is
parameterised using input from field sediment transport and hydrology data.
For the simulations detailed here, we examine the Corridor Creek catchment, which incorporates Pit 1, the
first part of the Ranger mine landform to be rehabilitated. The catchment area is derived from a proposed
design for a rehabilitated landform as supplied by ERA. The proposed landform is considerably different to
that of the present mining landscape surface (Figure 2). In this section, we describe the methods and input
parameters used to simulate periods of 1,000 years using the SIBERIA and CAESAR-Lisflood models.
Before SIBERIA can be used to simulate soil erosion, the sediment transport and area-discharge relationships
require calibration. The model also requires a DEM of the landform of interest (described in Section 3.3). The
fluvial sediment transport equation is parameterised using input from field sediment transport and hydrology
data. For this study, the SIBERIA model was calibrated using field data collected on the ERA Ranger Mine site,
the surrounding undisturbed (Koolpinyah) surface and an analogue soil site (Tin Camp Creek, approximately
50 km northeast of the Ranger Mine).
To calibrate the erosion and hydrology models, complete datasets of sediment loss, rainfall and runoff for
discrete rainfall events were collected, allowing calibration from field plots (Table 1). Parameters for the ERA
Ranger site were derived from plots on the batter slope of the waste rock dump (Evans et al., 2000; Moliere
et al., 2002). This plot (38 m long by 16 m wide [608 m2]) rose approximately 12 m above the surrounding
land surface and had an average slope of 20.7%. The site was covered with an armour of coarse material and
had negligible vegetation cover. Parameters for the vegetated surface condition were determined from a
similar plot covered in topsoil, ripped and vegetated with low shrubs and grasses that provided
approximately 90% cover (Evans et al., 1998).

Landform
The runoff and erosion characteristics of the surrounding natural or undisturbed (Koolpinyah) surface were
derived from a runoff and erosion plot (30 m long by 20 m wide [600 m2]) constructed within the ERA Ranger
Mine site on a natural undisturbed area (Moliere et al., 2002). The plot had an average slope of 2.7%. The
site was covered with small trees, low shrubs and, during the wet season, speargrass. The surface of the site
is representative of the landscape before the commencement of mining operations and is typical of natural
areas within the region.
SIBERIA parameter
Surface type Comparable site
m1 n1 3 m3 1
Mine pit and Ranger waste rock
waste rock dump (Moliere et 2.52 0.69 0.00016 0.81 27743
dump al., 2002)
Vegetated, ripped
surface (Evans et 1.59 0.69 0.000006 0.90 2088
Vegetation al., 1998)
Unlike the SIBERIA model, the CAESAR-Lisflood model is able to simulate individual storm events (driven by
the hourly rainfall record) and flood flows hydrodynamically. In addition, CAESAR-Lisflood is able to simulate
erosion and deposition over a range of grain sizes (as opposed to the erosional properties of a bulk sample
in SIBERIA). Consequently, in addition to a DEM of the study site, CAESAR-Lisflood requires soil/sediment
particle distribution data and rainfall data (mm hr-1).
Twenty-two years of complete (i.e. no gaps in the annual dataset) hourly rainfall intensity data are available
for the period between 1971 and 2006 for Jabiru airport, two kilometres northwest of the Ranger Mine site.
Additional rainfall data from an extreme rainfall event from March 2007 was available for the same site. In
the latter event, 785 mm of rainfall were recorded in the three-day period between 27 February and 2 March;
rainfall intensity in this period exceeded a one-in-100 year storm event.
The one-hour rainfall totals (mm) were used to create the following rainfall inputs and simulate a range of
different scenarios:
 A 1,000-year dataset in which the 22 years of complete data were looped out to a period of 1,000
years. This dataset did not contain the 2007 rainfall event.
 A 1,000-year dataset in which the 22-year rainfall dataset was looped twice, out to a period of 44
years. The data from the 2007 extreme were added to this dataset, and the resultant 45-year dataset
then looped out to a period of 1,000 years.
For the purposes of this study, the grain size data of two different surface materials were modelled. Grain
size data representing waste rock were obtained from an eight-hectare trial rehabilitated landform at Ranger
in the 2009 dry season. Grain size data for Koolpinyah surface (or natural, undisturbed material) were
obtained from the catchment of Gulungul Creek. Grain size distributions from these two surfaces are shown
in Figure 3. The sub–0.00063 m (i.e. 63 µm) fraction is treated as suspended sediment within CAESAR-
Lisflood.

Waste rock Koolpinyah

35 35
30 30
Percentage
Percentage
25 25
20 20
15 15
10 10
5 5
0 0
0.02
0.000125
0.001
0.002
0.004
0.095
0.0005
0.00025
6.3E-05
0.000125
0.001
0.002
0.004
0.016
0.064
0.0005
0.00025
6.3E-05
Grain size (m) Grain size (m)
A DEM representing a fully consolidated conceptual landform was supplied by ERA (Figure 2).
The conceptual landform DEM was generated through the integration of two-metre-interval contour data
produced from a LiDAR survey of the mine in 2011, with a conceptual landform design developed for ERA by
the University of New South Wales Water Resources Laboratory using one-metre interval contour data. These
datasets were used to generate a grid surface with a horizontal spatial resolution of 10 m.
The CAESAR-Lisflood and SIBERIA models used the same DEM for their simulations.
Comparisons of measured and predicted bedload from the Ranger trial landform demonstrated the need to
incorporate a weathering function into the CAESAR-Lisflood model (Lowry et al., 2014). Consequently, the
CAESAR-Lisflood model has been modified to incorporate a physical weathering function based on the
parameters developed by Vanwallegham et al. (2013). In this study, the default weathering parameters
developed by Vanwallegham et al. (2013) for bedrock lowering, physical weathering and chemical weathering
were applied to selected model scenarios.
SIBERIA does not have a specific weathering component.
CAESAR-Lisflood currently has a relatively simple vegetation component, which has the effect of restricting
erosion. Two variables are used: “grass maturity” and “vegetation critical shear.” Grass maturity relates to
the speed with which vegetation reaches full maturity in years. For the purpose of these simulations, a value
of two was used, implying that a mature cover of grass was established across the entire surface in two years.
Vegetation critical shear is the value above which vegetation will be removed by fluvial erosion. The lower
the critical shear, the more easily vegetation is swept away; the higher the value, the more resistant the
surface. For the purposes of these simulations, a critical shear value of seven was applied. This implies the
vegetation is resistant to erosion. In this approach to modelling, a vegetative cover essentially increases the
physical resistance of the DEM surface to the erosive force of flowing water by allowing the development
through time of a surface cover of grass. Importantly, this is a bulk parameter approach that does not account
for specific physical properties of vegetation. To date, the parameters used in model simulations on the
Ranger landform are not based on bush, grass or tree communities in the north Australian environment.
Instead, parameters have been determined through an iterative testing program to determine parameter
values that best match erosion figures for vegetated environments in the Ranger lease. A similar approach
was used in the development of the vegetated cover parameters for SIBERIA.

Landform
Table 2 summarises all the scenarios that were modelled using CAESAR-Lisflood. All scenarios were run for
simulated periods of 1,000 years.
CAESAR-Lisflood simulations were divided broadly into “best-case” (vegetated) and “worst-case”
(unvegetated) scenarios. In total, 12 scenarios were modelled using CAESAR-Lisflood.
SIBERIA was used to model two scenarios, each for a simulated period of 1,000 years:
 a landform modelled using only parameters collected from a vegetated surface;
 a landform modelled using only parameters collected from a waste rock surface.
Rainfall Weathering No weathering

scenario
Single surface Two surfaces Single surface Two surfaces
No
Vege- No vege- Vege- No vege- Vege- No vege- Vege-
vege-
tation tation tation tation tation tation tation
tation
2007
extreme
rainfall X X X X X X X X
every 45
years
2007
extreme
rainfall X X X X
event
absent
The simulations conducted to date indicate that CAESAR-Lisflood and SIBERIA predict similar results for the
same model scenarios in terms of the areas in which erosion is likely to occur. Specifically, CAESAR-Lisflood
and SIBERIA predict the formation of gullies in approximately the same positions in relation to the conceptual
Pit 1 landform.
Initial simulations using CAESAR-Lisflood focussed on assessing potential worst-case scenarios. These were
defined as scenarios in which the landform was unvegetated and composed of two surface classes (waste
rock and Koolpinyah) while being subjected to the 2007 extreme rainfall event at intervals of 45 years over a
simulated period of 1,000 years. While it is recognised that this scenario is very unlikely (in terms of both the
absence of vegetation and the high return interval of an extreme event), it was run to identify areas on the
landform most susceptible to erosion, and the maximum amount of erosion that could occur. As shown in
Figure 4, this simulation predicts the formation of several gullies to the north of the capped Pit 1 landform of
up to seven metres deep. The formation of these gullies is a potential concern if contaminated materials are
buried at depths less than the depth of erosion. The simulations predict initial gully formation to develop
within 10 years of the start of the simulation/completion of the landform and then to progressively expand
and deepen over time.

The incorporation of a vegetation component into the simulation is predicted to lead to a decrease in the
overall sediment produced (Table 3). Interestingly, the presence of vegetation, while reducing the total
sediment loads, appears to result in a greater degree of incision into the landform, resulting in the formation
of gullies up to eight metres deep to the north of Pit 1 (Figure 5). The predicted formation of a gully on the
southern edge of the Pit 1 cap up to three metres deep is a potential concern, as it appears to erode directly
into the cap. Further erosion of this gully may potentially lead to the exposure of buried contaminants, and
their transport further downstream.
Removing the 2007 extreme rainfall event from the model simulations appears to reduce both the total load
produced (Table 3) and the extent and distribution of the gullies predicted to form (Figure 6).
A summary of the total sediment loads and denudation rates from these scenarios is contained in Table 3.
The relative proportions of the total sediment load in all the scenarios modelled were similar, with the
majority (80–90%) composed of bedload and a much smaller proportion (10–20%) of suspended material.
The simulations that modelled the landform with both the waste rock and Koolpinyah surface conditions
produced higher total loads and denudation rates than those scenarios in which the entire landform was
treated as a single waste rock surface. The latter scenario, while producing lower sediment yield, is
unrealistic, as the entire catchment model area will not be composed of waste rock when the landform is
rehabilitated.
While a weathering function has been specifically developed within CAESAR-Lisflood to simulate the
weathering of materials on the landform, it has not been possible to distinguish the specific effects of the
weathering function in simulations to date by using default weathering parameters. Further sensitivity
testing of the weathering function is therefore required.

Landform
45-year recurrence No 2007 extreme event

Single surface Two surfaces Single surface Two surfaces
No No
Vege- No vege- Vege- Vege- No vege- Vege-
vege- vege-
tation tation tation tation tation tation
tation tation
Total load
(m3) 531,039 757,625 575,390.7 827,126 387,369 695,917 432,260 656,682
Denuda-
tion rate 0.12 0.18 0.13 0.19 0.09 0.16 0.10 0.15
(mm y-1)
Predicted denudation rates for the CAESAR-Lisflood simulations are higher than published rates (0.01 to 0.04
mm y-1) of natural denudation for the region determined using stream sediment data from a range of
catchments of different sizes (Cull et al., 1992; Erskine and Saynor, 2000). The higher rates predicted by the
CAESAR-Lisflood simulations are likely due to the very high “first flush” of sediment produced as the newly
constructed surface undergoes equilibration. That is, finer particles on the fresh surface are preferentially
eroded and irregularities in the model DEM surface are removed, resulting in initially higher sediment yields.
This is an important strength of the model and provides a representation of what the erosion behaviour is
likely to be when the landform is first constructed. This is borne out by the good agreement between field-
measured and CAESAR-predicted erosion rates for the Ranger trial landform (Lowry et al., 2011; Saynor et
al., 2012). As the length of the simulations increases, the denudation rates will continue to reduce, reflecting
the length of time for which the data were averaged.

SIBERIA, while not using the same rainfall files as CAESAR-Lisflood, nonetheless predicts the formation of
gullies in similar areas under both waste rock and vegetated parameters (Figure 7). As with the CAESAR-
Lisflood simulations, more extensive gully formation is predicted to occur under waste rock parameters.
However, SIBERIA appears to predict the formation of more gullies than does CAESAR-Lisflood. The predicted
sediment loads of the different SIBERIA scenarios are summarised in Table 4.
At present, it is not possible to examine the landscape under varying rainfall scenarios with SIBERIA as done
with CAESAR-Lisflood. The current parameter set represents average annual or typical long-term erosion.
While possible, the determination of extreme or high parameter sets has not been undertaken. The
parameterisation of high rainfall and extreme events in SIBERIA is currently being investigated.
Vegetation Waste rock
Total Load (m3) 247,270 287,655
Denudation rate
0.06 0.07
(mm y-1)
Overall, the results of the simulations indicate that the CAESAR-Lisflood model predicts higher total bedloads
and denudation rates than does the SIBERIA model. This is likely due to CAESAR-Lisflood utilising actual hourly
rainfall data, as opposed to average rainfall data over a simulated period.
A significant issue for LEMs is their calibration/parameterisation and validation of any results. Before use,
each model requires a set of data and parameter values for it to run reliably and, most importantly, produce
results with some measure of precision and accuracy. LEMs are now at the stage of development where, if
calibrated, they can provide meaningful and useful results. Lowry et al. (2014) demonstrated that calibrated
model predictions can produce comparable results to field measurements for periods of up to four years.
Crucially, however, these models are largely calibrated with parameter values determined from present or

Landform
comparatively recent surface conditions that are themselves a product of much longer-term geology-soil-
climate-vegetation interactions.
Gullies Gullies
The employment of different parameter sets produces geomorphically and hydrologically unique landscapes
throughout the evolution of a landform while it is being modelled. Specifically, changing the frequency of
major rainfall events, the type of surface cover and the presence and absence of a vegetation component
have all been shown to have an effect. Therefore, careful parameterisation of a model scenario does matter.
A significant issue here is that the parameters employed are spatially and temporally stable, and over longer
simulations, these may change. It is well known that soils continually evolve; in particular, the quartz-mica-
schist that largely makes up the waste material has a relatively rapid rate of weathering (Wells et al., 2008).
How this material weathers and forms soil is unknown except from modelling studies (Cohen et al., 2010).
SIBERIA does not simulate pedogenesis, and further work is required to calibrate the weathering functionality
of CAESAR-Lisflood. However, there is the potential to examine the effect of armouring. Neither model
currently incorporates spatially changing vegetation.
SIBERIA simulations using parameters determined from vegetated plots at the ERA Ranger Mine demonstrate
a clear reduction in erosion (Hancock et al., 2010). As with CAESAR-Lisflood, the introduction of vegetation
via an increased soil shear strength threshold effectively reduces erosion. While it is assumed that initial rates
of erosion from the landform will be high, as that is when the landform will be at its most exposed and
unprotected stage, it is considered unlikely that the high erosion rate will continue for longer than 10 years.
This is because it is anticipated that a vegetation community will have developed from planted shrubs and
trees (tubestock), some soil formation will have occurred and a surface armour typical of the host materials
will have developed within this timeframe.
It has been suggested that waste rock materials at the Ranger site may evolve to be similar to soils at Tin
Camp Creek (Uren, 1992). These soils have been shown to have high erosion rates and be subject to gullying
(Hancock et al., 2014b). The process and rate of this evolution and likelihood of gullies is speculative at this
stage. However, these are all possible pedogenic and landscape trajectories. Further considerations include
that of climate variability and how this affects weathering, pedogenesis and overall landscape evolution, and
the impact of any extreme events.
For northern Australia, climate models predict an increase in both intensity and frequency of extreme rainfall
events (CSIRO, 2007). The model results here suggest that any contaminated material buried at depths

greater than 10 m will not be exposed under the imposed boundary conditions employed here. However, it
is uncertain whether this will be the case if the catchment is subject to more intense and more frequent
rainfall events. Currently, the existing 22-year rainfall record from the Jabiru Airport is looped to generate
long-term rainfall sequences. This only represents the variability of rainfall events that have occurred over a
relatively short period and runs the risk of not accounting for the full range of rainfall variability that might
be expected over 1,000 years. An alternative approach would be to use “weather generators” that create
rainfall records based on statistical representations of measured rainfall in the region. This synthetic rainfall
sequence is stochastic or random, so no two simulations will be identical. This means that many (100+) repeat
simulations may be necessary with mean outcomes used. Importantly, such a probabilistic approach also
allows uncertainty in predictions to be better accounted for and provide a range or probability of likely
outcomes. This is the only method at which millennial scale quantitative assessments can be made within a
risk framework.
From an engineering point of view, plots of maximum depth of erosion as well as erosion and deposition
patterns can be developed. This is of particular relevance for encapsulated materials within the engineered
landscape.
The results of the simulations described in this study using the CAESAR-Lisflood and SIBERIA models have
identified a number of issues likely to have an impact on the stability and/or integrity of the conceptual
rehabilitated landform. A specific concern is the prediction of large-scale erosion producing channels several
metres deep in the catchment of Corridor Creek over a period of 1,000 years, regardless of whether
vegetation is present. While the magnitude of the predicted gully formation may be considered indicative of
likely erosion patterns, the consistent prediction of gully formation in this study indicates concerns with the
design and stability of the landform. Given the landform is required to be stable for 10,000 years, the
development of these features after a period of 1,000 years is a concern.
Using two different models that utilise different approaches to landform modelling gives increased
confidence in the model predictions. However, it also needs to be emphasised that several important caveats
exist in the model assumptions and scenarios. These include the assumption that additional different surface
treatments, such as rock-armoured drainage collection channels, erosion control structures and sediment
traps, did not exist, and that no maintenance was performed on the landform in the period after
rehabilitation. A second caveat is the question of the suitability of a 22-year rainfall record to simulate a
period of 1,000 years or longer, particularly if issues such as climate change are to be taken into account.
Vegetation has a major effect on the erosion potential of the landform surface. The vegetation parameter
values used in the CAESAR-Lisflood model need to be better defined to better account for the effects of
developing vegetation cover over the area of the Ranger mine site. Given its role in the north Australian
landscape, the role of fire, which may disrupt or prevent the development of specific vegetation
communities, will also need to be considered. One option may be the integration of a dynamic vegetation
model linking soil moisture to biomass growth, which could then be used to restrict surface erosion rates.
Earlier studies have demonstrated that LEMs have developed to the stage where, if calibrated, they can
reliably predict landform stability over periods of several years. However, these models are largely calibrated
based on parameters determined from present surface conditions, which are the product of much longer-
term geology-soil-climate-vegetation interactions. A significant issue for all models is this calibration and
parameterisation, particularly for applied environments such as mine sites, where successful landscape
rehabilitation is dependent on the short- and long-term erosional stability of the design landform.
While model input parameters are static, climate and the soil-vegetation interaction clearly are not. Similarly,
rehabilitated landscape will have different dimensions from the pre-mine landscape and be constructed of
essentially different materials. Therefore, how valid is any prediction at a time scale any longer than that of
the period at which the parameters were derived? The actual result is likely to be a mix of all parameter
datasets together with other complex unknown influences relating to vegetation-climate interactions that

Landform
influence pedogenesis. While the surface is unlikely to maintain its waste rock characteristics over the
modelled period, equally unlikely is that vegetation will remain constant. How likely it is the material will
evolve to an analogue-type landscape is unknown. Both vegetation change and the regular occurrence of fire
are likely to influence erosion and therefore landscape evolution in this environment. Another complicating
factor is climate change and how this affects soil evolution.
The authors wish to thank Mike Saynor, Lucy Lytton and Rick van Dam from eriss and Sharon Paulka of Energy
Resources of Australia for their assistance in accessing and providing advice on data used in the landform
assessment and reviewing this manuscript.
Bates P.D., Horritt M.S. and Fewtrell T.J. (2010) A simple initial formulation of the shallow water equation for efficient two-
dimensional flood inundation modelling, Journal of Hydrology, Vol. 387, pp. 33–45.
Bevan, K. and Kirkby, M.J. (1979) A physically based, variable contributing area model of basin hydrology/Un modèle à base physique
de zone d’appel variable de l’hydrologie du bassin versant, Hydrological Sciences Bulletin, Vol. 24, pp. 43–69.
Bureau of Meteorology (2014) Climate statistics for Australian locations – Jabiru Airport, site viewed 22 May 2014.
Cohen, S., Willgoose, G. and Hancock, G. (2010). The mARM3D spatially distributed soil evolution model: Three-dimensional model
framework and analysis of hillslope and landform responses, Journal of Geophysical Research–Earth Surface, Vol. 115, F04013
doi: 10.1029/2009JF001536.
Commonwealth of Australia (1987) Department of Arts, Sports, the Environment, Tourism and Territories, Code of Practice on the
Management of Radioactive Wastes from Mining of Radioactive Ores 1982, Guidelines, AGPS, Canberra.
Coulthard, T.J., Neal, J.C., Bates, P.D., Ramirez, J., de Almeida, G.A.M. and Hancock, G.R. (2013) Integrating the LISFLOOD-FP 2D
hydrodynamic model with the CAESAR model: implications for modelling landscape evolution, Earth Surface Processes and
Landforms,Vol. 38(15), pp. 1897-1906 DOI: 10.1002/esp.3478.
CSIRO (2007) Climate Change in Australia – Technical Report 2007, CSIRO, Australia.
Cull R., Hancock, G., Johnston, A., Martin, P., Marten, R., Murray, A.S., Pfitzner, J., Warner, R.F. and Wasson, R.J. (1992) Past, present
and future sedimentation on the Magela plain and its catchment, in Modern sedimentation and late quaternary evolution of
the Magela Plain, R.J. Wasson (ed), Supervising Scientist for the Alligator Rivers Region Research Report 6, AGPS, Canberra,
Australia, pp. 226–268.
Einstein, H.A. (1950) The bedload function for sediment transport in open channel flow, US Department of Agriculture Soil
Conservation Technical Bulletin No. 1026, USA.
Erskine, W.D. and Saynor, M.J. (2000) Assessment of the off-site geomorphic impacts of uranium mining on Magela Creek, Northern
Territory, Australia, Supervising Scientist Report 156, Supervising Scientist, Darwin.
Evans, K.G. (2000) Methods for assessing mine site rehabilitation design for erosion impact, Australian Journal of Soil Research, Vol.
38(2), pp. 231–248.
Evans, K.G. and Loch, R.J. (1996) Using the RUSLE to identify factors controlling erosion rates of mine soils, Land Degradation and
Development, Vol. 7(3), pp. 267–277.
Evans, K.G., Willgoose, G.R., Saynor, M.J. and House, T. (1998) Effect of vegetation and surface amelioration on simulated landform
evolution of the post-mining landscape at ERA Ranger Mine, Northern Territory, Supervising Scientist Report 134, Supervising
Scientist, Canberra, Australia.
Evans, K.G. and Willgoose, G.R. (2000). Post-mining landform evolution modelling: 2. Effects of vegetation and surface ripping, Earth
Surface Processes and Landforms, Vol. 25 (8) pp. 803-823.
Flanagan, D.C. and Livingston, S.J. (1995) Water Erosion Prediction Project (WEPP) version 95.7 User Summary, National Soil Erosion
Reseach Laboratory Report 11, 131 pp.
Hancock, G.R., Willgoose, G.R. and Evans, K.G. (2002). Testing of the SIBERIA landscape evolution model using the Tin Camp Creek,
Northern Territory, Australia, field catchment, Earth Surface Processes and Landforms, Vol. 27(2), pp. 125–143.
Hancock, G.R., Lowry, J.B.C., Coulthard, T.J., Evans, K.G. and Moliere, D.R. (2010) A catchment scale evaluation of the SIBERIA and
CAESAR landscape evolution models, Earth Surface Processes and Landforms, Vol. 35, pp. 863–875.
Hancock, G.R., Willgoose, G.R., and Lowry, J. (2014b). Transient landscapes: gully development and evolution using a landscape
evolution model, Stochastic Environmental Research and Risk Assessment, Vol. 28(1), pp. 83-98 DOI 10.1007/s00477-013-
0741-y.
Hancock, G.R., Lowry, J. and Coulthard, T. (2014a). Catchment reconstruction – erosional stability at millennial time scales using
landscape evolution models, Geomorphology, Vol. 231, pp. 15–27.
Laflen, J.M., Elliot, W.J., Simanton, J.R., Holzhey, C.S. and Kohl, K.D. (1991) WEPP soil erodibility experiments for rangeland and
cropland soils, Journal of Soil and Water Conservation, Vol. 46(1), pp. 39–44.
Loch, R.J., Connolly, R.D. and Littleboy, M. (2000) Using rainfall simulation to guide planning and management of rehabilitated areas:
Part 2, Computer simulations using parameters from rainfall simulation, Land Degradation and Development, Vol. 11(3), pp.
241–255.

Lowry, J.B.C., Coulthard, T.J., Hancock, G.R. and Jones, D.R. (2011) Assessing soil erosion on a rehabilitated landform using the CAESAR
landscape evolution model, in Proceedings of the 6th International Conference on Mine Closure, A.B. Fourie, M. Tibbett and
A. Beersing (eds), 18–21 September 2011, Alberta, Canada Australian Centre for Geomechanics, Perth, pp. 613–621.
Lowry, J.B.C., Coulthard, T.J. and Hancock, G.R. (2013) Assessing the long-term geomorphic stability of a rehabilitated landform using
the CAESAR-Lisflood landscape evolution model, in Proceedings of the 8th International Conference on Mine Closure, M.
Tibbett, A.B. Fourie and C. Digby (eds), 18–20 September 2013, Cornwall, United Kingdom, Australian Centre for
Geomechanics, Perth, pp. 611–624.
Lowry, J., Saynor, M., Erskine, W. Coulthard, T. and Hancock, G. (2014) A multi-year assessment of landform evolution predictions for
a trial rehabilitated landform, in Proceedings of the Life-of-Mine 2014 Conference, 16–18 July 2014, Brisbane Australia.
Australasian Institute of Mining and Metallurgy, pp. 67–80.
McQuade, C.V., Arthur, J.T. and Butterworth, I.J. (1996) Climate and hydrology, in Landscape and vegetation of the Kakadu Region,
Northern Australia, C.M. Finlayson and I. von Oertzen (eds), Kluwer Academic Publishers, Dordrecht, Netherlands, pp. 17–35.
Moliere, D.R., Evans, K.G., Willgoose, G.R. and Saynor, M.J. (2002) Temporal trends in erosion and hydrology for a post-mining
landform at Ranger Mine, Northern Territory, Supervising Scientist Report 165, Supervising Scientist, Darwin, Australia.
Onstad, C.A. and Foster, G.R. (1975) Erosion modelling on a watershed, Transactions of the American Society of Agricultural
Engineers, Vol. 26, pp. 1102–1104.
Renard, K.G., Laflen, J.M., Foster, G.R. and McCool, D.K. (1994) The revised universal soil loss equation, in Soil erosion research
methods, 2nd edition, R. Lal (ed), Soil and Water Conservation Society, Ankeny, USA, pp. 105–124. Riley, S.M. (1994)
Approaches to estimating the erosional stability of the Nabarlek tailings pit cover, In Proceedings of the AusIMM Annual
Conference, Darwin 5-9 August 1994, AusIMM, Melbourne, pp. 415–421.
Saynor, M.J., Lowry, J., Erskine, W.D., Coulthard, T., Hancock, G., Jones, D. and Lu, P. (2012) Assessing erosion and run-off performance
of a trial rehabilitated mining landform, in Proceedings Life of Mine Conference, Brisbane, Australia, 10–12 July 2012, pp. 123–
134.
Schumm, S.A., Costa, J.E., Toy, T.J., Knox, J.C. and Warner, R.F. (1984). Geomorphic hazards and uranium tailings disposal, in
Management of Wastes from Uranium Mining and Milling, International Atomic Energy Agency, Vienna, Austria, pp. 111124.
Supervising Scientist Division (1999) Environmental requirements for the ranger uranium mine, Department of Sustainability,
Environment, Water, Populations and Communities, viewed 3 April 2013, http://www.environment.gov.au/ssd/about/
legislation/pubs/ranger-ers.pdf, last updated 30 June 1999.
Van De Wiel, M.J., Coulthard, T.J., Macklin, M.G. and Lewin, J. (2007) Embedding reach-scale fluvial dynamics within the CAESAR
cellular automaton landscape evolution model, Geomorphology, Vol. 90(34), pp. 283–301.
VanWalleghem, T., Stockmann, U., Minasny, B. and McBratney, A.B. (2013) A quantitative model for integrating landscape evolution
and soil formation, Journal of Geophysical Research: Earth Surface, Vol. 118, pp. 331–347.
Wells,T., Willgoose, W.R. and Hancock, G.R. (2008) Modelling weathering pathways and processes of the fragmentation of salt
weathered quartz-chlorite schist. Journal of Geophysical Research: Earth Surface, Vol. 113 F01014. DOI 10.1029/2006JF000
714.
Willgoose, G.R., Bras, R.L. and Rodriguez-Iturbe, I. (1989) A physically based channel network and catchment evolution model, TR
322, Ralph M. Parsons Laboratory, Department of Civil Engineering, MIT, Boston, USA, 464 pp.
Willgoose G.R., Bras R.L. and Rodríguez-Iturbe, I. (1991) Results from a new model of river basin evolution, Earth Surface Processes
and Landforms, Vol. 16, pp. 237–254.
Wilcock, P.R. and Crowe, J.C. (2003) Surface-based transport model for mixed-size sediment, Journal of Hydraulic Engineering, Vol.
129, pp. 120–128.
Wischmeier, W.H. and Smith, D.D. (1978) Predicting rainfall erosion losses – a guide to conservation planning, Agriculture Handbook
No. 537.2, US Department of Agriculture, USA, 85 pp.
Uren, C. (1992) An investigation of surface geology in the Alligator Rivers region for possible analogues of uranium mine rehabilitation
structures, Internal report 56, Supervising Scientist for the Alligator Rivers region, Canberra, Australia (unpublished).

M.S. Theisen Profile Products LLC, USA
Successful rehabilitation of massive soil and vegetation disturbances from surface mining requires a
comprehensive and holistic approach. Those overseeing rehabilitation efforts must integrate several
considerations, supported by proper planning and execution. “Soil poor” sites associated with mining
activities offer considerable challenges, particularly when topsoil sources are scarce. Such sites will require
innovative thinking and implementation to overcome the absence of favourable growing conditions. This
publication offers some fundamental principles to facilitate successful mined land rehabilitation.
The first fundamental is employing creative methodologies to develop suitable growing media from less than
desirable soils or substrates. This can only be accomplished by first understanding the make-up of the soil or
substrate through comprehensive soil testing for agronomic potential and limitations. The second
fundamental requires an assessment of suitable plant species for achieving sustainable growth and effective
erosion control – while meeting the collective post-reclamation needs or requirements of regulatory agencies
and mine owners.
Once soil, agronomic and species selection considerations have been addressed, it is appropriate to begin
analysing site conditions or characteristics to assess and select necessary erosion and sediment control
measures – the third fundamental. Site conditions, such as soils, climate, seasonality, slope lengths, gradients
and aspects, ditch and channel flow hydraulics, pond and stream banks, wetlands and more, must be
examined and proper controls selected.
The fourth fundamental entails proper installation practices critical to the success of the rehabilitation
program. Complete guidelines and details must be developed and combined with onsite supervision to assure
proper installation. Finally, once the rehabilitation measures have been implemented all active sites should
be routinely inspected and maintained after each significant precipitation or other potentially damaging
climatic event. This is the fifth and final fundamental.
Successful mined land rehabilitation entails an inclusive approach to assess, address, manage and integrate
treatments or techniques to overcome the considerable challenges presented in post-mining environments
such as poor substrates, large unprotected areas with high erosion potential, difficult access, adverse climatic
or seasonal weather conditions and much more. Reclamation or rehabilitation managers must then balance
these challenges with other operational concerns such as budgetary constraints, cost of materials, availability
of labour, sequencing of earthmoving activities and required timing of completion for mine closure related
activities.
Five fundamentals for mine rehabilitation that are proposed here are neither novel nor revolutionary
advancements or methodologies for successful rehabilitation outcomes, but rather the assimilation of time
proven design considerations combined with proper execution and implementation in the field. Examples
are given of the use of software to facilitate the integration of erosion and sediment control engineering with
agronomic considerations. Appropriate software can develop holistic, sustainable solutions for cost-effective
erosion control, vegetative establishment and subsequent reductions in sediment and other pollutants from
leaving mined sites and entering water bodies.

Five fundamentals for successful mined land rehabilitation M.S. Theisen
The first fundamental is employing creative methodologies to develop suitable growing media from less than
desirable soils or substrates. This can only be accomplished by first understanding the make-up of the soil or
substrate through comprehensive soil testing for agronomic potential and limitations. Soil testing,
interpretation of the test results, and incorporating prescriptive remedies to improve soils should be a
fundamental part of any reclamation or revegetation project. Without a proper understanding of soils or
substrates considered for use as growing media to establish vegetation, it is difficult to predict potential
project success.
Prior to conducting and interpreting soil tests, it is important to understand test methods that are relevant
for reclamation and/or vegetation establishment projects. There are various ways to extract measurable soil
characteristics and analyse samples, but rarely do different soil testing methods produce identical results.
Further, it is important to properly collect and label soil samples prior to sending them to a reputable
laboratory. The various methods listed below give a relevant testing protocol for erosion control projects
that require vegetative establishment.
 Texture – Hydrometer Method
 Soil pH and Soluble Salts – 1:1 Soil/Water Slurry and Saturated Paste Extraction
 Buffer pH – Sikora Method
 Cations (Ca, K, Mg, Na) – Ammonium Acetate Extraction
 Phosphorus – Bray 1 Extraction or Olson Extraction
 Trace Elements (Zn, Mn, Cu, Fe) – DTPA Extraction
 Sulphur – Phosphate Extraction
 Boron – DTPA/Sorbitol
 Nitrate Nitrogen – Cadmium Reduction
 Particle Size Analysis – Hydrometer Measurement
 Salinity Evaluation – Saturated Paste Extraction
 All Soluble Nutrients – Saturated Paste Extraction
Consistency in testing methods allows for simplified and more rapid evaluations of the results. Table 1
provides optimal ranges for various soil parameters for a temperate North American context, and values
where deficiencies or excesses may compromise or limit vegetative establishment – using the test methods
identified above.

Landform
Low value Optimal range High value

Soil characteristic tested Unit
(deficiency*) (sufficiency*) (toxicity*)
Texture (Hydrometer Method) % N/A N/A N/A
OM (Organic Matter) OM mass/sample mass < 3% 3%-5% > 10%
pH 0-14 < 6.3 6.3 – 7.3 > 7.3
HC03 (Bicarbonate) ppm N/A < 50 > 50
Electrical Conductivity (EC) mmhos/cm = dS/m N/A < 0.75 > 7.0
Total Dissolved Solids (TDS) ppm N/A < 480 > 4480
Sodium Adsorption Ratio (SAR) N/A < 2.0 > 7.0
N (Nitrogen) ppm < 10 10 – 30 > 30
Bray 1 P (Phosphorus) pH<7.2 ppm < 20 20 – 40 > 40
Olsen P (Phosphorus) pH>7.2 ppm < 10 10 – 25 > 25
K (Potassium) ppm < 150 150 – 250 > 250
Mg (Magnesium) ppm < 60 60 – 300 > 300
Ca (Calcium) ppm < 400 > 400 N/A
S (Sulphur) ppm <5 5 – 20 > 20
Zn (Zinc) ppm < 1.0 1.3 – 3.0 > 5.0
Mn (Manganese) ppm < 2.5 4.1 – 12.0 > 50
Cu (Copper) ppm < 1.0 1.0 – 2.0 > 2.0
Fe (Iron) ppm < 4.5 7.1 – 20.0 > 70
B (Boron) ppm < 0.5 1.0 – 1.5 > 2.0
K (Potassium) CEC % 3.0 – 7.0
Mg (Magnesium) CEC % 15 – 20
Ca (Calcium) CEC % 65 – 75
Na (Sodium) CEC % 0–4 > 15
H (Hydrogen) CEC % 0–5

Cation Exchange Capacity
<5 10 – 30 > 50
(CEC)
Cl (Chloride) ppm < 10 10 – 20 > 800

Equally important to addressing soils is an understanding of plant species for sustainable growth and resulting
erosion and sediment control (E&SC) performance. The second fundamental requires an assessment of
suitable plant species for achieving sustainable growth and effective erosion control – while meeting the
collective post-reclamation needs of regulatory agencies and mine owners. Soil properties, climate, moisture
regimes, slope aspect, maintenance, future land use and a host of other considerations contribute to proper
species selection.
Perhaps the best resource for obtaining information and availability of suitable plant species are regional
growers, collectors and suppliers of locally adapted seeds and plant materials. Experienced botanic
professionals are well versed in seasonal pricing, quantities and availability of native or introduced seed
sources as well as containerised or bare root shrubs and tree species.
In the US there is a wealth of botanical information available from universities, specialised consultants and
contractors, as well as federal, state and local agencies such as Office of Surface Mining
Reclamation and Enforcement (OSMRE), US Natural Resource Conservation Service (NRCS), US Bureau of
Land Management (BLM), US Forest Service (USFS), US Army Corps of Engineers (USACE), state mining
agencies and departments of transportation (DOTs) and local soil conservation chapters.
Key questions to consider in species selection include:
 Where is the project located?
 What are the soil characteristics?
 Do you want to achieve permanent or temporary vegetation?
 When will the installation occur – seasonality?
 What are the desired plant materials?
o Native, introduced, drought tolerant, palatable, warm or cool season, legumes,
wildflowers, shrubs, trees, etc.
 What is the intended application?
o Slope, channel, riverine, shoreline, levee, cover system, etc.
 What are the site characteristics – such as elevation, topography, aspect, climatic conditions?
 Maintenance activities – irrigation, mowing, supplemental amendments or grazing?
Once soil, agronomic and species selection considerations have been addressed, it is appropriate to begin
analysing site conditions or characteristics to assess and select necessary erosion and sediment control
measures. Site conditions, such as soils, climate, seasonality, slope lengths, gradients and aspects, ditch and
channel flow hydraulics, pond and stream banks, wetlands and more, must be examined and proper controls
selected.
A relevant and widely accepted methodology is the Revised Universal Soil Loss Equation (RUSLE/RUSLE 2) for
predicting annual soil loss on slopes combined with erosion control effectiveness, an international rainfall
database, growth establishment rates, functional longevities and factors of safety to facilitate product
selection for slopes using the following equation (Renard et. al., 1997).

Landform
Revised Universal Soil Loss Equation:

A = R × K × LS × C × P
Where:
A = computed soil loss/unit area/year
R = rainfall factor
K = soil erodibility factor
L = slope length factor
S = steepness factor
C = vegetation or cover factor
P = erosion control practice factor
C-Factor is calculated as the soil loss ratio of a treated surface versus the soil loss of an untreated control
surface. For example, if the treated surface produces 5 kg of soil loss while the untreated surface produces
100 kg of soil loss, then the C-Factor would be 5 kg/100 kg = 0.05. The lower the C-Factor value, the more
effective the erosion control technique is at preventing erosion. Conversely, percent erosion control
effectiveness (PE) may be calculated from the following equation using the above example C-Factor of 0.05
as follows:
PE = (One minus C-Factor) x 100% or
PE = (1 – 0.05) × 100% = 95%
Although comparative low C-Factors or high PE values can at first blush appear to be impressive, the potential
soil loss values can be very surprising and perhaps alarmingly high without constructed sediment capture and
retention devices or practices. For example, as shown in Table 2, an erosion control technique demonstrating
a seemingly impressive 95% effectiveness could potentially lose 18,700 kg/ha/year of soil.
However, the C-Factor and erosion control effectiveness values are in an unvegetated condition and their
ability to foster or accelerate vegetation establishment must go into design and selection criteria. If a
designer is confident an erosion control technique can rapidly establish vegetation and still provide suitable
erosion control effectiveness, then he or she may opt to go with a lower cost, less robust technique. One
determination of the ability of a material to improve vegetation establishment is using ASTM D7322 –
Standard Test Method for Determination of Rolled Erosion Control Product (RECP) Ability to Encourage Seed
Germination and Plant Growth Under Bench-Scale Conditions. This test has been modified to accommodate
hydraulically-applied erosion control and other types of products used for revegetation.
Another key product characteristic is functional longevity – which can be determined using ASTM D5338 –
Standard Test Method for Determining Aerobic Biodegradation of Plastic Materials under Controlled
Composting Conditions, Incorporating Thermophilic Temperatures. This test determines predicted
longevities of various biodegradable erosion control techniques and is often correlated with field
observations to offer material functional longevity ranges. Functional longevity will be affected by a
material’s composition as well as site specific conditions such as temperature, moisture, sunlight, soil
composition, biological activity and other environmental factors. Thus, localised observations should prevail.
Erosion control effectiveness, growth establishment and functional longevity are the three pillars of product
performance and fundamental to selection criteria for erosion control techniques. Assessing these three key
properties can assist designers in their risk versus reward selection scenarios – balancing costs and ease of
implementation versus factors of safety.

Potential soil loss

C-Factor % Effectiveness
(kg/ha/yr)
0.01 99 3,740
0.05 95 18,700
0.1 90 37,400
0.2 80 74,816
0.3 70 112,224
0.5 50 187,040
0.75 25 280,560
As demonstrated in Table 3 below, the erosion control practice factor or P-Factor from the Universal Soil Loss
Equation can have a large impact on erosion control effectiveness. Using a default value of 1.0 for a loose
disked slope, relative P-Factors can be determined for various grading techniques. It is important to note that
roughened slopes demonstrate reduced erosion control potential versus slopes that have been smoothed to
accommodate and ensure intimate soil contact for techniques such as rolled erosion control products. For
example, a compacted and smooth surface has a 20% greater erosion potential than loose disked soil. Rough
graded or tracked sites demonstrate up to 42% less erosion potential versus compacted and smoothed soil.
More flexible techniques such as hydraulically-applied erosion control products or blown hay/straw can be
applied on rough or cat-tracked slopes – providing less costly and in some cases superior erosion control
effectiveness.
Potential soil loss relative to

P-Factor practice P-Factor value
P-Factor = 1.0 (kg/ha)
Compact and smooth surface 1.2 134,400
Loose – disked plough layer 1.0 112,000
Rough surface with tracks in all directions 0.9 100,800
Tracked up and down slope 0.7 78,400
Methodologies from the US Federal Highway Administration’s Hydraulic Engineering Circular Number 15
(HEC 15) – Design of Roadside Channels with Flexible Linings (Kilgore and Cotton, 2005) are used for both
unvegetated and vegetated channel design and product selection. Primary design formulas are Manning’s
Equation and the Shear Stress or Tractive Force Equation as follows:

Landform
Manning’s Equation for Permissible Velocity

𝐶 2⁄ 1⁄2
𝑉 = 𝑛𝑅 3𝑆
𝑓
Where:
V = average velocity (m/s)
C = 1.0 for SI Units
N = Manning’s roughness coefficient (dimensionless)
R = hydraulic radius (m); cross sectional area divided by wetted perimeter (A/P)
Sf = friction slope of channel (can be approximated as average bed slope for uniform flow
conditions) (m/m)
Permissible Shear Stress or Tractive Force Equation
𝜏 = 𝛾𝑑𝑆𝑜
Where:
T = shear stress or tractive force (Pa)
Ƴ = unit weight of water (9.80 kg/m3)
d = maximum flow depth (m)
So = average bed slope (m/m)
Proper installation practices are critical to the success of any rehabilitation program. Comprehensive and
detailed construction specifications with clearly delineated plans and drawings as well as complete
mixing/application guidelines and details must be developed and combined with onsite supervision to assure
proper installation.
Installation guidelines are available for erosion and sediment control techniques, including Hydraulically-
applied Erosion Control Products (HECPs), Rolled Erosion Control Products (RECPs), and Sediment Retention
Fibre Rolls (SRFRs). Examples are shown in Figures 1 and 2.


Landform
All installations should be overseen by qualified and experienced professionals who are intimately immersed
in and familiar with the construction requirements for the project. Experience, preferably site specific
experience, is always a desired prerequisite!
Once erosion and sediment control measures have been installed it is important to visually inspect and
maintain them on a regular basis. All active construction sites must be inspected and maintained after each
significant precipitation or other potentially damaging event. Inspections should be conducted by qualified
professionals whose expectations are consistent with the installer as well as owner and regulatory entity(s).
Initial inspections should insure that all installations are in accordance with plans and specifications with all
material quantities and activities fully documented. Subsequent inspections should be executed at pre-
determined time intervals and maintenance activities conducted after each significant precipitation or other
potentially damaging weather event.
Obvious examples requiring maintenance would be damaged silt fences or sediment control devices, rills
appearing on treated slopes, displacement or movement of rock check dams or slope interruption devices,
and excessive sediment being deposited at toes of slopes or near/into receiving water bodies. Timely
inspections and maintenance can prevent small problems from turning into major complications.
Regrettably this final fundamental is perhaps the most underappreciated and is frequently acted upon with
the “one and done” mentality that often is associated with construction bidding and contracting. Savvy mine
reclamation managers should consider incorporating performance requirements and/or maintenance

requirements into their construction contracts. Mines with onsite resources should always make inspection
and maintenance a standard operating procedure.
Subsequent to the initial “grow in” period, inspections should be conducted to assess agronomic aspects of
rehabilitation efforts. Beyond monitoring vegetation vigour, cover and species composition, soils or
substrates should be tested to determine if supplemental applications of seed, fertiliser, biological inoculants
or other soil amendments are warranted. Soil tests may also identify developing problems with soil pH,
excessive salts or upward migration of heavy metals or contaminants from underlying substrates.
This project in the southern hemisphere entails the rehabilitation of mine tailings from processing of
underground deposits of chrome and platinum. The black tailings material, as seen in Figure 3, are acidic,
essentially devoid of organic matter and nutrients with surface temperatures reaching 63° C during the
summer months.
An analysis of the tailings led to recommendations for a prescriptive biological soil treatment composed of
organic matter from composted manure, humic and fulvic acids, microbial cultures and a slow release rock
phosphate to be applied at a specified rate of 8,000 kg/ha. Lime was also specified at a rate of 2,240 kg/ha.
Prior to application, the 2H:1V, 8 to 10 m long slopes were roughened to slow surface runoff, increase
infiltration, and create pockets for germination and growth. The amendments, seed mix and a tracer amount
of the specified flexible growth medium were applied in December of 2011 in the first of a two-step
application process. The biodegradable fugitive green dye in the flexible growth medium assists in visually
metering coverage and application rates.
The following seed mix was developed for this site at an elevation of 1,600 m characterised by a long “winter”
dry season offset by the summer monsoonal wet season.
Eragrostis tef – “Tef “
Eragrostis curvula – Weeping Lovegrass/Oulandsgras
Digitaria eriantha – Smutsfinger/Common Finger Grass
Chloris gayana – Rhodes Grass
Cynodon dactylon – Bermuda or Kweekgras
A 100% biodegradable flexible growth medium was then hydraulically applied at a rate of 3,600 kg/ha in a
second step from two directions over the treated area. This technology was selected for its ability to resist
erosion from heavy downpours, to facilitate growth establishment and its functional longevity of up to 18
months to offset the long dry seasons.
Significant growth was obtained during the first few weeks following installation with a reasonable stand of
vegetation developed prior to the onset of the dry season. By the second summer the grasses were fully
developed, spreading and producing seed as shown in Figure 4. Meanwhile the adjacent untreated and sterile
tailings remained essentially devoid of any colonising vegetation.

Landform
This project is in the famous Sudbury nickel mining district in Ontario Province, one of the world’s largest
fully-integrated mining, milling, smelting and refining operations. Over the years the operator had produced
a mountain of slag, a by-product of the smelting process, which was threatening to overrun part of the city.
As a result, the company decided to study new ways to accelerate its ongoing reclamation of the slag piles
and quickly convert them into green landscape.
An analysis of the slag showed it to be highly acidic and very low in organic matter and nutrients. Thus, it was
deemed to be unlikely to sustain vegetation and 61,164 m3 of a clay soil cover (46 cm depth) was specified
to be placed over the slag material. In October 2006 through February 2007, the slag piles were reshaped to
create a series of 3H:1V slopes – each 30 m long and divided from the next by horizontal benches 6 m wide.
The clay cover was cat tracked (dozer walked) to create a roughened surface with mini-check dams to reduce
sheet flow, increase infiltration and provide pockets for water retention and enhanced growth. After
placement of the clay soil cover in October 2007; lime, synthetic fertiliser, two biostimulants and a prescribed
seed mix were hydraulically applied directly on the clay with a flexible growth medium tracer.
The following seed mix was applied at a rate of 252 kg/ha:
Grasses – Lolium perenne (Perennial Ryegrass), Poa compressa (Canada Bluegrass), Phleum pratense
(Timothy), Agrostis gigantea (Red Top), Festuca ovina (Hard Fescue), Festuca rubra (Creeping Red
Fescue), Festuca elatior (Meadow Fescue), Festuca rubra subsp. commutata (Chewing’s Fescue)
Nitrogen-fixing legumes – Trifolium hybridum (Alsike Clover), Trifolium pratense (Red Clover),
Trifolium repens (White Clover), Lolium corniculatus (Birdsfoot Trefoil)
Due to an anticipated late fall seeding and the long 3H:1V slopes, a flexible growth medium was specified to
be applied at the high rate of 5,100 kg/ha in a second application above the seed and soil amendments. The
flexible growth medium was applied from two directions and then, fibre filtration tubes were installed at 11
m intervals to serve as slope interruption devices to slow sheet flow from storm events and snow melt.
Despite the late fall seeding, germination occurred quickly and vegetation was growing in the areas first
treated prior to the onset of winter and a snow cover approaching 1 m in depth. The following spring
vegetation emergence continued and the site achieved exceptional cover with the combination of prescribed
soil amendments, seed mix and flexible growth medium.

The fundamentals for mine rehabilitation that have been set out here have been a time-proven model for
successful mined land reclamation on six continents working in many of the planet’s biomes, climates, and
environments; addressing multiple types of mining and substrates. Employing the discipline to work through
the discovery sequence of the first three fundamentals – to analyse soils and substrates, pick the right plant
materials for the site and select the most cost effective erosion and sediment control techniques, will
undoubtedly head a project in the right direction. These fundamentals must be followed by the development
of clear and comprehensive construction plans and specifications to effectively communicate the project
requirements to contractors and installers. Once construction commences, onsite oversight of acceptable
installations must be conducted by qualified inspectors. Then, the active rehabilitation sites must be regularly
inspected and maintained after each significant precipitation or other potentially damaging event.
Inspections should be conducted by qualified professionals whose expectations are consistent with installer
as well as the owner and regulatory entity(s).
Software programs can be used as a platform for coordinating a rehabilitation approach into a cohesive and
interconnected framework for the designer as well as other project stakeholders. This process is enhanced
with the continued input of environmental conditions, rainfall return frequencies and project information
supplied by users around the world.
Two case histories from different continents and contrasting environmental conditions were provided to
briefly illustrate the discovery (required information gathering) and implementation of the mine
rehabilitation fundamentals – leading to successful mined land rehabilitation. There are many more mining
and projects from other market segments around the world to substantiate the utility and legacy of this
approach.
I would like to thank Michael Robeson and Matthew Welch for their assistance in preparing this manuscript.
Erosion Control Technology Council (2014) Rolled Erosion Control Products (RECPs) General usage and installation guidelines for
Slope, Version 1.0, Roxborough, Colorado, USA, viewed http://www.ectc.org/assets/docs/ectcmarch2014_recp%20slope%
20usage%20install%20guidelines%20final.pdf.
Kilgore, R.T. and Cotton, G.K. (2005) Design of roadside channels with flexible linings, Hydraulic Engineering Circular No. 15 (HEC-15),
Publication No. FHWA-NHI-05-114, US Federal Highway Administration, Arlington, Virginia, USA.
Renard, K.G., Foster, G.A., Weesies, D.K., and Yoder, D.C. (1997) Predicting soil erosion by water: a guide to conservation planning
with the Revised Universal Soil Loss Equation (RUSLE), US Department of Agriculture, Agriculture Handbook No. 703, 404 pp.

Landform
Utah Water Research Laboratory (2007-2014) Joint research findings with profile products LLC to determine erosion control rates
using standardised slope testing protocol, Logan, Utah, USA.


N. Slingerland Department of Civil & Environmental Engineering, University of Alberta, Canada

G.W. Wilson Department of Civil & Environmental Engineering, University of Alberta, Canada
The benefits of life-cycle mine closure planning and sustainability are abundant and diverse, and they have
been discussed within the mining and reclamation industries at length over the past decade. However, when
faced with the iterative nature of mining and site constraints, closure planning at many open-pit sites still
tends to revert away from these discussions, which are termed “progressive” or “idealistic.” Research over
the last 18 years has consistently shown that, once closure activities are completed, closure and reclamation
goals are often only partially achieved or not achieved at all. As public license increasingly influences mine
application approval, a proven track record demonstrating the ability to leave behind a safe and sustainable
landscape can be an advantage. This is reliant on early identification and achievement of end land-use goals.
In order to optimise the mining process as a whole to achieve closure goals and productivity targets, the
authors propose that closure design be included as a fundamental element and initial step in mine design.
This paper builds on previous work regarding mine closure goal achievement in Canada and the United States.
It evaluates conventional and proposed open-pit mine design and closure planning frameworks in order to
identify the most favourable stage(s) for practical integration of early closure planning tasks and to outline
some key challenges. Closure tasks range from initial site inventory and regional analysis to the management
of waste rock and geomorphic landform design. This paper critically examines the synthesis of mine design
and closure design frameworks with respect to practicality, and develops a new “closure-first” optimised
mining framework that addresses mine closure from a landscape architectural and land-use perspective. We
argue that allowing the predesigned end land use — whether this is ecological or more anthropogenically
focussed — to inform mine works and material placement will mean that closure goals will have a greater
likelihood of being successfully achieved and efficiencies in mining operations will be realised. The proposed
“closure-first” mining framework provides insight as to how new mining operations may achieve sustainability
from both business and end land-use perspectives.
In What Is Philosophy?, Gilles Deleuze and Felix Guattari explain how art, science and philosophy continuously
overlap each other in different ways. These three focus areas weave almost imperceptibly through our lives
and through the intricate systems in which modern society is immersed. However, the mining world, which
has been dominated by (applied) science throughout history, continues to be an exception. The complex
remnant social, economic and environmental problems we see as a result of mining may be indicative of this
rudimentary imbalance. Arguably, the first art form of modern civilisation was architecture (Sagan, 2008), so
this paper seeks to explain how we can use architecture and philosophy to rebalance the land alteration and
reclamation process in mining.
Traditional mine planning, practiced at most mines around the world, seeks to maximise net present value
(NPV) generated from the ore body. Closure planning in an operating mine takes place as mine life nears its
end to attain a maximum amount of required information, to allow for remining of low-grade stockpiles and
to delay the resulting decrease in profit streams for as long as possible. This methodology is associated with
several problems: there is little time to identify and solve potential closure-related problems; extensive earth
moving is required to relocate waste; and mixing during initial deposition and subsequent movement can
compound geochemical and geotechnical issues, making reclamation works more complex and costly.

End land use as a guide for integrated mine planning and closure design N. Slingerland and G.W. Wilson
Additionally, success rates for closure goals are low when this methodology is used. In contrast, reclamation
planning for greenfield sites occurs much earlier and success rates are higher. While proof of early closure
planning is required prior to the start of production in most developed and developing nations, this is not
always the case, nor does it guarantee the work proposed will be carried out.
Although extractive industries have historically been pioneering, they seem to have neglected the clean-up
portion of their projects; other industries have created a culture of innovation, while the mining industry
continues to use this traditional model. Investment towards innovation averages 0.2% per year for mining
and 1.2% for oil and gas sectors, while extraction increasingly consumes more energy and resources become
depleted (Nordic Innovation, 2012). Social and environmental expectations of the mining industry have
increased over time, and this stagnation in mine planning methodology means that, as an industry, we are
less able to effectively and efficiently tackle the problems we face. Closure goals are commonly not fully
achieved using the traditional model (Slingerland et al., 2014), and it is illogical to expect drastically improved
outcomes while continuing to use the same approach. In an effort to break this cycle, we experiment by
turning the process upside-down and proposing a “closure-first” framework to determine whether better
results are likely using a drastically different approach.
The proposed closure-first mine design and closure planning methodology prioritises postmining land-use
design and allows this to dictate mine design and planning. Just as the builders of skyscrapers do not start
outlining their approach and construction schedule before they know the size, shape and design
requirements of the structure to be built, the closure-first approach means that mining does not start until
the general design parameters of the end landscape have been determined. Mine works, waste rock dumps,
tailings impoundments and infrastructure are to be optimally designed and located based on the entire life
of the mine, including the end landscape plans. As such, end land use informs mine operations and design.
For example, if a tailings impoundment is not desired in the final landscape in perpetuity, alternative tailings
processing, treatment and deposition methods will need to be investigated prior to completion of the
feasibility stages. Changes to the mine and closure plan, which may be required over time, must be evaluated
and made using the same goal of holistic life-cycle optimisation upon which the mine plan was originally
developed.
The purpose of this paper is to analyse the previously developed framework using results from case studies
to indicate the potential drawbacks and benefits of this new theoretical framework. In this way, the ideal
methodology is compared and contrasted with what is realistic, and case studies are drawn upon as
illustrations.
Drawbacks of the closure-first methodology are dominated by the realities of the original objectives and
motivations for mining, which will be difficult to avoid. These drawbacks are anticipated to include greater
expenditures early in the mine life, the need to move away from an objective that is strictly focussed on
maximising NPV and the necessity to make minor adjustments to the plan throughout the mine life due to
the iterative nature of mining. Benefits are anticipated to include greater efficiency over the life of mine, the
ability to design physical and chemical properties of the final substrate, longer time periods to work towards
closure objectives, early identification and resolution of complications and greater support from the public
as a result of earlier and more meaningful public and stakeholder engagement.
The list of mining companies, consulting firms, governments and nongovernmental organisations that have
developed sustainable life-of-mine approaches and integrated closure plans (ICMs) has grown significantly
over the last decade. They are typically driven by a desire to reduce “environmental, social, financial, and
reputation-related risks” and not from a desire to develop the most site-appropriate closure solutions
(Hattingh and Bothma, 2013) or to better achieve closure goals. Many of these schemes imply they can make
operations easier without making fundamental shifts in process or effort on the part of mine design and
operation staff, and the result is that core problems remain unresolved. In order for any of these approaches
to be successful in the industry, there must be a benefit to the owner and/or operator; and while a social

Landform
license to operate may be sufficient in some areas of the world, greater efficiency and financial incentives
may be required elsewhere to make such a process more appealing.
When placing end land use at the forefront of the mine planning process, it makes sense to look to industries
and professions that also do this. We have applied a framework from landscape architecture and cross-
referenced this with mine and mine closure planning phases in order to outline what a closure-first
framework might look like and what consequences could realistically be expected.
The closure-first framework discussed herein was created to shift the focus towards more publicly-accepted
mining stages and away from those less acceptable to the public. It makes no assumptions as to what the
end land use(s) will be, as land use is determined in the early stages of the landscape architectural design
process. Determining the potential functionality of the theoretical framework is the primary objective of this
methodology, and as such the consequences of specific uses are beyond the scope of this paper.
Conventional mining occurs in five phases: preplanning, which includes exploration; planning;
implementation; production and closure. These phases are further subdivided into concept studies,
preliminary studies, feasibility studies, design and construction, commissioning, production start-up,
operation, decommissioning and reclamation. Mine closure plans are required prior to operation in most
developed and many developing regions of the world. However, governments make exceptions in some
cases; many regions still do not require them, and where they are necessary, closure plans are typically vague
with little or no testing to reinforce their stated intentions for the postmining period. In order to eliminate
ambiguity, understand the scope of work and communicate this work clearly to all parties involved, both
written documents and spatial graphics should be developed.
A linear landscape architectural model exists in theory, but the often circuitous nature of real-world design,
coordination and implementation means that this model has become more of a framework for organising
activities and directing work towards end goal(s). This framework theoretically aligns with the needs of the
mining industry, as the iterative nature of mining forces divergence from the linear progression of mining
phases and stages previously outlined. Expansion of the original extraction footprint and unexpected closure
are two relatively common mining occurrences that can disrupt the linear process and previously optimised
mine plan.
The landscape architectural framework includes the following progression of stages: site and regional
inventory, site and regional analysis, conceptual site design, design development, working drawings and
construction document preparation, construction and warranty period. This is a framework for determining
and designing appropriate land use; but when outlined in detail it can also provide a mechanism for deciding
when to engage stakeholders, when closure teams need to collaborate and how to link interdepartmental
tasks while working towards an end product (the final landscape). This framework relies on both written
specifications and spatial graphics to develop and communicate plans.
In Figure 1, mining stages, beginning at the planning stage, and the landscape architecture framework are
listed in matrix format and considered with respect to the relative timing and effort allocated to each. An
important feature of the integration of the landscape architectural framework with the mining model is that
the resulting approach, by necessity, does not force a strictly linear progression from untouched land to a
mine and then to end-use; it embraces and allows for the orchestration of a dynamic process.
As Figure 1 indicates, the site inventory, site analysis, conceptual design and design development stages of
the landscape architectural framework and the closure-first approach are completed within the planning
stage in mining. This means that major closure and mine planning decisions can be made based on big-picture
thinking and will no longer solely be a result of short-term mine functionality. End land-use design
development is completed prior to the design and construction phase of mine planning, allowing mine

features such as tailings dams, processing facilities and waste rock dumps that may exist in perpetuity to be
located with the end-use, ecological, spatial and monitoring considerations in mind.
Community engagement typically begins at the conceptual design stage within landscape architectural
projects, although it has been known to start earlier on occasion, at project initiation or later in the process,
depending on the user group, the nature of the site and the stakeholders. For the general mining case, this
would mean that public and stakeholder engagement would be initiated during the prefeasibility stage of
mine development.
The combination of mine planning and the landscape architectural framework forms the basis for the closure-
first framework discussed in the next section.
Closure first is a design and planning framework that couples the critical forethought, systems planning and
creativity of an architectural proposal with the geotechnical and ecological rigor of the science and
engineering fields. The idea behind the closure-first principle is that the bulk of our energy in designing the
land should be towards that which will remain in perpetuity, ensuring its success; and since mining is a
temporary land use, its design should be secondary, albeit closely aligned. The priority is on that which will
remain, not on the short-term use. As such, we design for end land use first and integrate mining afterwards,
essentially turning the conventional model upside-down.
Recognising the ramifications of the closure-first principle changes our approach to both mining and design
in several ways. It requires that we do the following:
 Value alterations of the land from its original state: The physical and chemical composition of the
subgrade is drastically altered through mining, and what the land once supported is not likely to
be viable after mining. This is due to features such as bulking, reduced water-holding capacity,
nutrient availability and other microfeatures, but also as a result of broader alteration of socio-
economic aspects in the surrounding area. If a mine is in existence for an extended period of time,
a community may develop around it and an anthropogenic focus may be deemed more desirable
for the land. Designing large plots of land for a time 10 to 50 years into the future is a task that
requires an open mind and strong analytical skills. Urban and regional planners commonly
perform such tasks on a larger scale when designating places to grow the population or
greenbelts to preserve within a province, but they also do this on smaller scales through local

Landform
zoning for school, commercial or residential land use, for example. This skill needs to be applied
to mining areas and communities in order to plan well into the future.
 Accept a certain degree of variability in our design: The mine initially planned for is rarely, if ever,
the mine that is left once mining has ceased. Commodity prices may lead to early closure or
expansion, so our designs need to be focussed enough to achieve our closure goals but loose
enough to be altered when changes undoubtedly occur. In a more specific sense, designs may
need to be completed in typical cross-sections and plans applied across the site once the full
extent is understood.
 Build a truly diverse team of experts and professionals to collaborate, not just work together:
Unlikely partnerships (such as Evergreen Brickworks in Toronto teaming up with Walmart to build
a low-impact sustainability hub) can lead to benefits for all parties involved. The problems
surrounding reclamation extend beyond one person’s knowledge base, and developing solutions
that are fully encompassing require teamwork from a range of disciplines. Gord McKenna has
done extensive work on this and has made great strides in developing multidisciplinary teams to
work on reclamation and landform design projects for mine sites. His teams generally consist of
geotechnical engineers, geologists, hydrogeologists, hydrologists, ecologists and forestry
professionals, just to name a few, and they achieve impressive end results (McKenna, 2002;
McKenna and Dawson, 2014). McKenna also understands the value of communication across
these varied professions and beyond and regularly employs a graphic artist to draw his team’s
ideas and developments. This visual aid facilitates the communication of ideas, developments and
approaches to a range of people, regardless of their technical background or language.
The term “teamwork” is used in its most basic sense here: it means that specialists from both
scientific and creative sectors are not simply consulted; they are fully integrated into the planning
process, asked to adopt ideas and methods borrowed from the landscape architectural
framework and then asked to collaborate (Steinitz, 2012).
 Design for a changing landscape over time — likely over 100 or more years from closure: Both a
challenge and a benefit in designing postmining environments is that doing it according to the
closure-first principle presents designers with a blank canvas: they have to build from nothing and
design from below the ground to up above the ground they make. This means there will likely be
some form of settlement or surface adjustments to deal with, and there will certainly be
ecological succession to consider if naturalisation is the end goal. Mine reclamation according to
the closure-first principle requires designing in four dimensions, rather than three, as a result of
this new and important time component. Unlike a typical landscape architectural project, this
design needs to be set up to evolve, starting from the end of construction or installation and
moving forward as the landscape changes. There will never be a stagnant finished product as the
landscape will be constantly evolving. This is unique to mine reclamation and may be difficult to
conceptualise, but it is essential to a successful result that mimics nature.
The idea behind designing for future and permanent land use prior to temporary use is not new; the venues
for Barcelona’s 1996 and Vancouver’s 2010 Olympic Games serve as high-profile examples of this (CBS News,
2014). These sites were successful in the long term partly because they were planned far in advance and
financing was put in place to maintain the sites well into the future.
The same is required for mine sites, where evolution needs a helping hand for an extended period of time
and where built structures have associated monitoring and maintenance costs. Funds must be set aside not
only for closure implementation but also for postclosure monitoring and maintenance, and this is rarely done.
Reclamation is not complete at the end of construction works: to evolve into its final form, it requires nursing,
which can be expensive. Realising the full extent of the job to be undertaken in reclamation means also
realising the full extent of its drain on resources in terms of time and money, which is perhaps not something
mine planners want to think of at the start of an operation but is a necessary consideration nonetheless.

While designing primarily for long-term use and considering short-term use secondarily may not be a
universally new approach, allowing a landscape architectural framework to guide project design in
combination with this is novel with respect to the mining industry’s accepted norm. This framework allows
for testing and experimentation of proposed methods throughout the mine life and leads to a higher
likelihood of success upon final implementation. It also requires manipulation of mine planning and typical
mine procedures in order to amplify the success of the postmining landscape.
The closure-first framework is obviously an ideal we feel sites should work towards, but we also recognise
that this transition will not happen quickly and that there are barriers currently in place that may necessitate
changes to this framework in the early stages of implementation or trials. What this theoretical framework
translates to when put in place will be something of a hybrid, at least initially.
Many theories as to how mine sites can be transformed for appropriate land uses have been proposed;
however, when we look at the success stories, most of them stem from worst-case scenarios (Baida et al.,
2014). The vast majority of successes are not metal mines but mines that produce decorative or structural
building materials, such as clay or marble, and dimension stone; these are commonly referred to as quarries
or clay pits. Two internationally recognised reclamation projects frequently cited as mine reclamation
successes are Butchart Gardens in Victoria, British Columbia, Canada, which was built progressively within a
quarry in the early 1900s, and The Eden Project in Cornwall, United Kingdom, which was built within an
abandoned china clay pit. Many such sites result in postmining land uses that are more financially productive
and longer lasting than the mine ever was. Incentive to transform these landscapes into productive land after
mining is strong as a result.
Quarries have had greater success in repurposing their land compared to other kinds of mines for a number
of reasons. First and foremost, the waste streams generated from quarries are typically fewer and less
hazardous. Quarried material is used for building material and undergoes less processing. Waste rock dumps
or tailings ponds can be challenging and expensive to reclaim and pose long-term risks, but these are not
often associated with quarries.
Quarries are also, by necessity, located in close proximity to large populations or urban centres. This means
that there is an active market for adaptive reuses that may generate revenue, such as recreational areas or
residential development. Land also tends to have a higher property value as a result of this close proximity.
When extraction sites are closer to populated areas, there is greater pressure from government and society
to reclaim primarily for safety and aesthetic reasons. In contrast, other types of mines are more often located
in isolated areas at a distance from a potential user group or market, have lower property values and are
subject to an “out of sight, out of mind” attitude. The motivation for reclamation here is to return land to the
governing body and retrieve financial holdbacks, if any.
From a legal perspective, the transfer of mine sites to third parties for reuse represents a liability since mining
companies are not typically in the business of adaptive reuse. As McKenna (2002) explains, mine sites are
sensitive to change and if activities were to occur on the land that degrade landscape performance, including
vandalism, the mine operator may still be held responsible. As such, they are unlikely to convert to uses that
invite the public or third-party affiliates onto the land due to ongoing, long-term liability concerns.
Quarries are a natural fit for some of these more anthropogenically focussed, revenue-generating adaptive
reuses, while other types of mines located at a greater distance from urban centres, with greater material-
related challenges and associated risk, are more likely to be revegetated.
Nevertheless, there will always be exceptions to a generalisation. The Daybreak residential community at the
Kennecott Mine near Salt Lake City, Utah, is one of these, albeit located near an urban centre. The copper,
gold, silver, lead and zinc mine has been operational since the mid-1800s, and this new community overlies
land that formerly contained the mine’s evaporation ponds. Prior to development, the site required capping
of gypsum sludge and extensive remediation for heavy metal contamination, and it continues to undergo
groundwater treatment (Rio Tinto, 2012). Regardless of this ongoing effort, the community serves as a

Landform
beacon of sustainability for Rio Tinto and the Kennecott Mine. It includes several LEED Platinum-certified
houses and buildings on-site, and a light-rail transit system connects it to urban centres; it has also won six
Governor’s Quality Growth Awards and achievements in sustainability (Rio Tinto, 2012).
Daybreak is an outlier, unfortunately, even within developing countries where repurposing of land is most
likely to occur. More typically, we see mine closure goals listed simply as “revegetation,” which is not nearly
specific enough to measure as having been successfully achieved or not upon completion. For example, a
vegetative cover of non-native grasses may meet these criteria, benefit a handful of species and provide
short-term erosion control while, at the same time, it may actually be harmful to the more extensive native
ecosystem. There is still a lot we do not know about ecological restoration and plenty that is simply not yet
being implemented (Eaton et al., 2014).
We need to go one step further in our goal setting in this area and revegetate with intent. For example,
revegetation must have objectives such as providing biodiversity, ecosystem services (which should be
specified) or conservation for specific species, or it must more broadly provide ecosystem management and
functionality. Delineating goals that are specific, measurable, attainable, realistic and time-based (SMART) is
a vital component of early planning and in-depth consideration of the proposal at project outset.
Just as imprecise goals can lead to poor results, legal requirements with intentions to create habitat for one
at-risk species can often result in doing so at the expense of a number of others. Historic initiatives to
preserve a single species are slowly being recognised as too narrow, and species-centred approaches have
been steadily traded in for broader landscape management initiatives over the last decade (Quinlan et al.,
2003; Ehrenfeld, 2000). The species-centred idea was good in theory and was widely adopted, but it proved
detrimental to the ecological system as a whole in practice. Clearly, we are still learning.
We must start by analysing the concept. We can critically analyse our ideal closure-first framework and ask
ourselves what this relates to in reality.
The closure-first framework begins with an in-depth inventory of the landscape at three scales ranging from
the immediate site to the larger region. Elements of natural or cultural significance and patterns in
vegetation, topography etc. are outlined and considered as an overlay to identify related opportunities or
constraints. Funds are limited in the early mine planning stages, and while inventory and analysis typically do
not cost over 1% of an average landscape architectural construction budget, it may be tempting to cut costs
not associated with an immediate return on investment.
Inventory and analysis are completed in order to guide the concept development stage of design. It is
important to do this early because choices that are made earlier will have more influence over the mine,
mine costs and overall efficiency (see Figures 2 and 3) (Hustrulid and Kuchta, 1998). At this stage, the closure-
first approach shifts the focus of reclamation away from reaction as reclamation plans begin to dictate
appropriate locations for site entrances and other elements that will impact surrounding areas and the end
landscape.

Adapted from Lee, 1984, and Hustrulid and Kuchta, 1998
Source: The GARD Guide (INAP, 2009)
This is a good time to integrate public consultation or engagement sessions into the decision-making process.
If this is done earlier, there are too many unknowns to make this beneficial to either the mining company or
the public, but if engagement is delayed beyond this point, the public may not believe they are sufficiently
included in the process. A sense of public ownership throughout the design process is essential to the long-
term success and custodial transfer of the postclosure site, so including the public at the proper time is key.
With the increasingly vocal presence of nongovernmental organisations (NGOs) both locally and globally, the
public may be wary of new mining operations; this is where a conceptual design can help to transfer the focus
onto the end product and the development stages so that people know what to expect and when. Reinforcing
design decisions with details of the inventory and analysis and with implementation strategies can help build

Landform
public confidence in the process; however, it is important to use hypothetical terms to indicate that feedback
will be considered and integrated. Local inhabitants can act as invaluable resources by contributing expertise
and traditional knowledge on vegetation and local customs (Evans and Goodjohn, 2008). While interviews
are typically included in the regional and site inventory, public consultation forums can further strengthen
the target end landscape.
This is a point in the design process where many changes occur, so no final decisions should be made until
this stage is complete. Public opinion can be unpredictable and has successfully stopped mines from initiating
operations altogether, based solely on the poor environmental track records of other mine sites (Kahn et al.,
2001). Public consultation professionals can provide guidance and direction for particularly challenging sites
throughout the engagement process; clear explanation and reinforcement of the reasons behind proposed
plans can initiate productive dialogue that should ideally continue throughout the life of the mine.
The public engagement process outlined here can include round-table discussions, town-hall presentations
and charrette-style brainstorming in small groups, or it can take the form of question-and-answer games to
indirectly determine values and land-use patterns. Over the past ten years, the inclusion of public
consultation in mine planning has become a common practice (Baida et al., 2014), and yet end land use is
still often incompatible with community needs and desires, so more work is needed that focusses specifically
on what is left behind.
The landscape design for the site will evolve through several stages after closure over time. Once it is
developed, the mine design is adjusted in terms of processing methods, extraction techniques, location of
infrastructure etc. Mine planning decisions regarding both order of extraction and order of deposition should
be completed inconjunction to optimise progressive reclamation wherever possible. Design ensures there
are no remnant hazards in the landscape and that waste materials (postmining construction materials) are
easy to work with so no additional problems are created in terms of reclamation; however, mining is
realistically an iterative process, so this may happen repeatedly. In the ideal case, a mine is designed to
operate for a set number of years, and reclamation is ongoing over this time so that, at the end of mine life,
there is relatively little to be done and significantly less liability (see Figure 4). The reality of uncertain
commodity prices means that unforeseen expansions or closures may occur, and added expense early in
mine life and for each completed earth movement can be a financial risk. Having a plan in place that is flexible
enough to respond to these sudden changes can save money in these cases, given that closure and
reclamation works need to be completed whether a mine shuts down early or when expected. This can be a
difficult point to understand, and the industry would benefit from detailed cost analysis for a typical mine.
Assumptions made throughout the design phase need to be critically analysed before they can be relied
upon. Mineralogy and other parameters can change spatially, mine life will realistically change and regulatory
frameworks can change financing, environmental or other requirements. Typical geotechnical engineering
design assumes that the structures put in place will not change over time; the reality is that ecological
processes act on postmining landscapes and alter the performance of these structures over time (DeJong et
al., 2014). Diverse teams that work through each aspect of the proposed end landscape and closure plan will
inevitably discover such assumptions from other specialist areas and will be better able to identify and
address these issues as a collaborative team. The realistic implication is that working as a team can take more
time, and if your specialists are external consultants, this can dramatically increase consulting expenses early
in the planning stages.
In the ideal model, mine works and staff transition seamlessly from extraction to reclamation landforming:
as reclamation works increase gradually, mine works decrease. However, each site is different, and not all
allow for progressive reclamation, so there may be a sharp change in activities. Nonetheless, efforts should
be made to work towards this optimal model when possible because failures tend to occur during transition
periods, such as when ownership of a site changes or when there is a large change or turnover in staff, which
typically happens when mine decommissioning and closure tasks begin. Knowledge of these and other risks
can help operators prepare by training staff and planning operations beforehand.

Source: Bocking, 2010.
As we have seen, there are a range of pros and cons of reimagining the traditional mining model. One leading
deterrent applies Newton’s first law of physics to the human condition: unless change is externally imposed,
there is substantial resistance to change. There are, of course, reasons for this that extend beyond simple
reluctance to change. Some of the major reasons (or constraints) and our reasons for believing that such a
model is worth developing further (opportunities) are summarised below. Each of these opportunities
requires an incentive to bring about implementation; these incentives may take the forms of a strong
business case (economic incentive) or regulations (ability to operate).
The greatest opportunity to influence costs throughout mine life is during the planning stage; this is when
closure and end landscape decisions need to be made to make the greatest impact (see Figure 2). In
traditional mine planning, the ability to influence costs all but ceases at the end of the implementation stage;
however, when mine works are influenced by closure plans and efficiencies can be realised throughout
production, the ability to alter costs continues through to the end of production. Efficiencies throughout the
life of the mine include earth movement and optimal placement, which result in reduced emissions and wear
on vehicles over the mine life. These translate to financial and environmental benefits and provide added
incentive for early planning.
Major decisions such as where to locate tailings impoundments and waste rock dumps impact productivity
throughout the life of the mine, but they also influence the landscape in perpetuity. Knowledge about rock
characteristics and processing allow for an understanding of the characteristics of waste products; when we
work from the desired landscape backwards, we can begin to design the subgrade characteristics specifically
for future landforms. The ability to design soil properties is exciting, and the ramifications for the end
landscape are enormous.
This landscape architectural framework can be applied to existing operational mines in addition to new ones;
however, since most major decisions are made in the planning stage, the effect of doing so will be dampened.

Landform
Mergers, acquisitions, mine expansions and unforeseen closures are some of the unpredictable events that
frequently occur in the mining sector. These affect the mine plan, and if ownership changes, there is no
guarantee that the new owner or owners will continue in the same way. Expansion requires changes in mine
and closure plans, and this can defer landforming or reclamation of low-grade stockpiles in case they become
profitable at a later date. Alternately, unforeseen closure can mean that closure plans are not implemented
as required materials are not yet mined or funds to complete closure as planned do not exist.
The natural flux of resources over time can put financial pressure on companies to reduce unnecessary costs.
If pressure is sufficient, short-term thinking may appear to be more desirable than long-term financial and
landscape planning.
Increased spending on closure planning and design in early years means the payback period for capital
expenditures is longer, which can drastically affect the NPV of a project. Delaying these costs until 15 or 20
years later (as in the conventional model) means that reclamation and closure planning costs have less impact
on NPV. This is why so many mines have minimal closure plans until late in their operations. An alternative
approach to NPV optimisation that makes other factors (dependent upon the region) the guiding force should
be used (Botha, 2013).
Similarly, it can be difficult to estimate closure costs when the future political and economic climate is
uncertain, goals are poorly developed and materials have not been thoroughly characterised. Gaining the
most precise estimate possible is important for funding the project. Closure costs may be consistently
underestimated, but it is still cheaper to work towards proactive closure than retrospective remediation
(Goodbody, 2013). Historically, when inflation-adjusted profits have been compared to closure costs at the
end of mine life, some mines have barely broken even (Berger et al., 2012). There are countless examples of
mine sites where closure and reclamation estimates were prepared despite the absence of any closure-
related studies or research; unsurprisingly, these are regularly undervalued by orders of magnitude. Had such
projects undergone thorough mineralogical characterisations and detailed breakdowns of the work required
for closure, the resulting cost estimates surely would have raised red flags or halted development altogether.
Because of these unknowns regarding cost estimation, it can be advantageous from the perspective of mining
companies to underestimate the reclamation proposals they submit to regulators and to find ways of
reducing any bond or financial holdback required by regulatory authorities. Unfortunately, these companies
are equally likely to be financially unprepared to adequately reclaim their sites when the time comes, and
the financial holdbacks they provided are unlikely to cover reclamation or ongoing monitoring and
maintenance costs should the government be left with the responsibility.
Ongoing community and stakeholder engagement is becoming standard practice prior to large-scale land-
use change near populated areas. The ability to present stakeholders with rigorous information and graphics
that illustrate the possibilities for a particular site after mining can help to build the confidence of
stakeholders and community members. The closure-first model allows for ongoing engagement from project
start-up into the future, which can help to build positive relationships and productive discussions.
Increasingly, we are seeing that social license is directing investment decisions and profit margins. It is
particularly telling that mining companies have repeatedly chosen to address corporate social responsibility
at the same time as announcing expansions (Rio Tinto, 2014). A social license can mean open doors at each
turn and its absence can mean an up-hill battle to operate. This license is increasingly noted as a wise
investment for long-term mining operations.
Benefits of a closure-first approach, such as a positive reputation or replacement of ecosystem services, can
pay dividends to both mining corporations and communities. However, these types of benefits are not easily

quantified and are thus complex to weigh when performing cost-benefit analyses. This leads to the incorrect
assumption that they make a smaller impact than more tangible and measurable benefits (such as the total
hours of tire use for earth relocation over the life of mine, which can be equated with the number of tires
used and a dollar figure). There is no dollar figure for positive reputation or ecological services; perhaps this
is an area of research to be addressed in future mine closure conferences so that we may begin to understand
the full impact of intangible, or qualitative, decision-making factors.
Early, precise goal identification may seem to be a simple task, but it has proven to be a difficult one
historically (Botha, 2013; Fair Mining Collaborative, 2014; McKenna, 2002). Early identification of clear
closure goals provides something to work towards and rally around as professionals. If a team cannot agree
on precise end goals for a site, its work will lack direction and cohesion. Early identification of goals means
that, over longer time periods, more collaboration can occur and goals are more likely to be achieved once
closure is completed. A positive legacy of achievement from both a social and environmental perspective
leads directly to improved publicity and social license to operate.
The poor environmental record of select postmining cases has increasingly been used to prohibit new mines
from opening, especially in prosperous regions with environmentally conscious and strongly connected
citizens (Kahn et al., 2001). Mining has become just as much a social problem as it is an environmental one.
Early and precise goal identification can direct reclamation efforts and is likely to result in greater social,
technical and environmental successes.
Mine closure requires that poorly defined, interconnected and dynamic problems from a wide range of
specialised disciplines be analysed and solved. This is not an easy undertaking; planning for closure is too
extensive a task to be completed or overviewed by one single person, and it needs both productive
collaboration amongst disciplines and a method of organising that collaboration to be successful. The
proposed closure-first framework seeks to provide a mechanism to balance the scientific understanding of
materials, conditions and processes with future long-range plans and objectives for mine sites. The
framework does this through the use of a landscape architectural design methodology that inherently
focusses on achieving an end goal and manipulating a variety of disciplines to do so.
What is learned from the many alternative closure planning methodologies and approaches currently in
existence (including some that are not in practice), is that early and precise identification of goals is necessary
in order to succeed. Land use 20 or 50 years into the future is difficult to predict; however, urban and regional
planners regularly identify places to grow the Canadian population and delineate which sections of present-
day farmland are to be converted into future residential, industrial or commercial lands 20 to 50 years in
advance.
The same planning principles need only be applied to a mining-centric area to realise that not all mined land
is best converted to a greenfield or wildlife preserve; sometimes the best land use is human habitat, although
as we have noted, quarries tend to be a more natural fit for this than most mine sites. Regardless of what
postmining use is deemed appropriate through the inventory and analysis process, it is this end land use that
truly dictates the specific and measurable closure goals to be sought, and as such, land use needs to be
determined up front and used to guide the process as a whole.
Short-sighted emphasis on NPV has led us down the path of managing mining very well over the 20 to 50
year life of the mine, but it has done so at the cost of the residual landscape in perpetuity. While there are
constraints to this idealistic closure-first approach, when the reality of implementation is considered,the
potential long-term benefits and opportunities far outweigh the possible short-term constraints. Mining is
not a sustainable practice; but given the right design and implementation framework, mining provides a
unique opportunity to create a new landscape that is sustainable and may contribute to the further
development of a region.

Landform
Baida, M., Slingerland, N. and Wilson, G.W. (2014) The great divide between mine closure goals and outcomes and potential solutions,
in Proceedings 9th International Mine Closure Conference (Mine Closure 2014), 1–3 October 2014, Johannesburg, South
Africa, M. Tibbett, A.B. Fourie and C. Digby (eds), Australian Centre for Geomechanics, University of Western Australia.
Berger, A., Brown, C. and Ketellapper, V. (2012) Waste to place: a new model for reclamation in the American west, Booklet two, The
Project for Reclamation Excellence (P-REX), Cambridge, MA.
Bocking, K. (2010) Post-closure liabilities, Australian Centre for Geomechanics Newsletter, Vol. 35, pp, 20–23.
Botha, R. (2013) Mine closure toolbox, Anglo American PLC, Johannesburg, South Africa.
CBS News (2014) Olympic challenge: how do host cities fare after the games? viewed 1 April 2015, http://www.cbsnews.com
/news/olympic-challenge-how-do-host-cities-fare-after-the-games/, last updated 24 February 2014.
Dejong, J., Tibbett, M. and Fourie, A. (2014) Geotechnical systems that evolve with ecological processes, Environmental Earth
Sciences, Vol. 73(3), pp. 1067–1082.
Deleuze, G. and Guattari, F. (1996) What is philosophy? Columbia University Press, New York.
Eaton, B.R., Fisher, J.T., McKenna, G.T. and Pollard, J. (2014) An ecological framework for wildlife habitat design for oil sands mine
reclamation, OSRIN Report No. TR-67, Oil Sands Research and Information Network, University of Alberta, School of Energy
and the Environment, Edmonton, Alberta, 83 p.
Ehrenfeld, J.G. (2000) Defining the limits of restoration: the need for realistic goals, Restoration Ecology, Vol. 8(1), pp. 2–9.
Evans, H. and Goodjohn, M. (2008) Traditional knowledge – a resource for mining companies, Canadian Mining Journal, viewed 19
March 2015, http://www.canadianminingjournal.com/news/traditional-knowledge-a-resource-for-mining-companies/ 1000
226265/?type=Print%20Archives, last updated 01 October 2008.
Fair Mining Collaborative (2014) Mine closure and post closure: progressive reclamation, viewed 12 November 2014,
http://www.fairmining.ca/code/mine-closure-and-post-closure-3/progressive-reclamation/.
Goodbody, A. (2013) Closing the deal, Mining Magazine, 29 November 2013, http://www.miningmagazine.com/management/
general-management/closing-the-deal/.
Hattingh, R. and Bothma, J. (2013) Taking the risk out of a risky business: a land use approach to closure planning, in Proceedings 7th
International Mine Closure Conference (Mine Closure 2013), 18–20 September 2013, Cornwall, U.K., M. Tibbett, A.B. Fourie
and C. Digby (eds), Australian Centre for Geomechanics, pp. 15–24.
Hustrulid, W. and Kuchta, M. (1998) Open pit mine planning & design, Volume 1 - fundamentals (2nd ed), A.A. Balkema, Rotterdam,
Netherlands.
International Network for Acid Prevention (INAP) (2009) Global acid rock drainage guide (GARD Guide), International Network for
Acid Prevention, http://www.gardguide. com/.
Kahn, J. R., Franceschi, D., Curi, A. and Vale, E. (2001) Economic and financial aspects of mine closure, Natural Resources Forum, Vol.
25, pp. 265–274.
Lee, T. D. (1984) Planning and mine feasibility study – an owners perspective, in Proceedings of the 1984 NWMA Short Course, Mine
Feasibility – Concept to Completion, G.E. McKelvey (ed), Northwest Mining Association, Spokane, WA.
McKenna, G. (2002) Sustainable mine reclamation and landscape engineering, PhD thesis in Geotechnical Engineering, Department
of Civil and Environmental Engineering, University of Alberta, Edmonton, Canada, 660 p.
McKenna, G. and Dawson, R. (2014) Soft tailings capping and landform design, Hot topics in oil sands tailings management, IOSTC
short course, 7 December 2014.
Nordic Innovation (2012) The future of the mining industry: opportunities, challenges and responsibilities, viewed 11 January 2014,
www.nordicinnovation.org/news/the-future-of-the-mining-industry-opportunities-challenges-and-responsibilities/, last up-
dated 23 April 2012.
Quinlan, R.W., Downey, B.A., Taylor, B.N., Jones, P.F. and Clayton, T.B. (eds) (2003) A multi-species conservation strategy for species
at risk in the Milk River basin: year 1 progress report, Alberta species at risk report No. 72, Fish and Wildlife Division, Alberta
Sustainable Resource Development, Edmonton, AB, 229 pp.
Rio Tinto (2012) Daybreak’s environmental history, Viewed 1 April 2015, http://riotintokennecott.com/sites/kennecott.
com/files/daybreaks_environmental_history.pdf.
Rio Tinto (2014) Diavik diamond mine – development of the A21 pipe, (press release), 27 November 2014,
http://www.riotinto.com/media/media-releases-237_13678.aspx.
Sagan, D. (2008) Mines and design in their natural context, in A. Berger (ed), Designing the reclaimed landscape, Taylor & Francis,
New York, NY.
Slingerland, N. Baida, M. and Wilson, G.W. (2014) New teams for new schemes: diversification in mine closure, Canadian Reclamation,
Vol. 14(2), pp. 22–28.
Steinitz, Carl (2012) A framework for geodesign: changing geography by design, ESRI Press, Redlands, California, 224 pp.
McKenna, G. and Dawson, R. (2014) Soft tailings capping and landform design, Hot topics in oil sands tailings management, IOSTC
short course, 7 December 2014.


U. Barnekow Wismut GmbH, Germany

T. Metschies Wismut GmbH, Germany
G. Merkel Wismut GmbH, Germany
M. Paul Wismut GmbH, Germany
From 1960 to 1990, the former Soviet-German Wismut company milled uranium ore and processed a total of
about 110,000 t of uranium at the Seelingstädt mill in Thuringia, Germany. In total about 85 million m³ of mill
tailings were disposed from 1967 to 1990 into two partial ponds of the uranium mill tailings pond Culmitzsch
(pond area: 2.43 km2). Dry decommissioning in situ started in 1995. It included the following major
remediation steps: expelling pond water; collecting seepage and runoff, including water treatment;
temporarily covering tailings surfaces; reshaping dams; recontouring ponds; installing final covering,
including vegetation and diverting runoff from the tailings pond to the receiving streams. Interim covering
will end in 2016. Reshaping of the dams will continue until 2018 and reshaping of the ponds until 2027. In
2012, Wismut filed applications for final covering. Having received the first partial permit under the mining
law, it will begin the final covering on pond B in spring 2015 and shall end in 2028. On tailings beaches, the
final cover design consists of a multilayer-type cover consisting from top to bottom of 2.0 m storage and
recultivation layer, 0.3 – 0.5 m drainage layer, 0.3 m sealing layer made of cohesive clean soil (hydraulic
conductivity: kf ≤ 1E-9 m/s) and a sealing layer made of compacted waste rock dump material (hydraulic
conductivity: kf ≤ 5E-9 m/s). On fine tailings areas, the final cover consists of 1.0 m of clean soil or waste rock
dump material (mixed-grained soil from overburden rock layers: Ra-226 ≤ 200 Bq/kg) above a layer at least
1.5 m thick of compacted waste rock dump material (mixed-grained soil). The pond area is designed to be
recontoured in shallow hills and valleys. The multilayer-type final cover shall be constructed on short slopes
with a slope fall of v:h = 1:10 to 1:15 with slope length of less than 100 m. This design will allow for clean
surface runoff while reducing infiltration in sandy tailings as much as possible. The design of the final cover
may take some credit for the optimisation of the site water management. In February 2015, Wismut filed an
application for planning approval under water law to connect the surface runoff from the Culmitzsch tailings
pond to the receiving streams. This will include additional earthworks and construction of diversion ditches
and a runoff retention pond. Permitting procedures have been and will be influenced by stakeholders, NGOs
and further participating parties with concurrent interests. This paper presents the remediation progress
achieved, the status of the discussions and the major challenges ahead, in particular regarding final covering,
vegetation and landscape planning, runoff diversion and site water management.
From 1960 to 1990, the former Soviet-German Wismut company milled uranium ore and processed a total
of about 110,000 t of uranium at the Seelingstädt mill in Thuringia, Germany. In total, about 85 million m³ of
mill tailings were disposed from 1967 to 1990 into two partial ponds of the uranium mill tailings pond
Culmitzsch (pond area inside the dam crests: 2.43 km2). The location of the tailings ponds Culmitzsch and
Trünzig and of the Seelingstädt mill site as of 1991 are shown in Figure 1.

Remediation of Wismut’s uranium mill tailings pond Culmitzsch – progress and challenges U. Barnekow, T. Metschies, G. Merkel and M. Paul
The Culmitzsch tailings pond was erected in a former uranium open pit. From 1963 to 1966, rock-fill dams
were erected in the open pit and along the southern and northern rim; these separate two partial ponds for
separate disposal of mill tailings from soda-alkaline and acid leaching. Acid tailings were neutralised before
being disposed. During the operational phase, seepage and groundwater catchment systems were
successively installed. A row of pumping wells was installed in the north in the 1980s, and gravel-filled drain
trenches and other drains were installed north and south of the tailings pond. Immediately after the end of
the operational phase, initial securing measures were taken, including interim covering of tailings beaches to
reduce dusting, installation of monitoring systems and additional water collection systems.
The fundamental decision for dry-decommissioning in situ was agreed by the responsible authorities in 1995.
The site remediation included the following major remediation steps: interim covering of tailings surfaces,
reshaping of dams and recontouring of ponds, final covering, vegetation and landscaping and diversion of
runoff from the tailings pond to the receiving streams. The site water management has been crucially
important to progress during the entire remediation phase. Dry decommissioning in situ started while the
site water management from operational period was still in place. After the end of the yellow cake production
in 1996, part of the Seelingstädt mill was used for site water treatment. A new water treatment plant
(capacity: 300 m3/h, 330 m3/h since 2013) was erected in 2001. The site water management included
expelling supernatant pond water, collecting seepage and contaminated surface runoff including water
treatment and discharging this to the receiving stream, the Culmitzsch creek (see Figure 1).
This paper presents the actual remediation progress achieved to date and the challenges ahead regarding
future final covering, landscape management, the planning approval procedure for diversion of runoff to the
receiving streams and challenges regarding the long-term optimisation of site water management.

Case Studies
The locations of the Culmitzsch and Trünzig tailings ponds and the Seelingstädt mill are shown in Figure 1.
The Culmitzsch tailings pond is located in a mountainous region with altitudes from approximately 200 m
above sea level (m.a.s.l.) at the Weisse Elster river up to approximately 360 m.a.s.l. at the top of the waste
rock dumps around the Culmitzsch tailings pond. The topographic map in Figure 1 also shows the catchment
areas of the second-order receiving streams, the Fuchsbach creek and the Culmitzsch creek. The next first-
order receiving stream of is the Weisse Elster river. The Weisse Elster has its source near the Czech border
and flows northward to the city of Leipzig and further on to the Saale river, which is a tributary of the Elbe
river. Wismut´s former Ronneburg uranium mining area is located about 15 km north of Seelingstädt. There,
treated mine water is discharged into the Wipsebach creek. The confluence of the Wipsebach creek into the
Weisse Elster river is about 15 km north of the confluence of Culmitzsch creek, just upstream of the city of
Gera (population: approx. 100,000).
The Trünzig tailings pond and the southern smaller part of the Culmitzsch tailings pond are located in the
hydrological catchment area of the Culmitzsch creek (catchment area: 40.8 km²; creek length: ca. 11 km).
The major part of the Culmitzsch tailings pond is located within the regional catchment area of the Fuchsbach
creek (catchment area: 33.8 km²; creek length: ca. 10.7 km). The preferential land use in both catchment
areas is agriculture, which covers 75% of the land; 15% is used as forest land and 10% is residential areas.
The central European climate is characterised by an average temperature of 8.5°C and an average annual
precipitation rate of 700 mm.
Figures 2 and 3 show aerial photos of the situation in the early 1990s. Figure 2 provides an aerial view from
the northwest on the Culmitzsch tailings pond in 1991. Pond B is located in the foreground and pond A in the
background. One can see the north dam in the foreground, the south dam in the background and the
separating dam between both partial ponds. The tailings are surrounded by waste rock dumps placed during
ongoing uranium mining in the former open pit. Figure 2 shows the Lokhalde waste rock dump (WRD) on the
left side of the tailings pond in the background and the Jashalde WRD in the foreground. On the right side,
one can see the Waldhalde WRD in the foreground and the southwestern WRD in the background. Figure 3
shows the Culmitzsch tailings pond in 1993 (view from SE to NW).

The tailings ponds were constructed in former open pits mined out from 1948 to 1967. The Culmitzsch
uranium open pit was mined out from 1957 to 1967. Figure 4 shows a geological cross-section with the
Culmitzsch tailings pond on the left and the Trünzig tailings pond on the right. The geological cross-section is
oriented NW-SE parallel to the general groundwater flow directions. It shows the tailings embankments,
internal waste rock dumps and the major geological layers.
Two layers of grey mudstone or siltstone were mined as uranium-rich strata. The ore layers were part of
nearly horizontally bedded sedimentary rock layers of Permian age situated in a half-graben. The Permian
rocks are up to 70 m thick and underlain by Ordovician shales folded in the Variscan orogenesis. The Permian
sedimentary rocks consist from bottom to top of a base of conglomerate, mudstone, siltstone and sandstone
layers (in particular the “Culmitzsch sandstone”) overlain by a thin layer of weathered dolomite residues,
weathered mudstone and a sandstone layer of Lower Triassic age. Due to open-pit mining, overburden
materials were relocated onto the waste rock dumps in the surrounding area and inside the open pit. Three
rock-fill dams were erected from 1963 to 1966, enclosing two partial ponds; these include the north dam,
the south (and southeast) dam and the separating dam. The separating dam has a vertical sealing wall in the
dam centre.
Tailings disposal lasted from 1967 to 1990. Tailings from acid leaching were discharged into the southern
partial pond Culmitzsch A, applying the centreline method along the southern tailings beaches. Tailings from
soda-alkaline leaching were disposed from the northern perimeter into the northern partial pond B.
According to the historic discharge regime, sandy tailings beaches settled near the discharge spots while fine

Case Studies
tailings settled more distantly below the water table. For both the Trünzig and the Culmitzsch tailings ponds,
the hydraulic regime in the permeable surrounding or underlying geological layers, in particular in the main
aquifer layer of Culmitzsch sandstone, is closely connected with the hydraulic conditions in the permeable
sandy tailings. This led to significant contaminant seepage during and after the operational phase.
Geotechnical investigations on the situation in 1992 indicated a need for dam stabilisation with respect to
seismic dam stability.
Pore and seepage water composition is strongly determined by the chemical processing applied (Table 1).
While pore water from the acidic process deposited in pond A contains very high concentrations of iron and
magnesium, resulting in elevated hardness, the main characteristics of the pore waters at pond B are high
uranium concentrations. Sulphate and chloride concentrations are generally high in both ponds. Table 1
shows the average as well as a representative range of the concentrations of major water constituents based
on samples from sampling wells in the sandy and transition zones. Table 2 shows the same data for the fine
tailings. Pore water of the fine slime tailings contains even higher concentrations (Metschies et al., 2014).
2000–2013 pH U [mg/l] Mg [mg/l] Ca [mg/l] SO4 [mg/l] Cl [mg/l] Fe [mg/l] Ni [µg/l]

Pond A
Median 8.0 0.9 1,300 380 12,000 1,500 0.67 28
10th percentile 7.2 0.2 590 340 7,200 860 < 0.08 < 15
90th percentile 8.3 2.2 2,300 420 15,000 2,200 104 470
Pond B
Median 8.3 6.2 110 27 8,200 1,100 0.18 < 20
10th percentile 7.8 0.9 66 11 6,200 750 < 0.04 < 10
90th percentile 8.7 11.2 650 380 9,500 1,500 1.15 40
2000–2013 pH U [mg/l] Mg [mg/l] Ca [mg/l] SO4 [mg/l] Cl [mg/l] Fe [mg/l] Ni [µg/l]

Median 6.7 0.3 1,400 380 11,000 1,300 75 410
10th percentile 5.9 0.1 700 360 6,600 770 14 68
90th percentile 8.0 1.9 2,800 460 15,000 2,400 220 1400
The remediation has to ensure the safe long-term storage of the tailings by reducing the additional equivalent
dose to the population from all pathways to less than one millisievert per year (1 mSv/year). Requirements
for waste water discharge into the receiving streams were set by the authorities of Thuringia.
The first measures to secure against acute risks started in 1991. Dry decommissioning in situ started in 1995
and will last until 2028. It comprises the following major work steps: site water management during
remediation, interim covering of tailings surfaces, reshaping of dams, recontouring of ponds, final covering,
vegetation and landscaping, diversion of runoff connecting the tailings pond to the receiving streams and
optimisation of site water management during and after closure. Figures 5 and 6 show the status in 2014.

Figure 5 shows a view from SE to NW in December 2014. One can see the remaining air-exposed tailings
surface and the completed embankment fill along the dam crest of the southern dam. Dam buttressing has
just started at the bottom of the dam. On the tailings surface in the foreground, one can see the existing
storage pond of the water treatment plant, which is located in front of the dam toe on the right-hand side.

Case Studies
Figure 5 also shows the new iron treatment plant, including two partial ponds under construction. It is located
on the pond surface in the foreground. Both Figures 5 and 6 show dwellings in the immediate vicinity of the
tailings pond.
The interim cover is needed to avoid dusting on tailings beaches and to create a trafficable surface on fine
tailings for further remedial works. Tailings beaches were temporarily covered during the early 1990s. In
pond B, supernatant pond water was expelled by 2002. Interim covering of fine tailings included the use of
vertical wick drains (1.5 m triangular grid spacing and 5 m depth), the use of geotextiles and geogrid followed
by placement of thin layers of gravel-sand mixture overlain by 0.5 m of mixed-grained waste rock dump
material. On pond B, interim covering was completed in 2006. Figures 5 and 6 show the status in 2014. During
past years, the progress of the interim covering on pond A was affected by repeated heavy precipitation
phases. Nowadays, due to the stepwise connection of other remediated surfaces at Seelingstädt to the
receiving streams, the capacity of the treatment plant can continuously be used to grant a dry tailings surface
in the centre of pond A. Therefore the remaining air-exposed tailings surface is currently 1.6 ha. The interim
cover needs to be completed only on a remaining area of approximately 12 ha. This shall be achieved by
2016.
Dam reshaping is needed with respect to seismic dam stability. Reshaping of the north dam consisted of dam
flattening inside the pond area to a general slope fall of v:h = ca. 1:3.6. This lasted from 2007 until 2008.
Dam reshaping of the south dam started with preparatory measures in 2010. By 2013, an embankment
(volume: ca. 1 million m3) was placed on the pond area next to the dam crest in order to consolidate the fine
tailings located next to the dam crest. The removal of this embankment has recently started. In the future,
fine tailings shall be locally relocated from the dam crest area to inside the tailings pond. Locally, fine tailings
will be backfilled with proper waste rock dump material in order to grant seismic stability for the respective
parts of the dam crest. In addition, dam reshaping consists of buttressing and flattening the upper dam slope
inside the pond. Figure 5 shows the completed embankment along the dam crest. Reshaping of the south
dam to a general slope fall of v:h = ca. 1:3.6 shall be completed by 2018. Figure 6 shows the reshaped and
covered north dam and the recontoured northern slope and remediated subsurface of the northern part of
the Waldhalde waste rock dump in the foreground. This WRD is currently being partly relocated on pond B.
Recontouring of ponds is carried out with respect to geotechnical stability and functionality of the future final
cover, granting a stable surface runoff without ponding. Fine tailings consolidation is enhanced by combining
recontouring with the use of deep vertical wick drains. The surface contour of the pond areas was optimised
based on modelling of the time-dependent tailings consolidation with respect to the designed recontoured
surface and the time-dependent loading of the tailings surfaces during (future) recontouring works.
Recontouring of pond B shall be completed by 2018. Recontouring of the tailings beaches of pond A will be
more or less completed at the same time. Figure 6 provides an aerial view of the Culmitzsch tailings pond
(view to SE), which shows the remedial progress achieved in 2014. In the foreground, one can see the
Fuchsbach valley; Fuchsbach creek will be the main receiving stream for the diversion of surface runoff from
the major part of the Culmitzsch tailings pond. In the foreground, the north dam and the tailings area have
already been recontoured. On the right, one can see the ongoing relocation of the Waldhalde waste rock
dump. The waste rock material is being relocated onto the northern partial pond B for recontouring.
The design planning for recontouring the central part of pond A was completed in 2014. In January 2015, we
filed the respective applications for the necessary permits. Recontouring of pond A shall be completed by
2027. For this, we prepared by December 2014 the revised framework planning for the remaining
remediation of the entire site. This framework planning includes a total volume of 16.5 million m3 to complete

the interim cover and to construct the final contour and the final cover, including approximately
2.9 million m3 of clean cohesive soils and gravel-sand-mixtures. Starting in 2018–2019, the Culmitzsch tailings
pond will be the last remaining area for disposal of radioactively contaminated soils in Thuringia. Therefore,
an estimated 0.7 million m3 of contaminated soils from remediation of contaminated mining areas will need
to be disposed within the recontouring layer as well. According to the framework conception for the
Culmitzsch tailings ponds to date, about 18 million m³ of materials will be available on-site for the remaining
remediation of the Culmitzsch tailings pond. The Lokhalde waste rock dump (shown in the background on
the left-hand side of Figure 6) shall be completely relocated, including the remaining volume of 10 million m3.
About 1.8 million m3 of materials with a specific activity ≤ 200 Bq/kg Ra-226 in solids from this waste rock
dump should be separated for constructing the final cover layers in the fine tailings area.
The final cover design for the Trünzig tailings pond was completely permitted by 2006. The final cover design
depended on the underlying tailings types. On sandy tailings beach zones, the final cover consists of a 2 m
thick storage layer including a compacted bottom lift (0.5 m) above a compacted 1 m thick interim cover.
Immediately before final cover placement, the upper layer of the interim cover was compacted, granting a
hydraulic conductivity of kf ≤ 5 × 10 9 m/s. In addition, the lowest 0.5 m thick layer of the final cover was
compacted to at least 95% of Proctor density. The final cover material on the Trünzig tailings pond consists
of waste rock dump material (mixed-grained soil) from the Lokhalde WRD located in the vicinity of the
Culmitzsch tailings pond. On fine tailings zones, the functionality of the final cover shall be granted by a
minimum 2.5 m thick layer consisting of mixed-grained waste rock dump material compacted to 95% of the
Proctor density. The waste rock dump material of the Lokhalde WRD is characterised by an average specific
activity of 250 – 300 Bq/kg Ra-226. Depending on the geological origin, certain volumes of the Lokhalde WRD
can be characterised as clean soil, having a specific activity of Ra-226 ≤ 200 Bq/kg. Those mixed-grained soil
materials have to be used to construct the uppermost 0.5 m layer on the fine tailings and the entire 2 m thick
final cover on the tailings beaches.
Wismut filed applications for final covering of the entire Culmitzsch tailings pond in December 2012. The final
cover design was prepared with respect to the results of the conceptual site modelling on contaminant
transport from the sources (tailings) to the receiving streams. Beyond this and among other requirements,
the final cover was designed to grant erosional stability of the surfaces and to provide sufficient available
water for plants during growing seasons. The preferred final cover design was developed and discussed with
the authorities, taking into account the overall water management of the entire site with regard to the long
term. Wismut and the authorities agreed that, irrespective of the final cover design, the release of
contaminants will remain at a level that requires water management measures in the long term. A full
encapsulation of the contaminants could not be reasonably achieved due to missing natural or engineered
barriers. Water pumping and treatment will be required for the long term. Therefore, it was agreed in 2011
that Wismut may take credit for water management measures to be designed together with the final cover
and to be realised in advance to improve the water management system of the sites (Culmitzsch TP, Trünzig
TP and mill site) accordingly. The measures should be evaluated using the site model of contaminant
transport.
The subsurface of the final cover was recontoured in a shallow valley and the hills were structured with a
slope fall of at least v:h 1:15 to 1:10 and a slope length below 100 m. On the tailings beaches and transition
zones, covering approximately 100 ha, the designed final cover consisted of a 1.5 m storage layer above a 0.3
– 0.5 m drainage layer with a hydraulic conductivity of kf ≥ 1 × 10 4 m/s. The reason for this design was the
given suffosion stability of such a drainage material with respect to the overlying soil in the long term. Below
the drainage layer, a 1 m thick sealing layer of mixed-grained soil (waste rock dump material) should be
constructed with a hydraulic conductivity of kf ≤ 5 × 10 9 m/s. After initial establishment of grass for erosion
protection, afforestation was planned for the final cover on tailings beaches in order to assist in controlling

Case Studies
the percolation rate. Grassland was planned to be established on fine tailings areas according to
requirements set by the environmental authority.
During ongoing discussions with the authorities and due to stakeholder involvement, the first partial permits
under mining law and under radiation protection law prescribed an additional 0.3 m thick sealing layer with
a hydraulic conductivity of kf ≤ 1 × 10 9 m/s to be constructed underneath the drainage layer. This replaces
the uppermost 0.3 m of the originally 1 m thick compacted mixed-grained soil layer (kf ≤ 5 × 10 -9 m/s). For
this, we have to buy additional cohesive soil from outside. Because of the implementation of this tight layer,
which might be sensitive to root penetration, we were also additionally required to increase the thickness of
the storage layer from 1.5 m to 2.0 m. According to these first partial permits, this design shall be constructed
on the tailings beaches of pond B as a test field (area: 13.7 ha). In 2015, a lysimeter station (size:
approximately 20 m × 20 m) shall be constructed on this area inside the final cover.
On the fine tailings zones, the final cover was designed according to the final cover design on the Trünzig
tailings pond; this consisted of 0.5 m of clean waste rock dump material above a layer of at least 2.0 m of
compacted mixed-grained waste rock dump material. According to the first partial permits received in
February 2015, we are required to construct at least 1 m of clean soil and to grant 2.5 m of cover thickness
above tailings. With respect to the approximately 140 ha to be covered, we currently expect that there will
not be enough clean waste rock dump material available (Ra-226 ≤ 200 Bq/kg). Therefore, additional clean
soil might be needed. Final covering has not started yet. It shall start in spring 2015 and shall be completed
and vegetated by 2028. The final cover surface will be vegetated.
Based on a detailed ecological survey of the entire site, the landscape planning and vegetation of the
Culmitzsch tailings pond and its surrounding dams and waste rock dumps was prepared with respect to
balancing ecological intervention measures and compensatory measures with public interests. All ecological
intervention measures are balanced by compensatory measures. Some of the needed compensatory
measures including open grassland and lakes are to be realised about 15 km to the north, in the Ronneburg
mining area. Supervision of the construction of all earthworks at the sites is accompanied by ecological
construction monitoring, which prescribes in detail the preserving or compensatory measures to be taken
during ongoing construction works. In 2011, we received planning approval for the landscape management
plan for the entire site. It took several years to find appropriate sites for compensatory measures to fulfil the
requirements fixed by the landscape management plan.
It is expected that the landscape management plan will have to be further adapted to the design of the final
cover after we receive the final permits. Taking into account the requirements of the permits concerning the
cover design, we are obliged to review the vegetation concept with respect to cover functionality in the long
term. In this regard, we have to reconsider the distribution of open grassland and afforested areas.
Requirements from the future planning approval for the connection of the runoff from the facility to the
receiving streams may also have consequences for the landscape management plan. Figure 7 presents the
generalised landscape management plan, the actual plan for land use for the long term and the most recent
status of the design of the future runoff diversion system.

Pond B
Pond A

Case Studies
For the Culmitzsch tailings pond, the permitting authorities of Thuringia stipulated that the hydrological
situation downstream in the receiving streams before remediation (1990) must be preserved for the future
diversion of runoff with respect to flood water flow resulting from the 100-year precipitation event. The
planned ditches shall divert the runoff into the Fuchsbach creek upstream of the Wolfersdorf village.
Hydrological modelling was also carried out for the state before remediation (1990) and two states after
remediation: an interim state with only grass vegetation on the remediated areas and a long-term state with
grown-up trees established. The results show that the hydrological situation before remediation (1990) could
not be preserved for the Fuchsbach creek without a runoff retention basin. Therefore, a runoff retention
basin was planned for north of the Culmitzsch tailings pond. The location of the runoff retention basin was
chosen to avoid the use of foreign land, to avoid contact with contaminated areas and to meet environmental
criteria. This location also allowed costs to be reduced as much as possible.
In February 2015, we filed an application for planning approval under water law to connect the surface runoff
from the Culmitzsch tailings pond and certain surrounding areas to the receiving streams of Fuchsbach creek
in the north and Culmitzschbach creek in the south. Figure 7 shows the location of the runoff diversion ditches
and the runoff retention basins planned for the Culmitzsch and Trünzig tailings ponds. The remediation works
on the Trünzig tailings pond are nearly completed except for the planned construction of a runoff retention
basin in the Finkenbach creek valley (see Figure 7). The planning approval procedure regarding the
connection of the Trünzig tailings pond to the receiving streams is not yet complete. Figure 7 shows the
Seelingstädt mill site. This area is completely remediated. The water diversion system on the Culmitzsch
tailings pond connects the major area of the facility to the Fuchsbach creek in the north and the remaining
area to the Culmitzsch creek in the south. The entire runoff to the north will be diverted through this runoff
retention basin, while there will be no such basin needed in the south.
Major challenges to this planning approval procedure will arise due to concurring interests of stakeholders,
which require flood protection downstream to be improved and water quality standards in the receiving
streams to be met in the long term. However, the application documents demonstrate that the preservation
of the hydrological situation before remediation (1990) is ensured by the planned measures on the
remediation site. Furthermore, the cover design ensures clean surface runoff from the remediated site.
Water management at the Seelingstädt site includes the collection and treatment of surface, ground and
seepage waters at the tailings pond and in the surroundings. Various drainage systems and pumping wells
were built to contain contaminated waters. However, there is a small proportion of seepage water that still
infiltrates the local receiving streams, influencing the chemical composition of the surface waters. Apart from
the remediation objects managed under the Wismut remediation program, there are additional sources
nearby upstream, which also influence the water quality of the receiving streams.
In 2001, the new water treatment plant utilising a lime treatment was put into operation, replacing the water
treatment unit of the former mill. The new plant’s initial capacity of 300 m³/h was increased to a rate of
330 m3/h in 2013. The new treatment process reduces the concentration of metals, semi-metals and uranium
to below the discharge limits but does not decrease the salt and hardness content. The treated water is
released into the Culmitzsch creek.
Standards for waste water discharge into the receiving streams were set by the regulatory authorities of
Thuringia for metals, semi-metals and uranium. Limits for salt concentrations and water hardness are given
for a sampling point in the receiving Culmitzsch creek and further downstream in the Weisse Elster river. At
least temporarily, these limits restrict the discharge volume of treated waters and therefore curtail the water
management operations at the Seelingstädt site. As a result, the full technical capacity of the water treatment
plant might not be available throughout the year, depending on the discharge conditions in the respective
water bodies and additional sources of salts and hardness in the watershed.

During the ongoing remediation of the Culmitzsch tailings pond, additional major challenges for the water
management are as follows:
 Treat and release supernatant waters from pond A in order to allow sufficient work to progress on
the surface areas.
 Collect and treat the released, pumped or expelled pore waters.
 Continue to adapt the water treatment to the changing water composition to ensure the full
treatment capacity is available throughout the year.
Fine tailings consolidation is enhanced by recontouring and using deep vertical wick drains. The future
consolidation of fine tailings will release pore waters in the order of nearly 5 million m³ over 10 – 15 years.
To ensure that sufficient work progress is made, the water management at the site must do two things: (1)
reduce the formation of contaminant water and (2) install a reasonable storage volume for the period up to
the end of physical remediation works. Separately collecting uncontaminated surface waters from already
remediated areas and directly discharging them into the receiving creeks implemented over the past years
reduced the treatment volume by about 20%. This process shall be continued during the future remediation.
Long-term concentration trends for sulphate and uranium are influenced by two partly opposing processes:
(1) a continuous dilution of the outflow from the contaminant source and (2) the additional water release of
highly concentrated pore waters by tailings. Due to the discharge limits for sulphate, chloride and hardness
in the receiving creek may interrupt the discharge of treated waters, especially under low-flow conditions.
To help this, we were allowed to temporarily use water pumped from the Weisse Elster river to amend the
discharge in the receiving stream. This ensured continuous discharge of treated waters from the remediation
site.
In 2014, two additional pumping wells were installed in the tailings beaches of pond A. Since then about one
million m3of contaminated pore waters have been pumped out, treated and discharged. This measure
decreased the groundwater table inside the tailings beaches significantly thus reducing the contaminant
seepage via the groundwater pathway to the receiving stream, the Culmitzsch creek. .
In 2016–2017, the geotechnical stabilisation of pond A will lead to the release of pore waters from the fine
tailings, which will have significantly increased iron concentrations of up to 250 mg/L. The existing treatment
plant cannot cope with this changed chemical composition of the water to be treated. The necessary
additional iron treatment step of the treatment plant is currently under construction (see Figures 5 and 6).
Metschies, T., Laubrich, J., Müller, J. and Haupt, M. (2014) Challenges of water management during tailings remediation – site and
catchment-specific focus, in Proceedings of the 7th International Conference on Uranium Mining and Hydrogeology, 21–25
September 2014, Freiberg, Germany, pp. 401–408.
Barnekow, U., Roscher, M. and Merkel, G. (2011) Final covering and diversion of runoff from Wismut’s uranium tailings ponds at
Seelingstädt (Germany) – status achieved from concepts to realization, in Proceeding of the Conference on Tailings and Mine
Waste, 6–9 November, 2011, Vancouver, BC, Canada, pp. 803–814.

C.B. Brown AREVA Resources Canada Inc., Canada

D.M. Huffman AREVA Resources Canada Inc., Canada
B.J. Schmid AREVA Resources Canada Inc., Canada
The Cluff Lake project is a former uranium mine and mill located in the Athabasca Basin of northern
Saskatchewan, Canada, approximately 900 km north of Saskatoon. Uranium mining and milling operations
commenced in 1980. During operation, facilities included open pit and underground mines, a mill, a tailings
management area, a residential camp and various other support and site infrastructure activities. Mining and
milling activities ceased at the site in 2002 after more than 20 years of production. In the time since the
completion of mining and milling at Cluff Lake, AREVA Resources Canada Inc. (AREVA) has focused on
decommissioning activities at the site.
The Canadian Nuclear Safety Commission (CNSC) granted a decommissioning license in 2004, and the
majority of the physical decommissioning occurred between 2004 and 2006. This work included mill
demolition, decommissioning of the open pits and covering of the waste rock pile and tailings management
area. From 2006 to 2013, personnel remained on site for post-decommissioning activities and follow-up
monitoring. The Cluff Lake project reached a milestone in 2013, when the remainder of site clean-up work
was completed, a permanent site presence was discontinued and the site access restrictions were removed.
AREVA now conducts routine environmental monitoring of the Cluff Lake project through quarterly
campaigns, where a small team is mobilised to site from Saskatoon to collect the required environmental
data. The first full year of campaign monitoring occurred in 2014 and proved to be an effective method for
environmental monitoring. In addition to campaign monitoring, AREVA is completing an environmental
assessment follow-up program for the Cluff Lake project under the Canadian Environmental Assessment Act
(CEAA). This paper provides an overview of the decommissioning history of the Cluff Lake project site, with a
focus on the transition to campaign monitoring, the follow-up program and the next steps for the site.
The Cluff Lake project, owned by AREVA Resources Canada Inc. (AREVA), is a decommissioned uranium mine
and mill site located in the Athabasca Basin of northern Saskatchewan, Canada. The mine site is located
approximately 900 km north of Saskatoon, Saskatchewan, and approximately 75 km south of Lake Athabasca
(Figure 1).

Decommissioning the Cluff Lake project C.B. Brown, D.M. Huffman and B.J. Schmid
The Cluff Lake ore deposits were first discovered in the mid-1960s. Uranium mining and milling operations
commenced in 1980 and ceased in 2002. Over the 22-year operating life of the mine, five ore bodies were
extracted using either underground or open pit techniques. Cluff Lake produced 28 million kilograms of
uranium concentrate (U3O8) over its operational period and produced its final barrel of yellowcake in
December of 2002.
The decommissioning of the Cluff Lake project is governed by a number of federal and provincial acts,
regulations and policies. The federal government has responsibility for the regulation of matters relating to
nuclear energy, including uranium mines, under the Nuclear Safety and Control Act. Licensing for the various
stages of a uranium mining and milling project, including decommissioning, is the responsibility of the
Canadian Nuclear Safety Commission (CNSC). The Cluff Lake project is also subject to a number of provincial
acts and regulations and is regulated provincially by the Saskatchewan Ministry of Environment (SMOE).
Planning for the decommissioning of the Cluff Lake project began in 1998. Decommissioning requires a
licence from the CNSC, and the process of securing this licence triggered the requirement for a federal
environmental assessment (EA). AREVA submitted a comprehensive study report (CSR) for decommissioning
to regulatory agencies in December 2000. The decommissioning plan was subjected to intensive review and
discussion through a five-year assessment under the Canadian Environmental Assessment Act (CEAA). The
Cluff Lake project remained in care and maintenance mode from 2002 to 2004, throughout the review
process. During this time, several clean-up and reclamation activities were conducted to prepare for future
decommissioning. The CSR was approved in April 2004, and authorisations were received to proceed with
physical decommissioning activities.
The Cluff Lake project presently operates under a decommissioning licence issued by the CNSC, and an
approval to operate pollutant control facilities issued by the SMOE. During operations, the site was also
subject to the Metal Mining Effluent Regulations (MMER), which came into effect 6 December 2002. The
Cluff Lake project received recognised closed mine status on 16 January 2006, and no it longer falls under
the MMER.

Case Studies
Uranium mining companies in Saskatchewan are required by the SMOE and the CNSC to develop
decommissioning and reclamation plans, including financial assurance. Since the submission of the CSR,
AREVA has maintained and routinely updated a detailed decommissioning plan based on the estimated
remaining decommissioning and monitoring costs. The detailed decommissioning plan serves as the primary
reference in the decommissioning license from the CNSC.
In 2007, the Province of Saskatchewan legislated the Reclaimed Industrial Sites Act and the Reclaimed
Industrial Sites Regulations to put into effect an institutional control program (ICP). The ICP outlines the
process for long-term monitoring of sites where activities have ceased and decommissioning has been
completed and approved, and the eventual transfer of the site back to the province. AREVA’s
decommissioning activities at Cluff Lake are working towards the goal of transferring the site into the ICP.
The facilities at Cluff Lake included open pit and underground mines, a mill, a tailings management
area (TMA) with a two-stage effluent treatment system, a residential camp, an airstrip and various other
support and site infrastructure facilities.
The site consists of four major development areas, as outlined in Table 1.
Development area Associated infrastructure

Tailings management facility, primary
Tailings management area treatment system and secondary treatment
system, settling ponds, adjacent borrow area
Mill complex Mill infrastructure and warehouses
D, Claude, Dominique-Janine North (DJN) and
Dominique-Janine Extension (DJX) open pits,
Mining area OP, Dominique-Peter (DP) and Dominique-
Janine (DJ) underground, Claude waste rock
pile
Germaine camp Residential camp and kitchen facilities
AREVA’s key objective for decommissioning is to remove, minimise and control potential contaminant
sources and thereby minimise the potential for adverse environmental effects associated with the
decommissioned property. The decommissioning plan has been designed to reach a project end state where:
 the environment is safe for use by human and non-human biota;
 the reclaimed landscape is chemically and physically stable;
 the landscape is stable and self-sustaining, allowing for the use of the land for traditional
purposes such as trapping, hunting and fishing;
 potential constraints on future land use are minimised.
To evaluate the success of the decommissioning program, site-specific objectives were developed in
consultation with federal and provincial authorities as well as through public consultation processes. The
following criteria are indicators of decommissioning success:
 achieving decommissioning surface water quality objectives (DSWQO) and other accepted
decommissioning objectives at surface water and flooded pit locations;

 levels of gamma, radon and long-lived radioactive dust (LLRD) pose no unacceptable risk to
traditional land use;
 a stable, self-sustaining landscape;
 reduction of infiltration rates around the TMA and Claude waste rock pile (CWRP) to levels that
adequately restrict contaminant movement to groundwater and are suitably protective of
downstream surface water receptors;
 return of the site to an aesthetically acceptable state, similar in appearance and land capability to
that which existed prior to mining activities, and that poses no unreasonable risk to humans or
the environment.
The CSR outlined decommissioning objectives for surface water quality, sediment quality and site-specific
water quality standards for flooded pits.
The TMA was constructed in 1980 and received all tailings produced by the mill throughout the life of the
project. The TMA is an above-ground facility located in a topographic low with tailings solids retained behind
a clay core dam structure. The TMA was also used throughout the operational period as a receptor for all
contaminated mine water and site runoff. Water treatment occurred through a two-stage process primarily
designed for the removal of radium-226. Water was first treated in the primary treatment system prior to
discharging to two settling ponds and an area known as the liquids pond in the TMA for the settling of
precipitates. Water from the liquids pond was fed to the secondary treatment system for further treatment.
Treated water was held in settling ponds prior to final discharge to the outlet of Snake Lake. North and south
diversion ditches around the TMA convey uncontaminated water from the drainage basin surrounding the
TMA to Snake Lake, minimising infiltration of clean water into the TMA. The ditches are designed to safely
divert a probable maximum precipitation event around the tailings area.
Physical decommissioning of the TMA included a grading course, backfilling of the liquids pond, buttressing
of the main dam to 4:1 slopes for long-term stability, and construction of stormwater and runoff
management structures. An engineered soil cover of minimum one metre thickness was placed on the tailings
and revegetated with a native grass and legume seed mix. Cover monitoring stations were installed on the
TMA to evaluate performance of the cover system.
The mill operation at Cluff Lake consisted of primary and secondary crushing, grinding, acid leaching, counter-
current decantation, solvent extraction, yellowcake precipitation and drying. After approvals were received,
demolition began in November 2004 and was completed by December 2005. The area was graded and
planted with tree seedlings developed from tree species native to the site. Two warehouses were retained
for the post-decommissioning and follow-up monitoring period.
The D open pit (“D-Pit”) was mined from 1979 to 1981 and was the first orebody extracted at Cluff Lake. The
pit was flooded in 1983, and monitoring of the water column has been ongoing since 1987. The water column
is stable, and a chemocline has established in the pit. All DSWQO for D-Pit are being met in the upper 50% of
the water column. A waste rock pile was located adjacent to D-Pit, and this area was resloped and
revegetated shortly after pit flooding. Vegetation is well established on the D-Pit waste rock pile.
The Claude open pit was mined from 1982 to 1989. Decommissioning activities for Claude pit included
completion of dewatering, backfilling of the pit with waste rock, compaction of the waste rock, reclamation

Case Studies
of the ore pad, and demolition of the Claude shop area. The backfilled Claude pit contains waste rock from
OP/DP and DJ Underground, the Dominique-Janine open pits (DJN and DJX), and Claude, as well as demolition
waste. A soil cover was placed on the backfilled Claude pit and planted with tree seedlings developed from
native tree species. The adjacent waste rock pile was regraded and seeded with a native grass and legume
seed mix. A cover monitoring system was installed on the CWRP to evaluate performance of the cover
system.
The OP/DP and DJ underground mines operated from 1983 to 1999 and 1994 to 2002, respectively. The
underground mines have been allowed to flood naturally. The vent raises were backfilled, and reinforced
concrete caps were installed above the backfilled raises. A concrete plug was poured at the portal openings.
Infrastructure to support the underground mines has been demolished and the surrounding areas regraded
and revegetated.
The DJN open pit was mined from 1989 to 1991, and the adjoining larger DJX pit was mined from 1994 to
1997. Decommissioning of the DJN/DJX pit involved moving a portion of the existing waste rock in DJN pit to
Claude pit, flooding the pit, regrading the area, improving stormwater management and revegetating the
area. The water column of the DJX pit has been routinely monitored since 2006.
The Germaine camp was partially decommissioned during the physical decommissioning period, with a small
number of facilities retained to support post-decommissioning and follow-up monitoring.
With the majority of physical decommissioning completed in 2006, the Cluff Lake project moved into a period
of post-decommissioning and follow-up monitoring. AREVA currently conducts two distinct monitoring
campaigns: an environmental monitoring program (EMP) and a follow-up program (FUP). AREVA has also
conducted monitoring to demonstrate that radiological objectives have been met.
Figure 2 shows the EMP and FUP components in the TMA and mill areas, while Figure 3 shows the mining
area. The Germaine camp area, not shown in these images, is located south of the TMA and domestic landfill
in Figure 2.

AREVA conducted a routine EMP for the Cluff Lake project throughout operations, and a small permanent
staff was maintained on site from 2006 to 2013 to continue this monitoring in the post-decommissioning
period. The purpose of the EMP is to monitor the effectiveness of mitigative measures and demonstrate
compliance with licensing conditions and decommissioning objectives. The sampling program outlined in the
EMP is also a condition of the approval to operate pollutant control facilities issued by the SMOE.
The EMP specifies sampling locations and frequencies for surface water, groundwater, pit water quality,
water level monitoring, cover monitoring, meteorological data, air quality monitoring and routine
inspections. The EMP also contains requirements for monitoring environmental effects such as sediment,
benthic invertebrates, fish, water quality and limnology, which are collected every five years and evaluated
in a status of the environment (SOE) report.
An EA FUP is a key component of a decommissioning licensing application under the Nuclear Safety and
Control Act. FUPs are also an important component of the EA process under the CEAA. The purpose of an
FUP is to reduce uncertainties in the EA of a project, and determine the effectiveness of any measures taken
to mitigate the adverse environmental effects of the project.
Follow-up items for the Cluff Lake project can be broadly classified into three categories:

Case Studies
 establishing a post-operational baseline of aquatic conditions and wildlife use patterns to which
post-decommissioning monitoring can be compared;
 conducting follow-up studies to confirm and refine contaminant transport modelling downstream
of the mining area and TMA;
 conducting follow-up studies to confirm and refine ecological risk assessment (ERA) modelling.
Post-operational baseline studies to characterise the wildlife community were conducted in 2005. An
intensive study area, local study area, and regional study area were established at the Cluff Lake site along
with two control watersheds that were not impacted by mining and milling activities. A focus of the study
was determining the presence of moose and muskrat, which were identified as the two species most at risk
in the study area.
A number of aquatic baseline studies were completed in 2004. The Cluff Lake project area consists of two
major drainage areas. The Island Creek drainage area carried treated effluent from the secondary treatment
system, discharged at Snake Creek, through Island Lake, the Island Lake Fen, and Island Creek to Sandy Lake.
The drainage path can be seen in Figure 2. The Cluff Creek drainage area collects flow through a series of
tributaries in the mining area that drain towards Cluff Lake. Cluff Creek drains from Cluff Lake and passes
through several small lakes prior to discharging in the Douglas River. Figure 3 shows the tributaries in the
mining area (which include Claude Creek, Peter River, Earl Creek and Boulder Creek) and their drainage into
Cluff Lake.
The aquatic environment was characterised by evaluating differences in the physical components, such as
limnology, water, sediment, and fish tissue chemistry, with biological components such as benthic
invertebrate communities and fish populations. Aquatic baseline studies were conducted on several
exposure stations in the Island Creek drainage area and Sandy Lake. Reference stations were established at
four locations. Additional aquatic studies took place in 2005, 2006 and 2009.
A detailed hydrogeology and contaminant transport modelling study was completed in the CSR to assess
various decommissioning options. The CSR predicted two primary potentially adverse effects: seepage from
the mining area, which may adversely affect downstream groundwater and surface water quality in the
Claude Creek and Peter River systems draining to Cluff Lake, and seepage from the TMA, which may result in
adverse effects to groundwater and surface water quality in Snake Lake and downstream waterbodies in the
Island Creek drainage. The success of the preferred decommissioning approaches relies on the performance
of effective cover systems to reduce infiltration, passive removal mechanisms that will remain effective and
reliable in the long term, and the verification of source terms for tailings and waste rock pore water
concentrations.
In the mining area and TMA, a number of groundwater wells are sampled through the EMP for the purpose
of assessing source term trends. General groundwater flowpaths are indicated with arrows in Figures 2 and 3.
Groundwater levels are also collected from these wells and used in modelling. Drilling programs were
completed in 2010 and 2012 in both the mining area and TMA to collect additional data. Nickel is the primary
constituent of concern (COC) for the mining area, while uranium, selenium and molybdenum are the primary
COCs for the TMA.
Cover monitoring systems were installed on the CWRP and TMA in 2006 upon completion of a test plot
monitoring program. The cover systems are evaluated on an annual to biennial basis though field
investigations and performance reporting.
Passive removal mechanisms for the site have been identified and evaluated through FUP monitoring.
Sediments in Claude Lake, downstream of the Claude pit and CWRP, have been identified as having the
potential to intercept and remove contaminants in groundwater. Column studies were completed to

simulate groundwater discharge through the sediment and assess the attenuation of metals. The Island Lake
Fen, located downstream of the TMA in the Island Creek drainage, is known to sequester contaminants,
particularly uranium and molybdenum. A number of field evaluations were conducted in the fen from 2005
to 2010.
The CSR identified the potential for contaminant transport from the CWRP west to Claude Creek. Geophysical
surveys were completed in 2005 to delineate the solute front, and permeable reactive barriers (PRBs) were
installed in 2006 and 2007 as a removal mechanism. The performance of the PRBs is evaluated by routinely
sampling a number of groundwater wells in the EMP, as well as sampling stations in Claude Creek and Peter
River. Ongoing monitoring of the upper water column in the DJX pit and D-Pit is conducted routinely through
the EMP in order to determine if the pits will meet and remain within decommissioning objectives.
When the CSR was developed, there was no Saskatchewan or Canadian surface water quality objective for
uranium. Hardness-based DSWQO were developed through a review of available scientific literature and
consultation with regulatory agencies. SMOE has since developed a Saskatchewan surface water quality
objective (SSWQO) of 15 μg/L for uranium based on the most relevant scientific literature. A forthcoming
ERA will be conducted using the SSWQO of 15 μg/L as a screening criteria for completing the ERA at each
location.
Other follow-up studies relating to the ERA included sampling programs in 2009 and 2014 to assess the
implication of selenium for fish reproduction in the Island Creek drainage, and field investigations in 2004–
2006 to verify the assumptions made in the CSR ERA.
Decommissioning radiological objectives for the site are based on a need to keep radiation doses to the
general public below regulatory limits and compliant with the “as low as reasonably achievable” (ALARA)
principle. Objectives were developed with the goal of achieving a safe, stable property that could be used for
traditional purposes or occasional access.
Radon progeny (RnP) and LLRD levels were reduced through removal of source material or by covering with
clean soil material. Post-decommissioning LLRD and RnP levels are indistinguishable from background. The
potential exposure to gamma radiation is assumed to be the primary exposure pathway.
Gamma surveys were conducted on the areas that had undergone physical decommissioning. A submission
for surface gamma clearance of these areas was submitted to regulatory agencies in 2007 and subsequently
accepted following third-party verification.
The majority of the fieldwork relating to the FUP was completed by 2012. AREVA proposed a transition to
quarterly campaign monitoring to complete the EMP sampling. Campaign monitoring was identified as a cost-
effective method for completing the sampling, as a permanent site presence would no longer be required.
In April 2012, AREVA submitted an updated version of the EMP to regulatory agencies to prepare for
campaign monitoring. A small permanent staff remained on site to complete the first year of sampling and
advance site clean-up activities throughout 2013.
In the summer of 2013, site clean-up activities included demolition of the remaining camp and warehouse
buildings, decommissioning of the secondary treatment system plant and settling ponds, removal of site
culverts, and the disposal of demolition waste and any potentially contaminated materials.
Earthworks also occurred at the TMA and Claude pit areas. Recommended maintenance activities from the
2012 geotechnical inspection of the TMA were completed. Fill was placed in low areas observed on the TMA,
graded for adequate drainage and revegetated with the same seed mix originally applied to the TMA cover.
In the fall of 2009, some surface seepage had been observed on the Claude pit till cover. Field investigations

Case Studies
showed that the water was elevated in dissolved metals with a low pH. Mitigative measures were
implemented to prevent the surface seepage from directly entering Claude Lake. A seepage investigation
was conducted throughout 2010–2012 and consisted of field programs to install piezometers, sampling and
contaminant transport modelling to evaluate seepage improvement alternatives. The option selected was
the placement of three horizontal drains, which were installed in the summer of 2013.
Site clean-up activities were completed by September 2013. As of the fall of 2013, AREVA no longer has a
permanent site presence at Cluff Lake. The airstrip has been closed and site access restrictions have been
removed. The first campaign monitoring program took place in December 2013.
In its first full year, campaign monitoring has proven to be an effective method of collecting the EMP samples.
Campaigns generally occur with a field crew of two to four personnel and last one to three weeks. For short
programs, crews commute to the site from nearby exploration or outfitters’ camps, and for longer programs,
a temporary camp is set up on site. Campaigns generally occur in February, June, September and December
with the June and September programs being the largest.
SOE sampling was also scheduled for 2014. The SOE program is a comprehensive sampling and evaluation
program conducted in five-year cycles. In addition to the campaigns, three SOE programs took place at site
for the collection of sediment, fish, benthic invertebrate, terrestrial vegetation, surface water quality and
limnology data. These samples will be used in the upcoming preparation of the 2015 environmental
performance report (EPR), which includes the results of the SOE program as well as updates to groundwater
and contaminant transport modelling and the ERA.
Other programs that occurred on site in 2014 included a cover monitoring inspection, a revegetation
sampling program, a geotechnical inspection, a regulatory inspection and a LiDAR survey. A temporary camp
was set up on site to facilitate the various field programs and inspections in June, August and September.
In June 2014, a gamma survey was completed for the areas subjected to demolition and earthworks in 2013.
A submission demonstrating achievement of radiological end-state objectives was provided to regulatory
agencies in late 2014.
Aerial photos were collected in 2014 to assess the progress of decommissioning over time. Figure 4 shows a
comparison of the TMA from 2004 to 2014, and Figure 5 shows the Claude pit and CWRP for the same time
period.

Cluff Lake is a remote site, which presents challenges in planning field programs. The site has no cellular
service, and communication is through satellite phone and radio. With the closure of the airstrip, crews now
mobilise to the Cluff Lake project from Saskatoon, which is 900 km south of the site. The closest community
to the Cluff Lake Project is 260 km south of the site, and the final portion of the highway from this community
to the site is a gravel road. A detailed emergency response plan and health and safety plan are completed
prior to every program. Crews must be prepared with all necessary field and safety equipment and spare
supplies in the event of an equipment breakdown.
Since the removal of site access restrictions in the fall of 2013, challenges have also been faced with
vandalism to monitoring equipment, particularly dataloggers. Potential solutions for protecting or modifying
dataloggers are currently being explored.
Overall, the sampling program was successfully completed in 2014 with few issues. Campaign monitoring has
proven to be a successful and cost-effective option for completing the EMP.
Throughout 2015, AREVA is focused on preparing and submitting the EPR to the regulatory agencies. The EPR
will include an update to groundwater and contaminant transport models, the ERA and the terrestrial and
aquatic environment through the SOE program. A mid-term decommissioning licence hearing will occur with
the CNSC after submission of the EPR.
Following the EPR, AREVA anticipates the submission of an FUP report. Data collected through the EMP and
various follow-up field programs will be used to address and move towards closing off the items of the FUP.
Campaign monitoring will continue on a quarterly basis in 2015. AREVA intends to complete an optimisation
of the EMP upon completion of the EPR. This optimisation will rationalise monitoring requirements and
consider a reduction to monitoring program elements. Cost reduction measures such as automating data
collection will be explored.
AREVA’s eventual goal is to transfer the Cluff Lake project site into the provincial ICP. As monitoring results
demonstrate the achievement of decommissioning objectives, the EMP will be gradually optimised into a
long-term monitoring program (LTMP). The LTMP is envisioned to comprise periodic geotechnical inspections
to monitor stability, together with a minimal complement of environmental sampling to continue to assure
compliance with the decommissioning objectives. It is expected that the frequency of site monitoring will
diminish with time.

Case Studies
Over the 17 years since the planning for decommissioning began, lessons have been learned from the Cluff
Lake project. Starting with the end of a project in mind is essential for successful decommissioning of a mining
and milling site. The EA phase defines the evaluation of options for decommissioning, the selection of
preferred alternatives and the establishment of post-decommissioning objectives. Routine environmental
monitoring, as well as an FUP, verifies the compliance with post-decommissioning objectives while
addressing uncertainties from the EA. Decommissioning during the operational period and maintaining a
detailed decommissioning plan are also key contributors to success. Decommissioning in a jurisdiction with
an ICP in place is advantageous.
The decommissioning period can span many years, from planning to transfer into the ICP. Over this time, it
is essential to maintain corporate knowledge.
Maintaining a continuous site presence in the long term is costly. Campaign monitoring at the Cluff Lake
project has been shown to be an effective method of conducting EMPs in the later stages of post-
decommissioning and follow-up monitoring.
AREVA (2000) Cluff Lake project — comprehensive study for decommissioning, comprehensive study report, December.
AREVA (2014) Cluff Lake project — detailed decommissioning plan, version 3, revision 0, December.
AREVA (2014) Cluff Lake project procedure — environmental monitoring locations and schedule (EMLS), Version 6, Revision 0,
February.
Government of Canada (2012) Canadian Environmental Assessment Act, viewed 8 May 2015, http://laws-lois.justice.gc.ca/
eng/acts/C-15.21/index.html.
Government of Canada (1997) Nuclear Safety and Control Act, viewed 8 May 2015, http://laws-lois.justice.gc.ca/eng/acts/N-28.3.
Government of Saskatchewan (2014) The Reclaimed Industrial Sites Act, viewed 8 May 2015, http://www.publications.gov.
sk.ca/details.cfm?p=23009.
Government of Saskatchewan (2010) The Reclaimed Industrial Sites Regulations, viewed 8 May 2015, http://www.publications.gov.
sk.ca/details.cfm?p=23130.


L.M. Christoffersen SRK Consulting, Canada

S.R. Jensen SRK Consulting, Canada
T.R. Sharp SRK Consulting, Canada
M.W. Liskowich SRK Consulting, Canada
D.E. Allen Saskatchewan Research Council, Canada
The Lorado Mill processed uranium ore during the late 1950s near Uranium City in northern Saskatchewan,
Canada. Approximately 225,000 m3 of acidic tailings were deposited in and around Nero Lake and have been
a source of acidity and metal loads to the lake. In 2013, Nero Lake had a pH of 4.3 and exceeded Canadian
freshwater quality guidelines for aluminium, uranium and zinc. The Saskatchewan Research Council currently
manages reclamation of the site on behalf of the Saskatchewan Ministry of the Economy. Site reclamation
included in situ treatment of Nero Lake and the design and construction of an engineered cover for the tailings
on the lakeshore. In situ treatment consisted of neutralising the lake by adding lime. Neutralising the lake
with lime was expected to precipitate dissolved metals (aluminium in particular) and improve water quality.
Pilot-scale tests were used to develop the design basis of the in situ water treatment system, including the
lime dose required, the minimum volume fraction of the lake that should be dosed with lime slurry, the
expected post-treatment pH and the post-treatment alkalinity from different lime doses. The full-scale
treatment system was commissioned in July 2014 and operated in August and September. Four hundred
tonnes of lime were added to Nero Lake over two months. Approximately 12% of the lake’s total volume (1.3
million m3) was pumped through the treatment system and dosed with lime slurry. Two weeks after treatment
ended, the pH of the lake stabilised at 7.5 and total alkalinity was approximately 25 milligrams per litre.
Concentrations of all dissolved metals met Canadian freshwater quality guidelines, and concentrations of
most dissolved metals decreased by more than 90%. The results of the full-scale treatment were consistent
with the results of the pilot-scale tests. An acidity and alkalinity load balance developed for Nero Lake
demonstrated the lake is expected to remain neutral for the long term. Nero Lake will be monitored during
the summer of 2015 after vernal mixing to confirm that the post-treatment water quality has stabilised. In
situ treatment of reservoirs such as contaminated lakes, pit lakes or tailings ponds is a potential low-cost
alternative to conventional treatment, particularly in cold climates with short open water seasons and
relatively high operating costs during cold weather.
The Lorado Mill processed uranium ore during the late 1950s near Uranium City in northern Saskatchewan,
Canada (Figure 1). Approximately 225,000 m3 of acidic tailings were deposited in and around Nero Lake and
were a source of acidity and dissolved metal loads to the lake. Nero Lake is fairly shallow (approximately 10
m deep) with an area of about 1.7 square kilometres (Figure 2). In 2013, Nero Lake had a pH of 4.3 and
exceeded Canadian Council of Ministers of the Environment (CCME) freshwater guidelines for the protection
of aquatic life for aluminium, uranium and zinc. Dissolved aluminium and manganese were the two major
constituents of total acidity, and there was essentially no alkalinity.

In situ neutralisation of a lake impacted by legacy uranium tailings in northern Saskatchewan, Canada L.M. Christoffersen et al.

Case Studies
The Saskatchewan Research Council (SRC) currently manages reclamation of the site on behalf of the
Saskatchewan Ministry of the Economy. In 2013, SRC retained SRK Consulting (Canada) Inc. to design a
treatment system for neutralising Nero Lake. The concept for full-scale treatment was to withdraw water
from Nero Lake, dose it with lime slurry, and return it to the lake, where the treated water would mix with
and neutralise the untreated lake water. The design basis for full-scale treatment was developed using results
from a pilot-scale trial conducted during the summer of 2013. The pilot trial revealed that the two process
parameters that most significantly affected treatment performance were the lime dose and the fraction of
the total volume treated (Jensen et al., 2014).
The full-scale water treatment plant operated from 1 August to 29 September 2014 (60 days total), plus
commissioning and start-up from 28 to 31 July 2014. In total, 400 tonnes of lime were added to the lake and
approximately 12% of the total volume of the lake was pumped through the lime treatment system and
dosed with lime. The addition of lime caused the pH of the lake water to increase to 9.5. The water quality
monitoring results demonstrated that the water quality of Nero Lake improved significantly over the course
of the treatment campaign. Four weeks after treatment ended, the pH in Nero Lake had decreased to
approximately 7.5 as the lake equilibrated with atmospheric carbon dioxide. Removal of dissolved metals
exceeded 90% (with the exception of dissolved uranium, which was reduced by 50%). At the end of the
treatment campaign, the lake water met all applicable CCME guidelines (pH, aluminium, cadmium, uranium
and zinc). The 2014 water treatment campaign for Nero Lake achieved the goals of neutralising the lake,
improving the water quality (reducing concentrations of dissolved metals), and adding excess alkalinity.
Major project challenges were the short operating season and remote location. In northern Saskatchewan,
temperatures are typically above freezing only from June to September. Water treatment becomes
increasingly difficult during the rest of the year because of the low temperatures and ice cover. Uranium City
and the Lorado Mill site are only accessible by air the majority of the year. Equipment, lime and other heavy
items must be hauled to site over an ice road (in March and early April) or by barge in the open water months
from mid-June to September. This paper presents the results of the pilot- and full-scale treatment systems
and discusses the processes that affected their implementation and effectiveness.

Lime treatment is the most commonly used treatment process for acidic mine water with elevated metal
concentrations. When lime is added to water, it increases pH. Many metals are insoluble at high pH and
therefore precipitate, decreasing the concentration of dissolved metals in the resulting treated effluent (EPA,
1983; Aubé, 2004; MEND Program, 2013). Lime treatment typically proceeds in several basic steps. The first
step is the dissolution of lime in water to increase pH. Hydrated lime (Ca[OH]2) can be added directly to water,
whereas quicklime (CaO) is usually slaked first and added as slurry. The increased pH from lime dissolution
results in the formation of hydroxide ions. The hydroxide ions react with dissolved metal ions to precipitate
solid metal hydroxides, carbonates, oxyhydroxides or mixtures of these compounds. Other dissolved metals
may co-precipitate or adsorb to the surface of the precipitates. Other reagents can improve settling efficiency
by flocculating or coagulating these solids. The settled solids are referred to as sludge.
SRK developed an acidity and alkalinity load balance as part of the treatment system design to understand
the potential long-term fate of Nero Lake after the treatment campaign. The key consideration in the load
balance model was whether loadings of alkalinity that report to the lake exceed loadings of acidity from the
Lorado Mill tailings. This balance determines whether the lake will be acidic or alkaline in the long term.
The load balance was developed using historical water quality and hydrological data collected at the site.
Inputs of alkalinity to Nero Lake included direct surface runoff and runoff from surrounding lakes. Inputs of
acidity to Nero Lake included runoff and seepage from the tailings. The load of acidity to the lake was
calibrated using measurements of acidity in Nero Lake that had been collected in 1978, 1980, 2004 and 2013.
The benefit of this approach is that it relies on the relative difference between the annual alkalinity and
acidity loadings to predict the future water quality of the lake. This approach accounts for all sources of
acidity (e.g. tailings runoff, tailings seepage, diffusion from the lake sediments and others) that have
contributed to the lake over the past 35 years (Jensen et al., 2014).
A sensitivity analysis demonstrated that the average annual alkalinity load reporting to Nero Lake is
approximately 25 tonnes per year greater than the average annual acidity load reporting to the lake from the
tailings. The acidity (as CaCO3) of Nero Lake decreased from approximately 200 to 20 mg/L from 1978 to 2013
as a result of this net alkalinity load and dilution of the acidic lake water. Based on the historical trend, the
load balance suggests that the alkalinity concentration would likely exceed the acidity concentration within
approximately 10 years, resulting in a circumneutral lake. Based on the model predictions, the pH of Nero
Lake is expected to remain neutral for the long term because the annual alkalinity load exceeds the annual
acidity load. Under neutral conditions, the majority of the metals precipitated will not redissolve (Jensen et
al., 2014).
The alkalinity and acidity balance indicated that a single treatment campaign would hasten the return of
circumneutral conditions, which would persist for the long term. Because of the excess alkalinity reporting
to the lake every year, additional treatment of the lake may not be required.
A pilot-scale trial of the water treatment approach was conducted during the summer of 2013 to develop
the design basis for the full-scale treatment system. The pilot trial used three limnocorrals immersed in Nero
Lake to simulate the water column of the lake. The limnocorrals were cylinders made of impermeable high-
density polyethylene (HDPE) with closed bottoms, which allowed precipitates to be collected after
treatment. The limnocorrals were 3 m in diameter and 4 m deep. Floats were attached to the top of the liner
to provide buoyancy, and chains were attached to the bottom to keep them in place. The limnocorrals
provided a comparatively large volume (28 m3) to evaluate the treatment process and closely matched the
temperature of Nero Lake.

Case Studies
The limnocorrals were filled with water from Nero Lake. For each trial, a fraction of the volume of the
limnocorral was pumped out, dosed with lime slurry and pumped back into the limnocorral. SRK ran 14
different treatment scenarios in these limnocorrals with differing lime slurry concentrations, treated volume
fractions, residence times and flocculant doses. The treatment performance for each scenario was evaluated
by monitoring pH and dissolved metals concentrations (Jensen et al., 2014).
The pilot trial results demonstrated that the lime dose and treated volume fraction have the biggest effects
on treatment performance. A lime dose of 10 mg/L in the dosed water was sufficient to neutralise the lake
water in the limnocorral, but did not provide any excess alkalinity, which would provide long-term buffering
against future acidity loads. Larger lime doses provided additional alkalinity for long-term buffering capacity.
Increasing the volume of the limnocorral treated (from 5 to 25%) generally improved aluminium removal and
lime utilisation, but the potential performance benefits of treating a larger volume of water must be weighed
against the increased costs for equipment and pumping. The composition of the solids precipitated depended
on the lime dose, but was primarily magnesium, manganese and aluminium hydroxides. Neither flocculant
addition nor increased residence time were shown to significantly improve treatment performance or lime
utilisation (Jensen et al., 2014).
The concept for full-scale treatment was to withdraw water from Nero Lake, dose it with lime slurry and then
return the water to the lake, where the treated water would mix with and neutralise the untreated lake
water. The design basis for full-scale treatment was developed using information from the 2013 pilot trial.
The design basis for the water treatment plant is summarised in Table 1.
Process parameter Design basis Units Source

Total volume of Nero Lake 11,000,000 m3 Golder Associates (2013)
Volume fraction of lake to be treated 20 % Pilot trial
Volume to be treated 2,200,000 m3 Calculated
Treatment duration 60 days Seasonal window
Total quicklime mass to be added 400 tonnes Calculated from pilot trial
Quicklime (CaO) was slaked using water from Nero Lake to make 20% by weight lime slurry. Hydrated lime
was considered as an alternative to quicklime but was ruled out because of higher shipping costs to Uranium
City. Quicklime provides the same effective dose from approximately half the volume of hydrated lime,
thereby significantly reducing reagent volume demand and shipping costs.
Slaked lime was then transferred to two storage tanks, where the slurry was diluted to approximately 10%
using water from Nero Lake. From there, the slurry was injected at a set rate to the discharge lines by two
diesel-powered water recirculation pumps and discharged to two locations in the lake. Figure 3 is a schematic
of the water treatment process.

Raw Water Lime Quick

Slaker Lime
20% Lime
Slurry
Nero Lake Storage Storage

Tank Tank
Treated Raw Water

10% Lime
Water Slurry
Raw Water
Depth profiling of pH, temperature, specific conductivity and oxidation-reduction potential (ORP) at four
locations in Nero Lake was completed concurrently with the pilot trial in July 2013. The field parameters were
nearly identical at all four locations and at all depths. Specifically, temperature depth profiles demonstrated
that the lake is well mixed vertically and laterally. Because the lake is well mixed, equipment to facilitate
mixing within the lake during treatment was not required during the full-scale treatment campaign (Jensen
et al., 2014). Instead, the discharge locations were each approximately 1,000 metres away from the intake
line, which was intended to encourage circulation in the lake.
The contractor for the tailings cover and water treatment plant was a joint venture between Milestone
Environmental Contracting Inc. and PBN Nuna Contracting Ltd. (PBN Nuna Milestone JV). Milestone
completed the detailed engineering design of the water treatment plant; procured all equipment and
reagents; and commissioned, operated and decommissioned the plant. The plant operated from 1 August to
29 September 2014 (60 days total), plus commissioning and start-up from 28 to 31 July 2014. Process
parameters were measured and recorded twice every 12-hour shift. The recorded parameters included the
flow rate and pH of the discharged treated water and the quantity of lime added per shift.
The total cost of the water treatment campaign was approximately C$ 2.5 million, which included plant site
setup, equipment and materials, plant operation, decommissioning and plant site restoration.
The performance monitoring program for the 2014 treatment campaign included one sampling event before
water treatment started (18–22 July), monthly sampling during active water treatment (9–10 August and 11–
13 September), and one sampling event approximately two weeks after water treatment ended (11–12
October). In addition to the depth profiling completed during the four sampling events, field parameter
profiles were also recorded on 27 September (shortly before the water treatment plant was
decommissioned) and 31 October (approximately one month after the treatment campaign ended).

Case Studies
Surface water grab samples were collected from six locations in Nero Lake. Three locations were located in
the middle of the lake and accessed by boat, and three onshore locations were accessed from shore. The
sample bottles were rinsed with lake water three times before being filled. Field parameters were measured
in situ (MOE, 2013). An YSI multiparameter probe was used to measure field pH, conductivity, temperature
and ORP. Field parameters were measured before collecting samples at each monitoring location. For the
offshore Nero Lake monitoring locations, field parameters were measured at one-metre intervals from the
surface to the lake bottom. These field parameters were used for depth profiling. After measuring field
parameters in situ, samples were taken and then filtered and preserved in the field following laboratory
guidance. The samples were analysed for general chemistry parameters, major ions, nutrients, radionuclides
and dissolved and total metals.
The most important aspect of the water treatment campaign was to ensure that lime was consistently slaked
and added to meet the design concentration of approximately 200 mg/L in the dosed lake water. If the
concentration were higher, there would be a greater risk that unreacted lime would settle to the bottom of
the lake at the discharge points, which would reduce effective lime utilisation and the alkalinity load delivered
to the lake (Jensen et al., 2014).
SRK engineers visited the water treatment plant daily in August and September 2014 to measure pH in the
discharge lines to confirm that lake was being dosed uniformly. In addition to the profiling conducted during
the monthly monitoring events, SRK also completed pH profiling around the discharge points periodically
during the treatment campaign. The objective of these “spot checks” was to confirm that the lime in the
discharge lines had dissolved. A significant pH increase at the bottom of the lake would indicate that the lime
had not dissolved.
Depth profiles of field parameters were measured at one-metre intervals from the surface to the lake bottom
six times before, during and after the treatment campaign at three locations in Nero Lake. Profiles were
nearly identical at each location during the treatment campaign.
Figure 4 illustrates the nearly constant temperature of Nero Lake throughout the water column at one of the
sampling locations. The temperature of the lake decreased as summer ended and fall began, decreasing from
approximately 20°C in July and August to less than 3°C by the end of October. The temperature data confirm
that Nero Lake was well mixed throughout the treatment campaign.

Temperature (°C)
0 5 10 15 20 25
0
Depth (m) 2
10
October September August
Figure 5 illustrates the change in pH (measured in the field) during the treatment campaign for various
depths. The pH was less than 5 in July before treatment, and increased to greater than 9 during treatment
(August and September). The pH decreased to approximately 7.5 by the end of October, one month after
treatment ended, as the lake equilibrated with atmospheric carbon dioxide.
2m 4m 6m 8m 10 m Lime Added
12 400
11
Lime Added (tonnes)

10 300
9
pH
8 200
7
6 100
5
4 0
7/13 7/27 8/10 8/24 9/7 9/21 10/5 10/19 11/2
Surface water samples were collected from three offshore locations in Nero Lake. Concentrations were
nearly identical at all three locations during the treatment campaign. Water quality results for the
parameters of interest are presented in Table 2.

Case Studies
Parameter CCME guideline* July August September October

Field pH (s.u.) 6.5 to 9 4.57 5.35 8.54 8.80
Laboratory pH (s.u.) 4.83 5.07 6.92 7.52
Total alkalinity (mg/L) <1 <1 14 24
Bicarbonate (mg/L) <1 <1 17 29
Total dissolved solids (mg/L) 1,520 1,490 1,490 1,510
Calcium (mg/L) 208 217 232 235
Aluminium, dissolved (mg/L) 0.10 1.63 0.84 0.027 0.045
Cadmium, dissolved (mg/L) 0.00037 0.00012 0.00009 0.00001 0.00001
Manganese, dissolved (mg/L) 6.0 5.9 1.22 0.11
Uranium, dissolved (mg/L) 0.015 0.016 0.010 0.0017 0.0074
Zinc, dissolved (mg/L) 0.030 0.035 0.030 0.0016 0.0013
*Note: CCME guidelines are for total metals, but performance was evaluated using dissolved concentrations.
Four hundred tonnes of lime were added to the lake over 60 days of operation. The total volume of water
treated was approximately 1.3 million m3, representing 12% of the total volume of the lake (11 million m3).
The goal was to add 400 tonnes of lime to 2.2 million m3 of water, which represents 20% of the lake volume.
The target lime dose was achieved, but the target treated volume fell short owing to the lower-than-expected
capacity of the recirculation pumps.
However, the smaller volume of water treated did not affect the treatment performance or lime utilisation.
The spot checks (Section 2.4) of pH around the discharge points during active discharge of treated water
revealed that pH was consistent throughout the entire water column, indicating the lime in the discharge
lines had dissolved sufficiently and unreacted lime did not settle to the bottom. The design concentration of
200 mg/L was conservative, and the concentration of lime in the discharge lines could be increased without
negatively affecting lime utilisation.
The water quality results (Table 2) demonstrate that the water quality of Nero Lake improved significantly
over the course of the treatment campaign. Total alkalinity and bicarbonate increased from less than 1 mg/L
to about 25 and 30 mg/L, respectively. Calcium concentrations increased by approximately 25 mg/L during
treatment.
Concentrations of most dissolved metals decreased significantly during treatment, in most cases by more
than 90%. This decrease was maintained two weeks after treatment ended as the lake equilibrated with
atmospheric carbon dioxide, and the pH decreased from more than 9 to about 7.5. The only exception was
dissolved uranium, which decreased from about 16 to 2 µg/L from July to September (89% removal), but
increased after treatment ended to about 8 µg/L by the October monitoring event two weeks after treatment
ended (54% removal total). The cause for the increase of dissolved uranium concentrations is unknown, but
may be caused in part by the decrease in pH as the lake equilibrated with atmospheric carbon dioxide. The
post-treatment uranium concentration will be monitored during the summer of 2015.

Table 3 summarises the per cent removal of dissolved metals from the September and October monitoring
events, which were about halfway through treatment and approximately two weeks after treatment ended,
respectively.
Removal during treatment Removal after treatment

Parameter
(September) (October)
Aluminium, dissolved (mg/L) 98% 97%
Cadmium, dissolved (mg/L) 92% 92%
Manganese, dissolved (mg/L) 80% 98%
Uranium, dissolved (mg/L) 89% 54%
Zinc, dissolved (mg/L) 95% 96%
Dissolved aluminium, uranium and zinc concentrations for two of the offshore monitoring locations are
shown in Figures 6, 7 and 8, respectively.
Point 1 Point 2 Guideline Lime Added

2 400
Concentration (mg/L)
Lime Added (tonnes)

1.5 300
1 200
0.5 100
0 0
7/13 7/27 8/10 8/24 9/7 9/21 10/5 10/19

0.02 400
Lime Added (tonnes)
0.015 300
0.01 200
0.005 100
0 0
7/13 7/27 8/10 8/24 9/7 9/21 10/5 10/19

Case Studies

0.04 400
Lime Added (tonnes)

0.03 300
0.02 200
0.01 100
0 0
7/13 7/27 8/10 8/24 9/7 9/21 10/5 10/19
The pilot-scale trial results and the full-scale treatment results were comparable for most parameters. Initial
(i.e. before treatment) and final (i.e. one month after treatment) results from the pilot-scale trial and the full-
scale treatment are compared for pH, alkalinity, dissolved aluminium and dissolved uranium in Figure 9. The
full-scale treatment results are from the October monitoring event (two weeks after treatment).
Pilot Full-Scale Pilot Full-Scale

9 30
8 25
Alkalinity (mg/L)
7 20
Field pH (s.u.)
6 15
5 10
4 5
3 0
Initial Final Initial Final
Pilot Full-Scale Pilot Full-Scale

2.5 0.025
Uranium (mg/L)
Aluminium (mg/L)
2 0.02
1.5 0.015
1 0.01
0.5 0.005
0 0
Initial Final Initial Final
The total alkalinity in the pilot trial is lower than in the Nero Lake sample after full-scale treatment (15 mg/L
compared to 24 mg/L). This difference indicates that the full-scale treatment achieved higher lime utilisation
than the pilot-scale trial. The final dissolved uranium concentration after full-scale treatment (7.4 µg/L) is
higher than the final concentration in the pilot trial (1.4 µg/L). However, the dissolved uranium concentration
in Nero Lake in September (during treatment) was 1.7 µg/L, which is similar to the pilot result. The post-
treatment uranium concentration will be monitored during the summer of 2015.

As part of the acidity and alkalinity balance (Section 2.1), SRK predicted the net alkalinity from varying lime
doses. Figure 10 shows the predicted long-term net alkalinity trend of no lime treatment, treatment with 400
tonnes of lime and treatment with 800 tonnes of lime. Historical measurements of total acidity are included
for reference. Acidity and alkalinity were reported in terms of mg/L calcium carbonate (CaCO3) equivalents.
Total acidity was not included in the laboratory analysis in 2014. SRK calculated total acidity before and after
treatment using field pH and concentrations of dissolved aluminium, iron and manganese. These dissolved
metals form hydroxides at a pH of less than 8.3 and thus contribute to total acidity (Hem, 1985). The
calculated total acidity before treatment (in July) was 22 mg/L. This value compares favourably with the
measured total acidity in 2013 (20 mg/L). Total alkalinity (measured by the lab) was <1 mg/L in July 2014 and
was approximately 25 mg/L in October 2014. The difference between acidity and alkalinity was 22 mg/L
CaCO3 in July before the treatment campaign (i.e. total acidity exceeded total alkalinity). This difference
decreased to −23.5 mg/L CaCO3 in October after the treatment campaign (i.e. total alkalinity exceeded total
acidity). Figure 10 shows that the actual pre- and post-treatment values closely match the acidity and
alkalinity balance predictions for treatment with a dose of 400 tonnes of lime.
250
Historical Measured Acidity/Alkalinity Balance

200 Prediction: No Lime Treatment
Acidity minus Alkalinity Concentration (mg
Prediction: Lime Treatment (400 tonnes)

Prediction: Lime Treatment (800 tonnes)
150
Pre-Treatment Acidity/Alkalinity Balance
Post-Treatment Acidity/Alkalinity Balance
100
CaCO3/L)
50
-50 Alkalinity > Acidity
-100
1975
1985
1995
2005
2015
2025
2035
2045
2055
2065
2075
2085
2095
2105
2115
2125
Year
The concept for in situ neutralisation of Nero Lake was to withdraw water, add lime slurry and return the
water to the lake where the treated water would mix with and neutralise the lake. The design basis for full-
scale treatment was developed using information from a pilot-scale trial conducted during the summer of
2013. The pilot trial revealed that the two process parameters that had the largest impact on treatment

Case Studies
performance were the net lime dose and the fraction of the total volume treated. The full-scale water
treatment plant was commissioned in July 2014 and operated in August and September. In total, 400 tonnes
of lime were added to approximately 1.3 million m3 of lake water, representing 12% of the total volume of
the lake (11 million m3).
The monitoring program demonstrated that the water quality of Nero Lake improved significantly over the
course of the treatment campaign and that the full-scale treatment results were comparable to the results
from the pilot-scale trial. Before treatment, the pH of Nero Lake was 4.3 and concentrations of dissolved
aluminium, uranium and zinc exceeded CCME guidelines. There was no alkalinity, and dissolved aluminium
and manganese were the major components of total acidity.
The pH increased to greater than 9.0 during treatment (in August and September) and then decreased to
approximately 7.5 by the end of October, one month after treatment ended. Total alkalinity increased
significantly during treatment, from less than 1 mg/L to about 25 mg/L after treatment. Concentrations of
dissolved metals decreased, often by more than 90% (with the exception of dissolved uranium, which was
reduced by 50%). All CCME guidelines (for pH, aluminium, cadmium, copper, uranium and zinc) were met by
the end of the treatment campaign. The 2014 water treatment campaign for Nero Lake achieved the goals
of neutralising the lake, improving the water quality (reducing concentrations of dissolved metals) and adding
excess alkalinity.
An alkalinity and acidity load balance was developed to understand whether the lake would need to be
treated again in the future. This model, which was calibrated with site data, demonstrated that the annual
alkalinity load to Nero Lake exceeds the acidity load by approximately 25 tonnes per year. Consequently, the
lake is expected to remain alkaline after treatment. Nero Lake will be monitored during the summer of 2015
after vernal mixing to confirm that the post-treatment water quality has stabilised. Additional treatment is
not anticipated.
In situ treatment can be a low-cost alternative to conventional treatment, especially for implementation in
cold climates with short open water seasons, so high operating costs during cold weather are eliminated.
The total cost of the water treatment campaign was approximately C$ 2.5 million, which included plant site
setup, equipment and materials, plant operation, decommissioning and plant site restoration.
Aubé, B. (2004) The science of treating acid mine drainage and smelter effluents, InfoMine, Vancouver, Canada, viewed June 30,
2014, http://www.infomine.com/library/publications/docs/Aube.pdf.
British Columbia Ministry of Environment (MOE) (2013) British Columbia field sampling manual for continuous monitoring plus the
collection of air, air-emission, water, wastewater, soil, sediment, and biological samples, Water, Air and Climate Change
Branch, Ministry of Water, Land and Air Protection, Province of British Columbia, Canada, viewed July 15, 2014,
http://www.env.gov.bc.ca/epd/wamr/labsys/field-sampling-manual/pdf/2013/field_sample_man2013.pdf.
Golder Associates (2013) Environmental impact statement: Risk reduction plan for the inactive Lorado uranium tailings site, prepared
for the Saskatchewan Research Council for submission to the Saskatchewan Ministry of Environment and the Canadian
Nuclear Safety Commission. Available upon request from the Saskatchewan Research Council.
Hem, J.D. (1985) Study and interpretation of the chemical characteristics of natural water, 3rd edition, U.S. Geological Survey Water-
Supply Paper 2254, Alexandra, USA, 262 p.
Jensen, S.R., Sharp, T.R., Allen, D.E., and Wilson, I. (2014) Design of in-situ water treatment of acid contaminated lake, in Proceedings
B.C. Mine Reclamation Symposium 2014, 22–25 September 2014, Prince George, Canada, viewed October 7, 2014,
https://circle.ubc.ca/bitstream/handle/2429/51134/Jensen_et%20al_BC_Mine_2014.pdf?sequence=1.
Mine Environment Neutral Drainage (MEND) Program (2013) Review of mine drainage treatment and sludge management
operations, MEND Report 3.34.1, prepared by Natural Resources Canada for MEND, viewed August 20, 2014, http://mend-
nedem.org/wp-content/uploads/MEND3.43.1_ReviewofMineDrainageTreatmentSludgeManagementOperations.pdf.
United States Environmental Protection Agency (EPA) (1983) Design manual: Neutralization of acid mine drainage, EPA-600/2-83-
001, published by the Office of Research and Development, Industrial Environmental Research Laboratory, Chicago, USA,
viewed June 30, 2014, http://yosemite.epa.gov/water/owrcCatalog.nsf/9da204a4b4406ef885256ae0007a79c7/1827367
ed5f7e35c8525713b004db32e!OpenDocument.


L.K. Holder Golder Associates Inc., USA

F. Shuri Golder Associates Inc., USA
B. Nielsen Freeport-McMoRan Copper & Gold, USA
R. Jones Barrick Gold Corporation, USA
The White King / Lucky Lass Superfund Site consists of two former uranium mines in south-central Oregon. The
remedy included consolidating two large mining overburden stockpiles that contained low levels of
radioactivity and heavy metals by moving over 535,000 cubic metres (700,000 cubic yards) of material and
then capping the combined stockpile. An innovative armoured cap design was used to minimise the stockpile
footprint and thereby prevent disturbing existing wetlands. In addition, the stockpile was designed to merge
into the surrounding topography. The remedy also included stream restoration and development of new
wetlands. The stream was returned to its pre-mining channel, with the addition of hydraulic features to create
new wetlands. The wetlands have greatly enhanced the ecological value of the remedy and the site.
The White King and Lucky Lass mines Superfund site is one of only two uranium mine sites on the US National
Priorities List under the Superfund program. The site was remediated by the former owners/operators of the
site, primarily Tronox (formerly Kerr-McGee Chemical), Western Nuclear, Inc., and Fremont Lumber under the
oversight of the US Environmental Protection Agency (USEPA), the US Forest Service (USFS), the Oregon
Department of Environmental Quality (ODEQ), and the Oregon Department of Energy (ODOE).
The White King / Lucky Lass Mines Superfund Site (the Site) is located in south-central Oregon, approximately
27 km (17 miles) northwest of Lakeview, Oregon (Figure 1). The Site consists of two former uranium mines,
located within 1.6 km (1 mile) of each other (Figure 2). Portions of the Site are within the Fremont National
Forest, managed by the USFS, and portions are on private lands owned by Fremont Lumber and the Coppin
family trust.
Prior to remediation, major features at the White King Mine included the White King Pond (formed when
water collected in an open-pit mine), the so-called Protore Stockpile, and the Overburden Stockpile. Protore
is a mineral deposit containing sub-economic material that may become profitable with technological advance
or change in market value. However, both of the stockpiles at this site were actually overburden (soil and rock
removed to access ore deposits). Figure 3 shows the two stockpiles in the White King meadow prior to
remediation. The two stockpiles contained a combined volume of almost 765,000 cubic metres (one million
cubic yards) of material. The pit pond occupies approximately 5.2 hectares (13 acres) (Figure 4).
Augur Creek runs south through the eastern side of the White King Mine area and receives discharge from the
White King Pond.

Sustainable closure of two former uranium mines L.K. Holder, F. Shuri, B. Nielsen and R. Jones
Major features at the Lucky Lass Mine include the Lucky Lass Pond and the overburden stockpile (Figure 5).
The Lucky Lass overburden stockpile covers about 5.7 hectares (14 acres) and contains approximately 199,000
cubic metres (260,000 cubic yards) of material. This pond covers approximately 2 hectares (5 acres).

Case Studies

The White King and Lucky Lass uranium deposits were discovered in mid-1955. The local individuals who made
the discoveries conducted exploratory work and some initial mining. Mining began in earnest after September
1955. The Lakeview Mining Company began significant underground ore production in 1958 at the much larger
White King Mine. In April 1959, Lakeview Mining converted the White King Mine to open pit mining. Open-pit
mining at the White King Mine continued until December 1959. Open-pit mining commenced at the Lucky Lass
Mine in 1956.
Lakeview Mining discontinued commercial operations at both mines by early 1960. After 1961, sporadic small-
scale mining conducted by others continued at both mines. Exploration activities at the site occurred through
the early 1980s.
The Site was listed on the National Priorities List on 25 April 1995. The remedial investigation (RI) was
completed in 1997 (Weston, 1997), and the feasibility study (FS) was completed in 1999 (Weston, 1999). The
USEPA issued a record of decision (ROD; USEPA, 2001) on 30 September 2001.
Subsequent to the ROD, a group of potentially responsible parties (PRPs) agreed to take primary responsibility
for implementing remedial action and post-remediation monitoring at the site. The PRPs retained Golder
Associates Inc. (Golder) to perform remedial design, construction management, and construction quality
assurance (CQA) monitoring for the remedial action.
The stockpiles at both mines and some off-pile soils contained radioisotopes with radiation emissions above
background levels. The primary remedial action objectives (RAOs) for the site were, first, to prevent direct
contact with site soils having constituents of concern (COCs) above clean-up levels, and, second, to prevent
migration of COCs from site soils into groundwater.
In the ROD, the USEPA determined that the primary COCs were radium-226 (Ra-226) and arsenic (As). Uranium
isotopes were also COCs, but clean-up levels were not assigned to them because they were co-located with
Ra-226, and Ra-226 was the primary contributor to risk.
Clean-up levels were set as follows:
 White King Mine soil: As 442 mg/kg; Ra-226 6.8 pCi/g
 Lucky Lass Mine soil: As 38 mg/kg; Ra-226 3.6 pCi/g
The low pH of the White King pond, resulting from acid groundwater, was preventing development of natural
biological habitat within the pond and around the shoreline. Because of differences in the chemistry of the
ore bodies between the two mines, the Lucky Lass pond did not have a pH problem and had robust aquatic
habitat.
For the White King pond, the ROD (USEPA, 2001) determined, “sediment clean-up is not warranted at this
time. A sediment clean-up action, if determined necessary, will be documented in a future ESD [explanation
of significant differences] or ROD amendment.” The ROD did not require any remedial action for Augur Creek
or the Lucky Lass pond.
The remedial activities and associated design objectives for this project were:
 re-contour the White King Protore Stockpile so that it is no longer within the Augur Creek
floodplain;

Case Studies
 excavate the White King Mine haul road, a portion of the Lucky Lass stockpile, and certain off-pile
areas where there is evidence of mine-related waste, and place these materials on the recontoured
Protore Stockpile (called the Consolidated Stockpile after this additional placement);
 regrade remaining soil and waste rock to prevent erosion and promote vegetation, and cover
disturbed areas with 7.6 cm (3 inches) of soil;
 excavate the Overburden Stockpile, which consists predominantly of clay-like material, and place it
on the Consolidated Stockpile to form a low-permeability layer;
 place cover soil and topsoil on the Consolidated Stockpile surface sufficient to support vegetation,
and seed the topsoil surface;
 place topsoil, and reseed those areas where soil has been removed;
 adjust the pH of the White King Pond as necessary for beneficial use;
 perform a supplemental environmental project consisting of constructing wetlands in the Augur
Creek floodplain to enhance the environmental value of the project.
The approach to sustainable remediation of this site included:

 field determination of contamination, both before and during remedial action;
 minimising the footprint of the Consolidated Stockpile at the White King Mine;
 low-maintenance cap design;
 merging the profile of the Consolidated Stockpile into the surrounding topography;
 use of overburden materials as a low-permeability layer;
 direct measurement of sediment bioavailability;
 returning Augur Creek to its natural streambed;
 creation of wetlands in the White King meadow.
Normally in site remediation, determination of soil contamination and clean-up verification is accomplished
by laboratory analysis of soil samples. This process takes days if expedited, and weeks if not. For this project,
on the other hand, determination of clean-up by real-time field measurement using a gamma radiation
detector allowed the remediation construction to proceed without the typical delay of waiting for results. This
shortened the project duration and was a significant factor in allowing remediation to be completed in only
two construction seasons. This approach also allowed more precise delineation of removal areas than would
widely spaced soil samples, and thereby provided a high level of confidence that all impacted soil had been
removed, while minimising the area of disturbance.
First, a study was conducted to correlate COC concentrations to gamma radiation levels (Figure 6). It was also
demonstrated that, due to co-location of COCs, achieving the Ra-226 clean-up level was sufficient to meet the
arsenic clean-up level. Using these correlations, a gamma survey was conducted to determine the off-pile
areas where soil needed to be removed and placed in the Consolidated Stockpile (Figure 7) (Golder, 2004).
The precision of this survey allowed more precise design of the dimensions of the Consolidated Stockpile than
would have been possible with typical pre-design sampling and analysis.

Gamma radiation measurements were used to verify clean-up in real time during construction. Gamma
measurements were used immediately in the field to track clean-up progress. Under the agreed clean-up
criteria, clean-up verification required area averaging. To do this, field measurements were sent daily to the
office, where the data were analysed and mapped, and the maps were sent to the field the day after the
measurements.

Case Studies
The conceptual design of the Consolidated Stockpile in the ROD envisioned 5H:1V side slopes, with the slope
determined by the need for erosion control. However, using principles developed by the US Nuclear
Regulatory Commission (USNRC) for closure of uranium mill tailings sites (USNRC, 1990), Golder designed a
low-maintenance, erosion-resistant cap with steeper side slopes. Covers designed in this way have potentially
no maintenance. Golder had previously applied this methodology to two mine reclamation projects that
involve waste rock and tailings other than uranium, with functional success to date. At the White King site, use
of an armour gravel/topsoil mixture not only significantly reduced the need for ongoing maintenance, but
allowed steeper side slopes, up to 3H:1V. This in turn decreased the footprint of the stockpile, reducing both
the cost of the cover and the area of impacted land. As a result of the reduced stockpile footprint, it was
possible to avoid filling wetlands immediately adjacent to the stockpile; these wetlands would have been
destroyed if the ROD design approach had been used (Figure 8).
At the request of the USFS, the Consolidated Stockpile was designed to be visually compatible with the
surrounding topography (Figure 9). The goal is that, once natural vegetation has established on the stockpile,
the stockpile will not detract from the natural beauty of the area.
To minimise the potential for groundwater contamination from the Consolidated Stockpile, at least one meter
of low-permeability material were placed directly below the cover. However, instead of importing clay for the
cap, it was found that the overburden materials themselves could serve as the low-permeability layer.
Hydraulic conductivity was measured on undisturbed samples from the Protore and Overburden stockpiles
and from test pads constructed to evaluate soil placement methods. The results showed that the stockpile
soils could achieve low permeabilities on the order of 10-8 to 10-7 cm/sec over a wide range of moisture and
density conditions.
The agencies required evaluating the need for remediation of sediment in White King Pond. A common
approach would have been sampling and analysis of the sediments, followed by modelling of the bioavailability
of contaminants in the sediments. Instead, Golder proposed direct measurement of biological availability via
study of uptake of contaminants by freshwater mussels. The agencies accepted this approach. The uptake
study avoided many uncertainties inherent in the typical approach, and resulted in agency acceptance of the
conclusion (that sediment remediation was not required) more quickly than would normally be expected.
The White King pond formed when the mine pit filled with water following the cessation of mining activity. It
is not a natural lake or pond. The average water depth is approximately 5.5 metres (18 feet), with a maximum
depth of about 21 metres (70 feet). Groundwater impacted by acid rock drainage enters the pond through the
former mine shaft at the deepest point. The pH in the pond is monitored twice annually, during periods in the
summer when the pond is thermally stratified and again in the fall when the waters are mixed. This monitoring
produces pH profiles with depth at several locations in the pond. This monitoring generally shows a very slow
acidification of the pond, resulting in the need for periodic neutralisation. Lime was added to the pond in 1998,
2004 and 2009. Pond acidification appears to be slowing, either by reduced acidic input or from biological
neutralisation. Pond monitoring in 2014 showed no need for neutralisation at that time, and that the period
between neutralisation efforts can be longer than the five years originally estimated.


Case Studies
During mining, a portion of Augur Creek had been confined by berms and diverted around the White King
meadow. Once the off-pile removal and stockpile consolidation were complete, this berm was breached to
allow Augur Creek to flow again into the meadow. The intent was to allow Augur Creek to return to natural
flow using the historic, pre-mining channel locations in the meadow where possible.
As a “Supplemental Environmental Project”, wetlands were created in the White King meadow once the
overburden stockpile was removed (Golder, 2006a). The purpose and environmental benefit of the project
was enhancement of habitat associated with Augur Creek and White King Pond. The wetland areas were
developed by constructing low, broad soil berms where water could collect. The gently undulating meadow
surface produced small-scale mounds and ridges that projected 0.3 to 0.6 metres (1 to 2 feet) above the
wetlands, providing a greater diversity of habitat.
A wetland seed mix of sedges, rushes and grasses was applied to the ground surface within the wetland
boundaries. In addition, 100 willow cuttings and five whole willow bushes, collected from elsewhere on the
site, were planted around the perimeter of the wetlands where soils were expected to remain moist
throughout the year. An upland seed mix, also used elsewhere on the site for revegetation as part of remedial
action restoration, was spread on mounded areas within the wetlands, around the perimeter of the wetlands
and on the adjacent meadow surface. Twenty-eight large tree stumps, collected during clearing and grubbing,
were placed on the meadow to the east of White King Pond. The stumps were randomly placed in the meadow
and wetland areas to provide a diversity of habitat.
Water for the wetlands area has been sufficient to maintain the wetlands year-round since construction, and
is supplied by:
 Augur Creek, redirected to flow through the White King meadow;
 discharge from the White King Pond;
 a diversion ditch around the northwest portion of the White King highwall and runoff from the
Consolidated Stockpile cover and surrounding area, flowing through the White King meadow;
 natural springs in the White King meadow;
 precipitation falling directly on the wetlands.
The construction created approximately three combined hectares (seven acres) of wetlands at the normal
water level (Figure 10). The areal extent of the wetlands varies seasonally, and is greatest during spring. Other
areas of the meadow hold standing water because of the topography after excavation, which collects seepage
from natural springs, providing other wetlands in addition to the areas created by the wetlands berms.

Remediation of the site eliminated potential risks to human health and the environment and returned the
area to a condition of beneficial use. Wetlands were constructed at the site to enhance the ecological and
biological value of the site and to restore the area to pre-mining conditions.
The stockpile design has been highly successful at resisting erosion. The Consolidated Stockpile was completed
in 2006; as of 2015, only negligible erosion has occurred. Annual groundwater monitoring has been completed
at the site.
Using field measurements to determine clean-up and staying in close communication with the agencies
allowed the remedial construction to be completed in two seasons. In a typical remediation project, agency
approvals can significantly delay construction. At this site, those types of delays could easily have led to a third
field season with additional mobilisation/demobilisation costs and impact to the site and adjacent areas.
The biological study of White King Pond (Golder, 2006b) determined that sediment remediation was not
required. The USEPA accepted these results .
Cattle grazing in the White King meadow initially challenged the recovery of the meadow and threatened the
wetlands. However, improvements to perimeter fencing maintenance by the USFS were eventually successful
in keeping stray cattle out of the meadow, with the result that the meadow and wetlands are now thriving
(Figure 11).

Case Studies
Golder Associates Inc. (Golder) (2004) Gamma radiation survey report for the White King/Lucky Lass mines superfund site, Redmond,
USA, 23 September 2004.
Golder Associates Inc. (Golder) (2006a) Report on the supplemental environmental project at the White King/Lucky Lass mines
Superfund site, Redmond, USA, 8 December 2006.
Golder Associates Inc. (Golder) (2006b) Report on White King Pond and Augur Creek study at the White King/Lucky Lass mines
Superfund site, Redmond, USA, 12 September 2006.
Roy F. Weston, Inc. (Weston) (1997) Final remedial investigation report, White King/Lucky Lass mines site, Lakeview, USA, Kerr-McGee
Corporation, Oklahoma City, USA, 25 August 1997.
Roy F. Weston Inc. (Weston) (1999) Final feasibility study, White King/Lucky Lass mines site, Lakeview, USA, Kerr-McGee Corporation,
Oklahoma City, USA, 27 August 1999.
US Environmental Protection Agency (USEPA) (2001) Office of Environmental Cleanup, White King/Lucky Lass Superfund site record of
decision, USEPA Region 10, Seattle, USA, September.
US Nuclear Regulatory Commission (USNRC) (1990) Final staff technical position on design of erosion protection covers for stabilization
of uranium mill tailings sites, Washington, D.C., USA.


C. Ardito INTERA Incorporated, USA

J.M. Sigda INTERA Incorporated, USA
R. Blickwedel General Electric Company, USA
The St. Anthony mine is located within the Grants uranium mineral deposit, a west-to-northwest trending
mineral belt approximately 150 km (95 mi) long and 2.5–32 km (1.5–20 mi) wide (McLemore, 2007). In
general, these deposits are formed within aquifers as defined by the New Mexico Water Quality Control
Commission (WQCC) regulations. As these deposits create groundwater conditions that naturally exceed
WQCC standards for groundwater and there is little to no premining groundwater quality data, permitting
and closure of uranium mines in New Mexico require establishment of alternative groundwater concentration
limits.
The St. Anthony mine is a combined open-pit and underground uranium mine near the eastern margin of the
San Juan Basin in central New Mexico. Ore was mined from the Jackpile sandstone member of the Jurassic
Morrison Formation. The very fine-grained Jackpile sandstone has a low permeability, approximately
0.03 m/day (0.09 ft/day), which forced development of alternative water sources during mining operations.
Surface mining took place in two pits: the large pit, which penetrates most of the Jackpile sandstone and has
a small pit lake; and the small pit, which does not fully penetrate into the Jackpile sandstone. A fate and
transport analysis of uranium and several other constituents of concern was completed to evaluate the
potential for postclosure impacts to human health and the environment.
In its current state, the large pit of the St. Anthony mine captures groundwater via a cone of depression that
has developed in response to evaporation of pit water; however, reclamation alternatives that include
backfilling the large pit would eliminate the evaporation and ultimately result in the loss of groundwater
containment. A feasibility study using a multiple accounts analysis (MAA) (Robertson and Shaw, 2004)
resulted in the selection of both pit backfill and pit backfill waiver options as the highest scoring alternatives.
The pit backfill waiver alternative ranked high due to containment of mineralised groundwater. As pit closure
is the preference of key stakeholders, and regional groundwater flow conditions would result in the loss of
containment of mineralised groundwater, a detailed hydrologic analysis was necessary to ensure there was
no complete groundwater pathway resulting in risk to public health and the environment.
The climate of the southeastern San Juan Basin is arid to semi-arid with variable precipitation that is
consistently exceeded by evaporation and transpiration demands. Recharge into the subsurface is negligible
across most of the site because topography and the low permeability of the surficial materials severely
constrain recharge, leaving most precipitation to be lost to evapotranspiration. Water balance calculations
on the large pit revealed that evaporation is the primary outflow driver and that groundwater inflow through
the Jackpile sandstone is on the order of 26.5 l/min (7 gpm). Similar calculations revealed that the small pit is
a potential intermittent source of recharge.
Our hydrogeologic analysis and groundwater flow model revealed the following hydrogeologic conditions: (1)
the Jackpile sandstone is predominantly a confined system; (2) precipitation is far exceeded by transpiration
during the growing season and evaporation throughout the year, with negligible opportunity for areal
recharge of the Jackpile sandstone; (3) current groundwater flow through the Jackpile sandstone is controlled
by evaporation within the main pit, by evapotranspiration and seepage in the adjacent Meyer Draw drainage,

Evapotranspiration control on groundwater migration at a uranium mine in the high desert C. Ardito, J.M. Sigda and R. Blickwedel
and by outcrop evaporation; and (4) tamarisk transpiration in the Meyer Draw drainage is estimated to occur
at a maximum rate of approximately 53 l/min (14 gpm).
Predictive modelling of postclosure groundwater flow demonstrated that local groundwater discharge to
Meyer Draw via tamarisk transpiration or seepage into alluvium in contact with the Jackpile sandstone is
more than sufficient to capture groundwater migrating from the pits and surrounding site area. Sensitivity
analyses revealed that major changes in boundary conditions or Jackpile hydraulic conductivity did not change
the capture of groundwater by Meyer Draw by transpiration and seepage. Relatively low outflow rates are
required to keep any solute plume on site because of the Jackpile sandstone’s low permeability.
The St. Anthony mine is located in the Grants uranium mineral belt, a west-to-northwest trending mineral
belt approximately 150 km (95 mi) long and 2.5–32 km (1.5–20 mi) wide (McLemore, 2007) (Figure 1) in
central and northwestern New Mexico. In general, these deposits are formed within aquifers as defined by
the New Mexico Water Quality Control Commission (WQCC) regulations. As these deposits create
groundwater conditions that naturally exceed WQCC standards for groundwater, and there is little to no
premining groundwater quality data, closure of uranium mines requires development of alternative
groundwater concentration limits.
Records indicate that uranium was mined at the site as early as the mid-1950s, several years after the Grants
uranium region was discovered. Between the mid-1950s and the early 1960s, the St. Anthony Uranium
Company, a unit of Climax Molybdenum Company, operated an underground mine at the site.
More recently, the site property was leased by United Nuclear Corporation (UNC) from the Cebolleta Land
Grant. The St. Anthony mine was owned and operated by UNC from 1975 to 1981, and the lease was

Case Studies
terminated in 1988. Excavation of the small pit was initiated in November 1975, and excavation of the large
pit was initiated in January 1977. Both of the pits penetrated approximately 23 m (75 ft) of the Jackpile
sandstone (Baird et al., 1980). Figure 2 illustrates the historical site conditions: in this photograph, the small
pit appears to have been excavated to the current extent, and the large pit was being mined largely on the
south end with overburden (which was later removed) remaining on the north side of the pit.
A feasibility study to select remedial alternatives for managing groundwater at the St. Anthony mine was
conducted using a process called multiple accounts analysis (MAA) (Robertson and Shaw, 2004). MAA is a
systematic process for managing the concerns and issues raised by numerous stakeholders (in this case, nine
government, tribal and private entities) having differing positive and negative judgments about selecting
among multiple approaches to solve a particular problem. The MAA process provided a means to objectively
balance multiple and often competing goals and objectives among the stakeholders. The evaluation of
remedial alternatives compared three key remediation goals: (1) preventing exposure to water in the large
pit, (2) reducing the risk of future groundwater impact, and (3) stabilising groundwater conditions for the
long term. The task was to objectively rank the nine potential abatement alternatives for managing water
resources related to the St. Anthony mine large pit. The general objective of the MAA was to provide a means
by which evaluators could select the most suitable, or advantageous, alternative from a list of alternatives
developed specifically for the St. Anthony site.
The MAA process reduced over 20 remedial alternatives or subalternatives to six alternatives, which were
either pit waiver alternatives or pit backfill alternatives. As required by the New Mexico WQCC abatement
regulations, a remedial action plan must design and justify a single alternative. Based on additional field
investigations and bench-scale studies, the single remedy selected was a pit backfill remedy including
chemical stabilisation of pit water and sediments in advance of the pit backfill with the incorporation of

alternative groundwater concentrations limits. The key factors influencing the remedy selection are
summarised as follows:
 Backfilling of the large pit is preferred by all parties if the groundwater pathway is not complete
(i.e., there is no risk to public health and the environment).
 Groundwater in the Jackpile sandstone does not migrate offsite even after the large pit is
backfilled and regional groundwater gradients are re-established.
 Water quality in the Jackpile sandstone is not drinking-water quality due to the presence of
mineralised zones throughout this area.
 Water quantity and water quality in the Jackpile sandstone do not support water supply
development for any sustainable future use.
 As described further below, geochemical stabilisation prevents migration of the most hazardous
constituents of potential concern (uranium and radium) beyond the immediate vicinity of the
large pit.
After active mining ceased and the pit was no longer dewatered, perennial pit water ponding ensued. Over
time, as groundwater flowing into the pit has evaporated, water has increasingly concentrated its dissolved
components, in some cases leading to the saturation of mineral phases such as gypsum. Recent sampling
indicates that sediment with sufficient decaying plant matter to induce precipitation has concentrated
uranium at the sediment–water interface. If the pit were backfilled and no action was taken to immobilise
uranium, the uranium could be transported in groundwater away from the pit after regional gradient
conditions return. Similarly, the presence of secondary uranium mineralisation around the pit would be
subject to dissolution if groundwater levels were to rise to premining elevations. To mitigate this potential
migration of uranium, several options, including bioreduction, were evaluated in bench-scale microcosms of
zero-valent iron (ZVI), sodium tripolyphosphate (STPP), iron oxide and surfactant-modified zeolite. The
microcosms indicated that STPP immobilised uranium effectively and was resistant to remobilisation upon
exposure to oxygen-containing groundwater. Larger bench-scale studies replicated this result and also
indicated effective radium immobilisation.
The conceptual hydrogeologic model encompasses the St. Anthony mine, the JJ mine and the Jackpile–
Paguate mine, three neighbouring mines in this part of the Grants uranium mineral belt (Figure 1). The area
of interest is roughly 13 km (8 mi) wide and 6.4 km (4 mi) long, with an area of approximately 83 km2 (32 mi2).
Within it, only a small interval of the San Juan Basin’s stratigraphy is exposed. The Jurassic and Cretaceous
stratigraphic units dip slightly to the north-northwest at approximately 1 to 2 degrees.
The oldest unit exposed at the site is the Brushy Basin Member of the Morrison Formation, comprising
greenish-grey indurated mudstones with interbedded sandstones (Baird et al., 1980). Its thickness ranges
between 76 and 143 m (250 and 470 ft) (Zehner, 1985; U.S. Bureau of Land Management, 1986) at the site.
Overlying the Brushy Basin Member is the Jackpile sandstone, the source of uranium mined at the St.
Anthony, JJ, and Jackpile–Paguate mines. The Jackpile sandstone is a thick lense of arkosic sandstone that
occurs in the upper portion of the Brushy Basin Member of the Morrison Formation. Its thickness varies
between roughly 30 and 61 m (100 and 200 ft) near the St. Anthony mine and reaches up to 91 m (300 ft)
elsewhere (Risser et al., 1984; INTERA, 2006).
Overlying the Jackpile sandstone is the Dakota sandstone, which is a white, fine- to medium-grained, sugary-
textured, well-cemented sandstone. It ranges in thickness from 1.8 to 6 m (6 to 20 ft) at the site and can have
a thickness of about 15 m (50 ft) at the regional scale.

Case Studies
The Mancos shale, which overlies the Dakota sandstone, comprises interlayered beds of mudstones, shale
and three well-lithified sandstone units referred to as the Tres Hermanos (Kittel, 1963). Its thickness varies
with topography and erosion, but maximum thickness is 125 m (410 ft) at the St. Anthony mine.
Quaternary alluvium includes the sediments deposited by ephemeral flows within the erosional drainages,
Meyer Draw and Bohart Creek, and the sediments deposited by flow in the Rios Paguate and Moquino.
Sediments range in thickness from roughly 4.6 m (15 ft) in Rios Paguate and Moquino to 7.6 m (25 ft) in
Meyer Draw. Colluvium typically has a thickness of 0.3 to 1.5 m (1 to 5 ft).
The climate of the southeastern San Juan Basin is arid to semi-arid with variable precipitation that is
consistently exceeded by evaporation and transpiration demands (Zehner, 1985; INTERA, 2006). Long-term
data from three climate stations surrounding the St. Anthony mine show that average annual precipitation,
which ranges between 24.4 and 27.4 cm/y (9.6 and 10.8 in/y), is far exceeded by measures of evaporation,
with mean pan evaporation at the Laguna station equal to 160 cm/y (63 in/y), and potential
evapotranspiration falling between 129.5 and 134.6 cm/y (51 and 53 in/y).
The salt cedar, Tamarix ramosissima, also known as tamarisk, is an important transpiration agent at the site
and in nearby drainages. Tamarisk has a deep, extensive root system that extends to the water table (> 4.6 m
[15 ft] deep) (DiTomaso et al., 2013). Annualised transpiration rates are estimated to range between 2.0 x
10-3 and 2.7 x 10-3 m (6.6 x 10-3 and 8.9 x 10-3 ft) of water per day for a stand of tamarisk in central New
Mexico (Hansen and Gorbach, 1997). At the site and to the south, tamarisk trees are most commonly found
within Meyer Draw. INTERA estimated that the Jackpile sandstone is in direct contact with alluvial sediments
along a section of the draw comprising an area of approximately 37,161 m2 (400,000 ft2). Within this area of
the tamarisk stand, assuming an annualised transpiration rate of 2.1 x 10-3 m/day (6.9 x 10-3 ft/day) per unit
area, tamarisk transpiration can remove roughly 53 l/min (14 gpm) from the Jackpile sandstone along this
reach of Meyer Draw.
Perennial surface water bodies near the site are limited to the large pit and reaches of the Rio Paguate near
the Jackpile–Paguate mine. Ephemeral surface water bodies near the St. Anthony mine include ponded water
in the small pit, cattle tanks (ponds), and the Meyer Draw, Rio Moquino, and Bohart Creek drainages (north
of the site area).
The water in the large pit has persisted for roughly 30 years, beginning sometime between late 1981 and
early 1982 and continuing through 2013.
In the vicinity of the St. Anthony and Jackpile–Paguate mines, shallow groundwater flow is almost entirely
restricted to the Jackpile sandstone. The overlying Mancos shale and Dakota sandstone are in the vadose
zone in this area. Groundwater flow in the underlying Brushy Basin unit is very likely negligible because of its
low permeability.
Groundwater flows through the Jackpile from the north and from the west toward the south, where it
outcrops. This is demonstrated in Figure 3, which shows heads at the Jackpile–Paguate mine for premining
conditions. Figure 4 shows the best available estimate of present-day groundwater heads for the St. Anthony,
JJ, and Jackpile–Paguate mines by combining the most recently available head data. Groundwater heads are
higher in the north and west and lower in the south, whereas the Jackpile is tens to hundreds of metres below
the ground surface in the north and outcrops at ground surface in the south. Consequently, the Jackpile
sandstone is a confined flow system for its entirety except for its surface exposures at outcrops, in the mine
pits, in drainages and in the lower reaches of the Rio Paguate and Rio Moquino.


Case Studies
Groundwater inflow to the Jackpile is comprised of recharge from the Jackpile sandstone and Dakota
Formation to the west and north and localised recharge from saturated reaches of the Rio Paguate alluvium,
cattle tanks, a pond near the Jackpile pit, and the St. Anthony small pit.
Groundwater outflow includes discharge to the Rio Paguate and Rio Moquino, evaporation from the pit lake
in the large pit, and evapotranspiration along the drainages with tamarisk, where the Jackpile either outcrops
or is overlain by alluvium. Historical outflows include dewatering at the JJ, St. Anthony, and Jackpile–Paguate
mines.
A water balance calculation for the large pit revealed that groundwater inflow from the Jackpile sandstone
to the large pit is approximately 26.5 l/min (7 gpm). This inflow rate indicates that the Jackpile sandstone is
a low-flow, low-permeability unit. A water balance conducted for the small pit demonstrated that water in
the pit has recharged the Jackpile sandstone at a rate of about 3.8 l/min (1 gpm) over the last several years.
The geometric mean for the Jackpile sandstone K values from aquifer tests in the Jackpile–Paguate Mine area
is 0.028 m/day (0.091 ft/day). Aquifer testing yielded similar K values near the JJ and St. Anthony mines. The
resulting geometric mean K for the Jackpile sandstone across the area of interest is 0.027 m/day (0.09 ft/day).
The hydraulic conductivity of the alluvium at the confluence of Rios Paguate and Moquino is 7 m/day
(23 ft/day) (Risser et al., 1984). Sediments in Meyer Draw and Bohart Creek appear to be dominated by silty
sands so are not expected to have as high a hydraulic conductivity value as the river bed sediments.
The overall objective of the groundwater flow modelling is to determine groundwater flow patterns after
backfill of the two St. Anthony pits. A two-dimensional numerical flow model was constructed and calibrated
to represent present-day groundwater flow in the Jackpile sandstone. A predictive two-dimensional flow
modelling tool was then constructed from the calibrated flow model to simulate migration of groundwater
from the backfilled large pit. The numerical models represent all the key driving forces and hydrogeologic
features as the most important controls on current and future groundwater flow within the area of interest.
INTERA created a numerical geologic model of the Jackpile sandstone within the area of interest using
available geologic information and the Leapfrog Hydro software package (ARANZ Geo Limited, 2013). The
numerical geologic model was used to create the grid for the numerical flow model.
The model domain encompasses the area of interest, including the JJ, St. Anthony, and Jackpile–Paguate
mines (Figure 5). The domain extends approximately 9,100 m (30,000 ft) from west to east and approximately
8,200 m (27,000 ft) from north to south. A single model layer divided into a total of 300 rows and 469 columns
with a uniform grid cell size of 100 ft × 100 ft (approximately 30 m × 30 m) was used to represent the Jackpile
sandstone. The total number of active cells was 66,610.

Top and bottom elevations of the grid cells were set to match the top and bottom elevations of the numerical
geologic model for the Jackpile sandstone. Thus the top elevation of the model layer followed the dipping
Jackpile from ground surface in the south, to a maximum elevation of 1,868 m (6,130 ft) above mean sea
level (AMSL) in the southwest, to a few hundred metres below ground surface, with a minimum elevation of
1,750 m (5,740 ft) AMSL in the southwest, along the north domain edge.
The model domain edges were defined as a combination of steady-state no-flow and specified head
boundary conditions. A specified head boundary condition with head equal to 1,795 m (5,890 ft) was set
along the entire northern boundary (Figure 5). The high head to the west was represented as a specified head
boundary condition along 305 m (1,000 ft) of the western domain edge with a head of 1,836 m (6,025 ft)
(Figure 5). These boundary head values were determined through trial and error during the initial calibration
process. The remainder of the domain boundaries was set to be a no-flow boundary condition.
The MODFLOW river package was used to represent Rio Paguate and Rio Moquino. INTERA intersected the
grid cells containing the two rivers with the highest resolution digital elevation models available (National
Elevation Dataset, 10-metre) to determine riverbed elevations in each model grid cell. These riverbed top
elevations were used to set the stage height in each river boundary condition cell.
Transpiration of the tamarisk along Meyer Draw and its tributary was simulated using the standard
MODFLOW EVT package. The boundary condition was only applied to model cells where the tamarisk roots
were expected to be able to extract groundwater from the alluvial sediments overlying the Jackpile sandstone
(Figure 5). The length of the reach with tamarisk transpiration is approximately 1,130 m (3,700 ft) (Figure 5);
however, live tamarisk are present in the draw far downstream of this reach. The maximum
evapotranspiration rate was treated as a calibration parameter with the starting value equal to the
annualised transpiration rate.

Case Studies
Evaporation from the large pit at St. Anthony was simulated using the head-dependent boundary condition
called the general head boundary (GHB) condition as defined by the standard MODFLOW GHB package. The
stage elevation was set to 1,783 m (5,850 ft), which is 1.2 m (4 ft) above the minimum observed large pit
water level. This rate was compared to the rate estimated by the large pit water balance.
Aerial photographs were used to determine the areal extent of the ponded areas north of the Jackpile pit
and within the small pit. These areas were used to create small, local recharge areas in the model for which
the rates were treated as calibration parameters. The recharge rate for the small pit was compared to the
recharge rate estimated by the water balance calculation.
The hydraulic conductivity of the Jackpile sandstone was set everywhere to equal the geometric mean K of
0.027 m/day (0.09 ft/day). Initial head values for the start of the calibration simulation were created by
interpolating the head contours shown in Figure 4.
Groundwater heads from 2013 for the five wells at the JJ mine and the 10 wells at the St. Anthony mine were
used as calibration targets (Figure 6). Head values at 10 wells around the Jackpile–Paguate mine from 2010
and 2011 were also used as calibration targets because they were the most recently available data. Bore logs
were checked to ensure that the selected targets were screened within the Jackpile sandstone.
Figure 6 shows the resulting residuals and simulated head contours for the calibrated flow model. Barring
the target at MW-8 (which was a consistent outlier), simulated and observed heads have relatively small
differences (residuals) at the JJ and St. Anthony mines, with a good balance of overestimated and
underestimated heads (Figure 6). The simulated head contours provide a reasonably good match to the
contours interpolated from the observed heads (Figure 6).

Residual statistics confirm that the calibration is good. The mean head residual is −0.676 m (-2.22 ft) and the
mean absolute head residual is 2.17 m (7.13 ft). The normalised root mean square of errors is 0.07, which
indicates a good- to-acceptable calibration based on the commonly used guideline of values less than 0.10
(Spitz and Moreno, 1996). Given the available data, the simulated heads and head contours provide a very
good match to the observed heads and the contours of interpolated observed heads (Figure 6).
The outflow rate for the large pit GHB cells is 18.9 l/min (5 gpm), which closely matches the 26.5 l/min (7 gpm)
rate estimated by the large pit water balance calculation. The final calibrated recharge rate for the small pit
yields 5.7 l/min (1.5 gpm), which equals the recharge rate estimated from the small pit water balance. The
calibrated recharge rate for the pond and tanks north of the Jackpile pit is 22.7 l/min (6 gpm). Total outflow
from the evapotranspiration boundary condition cells in Meyer Draw is 3.4 l/min (0.9 gpm), which is much
less than the 53 l/min (14 gpm) calculated from the field transect and the annualised transpiration rate.
The calibrated flow model served as the foundation for constructing a predictive flow model to simulate
groundwater flow after backfilling St. Anthony’s large and small pits. The predictive flow model was identical
to the calibration flow model except for three important changes. These included removing the GHB cells
used to simulate the evaporative sink within the large pit, removing the recharge cells in the small pit, and
using the calibrated heads from the calibration model as the initial heads.
After solving for the final steady-state groundwater flow conditions, particle tracking was carried out to
demonstrate the groundwater flow pattern in and around the backfilled large pit. Particle tracking was
implemented using the U.S. Geological Survey MODPATH code (Pollock, 2012).
Results for the predictive simulation showed that the groundwater head at the large pit will eventually
rebound to 1,818 m (5,966 ft), whereas heads at the JJ mine will reach 1,822 m (5,977 ft) (Figure 7). Particle
tracks demonstrate that groundwater from the area around the large and small pits will be captured by
tamarisk transpiration within Meyer Draw (Figure 7). Outflow from transpiration is 6.8 l/min (1.8 gpm),
double the outflow rate simulated for current conditions. This indicates that a very small rate of groundwater
discharge to Meyer Draw through tamarisk transpiration is sufficient to capture all groundwater from the
area surrounding the large and small pits.
Boundary conditions and hydraulic conductivity were changed to determine whether Meyer Draw would
continue to capture groundwater from the large pit. The changes were conservatively chosen to investigate
the full range of uncertainty in the predictive model inputs.
For the sensitivity analysis, four different scenarios were selected to test the robustness of the predicted
future groundwater flow pattern. Each sensitivity scenario was carried out by revising selected model inputs,
running the flow simulation and particle tracking, and comparing the particle tracks to the particle tracks
from the predictive scenario.
For all sensitivity cases, groundwater from the large pit was captured through discharge to Meyer Draw. This
occurred whether the hydraulic conductivity was set to the maximum observed Jackpile K value of
0.13 m/day (0.43 ft/day) (Figure 8), or the minimum observed Jackpile K value of 0.0015 m/day (0.005 ft/day)
and proportionally scaling the calibrated recharge appropriately. All groundwater flow was captured by
Meyer Draw whether the transpiration area was doubled in size or whether the evapotranspiration boundary
condition was removed and replaced with drain boundary condition cells, which were based on the structural
analysis and represent discharge of Jackpile groundwater into the draw sediments. The outflow rate from
the drains simulated in Scenario 5 is 4.9 l/min (1.3 gpm), a value that is much less than the 53 l/min (14 gpm)
estimated for tamarisk transpiration along a roughly 1,130 m reach of Meyer Draw. In the absence of any
tamarisk transpiration along that reach, Jackpile groundwater would seep into the overlying vadose zone

Case Studies
sediments of Meyer Draw. Seepage flow will be driven downstream along the contact between the sediments
and the low-permeability Brushy Basin mudstone by topography. Tamarisk are also transpiring water from
Meyer Draw sediments downstream of the roughly 1,130 m (3,700 ft) reach, so any seepage into the vadose
zone sediments will likely be captured by the downstream tamarisk trees or other plants living with the draw.

The maximum area potentially affected by transport of solutes from the vicinity of the large and small pits is
an important input to the application for alternative groundwater concentration limits. This area can be
estimated by combining the particle tracks for all five predictive simulations and encompasses the large and
small pits and the area between them (Figure 9).
The findings from the predictive modelling demonstrate that local groundwater discharge to Meyer Draw via
tamarisk transpiration or seepage into alluvium in contact with the Jackpile sandstone is more than sufficient
to capture groundwater from the St. Anthony pits. The modelling results are very robust because major
changes in boundary conditions or hydraulic conductivity did not change the capture of pit groundwater by
Meyer Draw.
Uranium deposits in the Grants uranium mineral belt are mostly formed within aquifers. Groundwater at
mines within the Grants uranium mineral belt naturally exceeds WQCC standards. Therefore, to close a
uranium site in New Mexico, alternative groundwater concentration limits must be established, and risk
evaluations must confirm there are no complete groundwater pathways. The groundwater modelling
described in this paper confirms that evapotranspiration controls prevent the establishment of a complete
groundwater pathway. Based on this analysis, closure of the St. Anthony mine through backfilling of the pits
will not produce a complete exposure pathway.
ARANZ Geo Limited (2013) Leapfrog Hydro software package, Christchurch, New Zealand, viewed 28 April 2015, http://www.
leapfrog3d.com/products/leapfrog-hydro.

Case Studies
Baird, C.W., Martin, K.W. and Lowry, R.M. (1980) Comparison of braided-stream depositional environment and uranium deposits at
Saint Anthony underground mine, in Geology and mineral technology of the Grants uranium region, 1979, C.A. Rautman (ed),
New Mexico Bureau of Mines and Mineral Resources, Socorro, New Mexico, 401 pp.
DiTomaso, J.M., Kyser, G.B., Oneto, S.R., Wilson, R.G., Orloff, S.B., Anderson, L.W. … Mann, J.J. (2013) Weed control in natural areas
in the western United States, Weed Research and Information Center, University of California, 544 pp.
Hansen, S. and Gorbach, C. (1997) Middle Rio Grande water assessment: final report, US Bureau of Reclamation, Albuquerque Area
Office, 151 pp.
INTERA Incorporated (INTERA) (2006) Stage 1 abatement plan, revision 1, JJ No. 1/L-Bar Mine, Cibola County, New Mexico, INTERA,
Albuquerque, New Mexico.
Kittel, D.F. (1963) Geology of the Jackpile mine area, in Memoir 15 — geology and technology of the Grants uranium region, V.C.
Kelley (ed), New Mexico Bureau of Mines and Mineral Resources, Socorro, New Mexico, pp. 167–176.
McLemore, V. (2007) Uranium resources in New Mexico, in Proceedings SME Annual Meeting, 25–28 February 2007, Denver,
Colorado, Curran Associates, Inc., Red Hook, NY, pp. 653–665.
Pollock, D.W. (2012) User guide for MODPATH version 6 — a particle-tracking model for MODFLOW, Chapter 41 of Section A,
Groundwater, Book 6, Modeling Techniques, US Geological Survey, Reston, Virginia, 58 pp.
Risser, D.W., Davis, P.A., Baldwin, J.A. and McAda, D.P. (1984) Aquifer tests at the Jackpile-Paguate uranium mine, Pueblo of Laguna,
West-Central New Mexico, United States Geological Survey Water-Resources Investigations Report 84-4255, United States
Department of the Interior, Albuquerque, New Mexico.
Robertson, A. and Shaw, S. (2004) Use of the multiple accounts analysis process for sustainability optimization, in Proceedings SME
Annual Meeting, 23–25 February 2004, Denver, Colorado.
Spitz, K. and Moreno, J. (1996) A practical guide to groundwater and solute transport modeling, John Wiley and Sons, Inc., New York,
461 pp.
United States, Bureau of Land Management, Rio Puerco Resource Area (1986) Final environmental impact statement for the Jackpile-
Paguate uranium mine reclamation project: Laguna Indian Reservation, Cibola County, New Mexico, U.S. Department of the
Interior, Albuquerque, New Mexico.
Zehner, H.H. (1985) Hydrology and water-quality monitoring considerations, Jackpile uranium mine, northwestern New Mexico,
United States Geological Survey Water-Resources Investigations Report 85-4226, Albuquerque, New Mexico.


R.A. Peterson SNC-Lavalin Inc., Canada

S.A. Humphries SNC-Lavalin Inc., Canada
M.L. Unger Teck Resources Limited, Canada
The Sullivan mine in Kimberley, British Columbia, operated for almost 100 years and was one of the world’s
largest producers of lead, zinc and silver. Environmental and risk management controls were initiated in the
1960s, including construction in 1979 of the world’s first high-density sludge treatment plant for treatment
of acid rock drainage (ARD) associated with large impoundments of tailings, waste rock, other process wastes
and underground mine workings. Large-scale assessment of surface water and groundwater systems were
initiated in the 1990s in combination with the implementation of additional mitigation measures to control
ARD-related impacts. The Sullivan mine officially closed in 2001, and comprehensive site investigation and
risk assessment results were used to develop an adaptive site-wide risk management plan (RMP) that outlines
long-term monitoring requirements for groundwater, surface water, vegetation and aquatic biota, with
triggers for incremental assessment and mitigation as necessary.
Since closure, predicted improvements in groundwater and surface water quality have been realised;
however, in some locations, RMP action levels have been triggered and incremental assessment and
mitigation have been required. As an example, action level triggers were reached in surface water at the lower
mine yard, where acidic seepage high in metals from shallow groundwater discharges was detected in the
adjacent creek. This triggered additional investigation and implementation of interim measures to capture
the groundwater as a longer-term solution is being developed. Triggers were also reached in deeper
groundwater in the lower mine yard, and monitoring programs have been adapted to these changing
conditions. Other challenges have been administrative, resulting from a closure process spanning over a
decade and changes to regulatory policies and mine closure requirements during this time. For example, policy
changes under the British Columbia Contaminated Sites Regulation (BC CSR) now require protection of
aquifers for future drinking water use, which has affected applicable water quality criteria. Soil quality criteria
have also changed for a number of exposure pathways following active closure, changing risk assessment
assumptions and prompting additional site investigation activities to comply with the criteria under the BC
CSR. These evolving physical and administrative conditions, combined with variations in weather and
watershed land use, have reinforced the need for closure plans to include appropriate post-closure
monitoring, anticipate the potential need to revisit assumptions made during closure planning and set trigger
levels for investigating the need to adjust risk management processes.
The Sullivan underground lead-zinc-silver mine is located in Kimberley, British Columbia, Canada (Figure 1,
below). The mine operated for 92 years, from 1909 to 2001, and was one of the world’s largest producers of
lead and zinc, producing 8 million and 9 million tonnes, respectively. The Sullivan mine ore body averaged
6% lead, 5.7% zinc and 24.8% iron and consisted mainly of pyrrhotite and pyrite (7:3 ratio) as the most
abundant sulphide minerals, with galena and sphalerite as the principal ore minerals. Environmental issues
were unknown at the beginning of the mine’s life, but Teck started implementing environmental controls in
the 1960s. Reclamation research was initiated in the 1970s, and the world’s first high-density sludge
treatment plant was constructed to treat acid rock drainage (ARD).

Adaptive risk management and lessons learned: post-closure at the “mighty” Sullivan mine R.A. Peterson, S.A. Humphries and M.L. Unger
The mine closed in December 2001 owing to exhaustion of the ore body. The site has undergone
decommissioning, reclamation, extensive monitoring, water collection and treatment since 2001 based on a
decommissioning and closure plan approved by the Ministry of Energy and Mines. Post-closure monitoring
at the site has been conducted using an adaptive risk management plan (RMP) to ensure closure objectives
are continually being met. This paper discusses the post-closure management of the site and is a continuation
of Humphries et al. (2011).
Underground mine operations began in 1909, and a small open pit operated for a brief period of time.
Production was initially limited because the lead and silver were separated by hand. The zinc contained in
the ore was not a desired commodity at the time. In 1920, the Sullivan mine was the first to employ a newly
developed differential flotation process for sulphide minerals to produce lead concentrate on a commercial
scale. In 1923, a concentrator was constructed at a location remote from the mine site, as large flat surfaces
were required for the placement of tailings impoundments. The tailings were rich in iron and sulphur and
were used to support subsidiary industrial processes and buildings that operated for a short period of the
mine life and included fertiliser, iron and steel plants.
The mine site is located on property owned and operated by Teck Metals Ltd. (Teck) and its predecessors.
The mine property was developed in two distinct geographic locations (see Figure 1). The former
underground mine and waste rock dumps are located immediately north and west of the City of Kimberley.
The concentrator, subsidiary industrial processes and tailings ponds (referred to hereafter as the mill site),
are located south of Kimberley, approximately 6 km from the mine. The drainage water treatment plant
(DWTP) is located on the St. Mary River approximately 5 km south of the mill.
A total of 9.75 million metric tonnes of waste rock was generated and placed at four locations at the mine
site, the north and south waste dumps along the Mark Creek Valley in the lower mine yard (LMY) and the No.
1 Shaft waste dump and the open pit waste dump located above the city centre.
The Sullivan concentrator produced waste by-products of tailings and float rock, a coarse by-product from
ore separation through heavy media production (i.e. flotation process). Float rock was generated from 1947
until closure, with 4.3 million metric tonnes stockpiled at the concentrator. The tailings ponds were located

Case Studies
immediately adjacent to the southeast side of the concentrator. Approximately 122 million metric tonnes of
mill tailings averaging 0.6% lead, 0.6% zinc and 27% iron were produced over the life of the mine.
Ammonium phosphate fertiliser was produced between 1953 and 1987 using iron concentrate from the
concentrator and ammonia and phosphate rock from other producers. The fertiliser process resulted in
residual wastes including 7.0 million metric tonnes of phosphogypsum and 3.4 million tonnes of iron oxide.
The gypsum was initially discharged into the St. Mary River but later placed into storage ponds to minimise
the impact on receiving waters. The impoundments for the gypsum were constructed further east of the iron
calcine ponds and south of the concentrator tailings ponds. The stockpiled iron calcine was sold for cement
manufacturing until 2011.
Mine site
Mill site
ARD is the main environmental concern in the former mine workings, waste rock piles and tailings
impoundments. Acidic water with high concentrations of metals and sulphate has been found in both surface

and subsurface waters. High metals concentrations in soil and vegetation exists around the mine site.
Fertiliser plant by-products are also sources of fluoride and sulphate in groundwater and soils.
Programs to address environmental concerns were initiated in the late 1960s and focused on reduction of
waste discharge directly to watercourses and reclamation of waste disposal areas on land. A research
program was initiated in 1972 to develop revegetation technology to return waste disposal sites and
disturbed land to productive rangeland. In 1979, Teck commissioned the DWTP to treat collected acidic
waters from the mine, tailings decant water from the concentrator and seepage water collected
downgradient of the tailings and gypsum ponds.
The DWTP is a high-density sludge treatment process that uses lime to neutralise acidic water and precipitate
heavy metals in the form of hydroxides in an iron sludge. The effluent from the plant is discharged into the
St. Mary River, and the sludge is retained in a holding pond. The treatment plant is highly effective, lowering
total metal concentrations in influent by over 98% prior to discharge. The DWTP treatment plant operates
under an effluent permit from the BC Ministry of Environment.
Although site investigations were undertaken during mine operation, large-scale environmental assessment
in the form of comprehensive soil, groundwater, surface water and sediment investigations commenced in
the early 1990s, approximately coinciding with closure. Based on these investigations, a hydrogeological and
geochemical conceptual model was developed to allow predictions of post-closure groundwater and surface
water concentrations. A risk assessment for human and ecological (terrestrial and aquatic) receptors (HHERA)
was completed, which included biota collection and analyses.
Surface water and groundwater flow patterns in the area have been shaped by till/bedrock topography
caused by large permeability differences between till/bedrock and other, more permeable surficial deposits.
The majority of groundwater flow occurs through sand and gravel deposits within and above the till and
generally follows buried sub-basins defined by the till/bedrock topography. Groundwater flow in the mine
area is generally divided into four discrete drainages that discharge into three different creeks. In some areas,
such as the LMY, significant heterogeneity in proximate source materials has required detailed investigation
to obtain an adequate understanding of local conditions. For the mill site and surrounding area, extensive
characterisation and delineation of the bedrock/till topography identified eight distinct groundwater flow
paths that ultimately discharge to the St. Mary River (Figure 2).
At the mill site, groundwater travel times to receiving water bodies are relatively long; thus, geochemical
modelling was undertaken to predict geochemical evolution of ARD-impacted groundwater along flow paths
2, 4 and 6. The model predicted it would take between 150 to 480 years to consume the carbonate aquifer
buffering capacity between the waste impoundments from the mill site and the St. Mary River, the main

Case Studies
downgradient aquatic receiving body (Humphries et al., 2011). Results from the modelling were used to
predict future concentrations of parameters of concern that may affect aquatic receptors in support of the
HHERA, as well as to provide a broad timeframe for the prioritisation and implementation of mitigation
measures. The draft RMP, described below, was based on the HHERA.
The Decommissioning and Closure plan was finalised and approved in 2001. General closure activities
included building demolition, remediation of hydrocarbon contaminated soils, removal of hazardous
materials, filling of the open pit, closure of landfills, and revegetation of the site. The mine openings were
closed and sealed, and mine subsidence areas were fenced to prevent public and wildlife access. Signage and
fencing were installed to prevent trespassing.
A major aspect of site decommissioning and closure was to minimise the amount of water contacting mine
waste and to protect groundwater and surface watercourses. As such, soil cover systems and water
interception measures were installed. The soil covers at the site include an engineered complex cover for the
tailings waste impoundments and a simple soil cover for the waste dumps and plant sites. The engineered
soil cover system for the tailings waste impoundments was designed based on nearly two decades of research
and investigation using locally available materials at sites representative of those to be reclaimed (Gardiner
et al., 1997). In total, 1,000 hectares of mine-disturbed areas were reclaimed.
Teck developed a long-term water management strategy at both the former mine and the former mill sites
to reduce downgradient loading of contaminants associated with historical activities and comply with
applicable regulations. The strategy consists of a variety of mitigation measures, a number of which include
hydraulic interception for treatment at the DWTP. The hydraulic interception measures were installed based
on the comprehensive environmental assessment work conducted and generally include deep groundwater
aquifer wells and a network of trenches and ditches with associated sumps and pumps installed to intercept
shallow groundwater. The trenches and ditches are typically keyed into a low permeability feature (i.e.
till/bedrock) downgradient of the source areas. An example of the construction and final trench in the James
Creek catchment is shown in Figure 4.
Approximately 24 km of pipeline and 30 aquifer pumps collect contaminated seepage water. The seepage
water is stored in a large surface water reservoir referred to as the ARD pond, which has enough capacity to
allow the DWTP to operate biannual campaigns. An underground mine dewatering system was also

constructed to store and collect ARD water generated from the mine workings. This water is also treated on
a biannual basis. The DWTP will continue to operate as required.
In 1991, at the request of the BC Ministry of Energy and Mines, Teck developed a comprehensive
decommissioning and closure plan and presented it to a public forum for review. The review participants,
forming the Sullivan Public Liaison Committee (SPLC), included government agencies, local union
representatives, city councillors, members of the public and the local environmental society.
Based on the requirements of the Mines Act, the closure plan focused on the effects of ARD from waste rock
dumps and waste impoundments. The BC Environmental Management Act defines the mine workings and
waste areas as “core” areas and other areas, including industrial facilities and ancillary operations such as
maintenance shops and mills, as “non-core” areas. Independent environmental studies for the demolition of
buildings and remediation of non-core areas were initiated in 1995. As the studies progressed, it became
clear that the potential impacts of metals as a result of mining and non-mining activities also needed to be
investigated (Higgins et al., 2004). To adequately determine whether post-closure conditions were
acceptable and properly managed in the long term, Teck adopted a risk assessment/risk management
approach. This approach was taken in consultation with regulators and the SPLC, as the Sullivan mine was
the first in BC to use a risk assessment approach for closure. The risk assessment process began in 2001 and
was concurrent with the final reclamation phases of the site. The reclamation permit was amended in 2004
with endorsement from regulators and the SPLC; one of the conditions was approval of an environmental
risk assessment by the BC Ministry of Environment. The draft risk assessment and RMP were submitted in
2005/2006.
Since submittal of the draft risk assessment and RMP, there have been a number of changes in regulatory
policy and therefore regulatory end points (goalposts) that have affected ongoing management. For example,
an added protocol in the BC Contaminated Sites Regulation (CSR) resulted in the need to consider protection
of groundwater for future drinking water purposes in areas with no current or planned use. The new protocol
resulted in ongoing consultation with the BC Ministry of Environment as to how to manage areas where
drinking water standards may be exceeded. In addition, risk assessment techniques and therefore policies
have also advanced since submittal; associated CSR requirements include new methods for the application
of land use standards and evaluation of terrestrial receptors, changes to technical guidance related to site
investigation and aquatic assessment (including addition of new methods of assessment), and changes to
numerical toxicity reference values (TRVs).
The RMP is the primary tool used to guide the long-term environmental management of the site. The purpose
of the RMP is twofold: (1) to ensure that future activities do not pose unacceptable risks to persons spending
time at the site or to the environment, and (2) to monitor and manage environmental conditions at the site.
The first was achieved by establishing a framework for action that will be taken should certain “triggers” for
surface water quality, groundwater quality, vegetation performance and receiving environment testing be
exceeded. The second purpose was achieved by linking the RMP to requirements of the Sullivan reclamation
permit issued under the BC Mines Act.
The RMP for water quality is generally structured around the protection of aquatic life based on results from
the HHERA. The hydrogeological and geochemical conceptual model identified different timeframes for
aquatic life exposure; as such, the RMP is structured differently for groundwater intrinsically linked to
receiving water bodies. For groundwater in site 1, flow paths 7 and 8 at site 2 and receiving water bodies at
both sites, the triggers are based on numerical criteria developed from regulatory criteria and/or risk-based
benchmarks established in the HHERA. For groundwater in site 2, where contaminant migration is expected

Case Studies
to occur over longer timeframes, the RMP is designed to track the evolution of groundwater quality and obtain
the information needed to plan further mitigative action and/or obtain the data needed to demonstrate that
breakthrough is unlikely.
The RMP is ongoing and is managed by Teck. The RMP is a result of requirements for closure and involves a
component of regulator consultation. However, the BC Ministry of Environment has not officially approved
the RMP because at the time it was submitted, the process fell outside of the Ministry’s typical administrative
procedures and the risk assessment was not finalised. Additional discussion is presented in Section 4.
The monitoring program at Sullivan includes 69 groundwater sites, 18 surface water sites, benthos
monitoring sites and a number of vegetation sites as part of the RMP. As can be expected on a site with a
number of sizable contaminant sources spread over a large mountainous area with complex geology,
hydrogeology and hydrology, trends at these monitoring stations vary based on the preferential groundwater
flow paths. However, water quality improvements from mitigation measures have been observed at a
number of locations and are consistent with the hydrogeological and geochemical conceptual models
established during the site characterisation work (Humphries et al., 2011).
The triggers are based on the hydrogeological and geochemical conceptual site model and HHERA developed
for the site and differ depending on the timeframe in which groundwater will potentially impact receiving
water. For site 1, flow paths 7 and 8 at site 2, and receiving water, the surface and groundwater quality
triggers are based on exceedances of regulatory standards, maximum concentrations within five years and/or
rising trends over three years, which would trigger additional investigation or corrective action. For
groundwater in site 2, the trigger for mitigative action or additional studies is based on regular review of
historical trends every six years. As the site 2 dataset accumulates, it will be continually evaluated as detailed
in the RMP, and incremental assessment and/or mitigative action will be taken to achieve risk-based objectives.
These monitoring plans, trigger and actions are shown graphically in Figure 5.

The Sullivan mine will require perpetual monitoring and maintenance related to the existing sources at the
site and operation of the DWTP. Since closure, the RMP and related ongoing management have had to take
an adaptive approach to address both changes in regulatory policy and triggers for action from monitoring;
these are described below.
As described in Section 3, a number of changes in regulatory policy have occurred that were neither
anticipated nor incorporated into the RMP. This has created a need for further consultation with the BC

Case Studies
Ministry of Environment and additional assessment requirements outside of the RMP, and may ultimately
have implications for the RMP. The evolution of guidance and regulations over the span of a decade post-
closure has highlighted the need for an adaptive approach to generate certainty in end points (goalposts),
and allow for improved timing of regulatory approvals, resource allocation and planning.
Surface and groundwater monitoring have occurred across the site for decades, allowing for trends in water
quality to be measured before, during and after the closure process. Groundwater concentrations of key
contaminants (i.e. sulphate and dissolved metals) at several sampling stations downgradient from site 1
sources, such as the No. 1 Shaft waste dump and the open pit waste dump, have been decreasing over time.
Concentrations in receiving environments downgradient of these sources, including Lois Creek, have also
exhibited decreasing trends supporting the water quality improvement predictions associated with the
mitigation and closure activities. Groundwater and surface monitoring results from locations in the LMY
intended to assess potential impacts related to the north and south waste dumps also indicate improving
quality and confirm interception system performance; however, in recent years, concentrations of elevated
zinc and other indicator parameters have been identified in Mark Creek and have triggered further
assessment beyond the routine RMP monitoring, as discussed further below.
At site 2, monitoring results also support the predicted improvements in many areas, including reductions in
water levels in source areas attributed to water diversion and cover systems, and improved chemistry in
select groundwater and receiving water monitoring locations. In locations where improvements are not
expected (e.g. some source areas), chemistry and hydrogeological conditions are compared to modelling
predictions to assess potential requirements for supplemental mitigation as necessary.
In recent years, RMP action levels have been triggered in some locations, and incremental assessment and
mitigation have been required. Action level triggers were reached in Mark Creek surface water at the LMY,
where acidic seepage high in metals from shallow groundwater discharges was detected in the creek. This
discovery occurred through visual observations rather than by analysing quantitative results from RMP
monitoring and comparison to trigger concentrations. The seepage was discovered in an area where sand
drains were installed to direct shallow groundwater to the deeper aquifer for hydraulic interception and
treatment during periods of higher groundwater levels. Since the sand drains were installed, there have been
changes upslope of the waste dump, including the clearing of large forested areas for residential and ski hill
development, which may have resulted in additional groundwater inputs above the waste dump. Changes to
surface and groundwater flow regimes, along with different patterns of precipitation and temperature (key
factors identified as contributing to seepage), were conditions not anticipated during mitigation planning and
are outside of Teck’s control. Additional investigation was triggered by the seepage discovery and included
implementation of benthic invertebrate and surface water monitoring programs outlined in the RMP.
Detailed groundwater investigations were initiated alongside weekly surface water monitoring to gain
further understanding of the seepage mechanisms and seasonal groundwater/surface patterns and
interactions, and to identify long-term solutions. Results from the investigations were used to identify and
trigger implementation of interim mitigation measures, including a series of shallow collection wells and
dewatering pumps to capture the groundwater, while a longer-term solution is being developed. Results of
the investigation work were used to refine the RMP monitoring program, which has been adapted to assess
new areas of concern and the effectiveness of temporary mitigation features. The RMP monitoring program
revisions have included the addition of new sample collection points in the creek and for groundwater, as
well as incorporating visual inspections of the creek and monitoring pump operation to maximise
performance.
Action levels were also reached in deeper groundwater in the LMY, which trigged additional sampling and
geochemical review of historical and recent groundwater conditions. Monitoring programs have been
adapted to include more detailed review of select areas of the LMY ahead of planned comprehensive review

in the RMP. Additional mitigation is not considered necessary, as the trigger levels in groundwater are
attributed to variability in flow rather than a factor requiring further mitigation at this time. No receiving
water monitoring locations in groundwater discharge areas have exhibited changes in quality that would
require additional deeper groundwater mitigation.
Additional mitigation actions unrelated to RMP triggers are based on desire to enhance environmental
performance and reduce uncertainty and risk for the long term. In general, the seepage collection pumps
and associated collection infrastructure have been well maintained, and long-time employees and
contractors have a good working knowledge of the systems. This results in a consistent approach to
management of the systems and opportunities to preserve infrastructure; however, supplemental actions
have been implemented for continual improvement and to address gaps in the systems. These initiatives
have included developing an updated site-wide water balance for post-closure conditions, improving
monitoring instrumentation, enhancing seepage collection maintenance programs and a variety of
supplemental investigation and mitigation activities to improve water collection and treatment efficiency.
Water quality and general site conditions continue to evolve; as such, diligence and regular review of
operational and RMP activities are required to ensure appropriate assessment and mitigation activities, and
an early warning system to detect potential issues, are in place. Similarly, as water quality is improving in
many areas, reductions in RMP monitoring efforts and frequency in these locations will be warranted and
will be proposed during RMP update processes. In addition to non-climate-related conditions, potential
effects related to climate change must be considered; these pose challenges based on the uncertainty of
how, and to what extent, conditions will change in the future.
More than a decade of post-closure experience has provided valuable insight for future management of the
Sullivan. Following are several key lessons learned through closure.
 Build adaptive plans for the long-term
Where long-term monitoring and risk management is required, management plans should incorporate
adaptive features that clearly outline how change will be handled. This includes processes for identifying
when and why updates might be required, how updates will be managed, who will be responsible for
executing and reviewing changes to the plan, and what performance indicators could be managed to assess
adequacy of change. Where plans include monitoring and maintenance activities, triggers for action and for
reducing monitoring requirements should be clearly outlined to ensure programs are representative and
sensible for the long term. In addition, triggers may require regular review and adjustment to ensure
adequate management of risks. Site managers should be flexible and expect to change the plan as lessons
are learned and conditions change.
 Confirm goalposts wherever possible
As regulation and protocol for closure-related activities can change over the course of closure processes,
there is potential to have to repeat or update work and documentation to meet current requirements.
Wherever possible and reasonable, the regulatory/environmental requirements (i.e. the goalposts for the
work) should be confirmed with and accepted by appropriate agencies. This may also include less formal
agreements on various approaches, where confirming an understanding of regulatory acceptance is
warranted to ensure disruptions such as changes in personnel or changes in regulatory oversight are
managed practically.
 Anticipate and prepare for changes
Conditions at any given site, as well as in the surrounding area, will change over time and can include positive
change (e.g. realisation of predicted improvements in water quality) and negative change (e.g. changes at
neighbouring sites with adverse affects for the site). Conditions may also not appear as they seem, and

Case Studies
complexity and heterogeneity pose additional challenges for managing large sites that cover a wide area with
variable climatic, geological and ecological conditions. Managers should anticipate potential changes during
closure planning, including worst-case scenarios, and consider how these may be managed or even avoided
if consequences are negative. Site managers planning for change benefit from a combination of creativity,
technical knowledge and experience at the site and in the surrounding area.
 Consistent maintenance is required to preserve infrastructure and ensure efficiency
A solid understanding of site infrastructure and a robust, well informed maintenance program will help
ensure infrastructure performance and durability over the long term, and help maintain operational costs.
At the Sullivan mine, the consistency related to long-term employees and contractors intimately familiar with
the site infrastructure was beneficial leading up to closure and has continued to be so.
 Visual inspections detect issues
Visual observations made by site personnel and contractors have identified issues at the Sullivan mine that
warranted further investigation and/or implementation of corrective measures. Some issues might not have
been detected through routine monitoring based on their scale, or because the location is not monitored as
part of the RMP or other site monitoring. Issues have been detected through observations made by long-
term employees; less experienced or less attentive employees might not have identified a potential problem.
An alert and experienced eye can help identify and address issues in a timely manner, and, in general, visual
inspections are relatively easy and inexpensive.
 Improve continuously
Site knowledge will build over the course of mine operation and through closure, and both minor and major
adjustments will be identified that can benefit the site, and site managers, for the long term. New
perspectives, new information, new technologies and lessons learned will become available through any
post-closure period. This provides great opportunity for continual improvement, and site managers should
take advantage of these lessons wherever possible to enhance environmental, regulatory and financial
performance.
The “mighty” Sullivan mine in Kimberley, BC, operated for almost 100 years and was one of the world’s largest
producers of lead, zinc and silver. Environmental and risk management controls were initiated in the 1960s
to address ARD issues associated with large impoundments of tailings, waste rock, other process wastes and
underground mine workings, and included construction of the world’s first high-density sludge water
treatment plant. The Sullivan mine officially closed in 2001, and comprehensive site investigation and risk
assessment results were used to develop an adaptive site-wide RMP that outlines long-term monitoring
requirements for groundwater, surface water, vegetation and aquatic biota, with triggers for incremental
assessment and mitigation as necessary. Stakeholders involved with closure planning included the SPLC,
government agencies, local union representatives, city councillors, the public, and the local environmental
society.
An important aspect of risk management at the Sullivan mine is Teck’s long-term water management
strategy, which includes a wide array of hydraulic interception and storage features to reduce loadings of
contaminants from the mine and mill sites. Since closure, predicted improvements in groundwater and
surface water quality have been realised; however, RMP action levels have been triggered in some locations,
and incremental assessment and mitigation have been required. Action has also been triggered as a result of
visual observations and monitoring activities outside of the RMP. Changes in site conditions, as well as
changes in surrounding land use and climate, have posed challenges for site management and have required
adjustments to the RMP. Other challenges have been administrative, resulting from a closure process
spanning more than a decade and changes to regulatory policies and mine closure requirements during this
time. These evolving physical and administrative conditions, combined with variations in weather and
watershed land use, have reinforced the need for closure plans to include appropriate post-closure

monitoring, anticipate the potential need to revisit assumptions made during closure planning and set trigger
levels for investigating the need for adjustments to risk management processes. Lessons learned through the
post-closure process include the importance of building an adaptive plan, confirming regulatory expectations
and anticipating and preparing for changes; the benefits of a consistent maintenance program and
experienced staff; and the importance of making visual inspections and taking advantage of opportunities
for continual improvement.
Gardiner, R.T., Dawson, B.B. and Gibson, G.G. (1997) Application of ARD abatement technology in reclamation of tailings ponds at
Cominco Ltd., in Proceedings 4th International Conference on Acid Rock Drainage (ICARD), May 31 - June 6, 1997, Vancouver,
Canada, Volume 1, pp 47 – 63.
Higgins, C., Wilson, R., Sandstrom, N., Dawson, B., Mann, G., Allard, P. and Baker, R. (2004) Post-closure human health and ecological
risk assessment at Teck Cominco’s Kimberley operations, BC overview of regulatory process and findings of the problem
formulation, in Proceedings 2004 Technical and Research Committee on Reclamation (TRCR) Annual Mine Reclamation
Symposium, 23 June 2004, Cranbrook, Canada.
Humphries, S.A., Thomson, D.G. and Unger, M.L. (2011), Long-term post-closure surface water and groundwater risk management
and subsequent water quality trends at the former Sullivan Mine, Canada, in Proceedings 6th International Conference on
Mine Closure, A.B. Fourie, M. Tibbett and A. Beersing (eds), 18–21 September 2011, Lake Louise, Canada, Australian Centre
for Geomechanics, Perth, Volume 2, pp. 367–376.

M.T. Lord MMG, Australia

M.H. Adams MMG, Australia
T.J. Shearman MMG, Australia
Mining and processing at MMG’s Century Mine is scheduled to cease in late 2015. Detailed planning for
transition operations into the active closure phase commenced in early 2011. This paper briefly summarises
the seven aspects of closure that constitute closure planning at Century and provides some recommendations
on practical measures that can be adopted to increase the likelihood of achieving a successful closure
outcome.
These observations may be of benefit to mining companies that wish to reduce their closure liability and
increase the potential to achieve lease relinquishment.
MMG Century project (Century) is situated in the North West Queensland Mineral Province and includes the
Lawn Hill mine site, situated approximately 250 km north-northwest of Mount Isa, and a dewatering plant
and port facility located at the Port of Karumba on the Gulf of Carpentaria. A 304 km slurry pipeline transports
the zinc and lead concentrates from Lawn Hill to Karumba. The Gulf region experiences high temperatures:
over the summer months (October to December); the monthly average exceeds 38°C and maximum
temperatures can be as high as 46°C. The tropical wet season in the region has an average annual rainfall of
approximately 530 mm and an annual average evaporation rate of approximately 3,100 mm. Prolonged
drought conditions can be interspersed with heavy rainfall associated with cyclones or monsoons. Periodic
flood events can cut road access and disrupt operations.
Century Mine is geographically remote, approximately 390 km by road from Mount Isa, which is the nearest
major town. The Gulf region has an area of approximately 186,000 km3 with a population of <15,000 people.
Century is the world’s third largest zinc mine. It is scheduled to cease production by the end of 2015 due to
resource depletion within the Lawn Hill ore body. Detailed planning for the transition from production to
active closure commenced in early 2011 with the appointment of a full-time closure planning resource.
To be successful, closure planning must be incorporated into all aspects of the mining operation. At Century,
closure planning is based on the following aspects:
 workforce
 stakeholders
 environment
 legal and other obligations
 financial and assets
 operations
 project management and execution

Century Mine closure M.T. Lord, M.H. Adams and T.J. Shearman
This paper presents key insights on these aspects based on closure planning at Century and provides
examples of measures that can be employed to increase the likelihood of achieving mine closure objectives,
including the ultimate objective of achieving lease relinquishment.
The workforce at Century exceeded 900 employees and contractors at the peak of production and included
a significant number of Indigenous people from the lower Gulf region during the life of the project. The site
operates entirely on a fly-in/fly-out (FIFO) basis due to the remote location of the mine. Turnover of
employees has been low by comparison with other FIFO operations, and a significant number of employees
have worked at Century for more than 10 years.
Communication with the workforce regarding closure planning commenced several years before the
scheduled cessation of production. The objective was to create an environment where there were no
surprises and to make employees aware that the process was one of transition rather than closure (i.e., the
site would not be shutting the gate when production ceased). A timeline (Figure 1) was used to demonstrate
that the site’s active and passive closure phases would require some personnel on site for a period of
decades, and that retention of key personnel to manage the site through these phases was desirable in order
for the company to achieve a successful outcome.
Figure 1 demonstrates the concept of beginning closure planning before commencement of mining and
processing. The green blocks within the timeline show rehabilitation activities that were conducted
opportunistically as areas became available and demonstrate the escalation of rehabilitation efforts towards
the end of the mining and processing phase as waste rock dumps became available for rehabilitation. It is
anticipated that five years will be required to complete rehabilitation of disturbed and constructed landforms
prior to the site entering into the passive closure phase. Site infrastructure will be retained in a state of
operational readiness during this time in anticipation of one or more growth options that may enable
resumption of operations.
The passive closure phase, which involves site management and monitoring activities, may extend for a
period of 30 years due to the requirement to obtain data that demonstrates progress towards achieving
rehabilitation completion criteria. This period will also allow constructed landforms, mineralised waste
encapsulation systems and developing ecosystems to be exposed to a range of climatic events such as
droughts, extreme rainfall events, fire and cyclonic winds in order to demonstrate stability and resilience.

Case Studies
Graphics were also used to show that the company was actively exploring opportunities to resume operations
should one or more of the resource development options prove to be economically viable. The site will be
placed in a state of operational readiness for a period of three to four years upon cessation of processing
from the main pit to ensure that valuable infrastructure isn’t removed prematurely.
A range of media was used to provide regular updates on transition planning and progressive rehabilitation
work. These included face-to-face presentations with workgroups, articles in newsletters, slide presentations
delivered via visual display units in the village dining area and the foyer of the main administration building,
media interviews, “state of the nation” addresses from the general manager, employee induction sessions
and presentations to the workforce consultative forum representatives.
The site management team and a cross-section of employees were given an opportunity to rate transition
planning success by completing a closure balanced scorecard assessment every six months. The survey
respondents rated each of 25 elements that make up the seven aspects of closure at Century on a scale from
one to seven. The results were presented on A3-sized posters that were displayed around the work areas.
Detailed planning for the transition from production to active closure at Century was undertaken by the
human resources (HR) team. An important aspect of planning is to consider how an operation will retain the
key personnel required to achieve a successful closure outcome. It is also important to distinguish between
roles and skills, as some skills that will be critical to managing the site do not necessarily reside within a typical
role description. These skills and qualifications may include emergency response skills, radiation safety officer
certification, first aid qualifications, electrically appointed person accreditation, firearms licences, wildlife
handling certification, aircraft refuelling certification, forklift tickets, confined spaces authorisation,
stakeholder engagement competency, media relations competency and airport reporting officer
qualifications. Developing a skills matrix to align postprocessing requirements with the skill set of the existing
workforce can assist in retaining competencies and identifying training requirements to address skill gaps.
Mining companies that are planning to transition from operations should consult early and often with the
workforce. Individual presentations to workgroups using quality graphics and key messages are an effective
way to engage the workforce and raise awareness of the roles workers will be required to fill to support the
active and passive closure phases. A detailed HR transition plan that addresses issues such as communication
with the workforce, employee retention, postproduction role requirements, postproduction skills matrix,
proposed rosters, proposed commuter base support for FIFO operations and so on should be developed at
least two years prior to the planned date for cessation of mining operations.
Century has a significant and diverse range of stakeholders. The mine is the largest employer in the lower
Gulf region and provides direct and indirect support to Gulf communities, organisations, businesses and local
government authorities. The company developed a closure consultation strategy approximately four years
before ceasing operations to assist in managing consultations with stakeholders. Records of engagement are
captured in a commercially available stakeholder engagement database. Consultation with stakeholders
began several years prior to the scheduled date for cessation of production to provide communities, councils,
suppliers, state government agencies and businesses with enough time to develop other revenue and
employment streams and to plan for less reliance on the Century project for employment, training, revenue
and other support services.
Engagement with the four native title Indigenous groups with whom Century has formal agreements has
been a critically important aspect of transition planning. Regular updates have been provided at the quarterly
Century Environment Committee meetings. Field trips have been conducted to provide Indigenous groups
with updates on the progress of land rehabilitation activities on the site. Input into the development of the
site’s rehabilitation success measures (completion criteria) has also been sought from these groups on an
ongoing basis. This has included workshops and field trips with a group of elders who represent the
traditional owners of the land on which the mine is situated.

Recording all commitments made by company representatives and ensuring that this knowledge is
transferred when turnover in management occurs is critical to avoid mixed messages, perceptions of broken
promises, loss of knowledge or unrealistic expectations.
Landform visualisation boards that show the current and proposed time-lapse appearance of mining
landforms from five key observation areas have been helpful in consulting on the proposed appearance of
postmining landforms. The observation point images in Figure 2 show Magazine Hill, a site of cultural
significance to the Waanyi people, and an operational sediment dam in the foreground. The waste rock dump
that wraps around Magazine Hill was constructed in 2002–2003 and does not meet contemporary design
standards for encapsulating mineralised waste. Relocation of this dump commenced in late 2014 and will be
completed by the end of 2015. This project provides a positive outcome from a cultural heritage perspective
whilst removing a long-term source of acid metaliferous drainage (AMD).
Developing a detailed understanding of the impact of mine closure can assist in transition planning. Century
engaged the University of Queensland’s Centre for Social Responsibility in Mining (CSRM) to conduct a
comprehensive mine closure social impact assessment (SIA). This study has enabled Century to direct
resources and planning to issues that are of concern to stakeholders. The study also provides insights into
areas where the state government may be able to help stakeholders adjust to life after production at the
mine ceases.
It is recommended that mining companies consult early and often with stakeholders, develop landscape
visualisation materials to help stakeholders assess proposed final landforms, seek stakeholder input when
developing rehabilitation success measures and undertake a comprehensive social impact assessment to
better understand the potential impacts on communities and evaluate opportunities to mitigate adverse
effects where practicable. Whilst it may not be possible to achieve all objectives sought by stakeholders, it is
important that the mining company actively engages with stakeholders over a period of years in a way that
is viewed as fair, transparent and honest.

Case Studies
Ongoing adverse impacts on the environment often preclude lease relinquishment for metaliferous mining
operations, with resultant implications for the parent company in terms of cost and reputation. In order to
maximise the likelihood of a successful closure outcome, the Century operation formed a small team of
environmental professionals to identify risks, undertake studies, develop designs and execute a programme
of closure projects. In order to better evaluate risks, the team members visited several closed mines in North
Queensland where environmental objectives had not been met. Lessons learned from these sites assisted in
planning for closure. Key lessons included the following:
 Ensure that the site is configured so that management of AMD can be achieved without reliance
on human intervention.
 Maintain relationships with neighbours.
 Retain critical information (data, as-built drawings, management plans, etc.).
 Establish realistic completion criteria.
 Develop a nonshedding mineralised waste encapsulation system in semi-arid regions that are
subject to high-intensity rainfall events.
 Ensure that a dedicated closure team is in place to take responsibility for closure planning and
execution.
Senior mining engineers participated in a visit to one closed mine to raise awareness of the importance of
avoiding or minimising planning-related mine closure liabilities.
A significant risk associated with metaliferous mines is the potential for AMD to adversely affect surface and
groundwater systems. Mine owners who fail to adequately evaluate this risk throughout the life of the project
or to implement ongoing management solutions could bring about a scenario that leaves the owners
responsible for care in perpetuity. Century sought technical research and expert opinion over the life of the
project on the most suitable means of encapsulating mineralised waste to limit the potential for AMD
generation. The methodology and supporting quality assurance and quality control verification processes
have been refined through ongoing research and expert review.
Century has adopted an enhanced store-and-release cover system, which is well suited to sites in semi-arid
environments that have distinct wet and dry seasons. Placement of a two-metre-thick sponge of weathered
limestone over a highly compacted reduced permeability layer (RPL) absorbs rainfall, which is subsequently
returned to the atmosphere via evapotranspiration. This approach provides a nonshedding cover, thus
eliminating the requirement to construct drains to convey stormwater from the waste rock dump’s horizontal
surface to the adjoining natural topography. A layer of inert limestone rocks has been placed on the dump
slopes at a thickness of approximately 20 m. The horizontal surface of the dumps falls approximately 1–3
degrees towards the pit void. Should the storage capacity of the store-and-release cover be exceeded during
extreme rainfall events, as occurred following a 1:150 year event in 2011, water will migrate across the RPL
and infiltrate at low rates through the limestone rock cover on the outer batters without causing erosion.
Observations from other operating and closed mines suggest that engineered drainage systems that seek to
drain water from waste rock horizontal surfaces to the adjoining natural topography via a system of
constructed drainage structures are prone to failure during high-intensity rainfall events and do not achieve
the long-term objective of creating a stable landform.
The store-and-release mineralised waste encapsulation system will limit the ingress of water and air into the
dumps once it is installed; however, the dumps have absorbed rainfall prior to encapsulation, and this water
will emerge as contaminated seepage for a period of years as water drains down through the mineralised
waste. The Century team’s solution to this issue is to construct a gravity-fed AMD drainage network to direct
poor-quality seepage from waste rock dumps to the open pit for permanent storage. A preliminary trial of
this option involved designing and constructing a one-kilometre-long section of drainage network to capture

AMD seepage from the toe of a waste rock dump. The design prevents mixing of clean water with AMD and
eliminates the potential for wildlife or livestock to drink contaminated water. This trial has been successful;
therefore, construction of the AMD passive drainage network will be extended to include all sources of AMD
during the site’s active closure phase, and all contaminated water will be transported to the pit lake for
permanent storage. This solution avoids the “four Ps” that often prevent lease relinquishment at other
metaliferous mine sites: reliance on pumps, pipes, power and people in perpetuity. Construction of a passive
AMD system is contingent on suitable topography and a semi-arid climate, which ensure that the pit lake
remains a permanent sink. Modelling over a period of 700 years has verified that the pit will remain a sink as
evaporation exceeds rainfall by a factor of approximately 6:1. Sites considering adopting this approach should
model the relevant surface water, groundwater and geology to ensure that impacts on groundwater systems
off-site do not occur as a result of permanent storage of poor-quality water in the pit lake.
The main risk associated with an enhanced store-and-release cover is if it does not meet its design intention
of reducing infiltration into the potentially acid forming waste rock or tailings. Complex monitoring systems
have been installed to measure the performance of the encapsulation system for the waste rock dumps, and
instrumented trial sites have been established to assess the effectiveness of three cover options for the
tailings storage facility. Cover monitoring systems are installed throughout the enhanced store-and-release
cover profile to quantify their performance and the potential risk of failure of the cover. The monitoring data
is reviewed annually to evaluate the enhanced store-and-release cover performance. The cover monitoring
system has dual matric potential and volumetric water content sensor nests throughout the cover profile to
determine the net percolation through the system and into the underlying mineralised waste. Data provided
by the sensor nests are matric potential, soil temperature, volumetric water content and electrical
conductivity.
Information on the design and performance of mineralised waste encapsulation and management of AMD
seepage is discussed with the government regulator and other key stakeholders on an ongoing basis. It is
recommended that mining companies commence monitoring a site’s proposed mineralised waste
encapsulation system as soon as practicable within the life of a mine due to the long timeframe associated
with collecting adequate data to demonstrate to stakeholders that encapsulation objectives are being met.
Likewise, early trials to evaluate ecosystem restoration methodologies and postmining land use options
should also commence as soon as practicable following commencement of mining operations.
It is important to ensure that the government’s regulatory authority and other key stakeholders are informed
about the function and appearance of postmining landforms. Waste rock dumps at Century are constructed
to safely encapsulate reactive waste rock within constructed landforms indefinitely. Achieving physical and
chemical stability are the principle objectives. Whilst it is desirable to establish a diverse native ecosystem,
this function is secondary to the waste encapsulation function of these landforms. The principle function of
biota is transpiration of water stored in the cover system via establishment of tree species such as Eucalypts,
which have very high transpiration rates when water is in abundance. Vegetation must also have the capacity
to survive periods of drought. As such, the constructed landforms cannot and will not ever exactly replicate
surrounding natural landforms nor the complex ecosystems that have evolved within undisturbed land units
within the mining leases.
It is also important to discuss the performance of the waste encapsulation system with the regulatory
authority. Whilst minimising infiltration into the dump is a key encapsulation objective, it is likely that there
will be some seepage generated from the dumps for decades, if not longer. Discussions should focus on
ensuring that the site is managed in a way that avoids adverse environmental impacts. Taking an outcome-
based approach to the development of completion criteria can reduce the potential for dialogue to focus on
differences of opinion regarding technical aspects of the land rehabilitation process.
It is anticipated that Century will not be able to relinquish its lease for at least two decades after production
ceases, and a more conservative target of 30 years has been selected for the site. During this time, the site
will require management activities such as weed control, fire management, feral animal control, seed
collection and site access management. Personnel will also be required to conduct ongoing monitoring to

Case Studies
demonstrate compliance with rehabilitation success measures. The company has recommended that a
“caring for country” team be formed and trained well in advance of planned closure. Where practicable, the
team should comprise local people with traditional links to the land. The scope for work undertaken by
environmental contractors and consultants should include opportunities to engage local people to develop
capability and create employment opportunities wherever possible. A good example of this approach at
Century has been expanding seed collection activities to employ people from local Indigenous communities
under the supervision of an experienced and skilled seed collection contractor.
Companies seeking a successful closure outcome should commit to a detailed risk assessment covering all
aspects of closure, undertake regular reviews of the risk register and proposed mitigation measures with
content experts, engage with key stakeholders regarding postproduction environmental outcomes, invest in
closure-related research throughout the life of the project and develop rehabilitation completion criteria
early in the life of the operation.
Century Mine has developed a detailed searchable database that lists specific closure-related legal
obligations relating to environment, community and land tenure. The database lists the status of each
commitment; notes can be updated to track progress in fulfilling obligations. The database may be expanded
in the future to include health and safety and human resources legal obligations.
The Century project is the subject of a unique agreement with the four native title groups who have
traditional custodianship of land that has been used for development of the mine, concentrate pipeline and
port facilities. The Gulf Communities Agreement (GCA) outlines commitments made on behalf of the mine
operator, the state government and the four native title groups. The University of Queensland’s Centre for
Social Responsibility in Mining has undertaken five-year reviews to monitor community perceptions
regarding implementation of the GCA. Century also internally tracks progress in implementing GCA
commitments on an ongoing basis.
Closure of a large-scale mining operation occurs infrequently in any government jurisdiction, and it is likely
that the transition from operations to active and passive closure phases will be a relatively unknown process
for both the regulator and the mining company representatives. Closure of a large mine such as Century may
only occur once per decade in a mining region, by which time both key regulatory personnel and community
expectations are likely to have changed. It is also likely that a minerals-rich region will have many legacy sites
where closure objectives have not been met; this raises genuine concerns amongst regulatory authority
representatives and other key stakeholders that closure of a site will not comply with statutory obligations
or other commitments. Frequent and frank dialogue with key stakeholders is required to identify issues of
concern and develop mitigation strategies. Holding open and honest discussions with the regulatory
authority representatives and committing resources to closure projects will build trust and confidence with
stakeholders. Century conducts regular meetings with government representatives and other key
stakeholders to provide updates on projects and early advice on emerging issues and proposed mitigation
strategies.
Moving from operations to closure with the ultimate aim of lease relinquishment is a journey that requires
collaboration between the regulatory authority and other key stakeholders to establish and achieve a
common goal. On occasion, discussions regarding lease relinquishment have become protracted due to
differences of opinion on technical issues (e.g., the percentage of rainfall infiltration though a mineralised
waste capping layer). A focus on identifying and mitigating potential adverse environmental effects to comply
with licence conditions and agreed rehabilitation success measures is an effective method of working
towards agreement on closure outcomes.
Mining companies are advised to engage regularly with the regulator and other key stakeholders to reach
agreement on postmining land use(s), rehabilitation success measures, closure risks and appropriate
mitigation measures. Developing a comprehensive database to document and track close-out of legal and

other obligations (e.g., agreements with native title groups, port authorities, transport corridor regulatory
authorities, local government or other organisations) is also recommended.
Mining operations typically focus on three key performance measures: safety, volume and cost. An equally
important and often ignored fourth pillar of successful mining operations is managing liability. Managing
financial liability, environmental harm and damage to a company’s reputation are key considerations
associated with mining operations. Unless the closure liability implications of mining and processing
operations are identified and quantified when making operational decisions during the planning and
execution stage, an operation will certainly fail to realise the full value of the ore resource and may
inadvertently create unnecessary and avoidable long-term liabilities that erode the company’s value and
cause harm to the company’s reputation. A mine closure expert should be involved in key decision making
throughout the life of a mining asset to ensure that liability arising from operational decisions is evaluated
and mitigated where practicable. Furthermore, measures such as the area of available land rehabilitated on
an annual basis should be taken to track progress in managing a site’s liability; the management team’s
remuneration should be linked to this management of liability.
Operational personnel in planning roles need to be provided with ongoing advice and education on closure
risks, with particular emphasis on measures required to avoid or minimise AMD generation. Implications for
long-term liability should be carefully considered in all decisions regarding segregation of reactive (potentially
acid forming and non-acid forming) waste; preservation of topsoil resources; preservation of competent non-
acid forming rock reserves for use in closure earthworks; location of mineralised waste storage landforms;
diversion of clean water; capture and management of contaminated seepage; management of spontaneous
combustion; progressive rehabilitation; slope design; methodology for waste placement; and diversion of
water courses.
Mine planning must also include a regular review of the materials balance to ensure that waste rock dumps
can be constructed to final design as soon as practical; this reduces the period of time that dumps are exposed
to infiltration and permits early encapsulation to reduce the risk of AMD generation and spontaneous
combustion of mineralised waste.
External closure expertise should be sought to assist in planning for the short term, long term and life of the
asset to ensure that decision makers are fully informed about closure risks and avoid creating unnecessary
liabilities. Century uses advice from external consultants in a range of diverse disciplines including limnology,
modelling of surface water and groundwater, encapsulation of mineralised waste, monitoring and
interpretation of mineralised waste encapsulation system performance, fluvial geomorphology, terrestrial
and marine ecology, assessment of contaminated sites, decommissioning of infrastructure, hydrology and
geotechnical engineering. Ongoing education of key operations personnel is also required as turnover in
professional mining staff can be high. Management of all aspects of operations, including waste, water, air
and biodiversity should also be fully documented via management plans and procedures to ensure that
critical information is available to operations personnel.
Financial evaluation of the closure liability and management of assets are two key areas of transition
planning, yet these disciplines are often absent or under-represented at closure conferences, workshops and
industry forums. A better understanding of the complexity, processes and costs of closure at an executive
level would reduce the likelihood of unacceptable mine closure outcomes and assist in ensuring that closure
liability is fully considered by those managing mining operations during the life of a mining project.
Calculating the financial liability associated with closing a mine is a complex process, and closure liability is
often under estimated. The GARD Guide (Figure 6.1) demonstrates that costs increase and options decrease

Case Studies
over the life of a project. Estimates are often prepared to calculate a mine’s closure liability in the event that
the operation becomes insolvent or fails to meet rehabilitation obligations and the regulator is required to
undertake mine site rehabilitation. This may form the basis for a financial guarantee referred to in
Queensland legislation as financial assurance (FA). A second liability calculation is prepared to determine the
cost to a company to close a site and achieve lease relinquishment. This estimate is referred to as the financial
provision (FP). The FP estimate generally includes obligations that are absent from the FA estimate, including
annual costs such as rates, insurance, staff travel, transport to and from remote sites, camp management,
power, legal fees, lease payments, training, subscriptions, information technology, surveying, native title
agreement payments, earthmoving fleet mobilisation/demobilisation, etc. Some liability items may be
absent from both calculations (e.g., costs associated with workforce redundancy).
Century has developed a detailed closure schedule using first principle costs and contractor rates to estimate
the rehabilitation earthworks liability. Liability associated with items listed above has also been built into the
schedule. Where possible, unit rates have been based on costs obtained from closure projects executed at
Century or from technical studies. The liability requires ongoing review and revision as knowledge is acquired
from ongoing execution of closure works and technical studies. Technical studies and field trials required to
address knowledge gaps should be implemented well before closure to better refine closure options and
costs. Encapsulation of tailings and mineralised waste dumps is one area where liability can increase
significantly if trials dictate that a more complex cover is required or if unanticipated failure of the
encapsulation system necessitates rework. An accurate estimate of in situ encapsulation costs may prompt
a review of other options; for example, hydraulic recovery of tailings for placement in the final void may
present a lower cost and lower risk that encapsulation.
For a remote site such as Century, it is also necessary to consider whether the transition to a smaller
workforce when processing ceases will necessitate changes to the things such as the nature and frequency
of flights, the number of commuter bases that support FIFO charters, rosters, hours of work, night shift
requirements, etc. It is also necessary to consider which services will be supported at a corporate level or by
other company sites that operate in the vicinity of the site that is closing. It may be possible to reduce costs
by sharing some services or support or by performing tasks remotely. Contracting out some services
previously undertaken by onsite personnel may be another option when the scale of an operation reduces.
The estimate of a realistic timeframe to achieve lease relinquishment will also significantly influence closure
liability. There are few examples of large mining operations in Australia that have successfully surrendered
their mining leases in recent decades. Expectations from governments and other stakeholders continue to
change, and it is increasingly difficult to achieve relinquishment criteria. A site may require ongoing
maintenance, monitoring and remedial works for a period of 20 to 30 years prior to achieving lease
relinquishment. If the site is in a remote location, then costs escalate due to the requirement to maintain
onsite accommodation, stores, power, potable water supply, waste water management, communications,
emergency response capability, etc. Other costs such as rates, insurance, travel, freight and lease payments
may also have to be maintained over a protracted timeframe.
Century Mine’s infrastructure would cost in excess of $1.8 billion to replace. As such, it is important that
assets are preserved until a range of growth options have been fully assessed. Fixed infrastructure will be
placed in a state of operational readiness for three to four years pending the outcome of an evaluation of
growth options. Careful planning will ensure that assets and infrastructure do not deteriorate during this
period and that equipment and infrastructure required to support the site during the active closure phase
are fit for purpose and maintained in a safe state.
A change in the scale of the operations necessitates a change in use for facilities such as the accommodation
village, water supply, waste water treatment, power supply and transportation. At Century, a manager from
the asset maintenance and reliability team was seconded to integrate transition planning for assets and
infrastructure approximately 18 months before production was scheduled to cease. The involvement of a

senior engineer with extensive site knowledge enabled a thorough review of both operational readiness and
the asset-related risks, knowledge gaps, maintenance requirements, staffing levels, skills, competencies and
modifications required for active closure and safe operation of the site on a reduced scale. The transition
planning engineer worked closely with the closure team and representatives of departments across the
business to prepare the site for the transition from production to operational readiness. Active rehabilitation
of waste rock dumps will continue during this time, and both mobile and fixed assets will be required to
support the rehabilitation program. Asset-related transition planning and projects involve modifying or
replacing systems such as water, waste water, power and accommodation to cater for a workforce that may
be 10% the size of the operational workforce.
Whilst it is anticipated that the site infrastructure will be utilised in the future, contingency plans have been
made for decommissioning the site. Comprehensive studies were conducted by external consultants in 2011
and 2012 to provide cost estimates and methodology for removing all infrastructure at the mine site and the
port. This information is also required for calculating the site’s financial assurance (bond) and financial
provision estimates. No value was assigned to these calculations for sale of the fixed plant as industry
experience suggests that asset sales often fail to realise the anticipated financial return assigned to assets in
an asset register.
Disposal of a mobile plant should not be assigned value to offset closure costs. Century’s experience suggests
that a mobile fleet should be divested whilst it is still operating as parking mining equipment will result in
ongoing costs to preserve the equipment, whereas the value of the equipment inevitably decreases. Demand
for second-hand mining equipment fluctuates depending on commodity prices, and divesting mobile assets
may not be viable if the industry is in an economic downturn when the mining fleet becomes redundant.
Century decommissioned 19 redundant haul trucks in late 2014. The trucks were cut up for scrap metal, and
costs such as freight and crane hire were borne by the mining operation.
Century has a range of assets that are likely to be of value to other entities should growth project options
not materialise. These include the port facilities, concentrate pipeline, accommodation village, airport, roads
and groundwater bores. Consultation over the life of the project has raised expectations amongst
stakeholders that some assets will transfer to stakeholders upon cessation of mining operations.
Stakeholders have been informed that the existing infrastructure will be required to support the active and
passive closure phases of the operation. Statutory regulations apply to the site whilst it remains a mining
lease, and the mine site cannot be used for nonmining activities until lease relinquishment is achieved. Partial
lease relinquishment is not recommended unless a comprehensive risk assessment has determined that the
land will not be required in the future, that there is a sufficient buffer zone in place to manage potential
reverse sensitivity issues, and that the site has not and will not have an adverse effect on the land that is
identified for relinquishment.
The process of planning for and executing the transition of a large mining operation from the production
phase to the active and passive closure phases is as complex as opening a new mining operation. The project
outcome is more likely to be positive if a company dedicates resources to closure planning during the life of
the asset. Operators must develop a detailed project schedule, comprehensive long-term risk register,
closure strategy, plan of operations, annual business plan and supporting management plans to guide closure
project management and execution. Figure 3 shows the components of a mine closure planning system at
Century.
Project management phases that are typically applied to construction projects cannot easily be applied to
closure project management as a whole. When a closure project is broken into individual projects, it is likely
that some components may be at the concept, prefeasibility, feasibility, or project execution phases, and
some may be finalised well in advance of the proposed date for ceasing operations. Some closure projects,
such as the design of a tailings storage facility, must be completed prior to the development of a site. Other
aspects, such as topsoil harvesting, stream diversions, determining final land use options and selecting an

Case Studies
appropriate mineralised waste encapsulation strategy, must be undertaken prior to commencement of

mining operations.
Research and technical studies should be undertaken during the life of the project to address closure
knowledge gaps and refine the closure strategy. Century has used specialist consultancy services and
developed long-term relationships with consultants and universities to undertake closure-related research.
The arrangement has assisted in retaining site knowledge and building upon previous research.
The Century closure team recommends that mining companies that are planning to transition from
production to the active and passive closure phases employ a dedicated, site-based closure resource at least
five years before the planned cessation of production date. Additional personnel and resources should be
provided to the team as the scheduled closure date approaches.
Century’s closure earthworks projects commenced at a small scale and were managed by the small site
closure team. Available areas of disturbance such as old borrow pits, a disused airstrip, historic landfill sites,
the redundant construction camp, and other disturbance areas were rehabilitated by contractors under
direction from the closure team. More challenging projects were added to the program in the second and
third years of the mine closure and rehabilitation program. These projects included rehabilitating the historic
bulk sample tailings storage facility (TSF) and the bulk sample waste rock dump, which was constructed in
the mid-1990s prior to commencement of operations; constructing three complex TSF encapsulation trials;
constructing the passive AMD drainage system; and constructing permanent clean water diversion drains.
The rehabilitation programme was expanded further in 2014 with the formation of a dedicated bulk
earthworks rehabilitation crew comprising operators and equipment from the mining fleet. The majority of
this crew are Indigenous employees from the Gulf communities. This program runs in conjunction with the
civil earthworks program that uses a small-scale civil earthworks plant and civil contractor expertise. The civil
contractor works in partnership with a local Indigenous earthmoving contracting business.
The structure of the closure project team requires careful consideration. Century elected to appoint senior
team members from the mine technical services (MTS) team and the asset maintenance and reliability (A&R)
team to work collaboratively on closure planning and execution. This has been largely successful; however,
high staff turnover within the MTS team has caused some disruption to planning continuity and knowledge
retention. Personnel from the finance and communities teams also worked closely with the closure team to
assist with closure planning and execution.
Mine closure planning must be an integral component of the management of a mining operation if there are
to be realistic expectations that liabilities developed during mining operations can be managed and
addressed with the minimal cost and that the mining lease can ultimately be relinquished. Closure planning

must cover all aspects of a mine’s operation and be endorsed by senior management. The management of
all seven aspects of an operation have implications for the success of mine closure, and closure implications
associated with operational decisions must be considered and risks assessed. Failure to identify, evaluate and
mitigate closure-related risks will impact a company’s financial position and reputation.

R.C. Dickin AECOM, Canada

S.E. Dickin AECOM, Canada
L. Groat University of British Columbia, Canada
K. Jia AECOM, Canada
U. Mayer University of British Columbia, Canada
R. Mills Environment Group, AECOM, Canada
M. Sanborn AECOM, Canada
In the 1960s, mine tailings were produced from the processing of iron ore from a magnetite skarn deposit in
a near-shore intertidal marine area on the west coast of British Columbia. The mine and processing facility
were abandoned over 40 years ago. The tailings formed a beach that was used for recreational purposes for
many decades. Recent investigations of the tailings identified concentrations of total arsenic (As) (26 µg/g to
19,200 µg/g) and cobalt (Co) (50 µg/g to 1,820 µg/g) that exceed the applicable BC Contaminated Sites
Regulation (CSR) soil and sediment standards for parkland and marine aquatic life land use. Groundwater
from monitoring wells screened in the near-shore and intertidal tailings had concentrations of dissolved As
(125 µg/L to 1,860 µg/L) and dissolved Co (40 µg/L to 1,040 µg/L) that exceed applicable CSR standards for
the protection of marine aquatic life. However, water quality in monitoring wells, located more than 30 m
inland of the high tide level and screened in tailings, meet the applicable CSR groundwater quality standards.
A clear understanding of the tailings mineralogy and the geochemical reactions that are causing the elevated
As and Co concentrations in groundwater is required before remedial options can be developed and
evaluated.
Concentrations of As and Co in soil, intertidal sediments and groundwater were measured along a profile,
parallel to the groundwater flow path that extends from the uplands into the intertidal zone. Representative
tailings samples from various depths and subject to different oxidation-reduction environments were
collected, including the saturated and unsaturated zone in the uplands region, the beach, as well as the
shallow and deep intertidal zones. These samples were analysed for mineralogy, total CSR metals, and shake
flask extractable metals (using distilled water and seawater). Sequential extraction procedures were also
undertaken to identify metal concentrations associated with the following fractions: F1 exchangeable, F2
carbonates, F3 bound to iron (Fe) and manganese (Mn) oxides, and F4 bound to organic matter.
The study concluded that trace (<0.01) sulphide mineralisation (arsenopyrite and cobaltite) associated with
the skarn wall rock is the source of the As and Co. As and Co concentrations in the tailings are higher in the
shallower tailings near the shoreline, which suggests higher concentrations of the sulphide mineralisation
present in tailings deposited late in the mine life. Oxidation of these sulphide minerals in the shallow intertidal
and near-shore zones is the likely cause of elevated As and Co concentrations in groundwater and intertidal
porewater.
A conceptual site model was developed to assist in understanding the geochemical reactions governing the
release of As and Co, and to allow assessment of the geochemical effects that could occur after
implementation of various proposed remedial options. In the uplands freshwater zone, As and Co appear to
be sufficiently adsorbed and sequestered by secondary Fe and Mn oxides or other secondary minerals that
the groundwater quality standards are not exceeded. However, As and Co in the near-shore water table

Groundwater contamination due to historical magnetite mine tailings deposited in a near-shore, marine environment R.C. Dickin et al.
fluctuation zone are much higher, likely because the large daily water table fluctuations promote the
oxidation and release of As and Co from the sulphide minerals present in the tailings and/or inhibit the
attenuation potential for these contaminants.
During the 1960s, magnetite iron ore from an open pit mine located on the west coast of BC was milled
(magnetic separation) near a deep water port before being exported to Japan. An estimated 1.4 million m3
of sand waste (tailings) was deposited into the ocean on the adjacent property. Soon after mining ended in
the late 1960s, the site began to be used for recreational activities.
Recent environmental investigations have determined that soils and sediment, both composed of tailings,
exceed the applicable BC Contaminated Site Regulation (CSR) metal standards for residential/parkland land
use and for protection of marine aquatic life, primarily for arsenic (As) and cobalt (Co). Preliminary mineralogy
and geochemical sampling were conducted to identify the source of the As and Co but only found traces of
arsenopyrite (>0.1%), and did not clearly identify a Co source.
Groundwater samples from monitoring wells installed in the tailings more than 30 m inland of the current
shoreline met the applicable CSR standards. However, samples collected from monitoring wells installed near
the shoreline or in the intertidal zone exceeded applicable CSR standards for As and Co by factors of more
than ten times at some locations.
The tailings deposit consists of approximately 2.9 ha of the uplands zone above high tide, 3.1 ha of the
intertidal zone and 18.7 ha of the sub-tidal ocean floor. A thin layer of sub-tidal sediments with As
concentrations above the applicable sediment standard of 25 µg/g (but generally less than two times the
standards) extends about 1 km out into the ocean.
The investigations described in this paper were completed to answer the following questions prior to
completing the human health/ecological risk assessment:
What minerals in the tailings contribute As and Co concentrations to groundwater?
1. What geochemical reactions occur in the near-shore and intertidal tailings and cause As and Co to
be released from the tailings into groundwater?
2. How do subsurface environmental conditions (e.g., salinity, oxidation reduction potential [ORP]
and pH) affect the geochemical reactions that release or attenuate As and Co?
Field and laboratory data and interpretation were used to develop a geochemical conceptual model for the
tailings, to facilitate development and selection of appropriate remedial options, particularly options that
might cause changes in the oxidation-reduction conditions of the tailings.
Samples of magnetite ore and host wall rock were collected from the open pit mine walls and waste rock
dump for thin section analysis. Representative samples of the tailings were collected by hand from surficial
soils and sediments and with the use of a drill rig for deeper samples. The mineralogy of rock samples from
the mine (open pit wall and waste rock piles) and the tailings disposal site was studied using powder X-ray
diffraction, the Rietveld technique (six samples), optical petrography, and scanning electron microscopy with
energy-dispersive X-ray spectrometry (13 polished thin sections and 17 epoxy tailings mounts).

Case Studies
Soil and intertidal sediment samples were collected from the tailings disposal area at various depths using a
drill rig, or by hand, to delineate the lateral and vertical extent of tailings, and determine concentrations of
As, Co and other metals above applicable CSR standards. These delineation samples are shown as circles on
the geological cross section in Figure 1.
This paper focuses on a set of eight samples of soil and intertidal sediment (all tailings) collected for
geochemical analysis at various depths along a profile that is parallel to the direction of groundwater flow,
from the uplands to the bottom of the intertidal zone. These soil and sediment samples are shown as boxes
on Figure 1. Each of these samples was analysed for: (1) mineralogy; (2) total CSR metals in soil (CRC ICPMS
EPA 6020a); (3) sequential leachate tests (modified from Tessier et. al., 1979); and (4) shake flask leachate
tests (Price, 1998), employing both distilled water and seawater.
Groundwater monitoring wells were constructed of 51 mm diameter PVC wells installed using drill rigs at
various depths, both on land and in the intertidal zone. Stainless steel mini-piezometers were installed by
hand to depths of approximately 1 m in the intertidal sediments. Groundwater wells were developed using
inertial pumps and allowed to equilibrate for a period of two weeks prior to permeability testing and
sampling. Average groundwater water elevations and hydraulic gradients were determined using average
values of hourly water level measurements collected over five months using transducers and dataloggers in
select wells. Geodetic groundwater elevations were determined using surveyed well head elevations. Single
well rising head slug tests were conducted on a representative subset of more than 10 wells to determine
the mean hydraulic conductivity of the tailings and other geological strata.
Groundwater samples were collected on two or more occasions from upland and intertidal groundwater
monitoring wells using low-flow sampling techniques, employing peristaltic pumps and flow-through cells to
prevent contact with oxygen. Samples were collected after pH, conductivity, ORP and DO levels had
stabilised. To assess potential water quality changes over a 12-hour tidal cycle, hourly samples were collected
from ten wells along a representative cross section from the uplands to the intertidal zone. Groundwater
levels were monitored hourly with transducers over this same time period.
The PHREEQC (Parkhurst and Appelo, 1999, version 3.1.4 with database llnl.dat) hydrogeochemical model
was used to calculate mineral saturation indices to determine possible solubility controls for As, Co, Fe and
Mn in solution and, where possible, identify mineral phases governing the release of these contaminants.
The redox couple “O(-2)/O(0)” was used to control redox conditions. The measured ORP was not considered
in the simulation because most field-measured values were outside of the range of values that can be
considered reasonable. The model was able to provide information on mineral saturation indices, and model
outcomes were assessed by comparison with mineralogy data obtained from the XRD and SEM work.

Some basic understanding of intertidal groundwater flow and discharge in a marine intertidal zone is required
to understand the potential for tidal fluctuations to affect groundwater quality and aquatic geochemistry of
the intertidal zone over relatively short time periods.
Net average groundwater flow is from the uplands region to discharge zones in the intertidal area near
geodetic mean sea level, as determined from average hydraulic gradients calculated from hourly pressure
transducer measurements of groundwater and sea elevations over dry and wet season lunar cycles
(29.5 days). Instantaneous groundwater flow within and near the intertidal zone is strongly affected by tides
and highly dynamic. During rising tide periods, there is a short-term reversal of groundwater flow directions,
and seawater infiltrates into the upper part of the flooded intertidal zone.
Maji and Smith (2009) presented the results of numerical modelling of groundwater and contaminant
discharge in a theoretical intertidal zone that in many ways is similar to the subject site. Some of the
conclusions of Maji and Smith (2009) are provided below, and illustrate how temporal changes in
groundwater flow caused by tides can affect groundwater discharge quality in the intertidal zone over a tidal
cycle:
 “Volumetric discharge of groundwater in the intertidal zone dominates the discharge occurring
beyond the low-tide line.”

Case Studies
 “Within the intertidal zone, the recirculated saline component of the discharge is greater than the
freshwater component.”
 “The majority of the contaminant mass exits through the seepage face located above the
instantaneous tide line.”
 “The peak contaminant loading rate occurs around the time of the falling midtide."
 “At low tide, the contaminant loading rate is more than 2 orders of magnitude lower than the
peak rates, even though groundwater discharge rates are highest as low tide is approached.”
The mineralogy results showed no As, Co or bismuth minerals in the magnetite ore. However, common
arsenopyrite (FeAsS) and rare cobaltite (CoAsS) were found in skarn rock from the mine, and arsenopyrite
(to ~200 μm), cobaltite (to ~400 μm), and bismuthinite (Bi2S3, to ~30 μm) grains were found in tailings
samples. No skutterudite (Co,Ni,Fe)As3 was seen in any of the samples.
The tailings sample from the vadose zone contained magnetite and loellingite (FeAs2), but no apparent
arsenopyrite or cobaltite, which is indicative of complete oxidation. The tailings samples from the water table
mixing zone all contained magnetite, but pyrite, chalcopyrite, galena and sphalerite were also identified, as
were pyrrhotite, arsenopyrite (in three of four samples studied, to ~200 μm in size), and bismuthinite (to
~30 μm) grains. No cobaltite was seen in these samples. The tailings samples from the saturated zone contain
magnetite and pyrite, galena, arsenopyrite (in one of three samples, to ~200 μm in size), cobaltite (in two of
the samples, to ~400 μm), and bismuthinite were also identified.
Figure 1 presents a representative geologic cross-section through the tailings deposit from the uplands,
through the intertidal zone to the sub-tidal zone. The tailings are magnetic and visually distinguishable from
the underlying marine sediments, which contain more organic matter and shells. The tailings extend from
the surface to depths of 3 m to 5 m in the uplands, and are greater than 10 m thick in the intertidal zone. The
tailings are underlain by several metres of native marine sands and clay, which are, in turn, underlain by
granodioritic bedrock.
Figure 1 also shows the tailings, soil and sediment samples that were used for contaminant delineation as
dots that are shaded to illustrate the relative concentrations of As. Solid phase concentrations of As in the
tailings range from about 50 µg/g to over 5,000 µg/g. Arsenic has proven to be a reliable indicator of tailings
metal contamination at this site. The tailings samples with the highest As concentrations tend to occur in the
upper part of the tailings within 30 m of the shoreline, which suggests that a greater amount of sulphide-
containing skarn wall rock was processed during the later stages of the mine life.
A data set of 33 paired solid and aqueous phase arsenic concentrations was statistically analysed to assess
the relationship between total (solid phase) and dissolved (groundwater or porewater) arsenic
concentrations in intertidal sediments. Histograms and quantile-quantile plots were used to evaluate the
distribution of the solid phase and dissolved phase, and showed that the datasets are not drawn from a
normal distribution. Spearman’s rank correlation indicated a moderate correlation (rho) = 0.404), which is
statistically significant (p-value = 0.01962), although there was considerable spread around the linear model.
The analysis of this relationship is complicated by a number of confounding factors that are discussed below;
however, it is clear that elevated As and Co concentrations in groundwater almost always occur in close
proximity to soil or sediment with elevated As and Co concentrations. The tailings have a consistent grain
size and appearance, but the presence of trace sulphide minerals from the host rock appears to vary from
place to place due to variations in the mining, ore processing and tailings disposal activities which occurred
over a decade.

Table 1 presents concentrations of select metals (e.g., As, Co, Cu, Fe, Mn and Zn) in the eight tailings samples
analysed using several testing methods, including: CSR total solid phase metals, shake flask (distilled water
and seawater) and the sequential leaching. The locations of these eight geochemical tailings samples are
shown as black rectangles on Figure 1. Groundwater monitoring well locations and screens are also shown.
The total As concentrations of the eight soil and intertidal sediment samples included in the geochemical
analysis are shown as black rectangles on Figure 1. Normalised arsenic and cobalt sequential leach results
are shown at the top of the section. As and Co concentrations derived from the sequential leach tests were
normalised by dividing the aqueous concentrations (µg/L) by the solid phase concentrations (µg/g), which
are plotted on Figure 1 as a normalised mass fraction (%). The sequential leach tests are designed to identify
which of the following five mineral fractions the metals in the tailings are associated with:
 Fraction 1 – water available and exchangeable;
 Fraction 2 – bound to carbonates;
 Fraction 3 – Fe-Mn reducible;
 Fraction 4 – organic matter (oxidisable); or
 Fraction 5 – residual forms (not completed during the current study).
The shake flask leachable As, Co and Fe concentrations are shown above the cross section on Figure 2, which
also displays dissolved As, Co and Fe concentrations in groundwater. The shake flask tests are designed to
determine the amount of metals that leach from a solid phase after 24 hours of contact with distilled water.
For this study, the tests were also conducted using seawater to evaluate the effect of salinity on metal
leaching. The shake flask test results clearly demonstrate that Co is preferentially leached by the seawater,
with Co concentrations that were more than an order of magnitude higher than concentrations in distilled
water leachate. However, the shake flask As concentrations in seawater are slightly lower than the distilled
water concentrations in all but one of the eight samples tested.

Case Studies
Vadose zone 1 (BH14-77)

Sample depth (m BGS) = 0–0.5m
Unit Arsenic Cobalt Copper Iron Manganese Zinc
Total Metals in Soil µg/g 394 134 98 99,900 1,740 171
Distilled Water Shake Flask µg/l 321 <0.50 3 29 3 <5.0
Sea Water Shake Flask µg/l 222 25 6 9 49 9
Sequential leach Fraction 1 µg/l 109 25 8 1,080 93 <20
Sequential leach Fraction 2 µg/l 180 300 211 1,410 3,560 336
Sequential leach Fraction 3 µg/l 5,660 1,790 486 130,000 7,750 2,020
Sequential leach Fraction 4 µg/l 1,100 139 589 18,700 659 571
Upland water table fluctuation zone 2 (BH14-77)
Distilled Water Shake Flask µg/l 736 2 2 32 6 <5.0
Sequential leach Fraction 1 µg/l 142 38 2 350 124 <20
Sequential leach Fraction 4 µg/l 1,540 298 643 38,900 798 851
Total Metals in Soil µg/g 2,830 927 115 102,000 1,660 186
Distilled Water Shake Flask µg/l 1,400 5 2 41 4 <5.0
Sea Water Shake Flask µg/l 1,110 661 5 11 233 5
Sequential leach Fraction 2 µg/l 772 1,070 145 2,630 3,540 137
Sequential leach Fraction 3 µg/l 5,530 2,620 228 66,600 3,590 542
Sequential leach Fraction 4 µg/l 7,100 997 855 53,600 723 1,200
Sample depth (m BGS) = 2–3m
Sequential leach Fraction 3 µg/l 1,070 597 15 57,400 4,090 240
Sequential leach Fraction 4 µg/l 3,400 933 1,100 56,500 965 1,570
NOTES:
Sequential leach Fraction 1: 'Exchangeable', uses reagent: MgCl2
Sequential leach Fraction 2: 'Bound to Carbonates', uses reagent: NaOAc
Sequential leach Fraction 3: 'Bound to Iron and Manganese Oxides', uses reagents: NH4OH•HCl
Sequential leach Fraction 4: 'Bound to Organic Matter', uses reagents: HNO3, H2O2 and NH4OAc
BGS = Below Ground Surface

Shallow intertidal zone 3 (MW14-57)

Distilled Water Shake Flask µg/l 1,290 2 8 <10 3 <10
Sequential leach Fraction 2 µg/l 590 559 208 2,280 2,920 1,090
Upland saturated zone 4 (MW14-57)
Total Metals in Soil µg/g 51 35 22 27,900 636 65
Sequential leach Fraction 4 µg/l 336 98 157 16,300 281 671
Shallow intertidal zone 3 (MW14-59)
Distilled Water Shake Flask µg/l 116 5 25 15 58 <10
Intertidal saturated zone 5 (MW14-59)
Distilled Water Shake Flask µg/l 75 <0.50 0 75 10 <5.0
Sequential leach Fraction 4 µg/l 565 602 458 29,600 867 1,200
NOTES:
Sequential leach Fraction 1: 'Exchangeable', uses reagent: MgCl2
Sequential leach Fraction 2: 'Bound to Carbonates', uses reagent: NaOAc
Sequential leach Fraction 3: 'Bound to Iron and Manganese Oxides', uses reagents: NH4OH•HCl
Sequential leach Fraction 4: 'Bound to Organic Matter', uses reagents: HNO3, H2O2 and NH4OAc
BGS = Below Ground Surface

Case Studies

Concentrations of As, Co and Fe in groundwater are shown adjacent to monitoring well screens on Figure 2.
Shallow groundwater quality in the tailings near the shoreline exceeded the applicable CSR standards for
marine aquatic life for As (125 µg/L) and Co (40 µg/L), including two wells that exceeded the standards by a
factor of greater than 10. However, shallow groundwater in the upland tailings, located more than 30 m
inland from the high tide mark, did not exceed applicable CSR standards. Iron concentrations in groundwater
within the upland tailings typically range from 1,000 µg/L to more than 3,000 µg/L, but decrease dramatically
in the saline intertidal zone.
The elevated pH near the freshwater-seawater interface appears to cause precipitation of Fe and Mn oxides,
which are potential scavengers of As and Co. However, salinity in the intertidal area significantly increases
Co mobility, as shown by the shake flask test results and groundwater quality in the intertidal zone. Although
Co tends to be attenuated by ion exchange or surface complexation on Fe and Mn oxides, these attenuation
processes do not appear to be effective in the saline intertidal zone, likely because there are less available
sorption sites and/or due to competition with other dissolved ions.
The impact of salinity on As concentrations is small. The high concentration of As in the intertidal area
primarily results from the high concentrations in the tailings sediment, active sulphide oxidation promoted
by the oxidising conditions within the zone of tidally-induced water table fluctuation, and/or discharge of As-
contaminated groundwater.
Groundwater quality samples collected hourly from monitoring wells located along the groundwater flow
pathway during a tidal cycle indicated that tidal fluctuations can significantly affect water quality
concentrations within hours in shallow intertidal wells. The hourly groundwater sampling of wells screened
in the intertidal zone (depths of 0.7 m, 2.0 m and 4.0 m) over a 12-hour tidal cycle determined that, in a few
shallow wells (2 m or less depth) screened near mean sea level, the dissolved As, Co, Fe, and Mn
concentrations varied by factors of up to three times. Cobalt and Mn concentrations followed similar
concentrations trends. Iron concentrations were variable, but were opposite to trends in As concentrations.
However, water quality in inland monitoring wells, deeper wells and many of the intertidal wells did not
change significantly over a tidal cycle. Assessment of net impacts on groundwater quality in the shallow
intertidal area must consider samples collected at different times in the tidal cycle and at different well screen
elevations relative to mean sea level. Predicting groundwater quality and geochemical reactions is extremely
difficult in this constantly changing, biologically active zone.
The PHREEQC model calculated oxidation-reduction potentials that ranged from pe 11 to 14, which is
expected considering the DO measurements were used to control the redox state. Not surprisingly, under
these oxidising conditions arsenopyrite (FeAsS) and CoS are highly undersaturated, indicating a high potential
for oxidation of these minerals and release of As, Co and other metals into groundwater. Cobaltite is not
included in the PHREEQC database, but it is expected that cobaltite would be similarly undersaturated
because it is also a sulphide mineral. The fact that As- and Co-oxides (arsenolite (As4O6), As2O5, Co3(AsO4)2,
CoO) are also undersaturated indicates that the precipitation of these phases does not provide an effective
attenuation mechanism.
Co-bearing phases (spinel Co and Co2SiO4) are silicate minerals, which are unlikely to form under the low
temperature near surface conditions present at this site. The fact that no other Co-bearing phases were close
to saturation suggests that secondary mineral precipitation can be ruled out as an attenuation mechanism
for Co. It is most likely that Co will be attenuated by ion exchange or surface complexation reactions on Fe-
and Mn-oxyhydroxides or clays, or possibly via co-precipitation with these phases.
Although oxidation of sulphide minerals tends to generate acid and decrease pH, the circumneutral pH in site
groundwater suggests that carbonate minerals in the tailings play a role in maintaining pH within the neutral
range. The PHREEQC results show that calcite and dolomite are close to saturation (-0.02 to -0.56), indicating

Case Studies
that the presence of limestone and other carbonates from the skarn deposit are able to buﬀer the acidity
caused by oxidation of the sulphide minerals.
Iron and Mn oxyhydroxides are supersaturated at most locations, providing a strong driving force for
formation of these phases. These minerals can therefore be considered important sinks for As and Co via
surface complexation reactions, ion exchange, or possibly co-precipitation. The model results suggest that Fe
will tend to precipitate as Fe(OH)3, although goethite and hematite may also be present in the sediments.
Manganese oxyhydroxide like hausmannite, birnessite, MnO2 (gamma), and pyrolusite are oversaturated,
providing a driving force for their precipitation and formation of surface sites for metal attenuation.
The collected data and analysis were used to develop a conceptual geochemical model of the tailings to allow
evaluation of the potential geochemical effects of various remedial options on groundwater/porewater
quality. Figure 3 is a cross section that delineates the five geochemical tailings zones in the conceptual site
model which are based on: a) location and depth; b) degree of saturation and anticipated redox conditions;
c) freshwater or marine; and d) pH, As, Co and Fe concentrations in groundwater.
The unsaturated zone is only about 1 m thick at this site. Infiltrating precipitation has a slightly acidic pH due
to equilibrium with atmospheric CO2, but the pH in the tailings is likely buffered by the presence of limestone.
The vadose zone is in direct contact with the atmosphere and is therefore characterised by highly oxidising
conditions. Any sulphide minerals present in this zone are expected to oxidise or to have oxidised already.
The sulphide oxidation rate is either slow, or there is sufficient metal sequestration on soil surfaces, so that
metal concentrations in the upland groundwater do not exceed the applicable CSR standards for protection
of marine aquatic life. The mineralogy data suggests that oxidation in this region may be near completion, as
indicated by the absence of sulphide minerals. Water in the vadose zone will move downward under the
influence of gravity to the water table.
This zone ranges from 1 m to 2 m thick in inland areas to almost 4 m thick near the shoreline where the water
table is affected by tides. In this zone, measurable DO concentrations are present in groundwater and
conditions are oxidising, becoming more reducing with depth. The daily tidal cycles and changes in water
table create oxidising conditions over a greater thickness, which suggests the potential for more active
oxidation of sulphides near the shoreline. The pH of groundwater in this zone ranges from 6.2 to 7.3, and pH
increases towards the marine shoreline.
Solid phase concentrations of As in the inland tailings generally ranges between 100 µg/g and 500 µg/g, but
increases to >1,000 µg/g in the shallow tailings near the shoreline. It is hypothesised that a minimal amount
of the skarn wall rock and associated sulphide mineralisation was deposited in the tailings early in the mine
life, so the deeper tailings near the former shoreline tend to have lower As concentrations. Arsenic and Co
concentrations in the inland groundwater are higher than background levels, but do not exceed the
applicable CSR marine aquatic life standards for As (125 µg/L) or Co (40 µg/L).
Near the end of the mine life, it appears that more skarn wall rock was processed, resulting in a greater
amount of sulphide mineralisation and As in the tailings deposited in the shallow subsurface near the present
shoreline and intertidal zone. Groundwater in these areas has higher As and Co concentrations that exceed
the applicable CSR standards.

In this zone, solid phase As concentrations in the tailings often exceed 500 µg/g, and in some locations
exceeded well over 1,000 µg/g, which suggests that a significant amount of tailings rich in skarn containing
sulphide mineralisation was deposited in portions of this shallow zone. The shallow intertidal zone is
inundated twice daily by the high tides which mix seawater with discharging groundwater. The highly
dynamic nature of this zone causes at least temporarily oxidising conditions, implying that oxidation of
sulphide minerals is expected in this zone.
The pH of the groundwater/porewater of this zone reflects the higher pH of sea water, and is typically
between 7.5 and 8.0. As a result, dissolved Fe and Mn concentrations decrease dramatically as groundwater
moves from the freshwater environment into the marine environment (from 10,800 μg/L to 15 μg/L), which
suggests precipitation of iron and manganese oxides near the freshwater/saline water interface.
Concentrations of As and Co in groundwater/porewater within zone 3 are elevated above the applicable CSR
standards for marine aquatic life for As (125 µg/L) and Co (40 µg/L). The daily introduction of large quantities
of seawater with background As and Co concentrations of <5µg/L does not seem to significantly dilute As and
Co concentrations. The 24-hour shake flask test results suggest that seawater may effectively leach Co, or at
least inhibit Co attenuation, within a relatively short contact time. Moreover, the sequential leaching tests
suggest that attenuation mechanisms may be less important in the intertidal zone due to a lower abundance
of Fe and Mn bound in the oxide fraction, implying that there are less available surface sites for As and Co to
sorb on.
Zone 4 is the deep, freshwater, constantly saturated zone that is below zones 2 and 3. This zone is likely
characterised by reducing conditions due to the presence of sulphide minerals below the water table, and
little to no dissolved oxygen is present. The sediments are expected to become increasingly reducing with

Case Studies
depth. There is minimal potential for oxidation of sulphide minerals in this zone. The pH of groundwater in
this zone is slightly acidic, in the range of 6.0 to 6.5. Concentrations of solid phase As in the tailings in this
zone are typically between 100 µg/g and 500 µg/g, which suggests that less of the skarn sulphide
mineralisation is present in these deeper tailings deposits.
Dissolved Fe concentrations in groundwater are elevated in this zone due to the reducing conditions. Arsenic
and Co concentrations in groundwater are elevated above background levels, but do not exceed the
applicable CSR standards. The reducing conditions, which restrict sulphide mineral oxidation, and the lower
sulphide mineral content, result in reduced groundwater quality impacts in this zone.
Zone 5 is the deep intertidal zone below zone 3 and below the geodetic elevation of the lowest tide (-2 m).
This zone is constantly saturated, exhibits reducing conditions with little to no dissolved oxygen, and becomes
increasingly reduced with depth. There is minimal potential for oxidation of sulphide minerals in this zone,
despite the elevated As concentrations (200 µg/g to 700 µg/g), which suggests that sulphide minerals are
present in the most recent tailings deposits. The water is saline and the pH is high (7.5 to 8.0) and similar to
seawater. Dissolved Fe concentrations are low and typically less than 20 µg/L. As discussed above, most fresh
groundwater discharge is expected to occur at elevations above this zone.
Trace (<0.1%) amounts of the sulphide minerals arsenopyrite and cobaltite associated with the skarn wall
rock are the most likely sources of As and Co in the tailings. Oxidation of these sulphide minerals in both the
freshwater and marine environments is the most likely geochemical reaction causing release of As and Co to
the near-shore and intertidal groundwater. The highest concentrations of As and Co in groundwater occur in
portions of the tailings that have: 1) elevated total As and Co concentrations in soil/sediment; 2) oxidising
conditions associated with large, daily water table fluctuations due to tidal effects; and 3) for Co only, higher
salinity groundwater/porewater, which seems to inhibit Co attenuation by adsorption and ion exchange
processes. A geochemical conceptual model based on five gradational zones was developed to assist in
predicting the effects of remedial options on groundwater/porewater quality.
Maji, R., and Smith, L. (2009) Quantitative analysis of seabed mixing and intertidal zone discharge in coastal aquifers, Water Resources
Research, Vol. 45, W11401.
Parkhurst, D.L. and C.A.J. Appelo. 1999.
User's guide to PHREEQC (version 2). A computer program for speciation, batch-reaction, one-dimensional transport, and
inverse geochemical calculations: U.S. Geological Survey Water-Resources Investigations Report 99-4259, 312 p.
Price, W.A. and Errington, J.A. (1998) Guidelines for metal leaching and acid rock drainage at mine sites in British Columbia. Ministry
of energy and mines, British Columbia.
Tessier, A., Campbell P.G.C. and Bisson, M. (1979) Sequential extraction procedure for the speciation of particulate trace metals,
Analytical Chemistry, Vol. 51, No. 7, pp. 844–851.


J.C. Jacob BRGM, France

M. Save BRGM, France
The Bodennec mine site produces a low flow, near neutral mine drainage, which contains 8 mg/L of dissolved
ferric iron. The water quality objectives given by the French authorities are 3 mg/L of total ferrous iron in the
outlet in the Rudalveget River. The present water treatment installations already in use, comprised of three
settling ponds, are not effective enough. The average concentration of total iron in the outlet is around 4–
5 mg/L. A sodium hydroxide pilot plant has been built on site to assess the feasibility of a low maintenance
and low chemical consumption water treatment plant. This pilot plant is made of a small pump controlled by
a flow meter, which injects a small volume of concentrated sodium hydroxide solution into the drainage. The
system is supplied with electricity by a solar panel connected to a battery for night-time functioning. Sodium
hydroxide consumption is very low because, given the neutral pH, there is no need to decrease iron solubility
by significantly raising pH. At pH 6, a small increase in the pH is enough to greatly increase kinetic ferrous iron
oxidation. Chemically induced iron (oxy)hydroxide precipitates have a far better settling velocity than biogenic
precipitates. Precipitation and settling are thus far more efficient. Injecting a small amount of concentrated
sodium hydroxide solution allows the use of a small sodium hydroxide tank (10 m3) that is refilled only once a
year. This simple and robust system does not use a pH probe because pH in the drainage is quite steady. Low
maintenance and low chemical consumption make this system an economical solution that is of interest for
near neutral mine drainage on a site with limited surface available for a usual passive treatment plant. The
final plant is expected to be built in 2016.
Acid mine drainage is caused by the oxidation of sulphide minerals such as pyrite FeS2 and pyrrhotite
Fe(1-x)S (0 < × < 0.20) and also sphalerite ZnS or galena PbS exposed to atmospheric oxygen (Blowes et al.,
2003). This phenomenon liberates metals and associated trace elements in the water. If carbonate minerals
such as calcite or dolomite are present, their dissolution can lead to the neutralisation of acidity (Jurjovec et
al., 2002; Kirby and Cravotta, 2005). Research on mine drainage has generally focused on the treatment of
very acidic waters with high metals concentrations. Comparatively little attention has been paid to the
treatment of neutral mine drainage with low iron concentration (Warrender et al., 2011).Nevertheless, some
researchers have proposed promising innovative solutions for the issue of mine drainage on sites with low
available area. These solutions are either passive, such as vertical flow reactors (Sapsford and Williams,
2009); semipassive, like the Aquafixtm system (Watzlaf, 2004); or active coagulation-flocculation-
sedimentation systems (Tesser et al., 2011). Neutral mine drainage usually has low metals concentrations,
but it can still be considered an environmental issue. In fact, the Water Framework Directive of European
Union (2000/60/EC) stipulates that surface and groundwater must achieve “good status,” which means both
good chemical status and good ecological status. To that end, very low pollutant concentration limits have
been set. In the case of the Bodennec site, French authorities have given a water quality objective of
3 mg/L- of total ferrous iron in the outlet in the Rudalveget River. Other metals are present in the water (Zn,
Pb and Mn), but only in harmless concentrations.
The treatment of mine drainage can be either active or passive (Skousen et al., 1998). Active treatment
systems are commonly considered for active mines or closed mines with high flow and very acidic mine
drainages (Trumm, 2009), while near neutral mine drainage is usually well suited for classical passive

Treatment of mine drainage with high topographical constraints: case study of the Bodennec site (France) J.C. Jacob and M. Save
treatments such as oxygenating waterfalls, settling ponds and wetlands (Hedin et al., 1994). However, the
usual passive treatment solutions require large areas. Aerobic settling ponds and wetlands are often sized
on the basis of an iron removal rate of 10 g/m2/d. These removal rates decrease with decreasing iron
concentration, and removal rates as low as 1 g/m2/d at iron concentrations < 10 mg/L have been observed
(Watzlaf, 2004). Another issue is that limited topographic relief can prevent the construction of structures
that efficiently aerate the water. This decreases iron removal efficiency of these systems because, in many
aerobic passive treatment systems where the water is net alkaline, the removal of Fe appears to be limited
by dissolved oxygen concentrations (Hedin et al., 1994). Sometimes the required area and height difference
are not available, and these solutions cannot be implemented.
Active treatment of iron in mine water usually includes alkalisation of the water with chemicals, followed by
oxidation of ferrous iron, hydrolysis and precipitation from the solution as iron hydroxide and finally iron
precipitate removal through sedimentation (Coulton et al., 2003). The main goal of alkalisation is to decrease
metals hydroxide solubility and thus remove metals from water through precipitation. However, this is not
very relevant for iron in neutral mine drainage because, at pH 6, the solubility of ferric iron is already quite
low (< 1 mg/L). Iron oxidation is always implemented because ferric iron is insoluble at a much lower pH than
ferrous iron. Ferric iron reaches its minimum solubility at pH 8, while the optimum for ferrous iron
precipitation is pH 12 (Hustwit et al., 1992). Ferric iron hydrolysis and precipitation proceed more rapidly
than ferrous iron oxidation, which is the rate-limiting step (Kirby and Brady, 1998). Oxidation rates for Fe2+
strongly increase with pH. Therefore it is beneficial to raise pH prior to the oxidation step in treatment of acid
mine drainage.
In this paper we present, at pilot scale, an unusual use of sodium hydroxide: the treatment of a low flow near
neutral mine drainage with low metals concentration on a closed mine site. Furthermore, we demonstrate
that in a context of topographical constraints and tough regulatory objectives for water quality, sodium
hydroxide treatment can be a competitive solution for neutral mine drainage. Two key findings are presented
here: despite the already neutral pH of water (5.9), sodium hydroxide treatment has a significantly positive
impact on iron precipitation rates; and in contrast to other mines sites, in Bodennec, chemically induced iron
precipitates have a higher settling capacity compared to those that occur naturally.
The Bodennec mine site is located in Finistère, Britany, France. Bodennec is an old exploratory mine site for
lead and zinc. The Bodennec site is located in a narrow valley between a hill on the west and the low flow
Rudalveget River on the east. A near neutral mine drainage, which contains 8 mg/L of dissolved iron, springs
through the ancient decline, flows through the present water treatment system and then discharges into the
river. Water qualities of Bodennec mine drainage are shown Table . Total iron concentration in the water is
steady. The geochemical balance indicates that at the measured pH (5.9) the dissolved Fe must be ferrous
irons.

Case Studies
Parameters Average Min Max

Flow 20 m3/h 10 m3/h 40 m3/h
pH 5.9 5.7 6.1
Redox potential 154 mV 50 mV 250 mV
Dissolved oxygen 1.7 mg/L 0 mg/L 3 mg/L
Iron 8 mg/L 4 mg/L 12 mg/L
The present water treatment plan is composed of three settling ponds in a series, separated by two filtering
walls (gabions). The filtering walls have been clogged for years, and the water flows over them. The settling
ponds have a surface of 130 m², 242 m² and 70 m² respectively and a volume of 358 m3, 248 m3 and 117 m3
respectively. The height difference between the water spring and the outlet in the river is 1.6 m. A map of
the site is shown in Figure 1.
The average total iron concentration in the inflow is reduced by 51%, with an average iron concentration in
the outflow of 4.5 mg/L. The average iron removal rate of the system is 4.3 g/m2/d. Recently, the water
quality objective has been set at 3 mg/L of iron in the outlet in the Rudalveget River by French authorities.
During the two years of monitoring of the Bodennec mine site, this system did not reach this concentration
88% of the time (68 times over a campaign of 77 samplings). Poor performance of this system is probably
related to the insufficient residence time for oxidation and settling of iron and to relatively poor oxygenation
of the water. Due to topographic and legislative constraints, increasing the size of the treatment system was
not possible.

The pilot treatment plant was made simply of a lab micropump injecting concentrated sodium hydroxide
solution into the water from a 1 m3 tank. Mine water flow was measured by an ultrasonic flow meter on a V-
notch weir. The pump flow was adjusted to the mine drainage flow in order to get a steady NaOH to Fe
stoichiometric ratio of 2.0, assuming a steady iron concentration of 8 mg/L. The sodium hydroxide solution
concentration was 25% mass in order to reduce the tank volume and to avoid freezing of the solution during
cold weather in winter (to −20°C). The system was supplied with electricity by a solar panel connected to a
battery for night-time functioning. The pilot plant and NaOH injection were located on the mine water spring.
Adequate mixing of the NaOH solution was assured by injecting the solution in a small waterfall (0.5 m)
located right behind the V-notch weir. Following NaOH injection, water flowed normally through the three
settling ponds of the present water treatment plan. The results presented in this paper originate from one
pilot plant operation that lasted three months, from May to July 2014.
The performance of the system was monitored weekly. Water flow and the level of caustic soda in the storage
tank were measured. Water samples were taken from the inflow, the outflow of the first settling pond and
the discharge in the river on a weekly basis for 13 weeks post start-up. Field measurements were taken for
pH, temperature and redox potential. For each sampling point, two water samples were taken: the first was
immediately filtered and acidified for dissolved iron concentration, and the second was immediately acidified
but filtered for several days after for total iron concentration. Acidifications were realised by adding a drop
of HNO3 30%. Samples were filtered at 0.45 µm. Iron was measured by atomic absorption spectroscopy.
During the pilot operation, water underwent distinct visible changes in the first week. Originally bright red
and turbid, the water turned blue/green and clear. This can be attributed to a washout of previous biogenic
iron precipitates, which hardly settled, and to a good settling of the new chemically induced iron precipitates.
Over the course of the system operation, the influent flow rate varied from 20 m3/h to 15 m3/h. This is usual
for Bodennec water at the end of spring due to a decrease of rainfall. Water residence time in the three
ponds increased from 40 h to 60 h, which had a positive influence on the plant’s efficiency. Mean residence
time was 46 h. The total volume of water treated during the 13-week pilot program was 38,000 m3. The actual
quantity of NaOH added per mole of Fe was calculated and, during the pilot operation, the average NaOH/Fe
ratio was 2.9, which is higher than the expected ratio of 2.0. This is due to both an iron concentration lower
than expected (7 mg/L instead of 8 mg/L) and most importantly to a flaw in the setting of the pump control
system.

Case Studies
After the start of NaOH injections, field measurements of pH had indicated a slight alkalisation between the
inflow, the first pond and the discharge, as shown in Figure 2. Mean pH increase between inflow and
discharge is 0.4, which confirmed the NaOH overdosing previously observed. The fact that pH is higher in the
discharge than at the end of the first pond during most of the experience tends to indicate that mixing of
NaOH with water is insufficient. Field measurements did not show a redox potential behaviour change with
the use of NaOH. Redox potential increases slightly from 100 mV to 200 mV during water passage through
the ponds, which indicates water oxidation. From the first week of pilot operation and for the full duration
of the experience, total iron concentration in the discharge was below the regulatory limit of 3 mg/L. At the
end of the pilot operation, total iron concentration in the discharge was below 1 mg/L, which is an efficiency
higher than 90%.

During the first week of NaOH injection, the dissolved iron concentration in the discharge and in the first
pond decreased sharply, as shown Figure 3. In the first pond, it continued to decline over the course of the
pilot operation, and by end of the pilot operation, it was below 1 mg/L. In the discharge, dissolved iron was
below the detection limit after the third week of the pilot operation.
NaOH injections are effective to enhance oxidation, but not as effective as expected. During preliminary one-
litre on-site oxidation tests under the same conditions, ferrous iron precipitation was 100% in four hours.
During the pilot operation, in the first pond (average residence time 24 h), precipitation reached 50% after
one week of treatment, and it took six weeks to reach a precipitation of 75%. Several factors can influence
dissolved iron concentration: water residence time, temperature and NaOH mixing with water.
Dissolved iron removal and iron precipitates removal in the first pond in µg/L/h were calculated by taking
account of the water residence time in the system. Dissolved iron removal is assimilated to precipitation, and
iron precipitates removal is assimilated to settling. Figure 4 presents the evolution of iron precipitation and
iron settling during the pilot operation.

Case Studies
From the start of the NaOH injection, settling increased from 40 to 106 µg/L/h. After that, the settling rate
was more or less stable. The average settling rate in the first pond during the pilot operation was 105 µg/L-/h-.
During the first week of NaOH injection, precipitation increased sharply from 50 to 190 µg/L-/h-, then more
slowly from 190 to 260 µg/L/h for four weeks and finally, precipitation rate stayed approximately steady,
between 230 and 270 µg/L/h, during the last seven weeks of the pilot operation. The slow increase of
precipitation rate has a shorter duration and is less significant than the slow decrease of dissolved iron
concentration. It means that, although the increase of residence time does have an effect, it cannot be the
only cause of the slow increase of dissolved iron concentration observed. The increase of pH between the
first pond and the discharge tends to indicate that NaOH mixing with water is not yet completed after the
first pond. In these conditions, water in this pond is probably not well replaced, and the slow increase of iron
oxidation observed might be related to a slow washout of the first pond’s non-oxidised water. However, the
rise in temperature of 2°C, observed during this part of the pilot operation, may also explain it. Several
authors (Vollrath et al., 2013; Chen and Jiang, 2012) have reported a positive correlation between ferrous
iron oxidation rates and temperature through biotic and abiotic processes.
An increase of precipitated iron concentrations has been observed each summer since the beginning of close
monitoring of Bodennec drainage at the end of 2011, as shown in Figure 5. This has been associated with an
intense bacterial development in the form of floating red flakes and centimetric red biofilms on pond edges.
A microscopic picture of these bacteria is shown in Figure 6. Given the characteristics of Bodennec water and
their sheath form and twisted stalk form, these bacteria are probably related to the Leptothrix and Gallionella
genus. Unfortunately, this intense and naturally occurring iron precipitation is not associated with a better
than usual total iron removal because these iron precipitates hardly settle in ponds when they are in their
free form and do not settle at all when they are associated with floating flakes.

Figure 7 shows the iron settling rate plotted on the Y-axis and precipitated iron concentration on the X-axis.
Values have been divided into three sets, according to the period of monitoring and the presence of NaOH
injections.

Case Studies
Without NaOH injection, during the period of cold temperature and high flow, natural precipitation is low
and so are settling rates, while in summer the quantity of precipitated iron available for settling is very high,
but actual settling is still low. When NaOH is injected into the system, chemically induced iron precipitates
settle faster than those that occur naturally. It has been reported that iron precipitates produced by NaOH
treatment have a lower settleability than those produced in passive treatment (Dempsey and Jeon, 2001).
This apparent contradiction can probably be explained by the effect of the very low iron content of the
Bodennec mine drainage and the effect of biogenic iron precipitate. Given the low solid iron content of the
water, the opportunities of collision are probably not sufficient to induce coagulation of the particles, and
smaller particles have lower settling rates. Furthermore, iron oxide sheaths produced by iron oxidising
bacteria such has those of the genus Leptothrix are hollow and are has small as 1 µm in their smallest
dimension (Hashimoto et al., 2007); this does not increase settling either.
NaOH injections were effective to enhance oxidation and produce iron precipitates, which settle faster than
biogenic precipitates. Nonetheless, the effect of NaOH injections on ferrous iron precipitation was not as
efficient as expected. The possibility that the NaOH treatment is less efficient than expected could raise
concerns about the capacity of this system to reach the water quality objective in winter, when the flow can
be twice as high and the temperature can be a few degrees colder. As an example, at a water flow of 35 m3/h-,
a settling rate in the whole system of 200 µg/L/h is required to reach the 3 mg/L iron concentration objective.
However, a previous similar pilot test in November showed good efficiency of the treatment with settling
rates as high as 288 µg/L/h (230 µg/L/h on average). Data from this pilot were not shown because of a
premature shutdown due to a battery failure after one month of operation. Given the low quantities of iron
oxide retained in the pond and the usually very low solid content of sludge from mine drainage (Dempsey
and Jeon, 2001), removing the iron rich sludge is expected to happen every 15 years. Precipitated iron oxides
will be removed by a slurry pump, and a mobile filter press unit will be used to reduce sludge volume and
return clear water to the ponds.
In this paper, we describe a sodium hydroxide pilot operation, which was set up to treat a low flow, near
neutral mine drainage containing 8 mg/L of dissolved ferric iron, while the water quality objective given by

the French authorities was 3 mg/L of iron in the outlet. Our results show that low volume injections of NaOH
are effective to greatly increase ferrous iron oxidation and induce the formation of iron precipitates with a
better settleability than those naturally occurring. The pilot was able to reach the regulatory objective of
3 mg/L during the totality of the pilot program while treating 100% of the mine drainage flow. This simple
and robust system does not use any probe, requires little maintenance and has low chemical consumption,
which makes it an economical solution of interest for near neutral mine drainage on sites with limited surface
available for a usual passive treatment plant. The final plant is expected to be built in 2016.
The authors would like to thank the past and present project team members: Mickael Beaulieu, Florian Koch,
Jean-Pierre Jegou, Dominique Breeze, Jean Feraud, Gael Bellenfant, Bernard Lamouille and François
Blanchard. Financial support for these studies was provided by BRGM SA within the framework of the waiving
of its mining rights.
Blowes, D.W., Ptacek, C.J., Jambor, J.L. and Weisener, C.G. (2003) The geochemistry of acid mine drainage, Treatise on Geochemistry,
Vol. 9, pp. 149–204.
Chen, C. and Jiang, W.-T. (2012) Influence of waterfall aeration and seasonal temperature variation on the iron and arsenic
attenuation rates in an acid mine drainage system, Applied Geochemistry, Vol. 27(10), pp. 1966–1978.
Coulton, R. and Bullen, C. (2003) The design and optimisation of active mine water treatment plants, Land Contamination &
Reclamation, Vol. 11(2), pp. 273–279.
Dempsey, B.A. and Jeon, B.-H. (2001) Characteristics of sludge produced from passive treatment of mine drainage, Geochemistry:
Exploration, Environment, Analysis, Vol. 1(1), pp. 89–94.
Hashimoto, H., Yokoyama, S., Asaoka, H., Kusano, Y., Ikeda, Y., Seno, M., … Murakami, R. (2007) Characteristics of hollow microtubes
consisting of amorphous iron oxide nanoparticles produced by iron oxidizing bacteria, Leptothrix Ochracea, Journal of
Magnetism and Magnetic Materials, in Proceedings of the 17th International Conference on Magnetism, The International
Conference on Magnetism, 20–25 August 2006, H. Harima, H. Kawamura, Y. Kitaoka, H. Kohno, K. Miyake, Y. Suzuki, H.
Sakakima and G.-q. Zheng (eds), Vol. 310(2, Part 3), pp. 2405–2407.
Hedin, R.S., Nairn, R.W. and Kleinmann, R.L.P. (1994) Passive treatment of coal mine drainage, USBM IC 9389, US Dept of Interior,
Washington DE, p 35.
Hustwit, C.C., Ackman, T.E. and Erickson, P.E. (1992) The role of oxygen transfer in acid mine drainage (AMD) treatment, Water
Environment Research, Vol. 64(6), pp. 817–823.
Jurjovec, J., Ptacek, C.J. and Blowes, D.W. (2002) Acid neutralization mechanisms and metal release in mine tailings: a laboratory
column experiment, Geochimica et Cosmochimica Acta, Vol. 66(9), pp. 1511–1523.
Kirby, C.S. and Brady, J.A.E. (1998) Field Determination of Fe2+ Oxidation Rates in Acid Mine Drainage Using a Continuously-Stirred
Tank Reactor, Applied Geochemistry 13, no. 4, pp. 509–520.
Kirby, C.S. and Cravotta, C.A. (2005) Net alkalinity and net acidity 2: practical considerations, Applied Geochemistry Vol. 20(10),
pp. 1941–1964.
Sapsford, D.J. and Williams, K.P. (2009) Sizing criteria for a low footprint passive mine water treatment system, Water Research,
Vol. 43(2), pp. 423–432.
Skousen, J., Rose, A.W., Geidel, G., Foreman, J., Evans, R. and Hellier, W. (1998) Handbook of technologies for avoidance and
remediation of acid mine drainage, National Mine Land Reclamation Center, Morgantown WV, 131 pp.
Tesser, E., D’Amato, E., Gori, R. and Lubello, C. (2011) Conventional and innovative processes for neutral mine drainage treatment
and reuse of residuals, in Proceedings, 11th International Mine Water Association (IMWA) Congress, Mine Water – Managing
the Challenges, T.R. Rüde, A. Freund and C. Wolkersdorfer (eds), Aachen, Germany, pp. 485-490.
Trumm, D. (2009) Selection of active and passive treatment systems for AMD – flow charts for New Zealand conditions, in Proceedings
Securing the Future and 8th ICARD, 23-26 June 2009, Skelleftea, Sweden, pp. 908-917.
Vollrath, S., Behrends, T., Koch, C.B. and Van Cappellen, P. (2013) Effects of temperature on rates and mineral products of microbial
Fe(II) oxidation by Leptothrix Cholodnii at microaerobic conditions, Geochimica et Cosmochimica Acta, Vol. 108 (May),
pp. 107–124.
Warrender, R., Pearce, N.J.G., Perkins, W.T., Kay, M.F., Brown, A.R., Sapsford, D.J. … Dey, M. (2011) Field trials of low-cost reactive
media for the passive treatment of circum-neutral metal mine drainage in Mid-Wales, UK, Mine Water and the Environment,
Vol. 30(2) pp. 82–89.
Watzlaf, G.R. (2004) Treatment of high-flow, low-iron mine drainage with a semi-passive system, in Proceedings National Meeting of
the American Society of Mining and Reclamation and the 25th West Virginia Surface Mine Drainage Task Force Symposium,
18–24April 2004, ASMR, Lexington, KY , pp. 1232-1268.

Mine Closure 2015 – A.B. Fourie, M. Tibbett, K.G. Mercer and D. van Zyl (eds)
P. Kos Norwest Corporation, USA

M.J. Hart Nu-West Industries Inc., USA
J.B. Williams JBW Consulting, USA
The historic Georgetown Canyon industrial/fertiliser facility was built on fill material in Georgetown Canyon,
near the city of Georgetown in southeastern Idaho. Phosphate ore from nearby mines was processed at the
facility from 1957 to 1964. Initial closure activities, in 2001 and 2009–2010, resulted in removal of buildings,
tanks, structures and other features, but water in Georgetown Creek continued to flow under the site through
a 700 m (2,300 ft.) long corrugated metal pipe (CMP) culvert. Final closure at the site required
decommissioning the culvert and replacing it with a functioning stream channel, among other activities. Goals
for the closure project included constructing a stable channel for final restoration that, established fish
passage and habitat, avoided structural foundations and areas with known contamination, and minimised
construction volumes. This case study demonstrates successful reclamation in a mountainous environment
using universal engineering practices and fish habitat guidelines applicable to species present in the western
United States and Canada.
The channel design provides fish passage during low-flow periods and maintains overbank stability during
flood flows. The channel alignment includes meanders to avoid foundations and areas of contamination and
to reduce flow velocities by increasing the channel length and reducing channel slopes. The channel profile
maintains stable flow velocities and uses a series of rock weirs and plunge pools to provide gradient control
and fish habitat. Random boulders provide additional fish habitat. The channel substrate includes sand and
gravel for fish spawning purposes. Floodplain vegetation includes native grasses, shrubs, and trees that
provide habitat and erosion protection.
The channel was constructed in the summer and fall of 2010 and commissioned in October 2010. Pre-
construction activities included estimating construction cost using a risk-based approach to quantify
uncertainties. Construction challenges included managing site water, generating appropriately graded
materials from local sources, and intercepting unmapped facilities. Georgetown Canyon is widely used by
hunters and recreationalists, and Nu-West constructed campsites along the channel to benefit the local
community and protect reclaimed areas. Immediately following commissioning, the stream flow did not reach
the channel outlet due to low-flow conditions in the stream. The alluvial aquifer was depleted and needed to
be recharged. Water level monitoring demonstrated that the water level in the alluvial aquifer later increased,
and the stream resumed continuous flow during spring runoff in 2011. In the four years since commissioning,
the channel has remained stable, and reclamation cover on the channel slopes has met expectations. Activities
have been limited to minor maintenance and routine water level and water quality monitoring. The lessons
learned during design, construction, and monitoring activities at the Georgetown Canyon facility can be used
to aid future reclamation projects in their design and execution phases.
The historic Georgetown Canyon industrial/fertiliser facility was built on fill material in Georgetown Canyon,
near the city of Georgetown in southeastern Idaho. Georgetown Creek was routed under the site using a
700 m (2,300 ft.) long corrugated metal pipe (CMP) culvert. Phosphate ore from nearby mines was processed
at the facility from 1957 to 1964. Initial closure activities, in 2001 and 2009–2010, resulted in removal of

Final closure at the Georgetown Canyon phosphate ore processing facility – case study P. Kos, M.J. Hart and J.B. Williams
buildings, tanks, structures, and other features, but water in Georgetown Creek continued to flow under the
site. Final closure at the site required decommissioning the culvert and replacing it with an open stream
channel, among other activities. This case study reviews the closure process from design and permitting to
construction and post-construction maintenance and monitoring.
The Georgetown Canyon facility is located in Bear Lake County, in southeast Idaho. The ore processing facility
is located approximately 12 km (7.5 mi.) northeast of Georgetown, along Georgetown Creek in Georgetown
Canyon, within Caribou National Forest (Figure 1). The site elevation is approximately 2,120 m (6,950 ft.). The
surrounding mountain range is comprised of rugged terrain, with peaks reaching over 3,000 m (9,800 ft.)
elevation and vertical reliefs of nearly 1,000 m (3,300 ft.). There are several seeps and springs in the project
area. Multiple side drainages enter the facility from both sides of the canyon, with Phosphoria Gulch being
the largest. A United States Forest Service (USFS) road runs through the site, and the road receives a large
amount of traffic, particularly from recreationists, but also from hunters and logging trucks. Figure 2 shows
the site features.

Case Studies

Phosphate ore from nearby mines was processed at the Georgetown Canyon facility from 1957 to 1964. Site
features included an electric furnace and kiln, phosphoric acid plant, and fertiliser processing plant. The
facility was constructed on fill in the canyon bottom, and Georgetown Creek was routed under the facility
using a 700 m (2,300 ft.) long CMP culvert. The site was idle from 1964 to 1996, when initial clean-up activities
occurred. Underground storage tanks and petroleum-contaminated soils were removed in 1996 and 1997. In
2001, the site buildings, tanks, and structures were removed, but the foundations were left intact. In 2009,
contaminated areas were capped, remaining site features were graded, and remaining areas were topsoiled
and seeded. Also, drainage features were constructed to dewater the site prior to channel construction and
to convey side drainages to the CMP culvert. Phosphoria Gulch flowed onto the site and into a sedimentation
pond. The sediment pond outlet was a riser pipe that discharged to the CMP culvert. Georgetown Creek
continued to flow through the culvert.
Final reclamation of the Georgetown Canyon facility focused on removing all flow from and abandoning the
CMP culvert by constructing a stream channel through the site and connecting side drainages to the restored
stream channel. The project goals required designing a stable stream channel that included a riparian corridor
for wildlife habitat and in-stream features for fish habitat and passage. Pre-construction activities included
cost estimating, permitting, and advancing conceptual-level drawings to construction-level drawings. The
channel was constructed with few setbacks and no incidents, and post-construction activities include routine
monitoring and limited maintenance.
The CMP bypass channel design is the result of multiple meetings with various agencies including Idaho
Department of Environmental Quality (IDEQ), United States Environmental Protection Agency (EPA), USFS,
Idaho Department of Water Resources (IDWR), and United States Army Corps of Engineers (USACE). Each of
these agencies was involved to one degree or another. The stakeholders agreed that a permanent solution
to flow through the facility required abandoning the CMP culvert and constructing an open channel with
sufficient capacity to convey at least the 100-year flow of 1.8 m3/s (65 cubic feet per second (cfs)) through
the site without causing any flooding or erosion, particularly in contaminated areas (Berenbrock, 2002). Base
flows needed to have sufficient flow depth and limited velocities for fish passage.
The typical channel geometry includes a triangular channel for low flows and a vegetated bench for flood
flows. The low-flow channel was armoured with D50 = 203 mm (8 in.) riprap to provide erosion protection.
Riprap provides channel stability, but it does not provide fish habitat, and water can flow wholly through the
void spaces during low-flow periods. Consequently, the riprap was covered with a sandy gravel to provide a
suitable substrate for fish passage and habitat. This finer material also fills void spaces to limit interspatial
flows. This geometry maintains at least a 228 mm (9 in.) flow depth during annual low-flow periods; this flow
depth is required for passage of cutthroat trout (Bates, 1999). Fall low-flow rates have been measured at
0.05 m3/s (1.8 cfs) (GET, 2010). A 1 m (3 ft.) wide vegetated bench on each side of the low-flow channel
provides additional flow capacity for flood flows above the peak annual flow of 0.85 m3/s (30 cfs)
(Berenbrock, 2002). All slopes are 2H:1V, as shown in Figure 3.

Case Studies
The channel inlet and outlet locations and elevations had to match the CMP culvert inlet and outlet locations
and elevations, but the channel alignment and the profile between these locations were designed to meet
the project goals of minimising the channel gradient, minimising construction volumes, avoiding foundations,
and avoiding contaminated areas. The channel inlet and outlet locations and the channel alignment are
shown on Figure 2. The pre-construction fill depth at the culvert inlet was approximately 3 m (10 ft.), and the
pre-construction fill depth at the culvert outlet was approximately 10 m (30 ft.). The majority of the CMP
culvert was left in place and abandoned in order to allow construction to occur during the spring runoff peak
flows without diverting these higher flows. The channel alignment meanders through the site to maximise
the flow path and to avoid remaining site features such as foundations, which would be difficult and
expensive to remove, and capped contamination areas, particularly the slurry pit, where construction would
be hazardous or impossible. The radius of all channel curves was kept above 61 m (200 ft.) to maintain a ratio
of curve radius to channel width of greater than 10. This ratio minimises localised increases in velocity in
attempts to prevent lateral channel migration. The channel alignment had to avoid the USFS road easement,
site access roads, and the remaining sedimentation pond. Maintaining the USFS right-of-way and providing
a buffer around contaminated areas resulted in the segment past the slurry pit controlling the channel invert
elevation and the resulting excavation dimensions. A channel slope of 1.4% was needed to meet the design
criteria for flow depths, velocities, invert elevations, and necessary buffers, but this gradient was not
sufficient to meet the outlet elevation; a steeper channel segment was required once the channel was past
the slurry pit. Increasing the channel slope would not maintain the project goals for fish passage and habitat,
and a series of 17 rock weirs was installed to satisfy all the design criteria. The channel profile is shown in
Figure 4. Each rock weir consists of a 0.3 m (1 ft.) drop followed by a plunge pool for energy dissipation. The
rock weirs mimic a natural step-pool system (Rosgen, 2006). The drop height was limited to 0.3 m (1 ft.) to
allow fish passage, and the plunge pool provides a resting area for fish. As in the channel design, riprap in the
plunge pools is covered with a sandy gravel to provide a natural streambed and to limit interstitial flow. The
weirs are spaced approximately 30 m (100 ft.) apart, and the channel between the weirs slopes at 1.4%.

The design team recognised that there was a potential for failure caused by future erosion of the stream
channel and sequential failure of the rock weirs. Consequently, rock sills were embedded in the channel
substrate to limit erosion; if future failures occur, the erosion would be confined between the sills.
Vegetation on the bench provides long-term erosion protection, but supplemental erosion protection was
also needed for the period after construction and before vegetation was established. Type 3B erosion control
blankets extend across the bench and up the slope at least 0.6 m (2 ft.) to maintain at least 0.3 m (1 ft.) of
freeboard above the 100-year flood elevation. Type 2D erosion control blankets extended to the top of the
slope to provide erosion protection from overland flow. The blanket locations are shown in Figure 3. Type 3B
erosion control blankets provide a protection period of at least 24 months and shear strength resistance of
96 Pa, and Type 2D erosion control blankets provide a protection period of at least 12 months and shear
strength resistance of 84 Pa (FHWA, 2003). The blankets also allow vegetation to spread and grow. Flood
plain vegetation consists of woody species that protect stream banks from erosion, including willow,
dogwood, and alder trees and snowberry bushes (Beddoes et al., 2007). The cut slopes were hydroseeded
with a mixture of native grass seeds.
The construction planning phase included the typical steps of permitting, cost estimating, preparation of
construction drawings and specifications, and selection of a construction contractor. These steps also
revealed several site details that needed to be resolved before construction commenced. While the
subsurface conditions at the site had been assessed and were well understood, the conditions along the
channel alignment were not known to the level of certainty needed for construction and worker safety. In
particular, water levels in the construction areas, the presence of weak layers, and the presence of
contamination had to be assessed. Also, the mapped alignment of the existing CMP culvert was suspected to
be incorrect because the field conditions did not match the available maps. The culvert location had been
surveyed using ground penetrating radar, but the results were suspected to be inaccurate, possibly due to
the thickness of the cover (up to 10 m) and the high density of the cover material (mostly crushed slag).
Several test pits were dug to analyse the subsurface conditions and to verify the culvert location at
intersections with the channel. The field verification of culvert locations allowed construction of the designed
channel to continue with minor revisions. The channel cut slope intersected the culvert, and additional
segments of the CMP culvert would have to be excavated. The original design, with 2H:1V slopes, required
backfilling the void from the removed CMP culvert section, and the design team decided that a 3H:1V slope
was preferable for that section of channel.
Construction costs were primarily estimated using the current RSMeans handbook for unit prices, with
quantities calculated from the 3D design model (RSMeans, 2009). The design team evaluated the database
of subsurface conditions, including historic maps of site utilities, and decided to include line item costs for
environmental remediation to address the potential of encountering unmapped features or unknown

Case Studies
conditions, which was assumed to be high. To address these unknown elements, the team assumed an
underground storage tank (UST) would be encountered and included costs for sampling, air monitoring
equipment, disposal of impacted soils, and construction standby time. These costs were estimated based on
the size of the facility and the contaminants found in other areas of the site. Once the cost estimate was
completed, it was adjusted for unknown parameters using a risk-based approach. The data quality of each
line item was evaluated and a risk factor was applied. The factors ranged from 1.1 for items with a high
degree of certainty to 2.0 for items with unknown elements. The risk-based cost estimating approach and
inclusion of costs for unknown utilities increased the cost estimate by approximately 25%. The team felt this
increase amount was reasonable considering the number of unknown elements and the high probability that
there would be change orders during construction.
The majority of channel construction occurred as planned and followed a typical construction sequence. Prior
to construction, a site storm water pollution prevention plan (SWPPP) was developed and best management
practices (BMPs) were installed. Channel excavation started in locations away from the CMP culvert so that
higher spring flows could recede before it was necessary to divert the stream. Air quality was monitored
during all excavation activities to determine if any contaminants were encountered. Once all the channel
segments away from the culvert were constructed, it was necessary to divert Georgetown Creek around the
site, construct the remaining channel segments, and abandon the CMP culvert.
Several field fit changes were expected due to remaining unknown conditions. The design team understood
that many of the unknown conditions could be further evaluated, but the costs associated with additional
studies exceeded the risks. Also, the studies would cause undesirable construction delays.
Diversion channels constructed the previous year prevented water from ponding onsite and reduced shallow
groundwater levels locally so that the excavations could occur in dry conditions. However, several seeps were
encountered during excavation, and these had to be pumped out of the excavation area in order for
construction to continue. Also, water continued to flow from the CMP culvert outlet after Georgetown Creek
had been diverted at the inlet, suggesting that there were additional unidentified water sources such as
diversion pipes or elevated groundwater. Sections of the CMP culvert had been removed or unearthed at this
time, and multiple inlet pipes were observed. These pipes were abandoned if dry or routed to the existing
channel if flowing. The CMP culvert outlets from each remaining segment continued to flow even after all
identifiable sources were diverted from the pipe. The culvert pipe was inspected from the inlet and outlet
locations, but the pipe was never entered for safety reasons. These inspections showed the pipe was
deteriorated and had multiple holes; it was assumed that shallow groundwater flows could seep into the
pipe through the holes, causing some of the flow. Surface water quality monitoring suggests there is no
difference in water quality upstream and downstream from the site; therefore, seepage into the pipe did not
cause a water quality concern (GET, 2010). The inspections also confirmed the need to abandon the pipe. It
was also assumed that additional inlets existed, and these could not be diverted without removing the entire
pipe. The design and construction teams decided to reduce the culvert at each outlet from the 1.5 m (60 in.)
CMP to 305 mm (12 in.) high density polyethylene (HDPE) pipe, which would discharge the remaining waters
directly to the new stream channel. The constructed channel profile roughly followed the CMP culvert profile,
and the outlet pipes could discharge to the vegetated bench with riprap armouring at the outlet. Each of
these design revisions was simple to facilitate due to open communication between the construction and
design teams, and they resulted in the drainage network being fully connected to the restored channel.
Borrow sources were available onsite, but the material needed to be separated using a screen plant to
generate the necessary sizes and gradations of fill material. The borrow source contained a higher percentage
of fine-grained and unsuitable material than measured in initial testing, and the borrow source was depleted
more quickly than expected. Also, the borrow source did not contain sufficiently large material for the riprap
armouring. The design team identified another potential borrow source that could be used for riprap and

other course grained material; this was located onsite along a limestone bedrock outcrop. The outcrop was
drilled and blasted, and the material was screened to remove the fine-grained fraction. This screened
material provided a larger amount of material than the original borrow source, and the blasted rock was a
superior quality for riprap than the original material.
The site is located along an active USFS road, and construction activities had to be planned to accommodate
public traffic. The USFS road received daily traffic from the local community and from logging trucks. The
primary borrow source was on the west (opposite) side of the USFS road, and the screen plant and stock piles
were on the east side of the road. Signs were placed along the road at both ends of the site, but not all the
drivers passed through the site at safe speeds. Water bars were required as part of the SWPPP, and the
construction crews used these as speed bumps, which reduced speeds through the site. Spotters and flaggers
were used to warn equipment operators of traffic. There was also daily truck traffic to haul away phosphate
ore remaining onsite. This traffic used the same roads as the channel excavation equipment, causing
unexpected congestion. The congestion was unavoidable at the start of the project, but once the channel
was partially constructed, a secondary access road allowed the ore haul traffic to use a different road than
the channel construction traffic. Speed bumps on each of these roads helped maintain site speed limits for
traffic.
The original design for the Georgetown Creek cofferdam and diversion assumed it would be necessary to
convey the peak annual flow of 0.85 m3/s (30 cfs) to be conservative. The design also required pumping
operations. The construction sequencing resulted in much of the channel excavation occurring before it was
necessary to divert the stream flow. This resulted in the diversion being installed later in the summer, when
the actual flow rates were lower than the design flow rates. Consequently, a smaller HDPE pipe could be
used. The construction crews also recommended an alternative design using an inlet box and gravity flow
system, which removed the need for pumping operations. These changes made the diversion much easier to
install and more cost effective to operate.
The channel alignment avoided as many foundations from larger buildings as possible, but the construction
crews were prepared to remove foundations from smaller buildings that were within the excavation
boundaries. During construction, additional buried and unmapped foundations were encountered and
needed to be removed. The construction crews expended significant effort removing these foundations.
Progress was slow because there was a considerable amount of concrete and steel in the foundation
structures, and a pneumatic hammer was needed to remove them.
The excavation crews encountered three areas with unexpected environmental conditions. One area
contained hydrocarbon contaminated soil, and two areas contained phosphorus. In each case, construction
was halted in that area, IDEQ was notified, and a remediation plan was developed.
Hydrocarbon contaminated soils were detected by workers who noticed a petroleum odour. Air quality
monitoring did not indicate that the atmosphere was hazardous. Soils emitting the petroleum odour and the
surrounding soils were transported to a containment pad for sampling and remediation. IDEQ was notified
about the soils, and the construction team developed and implemented a sampling plan to quantify
contaminant concentrations. The laboratory results demonstrated that, while petroleum products were
present, they were not at concentrations that presented a significant risk to human health or the
environment. IDEQ recommended that the contaminated soils should be spread on the containment pad and
covered with at least 1 m (3 ft.) of compacted soils, and that the area should be reclaimed.
The excavation crews encountered a septic tank during channel construction. The tank was filled with soil
and could not be moved unless it was emptied. The construction crews had mostly emptied the tank when
they detected the odour of phosphorous oxidising. Air quality monitoring did not detect any gases from the
tank. Additional soil was placed in the tank to halt any reactions, and construction in the area was halted.
IDEQ was notified of the septic tank conditions and asked to determine a remediation plan. The construction
team and IDEQ determined the best disposal method was to construct a disposal pad for the tank and any

Case Studies
contents. The disposal pad location was selected to be close to the tank location to minimise transport. Also,
an elevated location was selected so that the pad would not be influenced by surface or ground waters. The
remediation plan included moving the tank contents in batches to the disposal pad and then agitating the
materials with an excavator until the phosphorous in each batch was completely oxidised. The remediation
plan required notifications to IDEQ, the local sheriff, the local fire department, and the local USFS ranger each
day the material would be oxidised. The oxidation process only occurred during favourable weather
conditions (clear with light wind) and when there were no fire restrictions. An additional area of phosphorous
was encountered while excavating the channel. These soils were remediated with the septic tank soils. The
phosphorous oxidation took several days to complete, and once oxidation was completed, the septic tank
was transported to the disposal pad and the material was placed back into the tank. The tank was grouted
with concrete for additional containment. The entire pad was covered with at least 1 m (3 ft.) of compacted
soil, graded for drainages, and then reclaimed.
The CMP bypass channel was commissioned in the fall of 2010, and post-construction activities have been
limited to minor maintenance and routine monitoring. Also, at the end of construction operations, Nu-West
requested that campsites be constructed in an attempt to limit site vandalism that could compromise the
reclamation goals.
Channel construction and site reclamation was completed in the fall of 2010, and since then the stream
channel has performed as intended with only minor maintenance required. Channel flow typically exceeds
the low-flow channel and utilises the vegetated floodplain during spring runoff before returning to the low-
flow channel in late summer. There has been no erosion to the channel or floodplain, and the rock weirs
continue to perform as designed. The filter gravel has been redistributed throughout the channel and had to
be removed from two of the plunge pools following the first year of runoff flow. Approximately four or five
areas of erosion blanket required maintenance by placing additional stakes.
Maintenance has been required at multiple locations outside of the Georgetown Creek channel where water
flows or accumulates in unintended areas. Most of these areas required small amounts of grading to remove
barriers between the actual flow path and the desired flow path. In one location, a spring emerged above an
existing channel and caused erosion on the hillside. The rill was partially excavated and then backfilled with
riprap to prevent further erosion.
When the bypass channel was commissioned, stream flows infiltrated so that no water reached the channel
outlet. Initially, water filling the void space in the riprap was believed to be the cause of the infiltrating water,
but these conditions continued until the first spring runoff. Georgetown Creek flowed through the culvert for
approximately 50 years, and this depleted the shallow aquifer water level over the downstream half of the
culvert. Water level monitoring throughout the site demonstrated that water levels in the shallow aquifer
rose immediately after commissioning, as shown in Figure 5. The wetting front continued to move
downstream as the surficial geology was saturated. Slope changes on the potentiometric surface coincided
with changes in geological formations. Spring runoff in 2011 fully saturated the surficial geology, and water
level monitoring since 2011 has shown seasonal fluctuations in water levels, with water levels higher during
spring runoff and lower in the fall months.

Access to the site is restricted by fencing, berms, the channel slopes, and boulder placement; however, public
vehicles could still access the site in select areas. Vehicles were crossing the protective berm and creating
ruts along the bypass channel. This was causing erosion and preventing successful reclamation. Additional
boulders placed along the access berm prevent vehicle access to blocked areas of the site.
Georgetown Canyon is popular with hunters, and Nu-West decided to construct five gravel camping pads at
the north end of the facility, away from restricted, critical areas. There are other camping spots in the canyon,
but the open, flat spaces of the facility were likely to draw trespassers if access to the entire facility was
restricted. The pads enable camping in the canyon, but they concentrate people in a desired location. Also,
the pads prevent damage to the site by campers moving boulders or creating ruts when accessing reclaimed
areas. Sheep grazing has been a minor issue, but the impacts have been effectively managed through
communication efforts and site signage.
The project resulted in a stable stream channel to replace the CMP culvert. The restored channel allows for
long-term final closure of the facility. Channel construction occurred as planned and budgeted. The delays
and change orders that occurred during construction, while not entirely quantified, were expected. The
channel has functioned as designed for nearly five years with only minor amounts of maintenance. The site
has been fully reclaimed, and vegetation continues to grow. Figure 6 shows the Georgetown Canyon facility
during operations circa 1960. Figures 7 and 8 show approximately the same view as Figure 6, but Figure 7
shows the site in October 2010, following the completion of reclamation, and Figure 8 shows the most recent
site conditions from August 2014.

Case Studies

The design, construction, and post-construction phases each resulted in multiple conclusions and lessons,
which can be applied to future reclamation projects whether they include a significant hydrologic component
or not. Primarily, it is possible to design and construct ecologically friendly structures at legacy mining sites.
In this case, the stream channel provides fish passage and habitat that did not exist with the CMP culvert.
Also, the riparian corridor provides wildlife habitat. Key lessons learned are listed below:
 Find common goals with the regulators and work with them to achieve those goals.
 Employ a quality construction manager and construction company so that everyone keeps
working together.
 Maintain open lines of communication between the client, designer, construction manager, and
construction company to resolve issues in a timely and efficient manner.
 Quantify unknown conditions during the site investigation. However, at some point, it is time to
stop studying and perform the work.
 Use institutional knowledge to qualify remaining unknown conditions by determining likely
conditions, costs, and resolutions.
 Design the construction using a three-dimensional surface to identify fatal flaws and accurately
calculate volumes.
 Perform costing and scheduling using realistic values, and then modify these using a risk-based (or
similar) multiplier.
 Plan for delays including poor weather, personnel issues, equipment down time, and additional
regulatory comments.
 Pay attention to remote site areas. In this case, minimal work was needed in these areas, but
water flowing to undesired locations has required maintenance.
 Identify all end users and plan for public use.
This paper was the result of many people’s efforts. In particular, the authors would like to acknowledge J.B.
Brown, of Global Environmental Technologies, for providing the conceptual design and construction
oversight and the construction crew of Conestoga-Rover and Associates.
Bates, K. (1999) Fish passage design at road culverts, Washington Department of Fish and Wildlife, Olympia, Washington, USA, p. 14.
Beddoes, J., Ogle, D., Sampson, R. and Hoag, J. (2007) Plant materials technical note number 5: Riparian buffer design and species
considerations, United States Department of Agriculture, Natural Resources Conservation Service, Boise, Idaho, USA,
pp. 17−18.
Berenbrock, C. (2002) Estimating the magnitude of peak flows at selected recurrence intervals for streams in Idaho, Water-Resources
Investigations Report 02-4170, United States Geological Survey, Boise, Idaho, USA, p. 12.
Global Environmental Technologies L.L.C. (GET) (2010) Remedial action completion report, Phase II remedial action construction
activities, Central Farmers Fertilizer Facility in Georgetown Canyon, Idaho, submitted to Nu-West Mining, Inc., Salt Lake City,
Utah, USA.
Federal Highway Administration (FHWA) (2003) Standard specifications for construction of roads and bridges on federal highway
projects, FP-03, United States Department of Transportation, Federal Highway Administration, Washington, District of
Columbia, USA, p. 649.
Rosgen, D. (2006) The cross-vane, W-weir, and J-hook vane structures (updated 2006), Wildland Hydrology, Inc., Fort Collins,
Colorado, USA, p. 19.
RSMeans (2009) RSMeans heavy construction cost data 2010, Reed Construction Data, Norwell, Massachusetts, USA.

L. Novy Dominion Diamond Ekati Corporation, Canada

G. Koop Tetra Tech EBA, Canada
M.E. Wen Environmental Resources Management (ERM), Canada
W. Petherbridge EcoSense Environmental Inc., Canada
The Ekati Diamond Mine is a surface and underground diamond mine operated by Dominion Diamond Ekati
Corporation. It is located near the Lac de Gras Northwest Territories, Canada approximately 300 km north of
Yellowknife and roughly 200 km south of the Arctic Circle. The Ekati Long Lake Containment Facility (LLCF) is
a five celled containment area for storage of processed kimberlite generated during the processing and
extraction of diamonds from kimberlite ore. The LLCF has been in operation since 1998 and deposition of
processed kimberlite has occurred within the three northern cells with the remaining two cells being used for
water quality “polishing” to help meet discharge criteria.
The Interim Closure and Reclamation Plan for Ekati outlines a plan to cover the LLCF kimberlite surface with
a combination of rock and vegetation. The cover system looks to fulfill the closure objective of physically
stabilizing the processed kimberlite and creating a landscape safe for wildlife and human use. Cell B of the
LLCF has reached its capacity and is being used as a reclamation research area. The purpose of the
reclamation research is to identify a long term cover design that can be expanded to the whole LLCF.
A winter drilling investigation in Cell B of the LLCF was undertaken in 2013. The objective of the investigation
was to characterize the processed kimberlite and its porewater chemistry. Results from the investigation
indicated that permafrost has aggraded into the kimberlite and surface zone pore water concentrations were
higher when compared to process plant discharge.
In fall of 2013 various areas of Cell B were seeded with annual and perennial vegetation ground covers.
Further seeding of Cell B was completed in the summer of 2014. Seed from a variety of sources that includes
locally harvested and commercially available native plants and farm crops was applied at different rates using
different seeding techniques. In the winter of 2013 rock was placed in various configurations within the
seeded areas to evaluate its effects on vegetation growth and erosion control. A total of 25 hectares has been
seeded in Cell B since the fall of 2013 and the results of initial monitoring are positive regarding establishment
of long term ground cover on the kimberlite.
1

Ekati Long Lake containment facility reclamation research L. Novy, G. Koop, M.E. Wen and W. Petherbridge

C.D. McCullough Golder Associates/Edith Cowan University, Australia

C. Harkin Yancoal Premier Coal Ltd., Australia
The Lake Kepwari flow-through trial has monitored both lake and river ecology and water quality for two
years flow-through, plus one year post-breach monitoring (2012). The flow-through trial involved both the
development of a management plan and application of the appropriate licensing framework through the
Department of Water state regulatory agency via an operating strategy document. The aim of this document
was to establish the suitability of flow-through as a closure strategy.
The key conclusion of the three-year flow-through trial is that that social and environmental risk and benefit
to the Collie River South Branch (CRSB) and Lake Kepwari are best managed using flow-through as a closure
strategy.
Lake water quality has been significantly improved through nutrient addition and acidity neutralisation.
Improved water quality, particularly in terms of elevated pH and reduced metal/metalloid concentrations,
has now significantly reduced the socio-environmental risk of closure of this landform.
This water quality improvement has also led to improvements in aquatic biota biodiversity and abundance,
and changed biota assemblages to those of a more functional lake more typical of a freshwater ecosystem.
This improvement also better enables the lake to meet the identified end use values of recreation/aesthetics
and livestock drinking than would the no flow-through strategy. River flow-through in particular remediates
pH above recreational guideline pH values in the upper 10 m epilimnion waters in which people would swim.
The objectives of the flow-through trial were first to maintain key CRSB values for aquatic ecosystems,
swimming and livestock watering, and then to provide for lake water quality improvements to meet
recreational criteria and to improve wildlife values. However, pit lake risk reduction and lake end use value
improvement do not appear to have occurred at the expense of CRSB water quality and aquatic ecology, which
have been maintained. Flow-through appears to have significantly increased both total volume in the lower
CRSB and flow peak events, and may further assist CRSB social and environmental values. Additionally, the
pit lake appears to be fulfilling a theoretically predicted function of reducing CRSB nutrient loads, which are
elevated as a result of farming land use practices within the shared catchment.
In line with adaptive management principles, water quality assessment of Lake Kepwari and the CRSB (inlet
and outlet) is continuing until the lake is relinquished back to the state. Owing to the low apparent risk,
monitoring is now undertaken through a less intensive operational monitoring strategy rather than the
previous more intensive trial program.
Mine pit lakes may form at mine closure when voids formed through mining extractions have extended below
groundwater (Castro and Moore, 2000). Internationally and within Australia (Kumar et al., 2013), acid and
metalliferous drainage (AMD) is a particularly common problem for coal pit lake water quality (Vandenberg
et al., 2015). Even if it is not acidic, pit lake water quality may become degraded gradually through dissolution
of contaminants and evapoconcentration (McCullough, Marchand et al., 2013; Castendyk, Balistrieri et al.,
2014). There are few long-term strategies to remediate AMD impacted pit lake water quality (McCullough,
2008).

Engineered flow-through closure of an acid pit lake: a case study C.D. McCullough and C. Harkin
Pit lake waters may discharge into surface and groundwater or present direct risks to wildlife, stock and
human end users. As a result, contaminated pit lake waters can present significant risk to both surrounding
and regional communities and natural environments (McCullough and Lund, 2006).
We propose that riverine lake flow-through may often be a valid mine closure strategy in order to mitigate
pit lake water contamination in pit lakes with poor water quality. Chemical and biological processes such as
dilution, absorption, flocculation and sedimentation reduce solute loads from river and lake.
The southwest part of Western Australia is regarded as highly biodiverse for aquatic ecology (Horwitz et al.,
2001). The town of Collie is approximately 160 km south southeast of Perth, and is the centre of the coal
mining industry in Western Australia (Figure 1). The majority land uses in the catchment are coal mining,
timber production, power generation and agriculture. Approximately 79% of the catchment is state forest.
The recreation and nature conservation values of the forest areas are highly regarded, along with the
recreational opportunities provided by the Wellington Reservoir and other surface waters, including some
pit lakes. These values have led to increased promotion of the area for wildlife and recreation-based tourism
by the local business associations and shire.
Underground and open cut coal mining has taken place in the Collie Basin since 1898 (Stedman, 1988) (Figure
7). More than 100 years of coal mining in Collie have resulted in the formation of at least 13 pit lakes that
range from <1–10 ha in surface area, <10–70 m in depth, 1–50 years in age and 2.4–5.5 pH (Lund et al., 2012).
These lakes all differ in the extent of rehabilitation and water quality, although all are acidic owing to AMD
(McCullough et al., 2010).
Lake Kepwari (WO5B) is situated in the Cardiff sub-basin. Mining began in pit WO5B with diversion of the
Collie River South Branch (CRSB) around the western lake margin and ceased in 1997. The lake was planned
as a regional contact aquatic recreation resource (Evans and Ashton, 2000; Evans et al., 2003). Prior to rapid
filling with water, reactive overburden dumps and exposed coal seams were covered with waste rock,
battered and revegetated with endemic flora by direct seeding (Figure 2). To further reduce wall exposure
and rates of resulting acid production, the lake was rapid filled with a seasonal diversion from the CRSB over
the winters from 1999 to 2008, omitting 2001, a low-flow year. The volume of Lake Kepwari is around 32 Gl,
with a maximum depth of 65 m and surface area of 1.03 km² (McCullough et al., 2010).
Although CRSB water initially raised water pH to above pH 5 (Salmon et al., 2008), pH then fell below pH 4
and displayed elevated concentrations for some metals and metalloids as a result of acidic groundwater
inflow (McCullough et al., 2011). Although the reasonably good water quality of the pit lake potentially lends
itself to a range of end uses, low pH and high Al concentrations remain a challenge to remediate (Lund and
McCullough, 2009; Neil et al., 2009).
During the third week of August 2011, 85.6 mm of rainfall in Collie within 48 h led to high flows in the CRSB.
The water level in the CRSB rose, which led it to overtop and cause failure of the dyke along the south branch
wall that separates it from Lake Kepwari (PMS, 2012). Approximately 30 m of wall failed, with another section
of wall also damaged. The water level in Lake Kepwari increased in height by 1.7 m over a surface area of 106
ha with a flood return period of only around a seven-year average recurrence interval (ARI) event. Around 2
GL of CRSB water then flowed through Lake Kepwari, decanting through both culverts in the northeast and
northwest sides of the Lake (Figure 3).

Case Studies
N Diversion Channel
Lake Outlet
500 m Lake Inlet

An investigation found slightly degraded CRSB water quality against a background of water quality
significantly degraded by deforestation and agriculture causing secondary salinisation and eutrophication
(McCullough and Pearce, 2014). CRSB key end uses of stock watering and contact recreation were not
considered significantly impaired by this altered water quality (McCullough, Ballot et al., 2013).
Significant lake water quality improvements were observed following lake breach as halocline development
reinforced by thermal stratification as the season warmed maintained a less acidic and fresher/warmer layer
over more acidic/colder waters representative of original lake water quality. The river flow-through further
remediated pH above recreational guideline pH values in the upper 10 m mixolimnion waters in which people
would swim (McCullough et al., 2012).
Following broad stakeholder engagement, including community presentations, a flow-through trial that met
leading industry practice was proposed. The Lake Kepwari closure strategy sought first to maintain key CRSB
values for aquatic ecosystems, swimming and livestock watering. It then pursued lake water quality
improvements to meet recreational criteria and to improve wildlife values. Trial development included a
monitoring and reporting plan agreed upon by regulators. Regular quarterly assessment was a key
component of this plan and was planned for a minimum of three years for the biota and water chemistry of
both Lake Kepwari and CRSB.
During 2012, CRSB flow was not sufficient to inflow into the lake or to cause decant. During this time, water
quality in Lake Kepwari deteriorated, and aquatic biodiversity and abundances declined. Prior to CRSB flow-
through into Lake Kepwari in 2013, the water column chemistry was very similar at surface and bottom
sampling locations. Following the 2013 flow-through, the surface and bottom water quality improved by
becoming elevated in pH, nutrients (C, N and P) and reduced metal concentrations.
The effects of the 2011 breach on lake (McCullough et al., 2012) and CRSB (McCullough, Ballot et al., 2013)
water quality have been discussed previously. This paper discusses the monitoring results of the 2012–2014
planned Lake Kepwari flow-through trial and provides advice and recommendations to proponents
considering flow-through as a mine closure strategy for acidic pit lakes.
The aim of this study was to determine if flow-though was the most appropriate (lowest risk, high benefit)
closure strategy for the Lake Kepwari pit lake. Specifically, this study sought to understand how the inflow of
circumneutral (and alkaline) CRSB water improved lake water quality, and how passage of CRSB through Lake
Kepwari affected CRSB hydrology. Information on how lake flow-through influenced lake and CRSB aquatic
biota and how CRSB water and sediment quality was affected is presented elsewhere.
Daily rainfall was sourced using the Bureau of Meteorology’s Collie East weather station (BOM, 2015).
No data smoothing or detrending was used. CRSB flow volumes were measured through three flow gauging
stations using bubble flow meters located at the lake inlet, lake outlet and at the outlet of the diversion
channel (Figure 1).
Four seasonal lake water quality samples were taken quarterly as part of monitoring following breach at
three sites across the river’s path flowing through the lake from March 2012 to December 2014. All water
quality analyses followed standard methods (APHA, 1998). Environmental response data for diatoms,
phytoplankton, zooplankton and benthic macroinvertebrate assemblages were collected simultaneously
during site visits (discussed elsewhere).
A Hydrolab Datasonde 4a (Austin, USA) multiparameter meter was used to make water column profiles at
every metre for oxidation-reduction potential (ORP), dissolved oxygen (DO), temperature, pH and electrical
conductivity/total dissolved solids (TDS).

Case Studies
A water sample was also collected for nutrients and metals from the lake surface waters as well as another
from approximately 0.30 m above the lake benthos with a Teflon trace metal Kemmerer bottle (Wildco, USA).
Upon collection, half of the water sample was filtered through 0.5 µm glass fibre filter paper (PAL Metrigard,
USA) and the other half was left unfiltered. Filtered samples were analysed for ammonia (NH3-N), nitrate-
nitrite (NOx-N) and filterable reactive phosphate (FRP), SO42-, Cl- and dissolved organic carbon. Water sample
aliquots for nutrients were stored in clean, high-density polyethylene bottles and frozen at −20°C until
analysis. Unfiltered samples were persulphate digested and then analysed for total P and Kjeldahl N.
Remaining filtered samples were acidified with reagent grade nitric acid (1%) immediately and stored at 4°C
until analysed for selected elements by inductively coupled plasma optical emission spectrophotometry and
inductively coupled plasma mass spectrophotometry (ICP-MS).
Water quality results were compared to Australian and New Zealand Environment and Conservation Council
and Agriculture and Resource Management Council of Australia and New Zealand (ANZECC/ARMCANZ)
(2000) livestock drinking water guidelines, 80% ecosystem protection guidelines and recreational water
quality and aesthetic guidelines.
In 2013, two heavy rainfall events, one in the second week of August and one in the third week of September,
provided heavy flows through the diversion channel, bypassing the lake flow-through (Figure 5, top). Not
unsurprisingly, these dates also coincided with the highest flows into the lake (Figure 5, middle). However,
these flood events also coincided with the highest flows out of the lake exceeding inflow volumes. Although
there was a heavy rainfall episode mid-May, the CRSB did not start flowing into the diversion in 2013 until
late July, when inflow gates were opened to divert CRSB flow into Lake Kepwari (Figure 4). River flow
monitoring began immediately.
The inflow gates were fully closed on 22 August, when pH differences between upstream and railway bridge
sites became greater than 20%. Inflow gates were then progressively opened until completely open on 16
September 2013. During high flow periods from 16 September until late September 2013, over-topping
occurred at the previous breach site.
The total discharge from Lake Kepwari at the lake outlet is a combination of CRSB diversion inflow volumes
and catchment contributions from the lake itself (Figure 5, bottom). As expected, flow-through of Lake
Kepwari appeared to slightly reduce mid-season lake outlet baseflow by smoothing flood peaks.
Nevertheless, high rainfall “storm” events still presented high flows at the lake outlet. In both base and flood
flow cases, it is unclear without either modelling or baseline data to know unequivocally how baseflow and
floods would present at these sites with CRSB flow still maintained through the diversion channel only.
In 2014, there were heavy rainfall episode in early June and mid-September (Figure 5, top). Although the
early rain event appeared to be largely absorbed by the dry post-summer catchment, sustained winter rainfall
led to a lake water level rise of only around 0.5 m in early July. The second heavy rainfall event lead to a
similar magnitude lake water level rise with a quicker response when the catchment was already saturated
at the end of winter. Lake water level fluctuation during 2014 was less than 1 m, and this presented similarly
at Cardiff Pool, where water level fluctuation was also less than 1 m.
The early June rainfall began in the CRSB flowing at upstream PML45 in early June 2014 (Figure 5, middle).
Flow then ceased in early November 2014, showing 153 days flow in total. Flow at the lake outlet commenced
in late May 2014 and ceased in late November 2014 for approximately 193 days in total, sustaining
downstream flow a further 40 days longer than flow into the lake. Diversion channel flow commenced around
three weeks later than lake inflow in late June 2014 and then ceased in mid-November 2014 after
approximately 176 days. Inflow and outflow gates were not closed in response to low pH trigger events at
any time during 2014 flow.

Flow rates to, and particularly from, the lake were notably bimodal, with peaks in late July and early
September once flow was well established. Showing both the increased catchment afforded by inclusion of
the lake as well as the very high runoff coefficient of this catchment dominated by Lake Kepwari, flow from
the lake of up to 7 m3/s was around twice as high as flow into the lake at both times.
Cumulative flow volumes had inflection points coinciding with these peak flow periods (Figure 5, bottom). At
14 Gl, total flow volumes out of the lake exceeded inflow volumes at around 40% greater than the input flows
of 4 Gl. Less than 3% of lake discharge occurred through the diversion channel.
The total discharge from Lake Kepwari at the lake outlet (PML46) is a combination of CRSB diversion inflow
(PML45) volumes as well as catchment contributions from the lake itself (Table 1). It is not clear how flow-
through of Lake Kepwari influences the downstream hydrograph at PML46 relative to the previous flow at
this point attributed to the diversion and associated catchment. For example, either modelling or baseline
data is required to know unequivocally how baseflow and floods would present at PML46 with CRSB flow
maintained through the diversion channel only. However, owing to the small diversion catchment and lower
runoff coefficient relative to the lake’s surface, it is likely that increased flow-through has slightly extended
flow initiation and cessation timing with sightly increased mid-season baseflow and increased flood peaks.
Site 2013 2014

PML 45 11.682 10.231
PML 46 17.299 14.382
Difference 5.617 4.151
PML 51 2.695 0.486

Case Studies

Lake Kepwari water quality monitoring data from first filling events indicated improvement of pH when
diversion occurred (Figure 6, Figure 7, top). River diversion ceased when the lake was almost full, and pH
significantly declined; it continued to do so for a further three years when the lake was completely isolated
from the CRSB by the diversion channel embankment. When this embankment failed and fresher
circumneutral waters flowed through the lake, surface water quality markedly improved. When the trial
began flow-through of CRSB flows in 2013 and 2014, significant further water quality improvements were
made, including to deeper lake waters, which represent the bulk of lake volume. Lake pH in 2014 was higher
than any historic levels and, importantly, on a reversed and upward trend to that prior to riverine inflow.
During filling, lake water was brackish, reflecting the first flow diversion of the seasonal CRSB waters (Figure
7, bottom). Following flow-through, lake salinity has consistently declined. Although current lake salinity (as
electrical conductivity, EC) is elevated, this is only slight and is not expected to impair lake values significantly.

Case Studies

The Lake Kepwari flow-through trial monitored both lake and river ecology and water quality from two years
flow-through, plus one year post-breach monitoring (2012). The initial breach event led to improved water
quality localised to thermally and/or chemically stratified surface waters, which were then compromised by
mixing with more acidic lower water column waters. However, with sustained flow-through, the water
column now appears to be well mixed such that, although highest water quality still occurs primarily in
surface waters, improved water quality is now found though the lake water column (Figure 8).
Lake water quality in 2014 was significantly improved through nutrient addition and acidity neutralisation
above and beyond the flushing afforded by successful flow-through in 2013 after the no flow-through
conditions in 2012. Improved water quality, particularly with regard to elevated pH and reduced
metal/metalloid concentrations has now significantly reduced the socio-environmental risk of closure of this
landform.
This water quality improvement has also led to improvements in aquatic biota biodiversity and abundance,
and has changed biota assemblages to those of a more functional lake more typical of a freshwater
ecosystem. This improvement also better enables the lake to meet the identified end use values of
recreation/aesthetics and livestock drinking than would the no flow-through strategy. River flow-through in
particular remediated pH above recreational guideline pH values in the upper 10 m epilimnion waters in
which people would swim (Hinwood et al., 2012).

Case Studies
Pit lake risk reduction and lake end use value improvement have not occurred at the expense of CRSB water
quality and aquatic ecology, which have been maintained. There is also a possibility that greater flow volume
may further assist CRSB social and environmental values. Additionally, the pit lake appears to be fulfilling a
theoretically predicted function of reducing CRSB nutrient loads (von Sperling and Grandchamp, 2008;
McNaughton and Lee, 2010) elevated as a result of farming land use practices within the shared catchment
(Wetland Research and Management, 2009).
The Lake Kepwari flow-through trial has demonstrated that social and environmental risk and benefit to the
CRSB and Lake Kepwari are best addressed by flow-through as a closure strategy. In line with adaptive
management principles, continuing water quality assessment of Lake Kepwari and the CRSB (lake inlet and
lake outlet) will be undertaken until the lake is relinquished back to the state. Owing to the low apparent risk,
continued monitoring will be made through a revised operating strategy that recognises the learnings from
the first phase of the trial and that is a less intensive operational monitoring strategy rather than the current
more intensive trial program.
Existing riverine system values must be maintained first and foremost. We further suggest that decant river
water quality may, in some circumstances, be improved, notably so in examples of meso-eutrophic river
waters flowing through slightly acidic pit lakes.
Flow-through closure proposals for pit lakes must be scientifically justifiable and follow a risk assessment
approach (McCullough and Schultze, 2015). This approach should consider both null-positions of no flow-
through, as well as variations on flow-through hydrology, such as complete or partial flow-through, or flow-
through only for certain seasons and/or flow events. In addition to explicitly considering an assessing risk,
benefit and opportunity should be considered, as pit lakes offer an unusually high utility for a region relative
to other closure landforms (McCullough and Van Etten, 2011; Gerner and McCullough, 2014). Owing to the
high uncertainty, the biotic and physico-chemical attributes of both the upper and lower river and lake should
be well monitored (Castendyk, Eary et al., 2014b). Monitoring should directly feed into an adaptive
management framework approved by key stakeholders (McCullough et al., 2009).
Some of the Lake Kepwari data collection was undertaken in collaboration with Mark Lund of Edith Cowan
University under state government Centre of Excellence and industry Australian Coal Association Research
Program (ACARP) funded projects.
Australian and New Zealand Environment and Conservation Council and Agriculture and Resource Management Council of Australia
and New Zealand (ANZECC/ARMCANZ) (2000) Australian and New Zealand guidelines for fresh and marine water quality.
National Water Quality Management Strategy Paper No 4, Canberra, Australia, 1,500 pp.
APHA (1998) Standard methods for the examination of water and wastewater, 20th edition, American Public Health Association,
American Water Works Association, Water Environment Federation, Washington, DC, USA, 1,220 pp.
Bureau of Meteorology (2015) Collie climate averages, viewed 10 April 2011, www.bom.gov.au/climate/averages/ tables/
cw_009628.shtml, last updated 10 April 2015.
Castendyk, D., Balistrieri, L.S., Gammons, C.H. and Tucci, N. (in press) Modeling and management of pit lake water chemistry 2: Case
studies, Applied Geochemistry.
Castendyk, D., Eary, L.E. and Balistrieri, L.S. (in press) Modeling and management of pit lake water chemistry 1: Theory, Applied
Geochemistry
Castro, J.M. and Moore, J.N. (2000) Pit lakes: Their characteristics and the potential for their remediation, Environmental Geology,
Vol. 39, pp. 254–260.
Evans, L.H. and Ashton, P.J. (2000) Value-added closure planning, in Proceedings 4th International and 25th National Minerals Council
of Australia Environmental Workshop, 29 October – 2 November 2000, Perth, Australia, Minerals Council of Australia, pp.
393–409.

Evans, L.H., Rola-Rubzen, F. and Ashton, P.J. (2003) Beneficial end uses for open cut mine sites: Planning for optimal outcomes, in
Proceedings Minerals Council of Australia Sustainable Development Conference, 11–14 November 2003, Brisbane, Australia
Minerals Council of Australia.
Gerner, M. and McCullough, C.D. (2014) Planning for the future: Development of beneficial end use from a quarry pit lake, Victoria,
Australia, in Proceedings Life of Mine Conference, 16–18 July 2014, Brisbane, Australia, pp. 99–112.
Hinwood, A., Heyworth, J., Tanner, H. and McCullough, C.D. (2012) Recreational use of acidic pit lakes — human health considerations
for post closure planning, Journal of Water Resource and Protection, Vol. 4, pp. 1,061–1,070.
Horwitz, P., Lindsay, M. and O’Connor, M. (2001) Biodiversity, endemism, sense of place, and public health: Inter-relationships for
Australian inland aquatic systems, Ecosystem Health, Vol. 7, pp. 253–265.
Kumar, R.N., McCullough, C.D. and Lund, M.A. (2013) Pit lakes in Australia, in Acidic pit lakes — legacies of surface mining on coal and
metal ores, W. Geller, M. Schultze, R.L.P. Kleinmann and C. Wolkersdorfer (eds), C. Springer, Berlin, Germany, pp. 342–361.
Lund, M.A. and McCullough, C.D. (2009) Biological remediation of low sulphate acidic pit lake waters with limestone pH neutralisation
and amended nutrients, in Proceedings International Mine Water Conference, 19–23 October 2009, Pretoria, South Africa,
International Mine Water Association, pp. 519–525.
Lund, M.A., McCullough, C.D. and Kumar, N.R. (2012) The Collie Pit Lake District, Western Australia: An overview, in Proceedings
International Mine Water Association (IMWA) Congress, 30 September - 4 October, Bunbury, Australia, IMWA, pp. 287–294.
McCullough, C.D. (2008) Approaches to remediation of acid mine drainage water in pit lakes, International Journal of Mining,
Reclamation and Environment, Vol. 22, pp. 105–119.
McCullough, C.D., Ballot, E. and Short, D. (2013) Breach and decant of an acid mine lake by a eutrophic river: River water quality and
limitations of use, in Proceedings Mine Water Solutions 2013 Congress, 30 September - 4 October, Lima, Peru, Infomine Inc.,
pp. 317–327.
McCullough, C.D., Hunt, D. and Evans, L.H. (2009) Sustainable development of open pit mines: Creating beneficial end uses for pit
lakes, in Mine pit lakes: Characteristics, predictive modeling, and sustainability, D. Castendyk and T. Eary (eds), Society for
Mining, Metallurgy, and Exploration (SME), Colorado, USA, pp. 249–268.
McCullough, C.D., Kumar, N.R., Lund, M.A., Newport, M., Ballot, E. and Short, D. (2012) Riverine breach and subsequent decant of an
acidic pit lake: Evaluating the effects of riverine flow-through on lake stratification and chemistry, in Proceedings International
Mine Water Association (IMWA) Congress, 30 September - 4 October, Bunbury, Australia, pp. 533–540.
McCullough, C.D. and Lund, M.A. (2006) Opportunities for sustainable mining pit lakes in Australia, Mine Water and the Environment,
Vol. 25, pp. 220–226.
McCullough, C.D., Lund, M.A. and Zhao, L.Y.L. (2010) Mine voids management strategy (I): Pit lake resources of the Collie Basin,
Department of Water Project Report MiWER/Centre for Ecosystem Management Report 2009-14, Edith Cowan University,
Perth, Australia, 250pp. Unpublished report to Department of Water.
McCullough, C.D., Marchand, G. and Unseld, J. (2013) Mine closure of pit lakes as terminal sinks: Best available practice when options
are limited?, Mine Water and the Environment, Vol. 32, pp. 302–313.
McCullough, C.D., Müller, M., Eulitz, K. and Lund, M.A. (2011) Modelling a pit lake district to plan for abstraction regime changes, in
Proceedings Sixth International Conference on Mine Closure, A.B. Fourie, M. Tibbett and A. Beersing (eds), Lake Louise,
Canada, Australian Centre for Geomechanics, 30 September - 4 October, Perth, Australia, pp. 581–592.
McCullough, C.D. and Pearce, J.I. (2014) What do elevated background contaminant concentrations mean for AMD risk assessment
and management in Western Australia?, in Proceedings 8th Australian Workshop on Acid and Metalliferous Drainage, 28 April
– 2 May 2014, Adelaide, Australia, pp. 147–158.
McCullough, C.D. and Schultze, M. (2015) Riverine flow-through of mine pit lakes: Improving both mine pit lake and river water quality
values?, in Proceedings 9th International Mine Water Association (IMWA)/International Conference on Acid Rock Drainage
(ICARD), 30 September - 4 October, Santiago, Chile.
McCullough, C.D. and Van Etten, E.J.B. (2011) Ecological restoration of novel lake districts: New approaches for new landscapes, Mine
Water and the Environment, Vol. 30, pp. 312–319.
McNaughton, K.A. and Lee, P.F. (2010) Water quality effects from an aquaculture operation in a meromictic iron pit lake in
Northwestern Ontario, Canada, Water Quality Research Journal of Canada, Vol. 45, pp. 13–24.
Neil, L.L., McCullough, C.D., Lund, M.A., Tsvetnenko, Y. and Evans, L. (2009) Toxicity of acid mine pit lake water remediated with
limestone and phosphorus, Ecotoxicology and Environmental Safety, Vol. 72, pp. 2,046–2,057.
PMS (2012) Lake Kepwari breach — causes and remediation options, PSM Australia Pty Ltd., Brisbane, Australia.
Salmon, S.U., Oldham, C. and Ivey, G.N. (2008) Assessing internal and external controls on lake water quality: Limitations on organic
carbon-driven alkalinity generation in acidic pit lakes, Water Resources Research, Vol. 44, W10414.
Stedman, C. (1988) 100 years of Collie coal, Curtin Printing Services, Perth, Australia, 363 p.
Vandenberg, J., McCullough, C. and Castendyk, D. (2015) Key issues in mine closure planning related to pit lakes, in Proceedings Joint
International Conference on Acid Rock Drainage ICARD/International Mine Water Association (IMWA) Congress, A. Brown, L.
Figueroa and C. Wolkersdorfer (eds.), 21 April – 24 April, Santiago, Chile, International Mine Water Association (IMWA).
von Sperling, E. and Grandchamp, C.A.P. (2008) Possible uses of mining lakes, in Proceedings 33rd WEDC International Conference,
7– 11 April, Accra, Ghana, Water, Engineering and Development Centre, Loughborough University, pp. 75–80.
Wetland Research and Management (2009) Collie River ecological values assessment 2008, Wetland Research and Management,
Perth, Australia, 72 pp.

A. Chute Klohn Crippen Berger, Canada

N. Dombrowski Kirkland Lake Gold, Canada
A study of the aquatic ecosystems of Kinross Pond and the Eastmaque tailings basin were conducted in the
summer of 2012 to assess the recovery of these former tailings facilities. The objective of the study was to
evaluate the potential for a closed tailings basin, proposed as a closure lake, to support aquatic life once it is
flooded at closure. A study of water quality, sediment quality and aquatic communities was conducted on the
Eastmaque basin and Kinross Pond as analogue representation to the proposed closure lake. Fish populations
were sampled using minnow traps, seine nets and gillnets. Benthic invertebrate, sediment and water quality
sampling were also completed in each pond. Kinross Pond and the Eastmaque basin support different
productive fish communities. Kinross Pond supports a fish community composed of four small-bodied fish
species: northern redbelly dace (Chrosomus eos), finescale dace (Chrosomus neogaeus), golden shiner
(Notemigonus crysoleucas) and brook stickleback (Culaea inconstans). Evidence of recruitment between size
classes of each species (i.e. the presence of young-of-year) was documented. The Eastmaque basin supported
five species of fish: yellow perch (Perca flavescens), northern pike (Esox lucius), spottail shiner (Notropis
hudsonius), blacknose shiner (Notropis heterolepis) and common shiner (Luxilus cornutus). Young-of-year
northern pike and yellow perch were captured along with large adult individuals of each species. The
ecological recovery of Kinross Pond and the Eastmaque basin assessed during this study indicates that in the
years after the Lakeshore basin becomes part of a closure lake, it will potentially sustain a relatively complex
community of fish and benthic invertebrates. This study has shown that viable aquatic ecosystems can be
restored in bodies of water previously used for industrial purposes, specifically for the storage of gold mine
tailings. Removal of these tailings, and the passage of time, can return a tailings storage basin to a pond or
lake with a functioning aquatic ecosystem with few additional rehabilitation measures.
The Lakeshore basin originally formed the southern lobe of Kirkland Lake prior to its use as a tailings storage
facility (TSF). It is located immediately to the south of the Eastmaque TSF, which was developed in the
northern lobe of the former Kirkland Lake. The Lakeshore basin TSF was dewatered in the mid-1990s to allow
reprocessing of the majority of the impounded tailings. Tailings removal and the mining of the near-surface
crown pillars within the Lakeshore basin are believed to have improved the hydraulic connection between
the former lake and the active underground mine workings. Accordingly, Kirkland Lake Gold (KLGold)
seasonally pumps water accumulating in the Lakeshore basin to reduce infiltration and water management
requirements underground in the Macassa Mine.
In 2011, Klohn Crippen Berger (KCB) assessed the potential for the Lakeshore basin to flood following closure.
It was concluded based on these hydrologic and hydrogeologic studies that re-establishing a lake following
closure would be practical and would have the benefits of improving the aesthetics of the area and
preventing access to mine openings and the crown pillar within the footprint of the former lake.
The purpose of this study was to evaluate whether the Lakeshore basin closure lake could support aquatic
life in the future once flooded. The evaluation was based on data collected from two nearby “analogue” sites
(i.e. two historic tailings facilities that have been allowed to recover naturally over several decades), namely,
the Eastmaque basin and Kinross Pond (formerly the Don Lou tailings basin).

From tailings basin to aquatic ecosystem: the ecological recovery of two waterbodies in Kirkland Lake, Canada A. Chute and N. Dombrowski
Limited data are available describing tailings removal from Kinross Pond; however, according to a tailings re-
milling report, two-thirds of the tailings contained in Kinross Pond had been removed by 1993. The dredged
area was then backfilled with sand and gravel (Golder Associates Ltd., 1993). It is understood that the
remaining tailings were removed in the early 1990s.
Eastmaque Gold Mines mined tailings in the Eastmaque basin commercially from 1989 to 1992. Three diesel
power dredges were used to mine the sulphide-rich tailings, which were then processed in the mill (Kinross
Gold Corp., 1995).
Kinross Pond and the Eastmaque basin were allowed to re-fill with water and naturally revegetate; that is,
no active re-vegetation efforts were initiated. Kinross Pond and the surrounding area are now owned by the
Town of Kirkland Lake. A walking path surrounds the pond, and the land past the northern shore is a
landscaped park area that was sodded and outfitted with recreation structures at the time of the field
investigations in summer 2012. The Eastmaque basin is publicly accessible via Goodfish Road. A city snow
dump area is located east of the basin.
Field sampling was completed in the Eastmaque basin and Kinross Pond to assess the fish community, benthic
invertebrates, sediment quality and water quality.
Field sampling completed at the Lakeshore basin included the collection of in situ water chemistry data,
aquatic vegetation presence and distribution, and the documentation of fish presence.
A mark-recapture study was planned for both the Eastmaque basin and Kinross Pond. A permit to collect fish
in these water bodies for scientific purposes was obtained from the Ontario Ministry of Natural Resources
(MNR) (Permit 1069429). The mark-recapture study methodology was preferred because it generates an
estimate of common fish species abundance (as defined by the statistical precision of the population
estimate, where the 95% confidence interval is ≤ 20% of the estimate), biomass and fish community
composition. Minnow trapping was the primary method of capture in both ponds. Minnow traps with mesh
sizes of 6 mm and 3 mm were used to capture fish. The majority of traps had a mesh size of 6 mm.
Minnow traps were set in Kinross Pond during the mark portion of the study (August field investigation) for
approximately 24 hours per day over three days. Eleven traps were set on the first and second nights, and 21
traps were set on the third night. Traps were secured to shoreline emergent vegetation and tossed to a
distance of approximately one to four metres from shore. Fish caught in the traps were kept in a bucket,
measured for fork length to the nearest millimetre, marked by partial excision of the anal fin and moved to
a recovery bucket. Fish were then released back to the pond. Sufficient numbers of fish were caught during
the marking phase to justify completion of the mark-recapture study.
The recapture portion of the study occurred in Kinross Pond during the September field investigation.
Minnow traps were also used as the capture method. Five clusters of traps were set in groups of five,
approximately one metre apart from each other and approximately one to two metres from shore. Trap
locations were chosen to cover each type of habitat. Traps were set for a period of approximately 20 hours
per day, over two days. Representative subsamples (five fish per 10 mm size class of each species) were
weighed to the nearest 0.1 gram using a calibrated electronic balance.
Catch per unit effort (CPUE) was calculated for fish catches from Kinross Pond. Fish catches were calculated
by hour per trap, and then multiplied by 24 hours to obtain a standard catch of fish per trap per day (24-hour
fishing period).

Case Studies
Mark-recapture data were analysed according to the Lincoln-Petersen method (Ricker, 1975). This method
depends on the site being visited two times: once to mark fish and once to recapture a portion of those fish
after they have mixed back into the population. The Lincoln-Peterson method assumes that no fish leave the
population, immigrate to the population, lose their markings, die or are born. The estimated population size
would be:
𝑀𝐶
𝑁= 𝑅
(1)
A more representative estimate of the population size is as follows:
(𝑀 + 1)(𝐶 + 1)
𝑁= (2)
𝑅+1
N = the number of individuals in the population

M = number of fish marked and released in period one
C = the number of fish captured in period two
R = the number of fish marked in period one that are recaptured in period two
The number of recaptures, R, is treated as a Poisson distribution when the number of individuals recaptured
is less than 30. Ricker (1975) includes multipliers to calculate the fifth and ninety-fifth confidence intervals
for small recapture numbers. These values allow confidence intervals to be calculated while accounting for
the likelihood that the population has been underestimated, which tends to occur when few individuals are
recaptured. The size range of fish smaller than the minimum size of the recaptured individuals was excluded
from the calculation of population abundance to account for growth between the mark and recapture
periods.
A population estimate using the above method was calculated for each species of fish recaptured in the
September field investigation: dace spp. and brook stickleback. The population estimate represents the
number of fish present in Kinross Pond at the time of marking (August field investigation).
A biomass estimate was calculated using the truncated length frequency of the marked fish from period one,
the population estimate and the length-weight relationship. The length frequency was truncated to ensure
that the biomass in each size class reflected the fish captured in that size class. The equation used for this
calculation is:
(𝑛𝑢𝑚𝑏𝑒𝑟 𝑖𝑛 𝑠𝑖𝑧𝑒 𝑐𝑙𝑎𝑠𝑠)
𝐵𝑖𝑜𝑚𝑎𝑠𝑠 = ∑ (3)
(𝑡𝑜𝑡𝑎𝑙 𝑚𝑎𝑟𝑘𝑒𝑑)(𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑒𝑠𝑡𝑖𝑚𝑎𝑡𝑒)(𝑝𝑟𝑒𝑑𝑖𝑐𝑡𝑒𝑑 𝑤𝑒𝑖𝑔ℎ𝑡 𝑓𝑜𝑟 𝑚𝑖𝑑−𝑝𝑜𝑖𝑛𝑡 𝑜𝑓 𝑠𝑖𝑧𝑒 𝑐𝑙𝑎𝑠𝑠)
Minnow traps were deployed in the Eastmaque basin during the mark portion of the study (August field
investigation) for approximately 24 hours per day over two days. Fourteen traps were set on the first night,
and six traps were set on the second night. One trap was lost in deep water after the first night of deployment.
Traps were deployed from a boat along the shoreline. Traps were set approximately one to two metres from
the shoreline and tied to emergent vegetation or structures along the shoreline. The first 24 hours of fishing
caught one fish in one minnow trap. Traps were set for a second night and only two individuals were
captured. A mark-recapture study was ruled out, and other sampling methods were chosen to document
common species occurrence and relative abundance.
Floating gillnets and seine nets were deployed in an effort to capture fish in the Eastmaque basin. The gillnets
were 75 m long and 2.5 m in depth, with a float line and a lead line. Gillnets were composed of three 15 m
panels with a 50 mm stretched mesh size, and two 15 m panels with a 25 mm stretched mesh size. Two
gillnets were deployed once per day over two days of sampling. Gillnet sets occurred in the largest basin of
the lake, where the depth was consistently greater than 2.5 m. Nets were secured on or near shore (i.e. to a
tree stump, protruding log, etc.) and set perpendicular to shore. Gillnet set times were kept short
(approximately 1.5 hours) to reduce fish mortality. Fish caught by gillnet were identified, measured for fork
length (mm) and released from the boat. A seine net was deployed in two areas: one in a small bay where

fish had been observed from the boat, and the second in a relatively shallow area (approximately 0.5 m to 1
m deep) with a shoreline composed of cattails. The seine net was 9 m long and 1.5 m deep. Seine netting
locations were chosen based on the ease with which sampling could occur using one individual in a boat and
one in the water, and the presumed likelihood that fish would be caught there. Fish captured by seine net
were collected in a bucket and roughly enumerated and identified before being returned to the water. Seine
catches were large (> 200 fish), and fish were too numerous and small to justify length measurements.
Fish catch information from the Eastmaque basin was used to generate CPUE for each method used, and
length frequencies were generated for each species and fishing methodology.
General observations of fish presence were made during a walk-around survey of the Lakeshore basin.
Shallow, accessible areas were investigated visually, including the west, north and east shorelines of the east
basin, and the east and west shorelines of the west basin. Fish presence was noted where observed.
Benthic invertebrates were collected using an Ekman dredge (23 × 23 × 23 cm) at locations in Kinross Pond
and the Eastmaque basin. Samples were collected at stations chosen based on observed vegetation and
likelihood of encountering benthic invertebrates. Sampling was biased owing to the nature of the pond
substrates. Very few benthic invertebrates were collected in the dredge in an unbiased sampling trial;
therefore, the sampling methodology was altered to ensure that a variety of benthic invertebrates could be
sampled.
Two invertebrate samples in Kinross Pond were collected by dredge sampler in deeper sediments (1–1.5 m
in depth). Dredge sampling in the Eastmaque basin occurred at one site located in a shallow area (1 m deep),
which had abundant submerged aquatic vegetation, and one site in the deepest section of the lake (8 m
deep). These samples were kept for further analysis; however, additional sampling was completed in an effort
to capture more of the invertebrate diversity present in each pond. One sample was collected in each pond
at a location close to shore among shallow submerged and emergent aquatic vegetation. Invertebrates were
sampled by collecting and picking through submerged plants and shallow sediments. A canoe paddle was
used to lift vegetation and sediment into the boat (in the Eastmaque basin) or onto the shore (in Kinross
Pond), and invertebrates observed were collected and placed in a sampling jar for further identification.
Invertebrates caught as by-catch in minnow traps in Kinross Pond were also kept and preserved for further
identification. Four benthic invertebrate samples were taken from Kinross Pond, and three samples were
taken from the Eastmaque basin. Samples were sent to ALS Environmental Laboratories in Winnipeg for
sorting, and invertebrates were identified to the lowest taxonomic level possible (generally genus or species).
Sediment samples were collected during the September field trip. Samples were collected at the deepest
points in both Kinross Pond and the Eastmaque basin. A 23 × 23 × 23 cm Ekman dredge was used for sampling.
Samples were analysed for metals, total organic carbon (TOC) and particle size at Accuracy Environmental
Laboratories Ltd. in Kirkland Lake.
Sediment quality results were compared to the provincial sediment quality guidelines (PSQG), which have
two effect level values: lowest effect level (LEL) and severe effect level (SEL) (Ontario Ministry of the
Environment, 2008). Results were also compared to the Canadian sediment quality guidelines (CSQG), which
includes the interim sediment quality guideline (ISQG) and probable effect level (PEL) (Canadian Council of
Ministers of the Environment, 2001). These sets of guidelines outline sediment contamination levels that, if
exceeded, can present a risk to aquatic life. Results occurring below method detection limits (MDL) were
halved to estimate their concentration before being compared to guidelines.

Case Studies
Water samples were collected from each pond during the August field trip to provide a snapshot of the water
chemistry. Parameters analysed included free cyanide, specific conductivity, hardness, ammonia, total
suspended solids (TSS), total dissolved solids (TDS), pH, alkalinity, acidity, TOC, dissolved organic carbon
(DOC), dissolved oxygen (DO), oxidation reduction potential (ORP), total phosphorus, ferrous and ferric iron,
nitrate and nitrite, phosphate, sulphate, anions (Br, Cl, Fl), total metals and dissolved metals.
Water samples were collected from the surface water of the Eastmaque basin near the deepest part of the
waterbody. Water samples from Kinross Pond were collected from the north end of the pond. Water samples
were analysed by Accuracy Environmental Laboratories Ltd. in Kirkland Lake.
Surface water samples, including those collected on other occasions by KLGold staff, were compared to
provincial water quality objectives (PWQO) (Ontario Ministry of Environment and Energy, 1994). Results
measuring below MDL were halved to estimate their concentration before being compared to the guidelines.
In situ physical and chemical parameters of water were measured using a multi-parameter meter during the
August field investigation and a pH/temperature probe during the September field visit. The multi-parameter
meter measured temperature, specific conductivity, TDS, salinity, DO and pH. The pH probe measured pH
and temperature.
A DO and temperature profile was measured in the Eastmaque basin during the August field investigation.
The profile was measured in the deepest part of the lake; temperature and DO readings were taken at each
meter throughout the water column. A DO and temperature profile was not taken in Kinross Pond; the multi-
parameter meter was not available when boat work occurred on the pond.
The fishing efforts that took place in each pond during each field investigation is presented in Table 1.
Mean set Standard deviation

Total fishing No. of
Field visit Waterbody Gear1 duration of set duration
time (hours) traps2/sets
(hours) (hours)
August Kinross Pond MT 1079 43 25 3.8
Eastmaque MT 408 19 22 1.9
Eastmaque GN 6 4 1.5 -
Eastmaque SN - 2 - -
September Kinross Pond MT 1010 1010 21 2.7
Notes: (1) MT = Minnow traps; GN = Gillnets; SN = Seine nets. (2) Traps tampered with or lost were excluded from analysis.
A total of 482 fish were caught in Kinross Pond during the August field investigation. The fish species caught
were northern redbelly dace (Chrosomus eos), finescale dace (Chrosomus neogaeus), golden shiner
(Notemigonus crysoleucas) and brook stickleback (Culaea inconstans). Differentiating between finescale and
northern redbelly dace was difficult because they have similar traits and they tend to hybridise; therefore,

analysis of mark-recapture data was based on combined dace spp. numbers. A total of 686 fish were caught
in Kinross Pond during the September field investigation. The same fish species were captured during both
field investigations.
CPUE in Kinross Pond during the August field trip was 11 ± 15 fish/trap/day using 6 mm mesh traps. Fish
catches during the September trip were 18 ± 67 fish/trap/day when one 3 mm mesh trap was used during
the first day of trap sets. The 3 mm mesh trap was not used during the August field investigation, and was
removed from the study after the first day of use in the September field investigation owing to an inflated
catch of small-bodied fish (30 to 45 mm). CPUE during the September trip, when the small-meshed trap was
removed from the analysis, was 8 ± 12 fish/trap/day.
Population estimates based on the mark-recapture data could be generated for dace spp. and brook
stickleback. No marked golden shiners were recaptured; therefore, no population estimate or biomass
estimate could be generated for this species. Five fish that had been marked were caught during the
recapture portion of the study: four dace spp. and one brook stickleback. The population estimate generated
for dace spp. represents the fish population in the 45 mm to 85 mm size classes (truncated length frequency
due to recapture lengths). The population estimation for brook stickleback represents the fish population in
the 50 to 55 mm size classes (truncated length frequency). Population estimates and 95% confidence
intervals are presented in Table 2.
Species Population estimate Lower 95% confidence limit Upper 95% confidence limit
Dace spp. 27,900 12,500 69,750
Brook stickleback 550 170 1,000
The abundance and biomass of dace spp. and brook stickleback per size class are presented in Table 3 and
Table 4.
Size class (mm) Number marked Abundance Predicted weight (g) Biomass (kg)
45 105 8,002 0.7 6
50 135 10,288 1.0 11
55 54 4,115 1.4 6
60 44 3,353 1.8 6
65 18 1,372 2.4 3
70 6 457 3.0 1
75 3 229 3.7 1
80 1 76 4.6 0
85 0 0 5.6 0
Total abundance 27,892 Total biomass 34

Case Studies
Size class (mm) Number marked Abundance Predicted weight (g) Biomass (kg)
50 15 518 0.9 0.5
55 1 35 1.3 0.0
Total abundance 553 Total biomass 0.5
The population and biomass estimates for dace spp. and brook stickleback present in Kinross Pond are based
on a limited number of recaptures from a limited number of size classes. The estimates do not represent
precise or accurate population and biomass numbers. Large 95% confidence intervals reflect the conservative
nature of the population estimates. Recapture rates were very low, which may be due to mortality or
decreased fitness of marked fish caused by handling and fin clipping. These activities could make them more
susceptible to predation. Fisheries management decisions should not be made based on the aforementioned
population estimates.
Length frequencies for dace spp., brook stickleback and golden shiner are shown in Figures 1, 2 and 3,
respectively.

A total of 31 fish were caught in the Eastmaque basin in the minnow traps and gillnets, and approximately
400 fish were captured during two deployments of a seine net. Species caught included yellow perch (Perca
flavescens), northern pike (Esox lucius), spottail shiner (Notropis hudsonius), blacknose shiner (Notropis
heterolepis) and common shiner (Luxilus cornutus). Table 5 includes a summary of fish catch and fish length
associated with each methodology used in the Eastmaque basin.
Total Average length Standard deviation

Gear Fishing effort (h) Fish species
catch (mm) (mm)
MT 408 Yellow perch 1 51 -
Northern pike 2 112 18
GN 6 Yellow perch 25 111 20
Northern pike 4 573 91
Common shiner 2 142 3.5
SN 2 deployments Spottail shiner 8 40–50 -
Blacknose shiner 400 30–50 -
Yellow perch 40 40–50 -
Note: MT = Minnow traps; GN = Gillnets; SN = Seine nets.
The length frequency of fish caught in the Eastmaque basin is displayed in Figure 4.

Case Studies
The CPUE of gillnets deployed in the Eastmaque basin was 6.1 ± 5.5 fish/net/hour. The CPUE of the minnow
traps was 0.18 ± 0.63 fish/trap/day.
Observations along the eastern shoreline of the west Lakeshore basin revealed a deceased dace spp. minnow.
Unidentified small-bodied fish were observed swimming in the shallow portions along the east shoreline of
the east Lakeshore basin.
The sediments sampled using the dredge in each waterbody were very fine and did not appear to provide
high-quality habitat for invertebrates upon inspection in the field. Abundance and diversity of invertebrates
in the sediment samples were low. Samples taken in each waterbody from near-shore areas among the
submerged and emergent aquatic vegetation displayed higher numbers of organisms and more orders.
Invertebrates captured by the dredge sampler in Kinross Pond included genera from the following orders:
Odonata, Diptera, Basommatophora, Trombidiformes, Oligochaeta and Veneroida. A total of 31 individuals
were sorted out of these sediments. The remaining invertebrates were recovered either from minnow trap
catches or from the shoreline sampling. These invertebrates belonged to genera from the following orders:
Ephemeroptera, Trichoptera, Odonata, Amphipoda, Diptera, Hemiptera, Homoptera, Basommatophora,
Oligochaeta, Hirudinea and Veneroida. A total of 110 individuals were captured in the minnow trap and
shoreline samples.
Invertebrates captured in the Eastmaque basin by dredge were composed of 50 individuals from genera of
the following orders: Ephemeroptera, Odonata, Diptera, Basommatophora, and Oligochaeta. The sample
taken by dredge at the deepest part of the lake (a depth of 8 m) contained 66 individuals from two genera of
the Diptera order. The sample collected from the shoreline contained 158 individuals from genera of the
following orders: Ephemeroptera, Trichoptera, Odonata, Diptera, Basommatophora, Isopoda, Amphipoda,
Oligochaeta and Hirudinea.
Sediment analysis results revealed some elevated metal(loid)s in the sediments in Kinross Pond. Arsenic,
copper, lead and mercury exceeded CSQG ISQG and PEL, and the PSQG LEL and SEL. Cadmium, chromium,
nickel, zinc and TOC also exceeded the PSQG LEL. The sediment sample from the Eastmaque basin had
exceedances of the CSQG ISQG and PSQG LEL for chromium, copper and lead. Nickel exceeded the PSQG LEL.
Sediment chemistry results are displayed in Table 6.

Parameter Units CSQG PSQG Kinross Eastmaque Macassa

Pond basin tailings1
ISQG PEL LEL SEL
Arsenic (As) µg/g 5.9 17 6 33 47 4.3 6.1
Cadmium (Cd) µg/g 0.6 3.5 0.6 10 0.87 0.57 0.17
Chromium (Cr) µg/g 37.7 90 26 110 28 65 50
Copper (Cu) µg/g 35.7 197 16 110 393 55 59
Lead (Pb) µg/g 35 91.3 31 250 1,240 74 93
Manganese (Mn) µg/g 460 1,100 434 763 990
Mercury (Hg) µg/g 0.17 0.486 0.2 2.0 37 1.3 0.10
Nickel (Ni) µg/g 16 75 52 45 28
Zinc (Zn) µg/g 123 315 120 820 149 98 69
Note: (1) Typical tailings chemistry in the Macassa Tailings Storage Facility (Average values from July and January 2011 and June
and August 2012).
The metal concentrations of the Eastmaque basin sediment sample are typical of tailings produced from
Kirkland Lake camp mines. Concentrations of chromium, copper, lead and nickel were elevated but did not
exceed probable or severe effects limits. The sediment sample from Kinross Pond had elevated levels of
arsenic, cadmium, chromium, copper, lead, mercury, nickel and zinc. Probable and severe effects limits for
arsenic, copper, lead and mercury were exceeded. A single sediment sample was analysed from each pond,
and observed chemistry may be limited to a small area.
The PSQG SEL indicates a level of contamination expected to be detrimental to the majority of sediment-
dwelling organisms. Sediment chemistry alone, however, cannot be used to predict biological effects (Ontario
Ministry of the Environment, 2008).
One PWQO exceedance occurred in the Eastmaque basin water sample taken in during the August field
investigation. Total cyanide exceeded the PWQO in the Eastmaque basin water sample. Previous sampling of
water quality in the Eastmaque basin occurred in April and May of 2012, and in September (two sampling
events), October and November of 2011. During these sampling rounds, there were exceedances of ammonia
(November 2011), and total cyanide, total copper, total iron, total molybdenum and total silver (September
2011).
Water quality in the study basins was good, with only one exceedance of PWQO observed for each. The
concentration of water parameters in Kinross Pond, indicated by PWQO exceedances, did not correlate with
the high levels of mercury, lead, copper and arsenic found in the sediment.
Physical and chemical parameters collected during the August field investigation using a multi-parameter
meter are shown in Table 7.

Case Studies
Temp Conductivity TDS Salinity DO pH

Site Date
(°C) (µS/cm) (g/L) (ppt) (mg/L) (units)
Kinross Pond 13 August 20 490 0.32 0.24 0.17 7.0
2012
Eastmaque tailings basin 14 August 21 310 0.20 0.15 7.6 8.1
2012
Lakeshore west basin 14 August 21 870 0.56 0.43 11 8.2
2012
Lakeshore east basin 14 August 22 380 0.25 0.18 8.4 9.2
2012
Note: Probe of multi-parameter meter was deployed within a depth of one metre from the surface.
The DO and temperature profile from the Eastmaque basin are shown in Figure 5.
Kinross Pond represents a system that has progressed from a tailings storage area to a productive ecosystem
inhabited by a variety of aquatic organisms. Kinross Pond has a high abundance and biomass of small-bodied
fish species. The Kinross Pond community is diverse for a waterbody of its size. The four fish species found in
the pond feed on plankton and insects.
Kinross Pond is not a deep waterbody and contains dense submerged and emergent aquatic vegetation. It is
likely that these factors will limit the productivity of the pond, and it is unlikely that the pond could provide
habitat for larger bodied fish. The limited depth and excess of vegetation observed in Kinross Pond likely limit
DO, and possibly deplete it beyond levels suitable for some fish to survive. A low DO value was observed
during in situ chemical analysis using the multi parameter meter (Table 7). Since the study, a fountain has

been installed in the deeper part of the pond. A fountain may have the effect of increasing mixing in the pond
and therefore increasing the concentration of DO.
High sediment metal(loid) concentrations were observed in Kinross Pond; however, these may remain
contained in the sediments and have limited bioavailability. The presence of aquatic vegetation such as
cattails may help absorb heavy metals and aid in the remediation of the pond (Padmavathiamma and Li,
2007). Benthic invertebrates captured in Kinross Pond were composed of pollution-sensitive taxa such as
Trichoptera, Ephemeroptera and Odonata, which indicates that despite possibly harmful levels of
metal(loid)s in sediments, invertebrates are present and can provide an abundant food source for fish. The
community of fish and amphibians in the pond was also diverse and abundant despite the measured metal
concentrations in the single sediment sample.
Kinross Pond appeared to provide a good environment for fish: the shift in the length frequencies (Figures 1,
2 and 3) indicates that fish grew (recruitment occurred) between sampling periods.
The Eastmaque basin has shown evidence of recovery since a large proportion of tailings was removed from
the basin. The Eastmaque basin has a more diverse fish community than Kinross Pond; it is composed of five
fish species. The fish community contains multiple trophic levels, including those that feed on invertebrates
(common shiner) and plankton and invertebrates (spottail shiner and blacknose shiner), as well as omnivores
(yellow perch) and carnivores (northern pike). Large-bodied perch (>150 mm) and northern pike (>650 mm)
were captured in the Eastmaque basin. The presence of large-bodied omnivores and carnivores may explain
the low capture rates of minnow traps. Minnow traps are passive gear and rely on fish movement. The
presence of large predators in a lake would cause smaller prey species to remain in cover, reducing their
encounter rate with minnow traps. The minnow traps did, however, catch juvenile northern pike, and the
seine net deployments caught juvenile yellow perch. The capture of these juveniles indicates that
reproduction of the large predators in the lake is occurring. Reproduction of the species in the lake is a
positive sign of recovery; however, when the length frequency of fish caught in the Eastmaque basin was
generated, a gap appeared in the middle size classes (Figure 4), from the 180 mm to 430 mm size classes.
This gap may have been due to gear selectivity, because minnow traps catch smaller fish and gillnets catch
larger fish.
Effects of metal exposure on yellow perch have been investigated along contamination gradients in Sudbury,
Ontario, and Rouyn-Noranda, Quebec (Campbell et al., 2003; Rasmussen et al., 2008; Pyle et al., 2008). These
studies indicate that while yellow perch are generally metal-tolerant, they experience indirect and direct
effects caused by metal contamination by copper, nickel and cadmium. Direct effects included those on a
cellular level, which influenced organs, tissues, individuals and populations. Indirect effects included food
web–mediated effects caused by the lack of preferred food sources (large metal-sensitive benthic
invertebrates) and replacement by smaller metal-tolerant invertebrates. The extent of potential metal
exposure on yellow perch in the Eastmaque basin is unknown; however, it should be noted that
concentrations of nickel and cadmium were often below detection limits, and copper rarely exceeded PWQO.
The present study investigated the ecology of analogue sites in an effort to assess the likelihood that the
proposed Lakeshore closure lake will be able to support an aquatic community. Observations of the
Lakeshore basin were limited to in situ water chemistry, fish presence, vegetation type and distribution, and
presence of other organisms. These observations revealed that acceptable levels of DO were present in the
east and west basins, and that fish were present in the west basin. The closure lake will be created when the
barrier between the Lakeshore (west) and Eastmaque basins is removed, and the Lakeshore basin is allowed
to fill with water over time.
It is anticipated that the fish species present in each basin will mix, and that the resulting fish community will
be composed of multiple feeding levels, similar to the community of the Eastmaque basin. Benthic

Case Studies
invertebrates, which are likely already present in the Lakeshore basin, will continue to colonise the basin as
it floods to become the closure lake. An abundance of submerged and emergent vegetation is currently found
in and around the Lakeshore basin. The vegetation will likely recolonise after the water level for the closure
lake is established. Sediment quality in the Lakeshore basin area of the closure lake will depend on the current
sediment quality in the Lakeshore basin, which was not assessed in this study. Chemical parameters in the
sediment of the Eastmaque portion of the closure lake are not expected to be above the any severe or
probable effect levels. Water quality in the closure lake will also likely be similar to that found in Eastmaque,
with minimal exceedances of PWQO concentrations.
This study has shown that viable aquatic ecosystems can be restored in bodies of water previously used for
industrial purposes, specifically for the storage of gold mine tailings. Removal of these tailings, and the
passage of time, can return a tailings storage basin to a pond or lake with a functioning aquatic ecosystem
with few additional rehabilitation measures.
The Lakeshore basin, which has been dredged of tailings, is predicted to form a hydrologically stable lake
when combined with the Eastmaque basin upon closure (KCB, 2011). Based on the work completed during
this study, the closure lake will have a high potential of supporting benthic invertebrate production, a
complex fish community with multiple trophic levels and other organisms such as amphibians, mammals and
birds. It is likely that benthos and fish currently reside in the Lakeshore basin, and the creation of a closure
lake will enhance the amount and quality of habitat available to these aquatic species.
This study has raised some further questions and possible paths forward with respect to waterbody recovery
and ecosystem health:
 Kinross Pond sediment contamination: The contaminated sediment in Kinross Pond revealed that
even though a pond can support an abundance of fish and invertebrates, it can have high levels of
metals in some habitats. The extent of this contamination is not known, based on the scope of
this study. Similar levels of contamination were not observed in the Eastmaque basin. An
investigation directed towards identifying the extent of sediment contamination in the pond
would be beneficial to understanding the pond recovery. Contaminant testing on fish living in the
pond would also reveal if fish are being noticeably affected by this sediment contamination.
 Fish population estimates: The population estimates of fish from Kinross Pond calculated based
on the data collected in this study may not accurately reflect actual numbers of fish in Kinross
Pond. An additional season of fieldwork with a larger scale mark-recapture experiment using a
more diverse array of gear would be required to generate a usable population estimate. A larger
scale mark-recapture experiment could also be completed in the Eastmaque basin to determine
the population size and biomass of fish present in the basin.
Information collected about the Eastmaque basin may make the basin a possible candidate for fish stocking
in the Kirkland Lake area. The lake has public access, and information about the fish community, benthos,
water chemistry, sediment chemistry and DO profile is now available. The process to determine whether a
lake can or should be stocked can take up to two years (Vascotto, 2012).
The multiple feeding groups documented in the Eastmaque basin may have implications for the level of metal
bioaccumulation and biomagnification that can occur in the aquatic food web. An assessment of metal levels
in fish tissue would provide an indication of the extent to which metal concentrations in the system are
affecting fish, and would be recommended prior to evaluating the stocking potential of the lake.
Campbell, P.G.C., Hontela, A., Rasmussen, J.B., Giguère, A., Gravel, A., Kraemer, L., Kovesces, J., Lacroix, A., Levesque, H., and
Sherwood, G. (2003) Differentiating between direct (physiological) and food-chain mediated (bioenergetic) effects on fish in
metal-impacted lakes, Human and Ecological Risk Assessment, Vol. 9(4), pp. 847–866.

Canadian Council of Ministers of the Environment. (2001) Canadian Sediment Quality Guidelines for the Protection of Aquatic Life,
Publication No. 1299; ISBN 1-896997-34-1.
Ontario Ministry of the Environment. (2008) Guidelines for identifying, assessing and managing contaminated sediments in Ontario:
An integrated approach. Queen’s Printer for Ontario, PIBS 6658e.
Ontario Ministry of the Environment and Energy. (1994) Water management policies, guidelines, provincial water quality objectives
of the Ministry of Environment and Energy, Queen’s Printer for Ontario, PIBS 3303e.
Padmavathiamma, P.K. and Li., L. (2007) Phytoremediation technology: Hyper-accumulation metals in plants, Water, Air, and Soil
Pollution, Vol. 184, pp. 105–126.
Pyle, G., Busby, P., Gauthier, C., Rajotte, J. and Couture, P. (2008) Seasonal and regional variations in metal contamination and
condition indicators in yellow perch (Perca flavescens) along two polymetallic gradients. II. Growth patterns, longevity, and
condition, Human and Ecological Risk Assessment, Vol. 14, pp. 126–145.
Rasmussen, J.B., Gunn, J.M., Sherwood, G.D., Iles, A., Gagnon, A., Campbell, P.G.C. and Hontela, A. (2008) Direct and indirect
(foodweb mediated) effects of metal exposure on the growth of yellow perch (Perca flavescens): Implications for ecological
risk assessment, Human and Ecological Risk Assessment, Vol. 14, pp. 317–350.
Ricker, W.E. (1975) Computation and interpretation of biological statistics of fish populations, Bulletin of Fisheries Research Board of
Canada, Vol. 191, pp. 1–382.
Golder Associates Ltd. (1993) Geotechnical investigation for tailings re-milling: Lakeshore Mine, Kirkland Lake, Ontario, prepared for
Lac Minerals Ltd., May 1993.
Kinross Gold Corporation. (1995) Revision 11-95 Closure Plan: Macassa and Lakeshore Properties, November 1995.
Klohn Crippen Berger (KCB) (2011) Lake shore property closure lakes phase II assessment – FINAL, prepared for Kirkland Lake Gold
Inc., 30 November 2011.
Vascotto, S. (2012) Presentation: The science behind stocking – why we do what we do, prepared and presented on behalf of the
MNR, 27 November 2012.

G.S. Piorkowski Stantec Consulting Ltd., Canada

G. Price Department of Engineering, Faculty of Agriculture, Dalhousie University, Canada
N. Tashe Stantec Consulting Ltd., Canada
Soil amendments are often required to facilitate vegetation growth on mine spoils. Determining the rate of
amendment application required to maximise plant growth is challenging since plant responses depend on
the plant species used and the combined properties of the mine spoil and soil amendment. Experimentation
remains the best method for evaluating the type and rate of amendments required to promote plant growth
in mine soils. This study investigates the utility and usage of optimisation-centric experimental design
methods, based upon response surface methodologies (RSM), to determine optimum amendment blends and
application rates; it focuses specifically on using the mixture-amount (MA) subset of RSM. The optimum blend
of municipal solid waste (MSW) compost, alkaline-stabilised biosolids (ASB) and wood shavings (WS) was
determined for four distinct spoil substrate types of an operating quarry under greenhouse and field plot
settings. The amendment blend was found to be relatively consistent among the different substrate types and
comparable between greenhouse and field studies, indicating that the synergistic properties of the
amendment blend were more important for plant growth than particular soil or meteorological factors.
Combining the different soil types within a whole-site model allows the statistical model to better explain site-
scale variance in the parameter optimisation. In this study, the optimum amendment mixture was determined
to be 59% ASB, 40% MSW compost and 1% WS in the greenhouse and 57% ASB, 42% MSW compost and 1%
WS in the field plots. The optimum amendment application rate, however, was lower in the greenhouse than
in the field investigation: the optimum rate in the greenhouse was 82 Mg/ha, while the optimum rate in the
field was 104 Mg/ha. RSM approaches, particularly MA experiments, offer a valid alternative to optimising
mine soil amendment programs: they provide an optimisation-specific experimental design and model
development process; when comparing multiple amendment types, they require fewer experimental units
than standard treatment-comparison experimental designs. Amendment blends can be readily optimised in
a controlled greenhouse setting, but field plots are required to determine the rate of blended amendment
that should be added for optimised plant growth.
Mining projects generate large volumes of waste materials, principally mining effluent, waste rock and
tailings, which must be stabilised and integrated into adjacent landscapes as part of reclamation and closure
plans. Although fundamental to modern mining practice, reclamation and closure plans were not required
for historic mining projects, and an estimated 10,000 abandoned mine sites exist in Canada. Reclamation
costs for these are borne by the Crown (Cowan et al., 2010). Reclamation costs remain a significant economic
constraint to proposed projects and a burden to public agencies tasked with rehabilitating abandoned mines.
Realising efficiencies in reclamation costs can aid with project feasibility and reduce the public burden of
reclaiming abandoned mines.
Establishing a vegetative cover that stabilises mine spoils and contributes to planned postclosure land uses
is an integral part of mine reclamation. Mine waste rock and tailings materials are stressful substrates for
plant growth because of poor soil development, low water-holding capacity, lack of major plant nutrients
and the presence of phytotoxic compounds such as heavy metals and metalloids (Bolan et al., 2014; Zanuzzi
et al., 2009). Mine substrates can be improved to support plant growth through the application of

Optimising application rates of waste residuals in mine soil reclamation programs G.S. Piorkowski, G. Price and N. Tashe
amendments (Brown et al., 2007; Solis-Dominguez et al., 2012). Organic amendments, in particular, are
important for enhancing plant growth by improving water retention, providing an organic nutrient source,
elevating soil cation-exchange capacity and stimulating soil microbial communities that enhance nutrient
cycling (Larney and Angers, 2012; Drozdowski et al., 2014). Waste by-products from local municipalities and
industries are increasingly used as soil amendments in reclamation programs. Using waste by-products helps
operators realise economic benefits by reducing materials costs and providing a cost sharing mechanism
between the mine authority and organic waste producers (Brown et al., 2003; EPA, 2007; Kumpeiene et al.,
2008). For example, at the East Sullivan mine site in Quebec, the utilisation of wood waste from a local mill
to cap a tailings impoundment instead of a traditional cap dropped the project cost from C$ 50 million to
C$ 9.7 million (Cowan et al., 2010).
The costs of mine soil amendment can range between approximately US$ 400/ha to approximately
US$ 40,000/ha, depending on the logistics of materials acquisition and application (EPA, 2007). Optimising
soil amendment applications in terms of their effect on vegetation growth could improve projects by reducing
unnecessarily high application rates on mine soils. Amending mine soils with waste residuals can cause
unintended environmental impairment. For example, adding organic by-products alone can elevate the
leaching and bioavailability of some metals (e.g., lead and arsenic), and high application rates of inorganic
by-products can impair vegetation growth typically through the co-application of salts and/or phytotoxic
compounds (Brown et al., 2005; Zanuzzi et al., 2009; Houben et al., 2012; Bolan et al., 2014). The most certain
approach to formulating soil amendment mixes for contaminated soils and for optimising soil amendments
for inert substrates is to experimentally validate the amendment’s effects on vegetation growth and metal
transport.
Full-factorial designs and analysis of variance (ANOVA) statistics remain the standard experimental design in
reclamation sciences. While useful for hypothesis testing, these methods suffer from high experimental
effort and do not produce the most suitable statistical models for optimising amendment application rates.
Response surface methodologies (RSM) are a set of experimental designs that are explicitly used for process
optimisation, most commonly in the process engineering and manufacturing sectors. Multiple response
optimisations are also possible using these methods, whereby a compromise optimum can be attained
among diverse responses, such as in soil reclamation where vegetation growth must be balanced with project
costs and contaminant transport (Schrevens and Cornell, 1993; Ascough et al., 2013). Despite the benefits of
lower costs, reduced experimental effort and the potential for multiple response optimisations, response
surface methodologies have not been adopted within the reclamation field. However, recent studies have
found success with RSM designs for studying soil physicochemical properties such as carbon and nutrient
mineralisation (Sevilla-Perea et al., 2013; Sevilla-Perea et al., 2014), hydrocarbon remediation (Kalali et al.,
2011), soil phytotoxicity amelioration (Koo et al., 2011), and soil arsenic immobilisation (Naseri et al., 2014).
Considering vegetation growth in mine substrates has been observed to yield quadratic relationships or
saturation effects with respect to amendment applications (Gardner et al., 2010; Fan et al., 2012), RMS may
be useful for optimising amendment applications.
The purpose of this research program is to evaluate the utility of RSM for optimising amendment application
rates and combinations for mine soil reclamation programs. In this experiment, mixture-amount (MA)
designs (a subset of RSM) were specifically evaluated due to logistical and theoretical constraints on
amending quarry substrates with multiple organic waste by-products. The objectives of this research
program were (i) to evaluate whether MA experiments are effective for determining optimised amendment
combination and rates of application with respect to the inherent variability in vegetation growth; (ii) to
determine whether standard MA designs were sufficient for optimising plant growth or should be expanded
to include third-order polynomials and third-degree interactions; and (iii) to contrast the optimised
amendment combination and application rate determined in a greenhouse versus field setting.

Case Studies
The study site is a working aggregate quarry located near Dartmouth, Nova Scotia, at geographic coordinates
of 44.73497° N and −63.537252° W and 72 m elevation. According to climate normals (1981–2010), the area
experiences approximately 1,400 mm annual precipitation, has an average annual daily temperature of 6.6°C,
and has 163 frost-free days typically from May 7 to October 18. Four separate areas within the quarry were
targeted, including two areas within a rock fines disposal area and two waste rock storage areas. Rock fines
are periodically dredged from settling ponds that receive runoff from the rock crushing and sorting area of
the quarry, dewatered and deposited as infill against the north cliff face of the quarry. The rock fines disposal
area is periodically reclaimed through terracing, covered with approximately 30 cm of soil and seeded with
Nova Scotia highway mix (Table 1). Block 1 was situated in an area of old rock fines deposits (> 10 years) and
Block 2 was placed in an area of new deposits (< 5 years). Prior to the onset of the field study, the rock fines
areas were terraced and covered with topsoil, except for a 30 m × 20 m experimental area located near Block
1, which was left uncovered. The Block 2 substrates were collected for the greenhouse study but were not
available for the field investigation. The rock fines substrates were silt loam textured and contained 5% –
10% rock fragments.
The overburden stockpiles are located in the northeastern section of the property and away from active
mining. The majority of the site overburden was used to form the berms surrounding the property, then
contoured and seeded with Nova Scotia highway mix. The overburden stockpiles chosen for this study are
sparsely vegetated and fairly level in contour. Block 3, located near an equipment disposal area, is gently
sloping (1% – 2% slope) into adjacent forest. Block 4 is located at the top of a berm and is relatively flat (~1%
slope) but exposed. No substantial vegetative cover was present in either block at the onset of the study.
Both blocks were sandy loam-textured material with high rock content (> 30%). Soil nutrient contents are
presented in Table 2.
Common name Species name Percentage in mix

Perennial ryegrass Lolium perenne 35%
Creeping red fescue Festuca rubra 25%
Timothy Phleum pratense 15%
Tall fescue Festuca arundinacea 10%
Alsike clover Trifolium hybridum 10%
Kentucky bluegrass Poa pratensis 5%
The aggregate quarry used in this study is located within the Halifax Regional Municipality, approximately
21 km from the city centre. An amendment-based reclamation approach is viable for this site due to its
proximity to organic waste sources. Two organic waste by-products generated in Halifax are municipal solid
waste (MSW) compost and alkaline-stabilised biosolids (ASB). MSW compost is derived from source-
separated food and yard waste that is collected at residential households and commercial institutions,
subjected to in-vessel composting and cured for a minimum of 200 days at a local landscaping company. Class
A municipal wastewater biosolids used in this study were subjected to alkaline stabilisation, whereby the
biosolids were mixed with fly ash and lime to achieve a pH greater than 12 for at least 72 hours and heated
to 52°C for at least 12 hours. The wood shavings used in this study were from softwood sources and were
sized at 1” or finer. Relevant chemical constituents for the amendments are presented in Table 2.

Organic CEC* Base

C N P K
Substrate pH matter C:N (meq/ saturation
(g/kg) (g/kg) (g/kg) (g/kg)
(%) 100g) (%)
Mine soils
Block 1 8.1 0.9 1.26 0.02 70:1 0.03 0.25 20.1 99.6
Block 2 8.3 0.8 0.76 0.01 54:1 0.03 0.24 14.9 99.5
Block 3 4.9 4.3 26.5 0.97 27:1 0.05 0.05 4.0 21.7
Block 4 4.6 1.7 4.09 0.13 33:1 0.02 0.05 3.5 14.1
Amendments
ASB 7.8 9.3 174 9.16 19:1 0.45 13.0 31.2 99.7
MSW 7.4 19.0 214 22.7 9:1 1.39 2.71 184 99.8
WS 5.2 81.2 471 1.2 392:1
*Cation-exchange capacity
Two forms of RSM suited for soil reclamation programs are component-amount and MA experiments. In
component-amount (CA) designs, the response variable is assumed to be separately and interactively
influenced by the amendments, and the rate of each amendment is therefore manipulated individually. In
soil reclamation programs, CA experiments are more suited when amendments are added for differential
effects to soil physicochemical properties or plant growth, as in the case of adding one organic amendment
as a nutrient source and one alkaline amendment to ameliorate soil pH. In contrast, MA experiments assume
that the response variable is influenced by the total rate of application and the blending properties of the
amendment, whereby the effect of each amendment individually cannot reasonably be separated out of the
total response. Adding a combination of organic amendments to mine soils is better represented by MA
designs since the direct benefits of organic nutrient provision and indirect benefits of soil microbial
community stimulation, soil aeration improvement and soil water enhancement cannot reasonably
attributed to one organic amendment over the other. Thus, MA experiments are theoretically more
appropriate for establishing an optimised blend of organic amendments in a mine soil reclamation program.
In the experimental design, the four blocks were used in a randomised complete block design. The
experimental units within each block were determined by a D-optimal MA design algorithm computed in
JMP® (v. 9.0) software (SAS Institute Inc., Cary, NC, USA). Individual blocks were used as replicates within the
greenhouse and field studies in order to account for the variability of soil types at the site and the expected
variation in plant biomass (Gilmour and Trinca, 2006). Models were created and evaluated for each individual
block and for the composite data set (i.e., inclusive of all blocks together) to determine how variability in the
data affects the optimisation outcome and whether the optimised amendment mixture and application rate
changes among substrate types.
From each block, approximately 80 L of soil substrate was retrieved from a 25 m × 15 m area. Approximately
4 L of soil substrate was retrieved from the upper 15 cm of the soil profile at 20 randomly selected locations
within the block area; this was homogenised via mixing in the soil collection container and again prior to
being distributed into the greenhouse pots. Standard plastic greenhouse pots were used with 15 cm diameter

Case Studies
× 10 cm height. Within the pots, 1,000 g of soil was blended with up to 15% w/w of amendment, which was
composed of up to 100% biosolids, 100% MSW compost and 10% wood shavings. Wood shavings were
included only as a C:N ratio adjustment (Brown et al., 2007) and were therefore added at a lower proportion
of the overall blend. The rates and ranges of amendment composition were input as in Table 3; up to third-
order polynomial and third-degree interactive effects were included. For each substrate block, 27
experimental pots with varying composition and rates of amendments were created. Once the substrate-
amendment mixture was measured and homogenised, the pots were filled, hand packed and seeded with
the Nova Scotia highway mix (Table 1) at a rate equivalent to 60 kg/ha. The pots were then labelled and
randomly distributed in a greenhouse at Dalhousie University’s Bio-Environmental Engineering Centre (BEEC)
in Truro, Nova Scotia. The pots were irrigated via an overhead system at four-hour intervals and subjected to
ambient lighting and temperatures. The pots were incubated over 50 days, from May 22 to July 7, 2014. After
the incubation period, the biomass was harvested, dried for 48 h at 60°C and weighed. Biomass modelling
was performed, including up to third-order interaction and third-degree polynomial terms and the standard
second-degree and second-order models, to evaluate the performance of more advance models over the
simpler models.
Percentage Rate per field Field equivalent Field equivalent

Percentage of
Amendment in pot mix plot – wet weight – dry weight
blend (w/w)
(w/w) (kg/4 m2) (Mg/ha) (Mg/ha)
Alkaline-
stabilised 0% – 100% 0% – 15% 0 – 60 0 – 150 0 – 100
biosolids
Municipal solid
0% – 100% 0% – 15% 0 – 60 0 – 150 0 – 82
waste compost
Wood shavings 0% – 10% 0% – 1.5% 0–6 0 – 15 0 – 11

*Mine soils represent rock fines (Blocks 1 and 2) and overburden (Blocks 3 and 4) stockpiles. Amendments used are alkaline-stabilised biosolids
(ASB), municipal solid waste compost (MSW) and wood chips (WC).
At each block location, the same 25 m × 15 m area used for the greenhouse material collection was ripped
to approximately 20 cm depth and graded prior to the study to alleviate subsurface compaction and prepare
a seed bed. Within each block, 30 plots of 2 m × 2 m were demarcated. Each plot represented one
experimental unit under the MA design. In total, 14 amendment combinations were generated under D-
optimal design criterion for MA designs, and the set of 14 was duplicated within each block to generate
goodness of fit statistics. Two unamended controls were included in each block. Amendment materials were
weighted in the field, mixed in a wheelbarrow and raked into each plot by hand in the upper 5 – 10 cm of the
soil profile. After incorporation, the plots were seed with the Nova Scotia highway mix (Table 1) at the
equivalent of 60 kg/ha and covered with 3 – 5 cm of straw mulch. The plots were established on August 14,
2014, and aboveground biomass was harvested from a randomly placed 30 cm × 30 cm quadrant on
November 12, 2014. The collected biomass was dried for 48 h at 60°C and subsequently weighed.

Block 1 Block 2 Block 3 Block 4 Combined

Parameters
GH Field GH Field GH Field GH Field GH Field
Model fit
R2 0.75 0.46 0.80 – 0.79 0.58 0.81 0.63 0.45 0.45
RMSE 0.37 1.36 0.28 – 0.67 1.06 0.29 1.43 0.66 1.25
Model (p) 0.008 0.017 0.002 – < 0.0001 0.036 < 0.0001 0.012 < 0.0001 < 0.0001
Lack of fit
–
Max. R2 0.67 – 0.71 0.76 0.54 0.54
–
Fit (p) 0.13 – 0.11 0.16 0.37 0.08
Prediction profile
Rate 88 94 79 – 81 125 84 95 82 104
(Mg/ha)
Biosolids 63 57 58 – 60 57 57 52 59 57
(%)
MSW (%) 36 43 42 – 40 40 41 48 40 42
WW (%) 1 0 0 – 0 3 2 0 1 1
Est. yield 2.21 1.64 1.89 – 3.91 1.73 2.21 2.80 2.52 2.21
(Mg/ha)
Desirability 0.77 0.79 0.75 – 0.70 0.67 0.77 0.86 0.48 0.71
Significant model terms*
Term B B B – B B B B B B
p-value < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001
M M M – M M M M M M
0.001
B*M 0.0002
B*M <0.0001
B*M – 0.02
B*M <0.0001
B*M 0.004
B*M <0.0001
B*M <0.0001
B*M <0.0001
B*M
0.0007 0.047 0.0036
R*B – 0.007
R*B 0.025
R*B 0.020
R*B 0.008
R*R 0.003
R*B <0.0001
R*B
0.049 0.020 0.046 0.045 0.044 0.01 0.02
*RMSE = root mean squared error; R = Rate (Mg/ha); M = municipal solid waste compost (%); B = biosolids (%); W = wood shavings (%)

Case Studies
Aboveground biomass data from the pot and plot studies were used for model building to determine the
optimum amendments required for maximum growth. The model fit statistics, prediction profile and
significant model terms are presented in Table 4. For the greenhouse study, the inclusion of third-order
polynomial terms and third-degree interaction effects appeared to result in model overfitting since the
models included numerous terms that varied substantially among the blocks. However, overfitting was not
explicitly tested with a validation data set. Restricting the greenhouse data to second-order polynomials and
second-degree interaction terms resulted in similar predictive model performance and similar significant
model terms (Table 4). Subsequently, the field plots were designed to include only up to second-order
polynomials and second-degree interaction terms, which allowed for duplicate experimental units to be
established at each block location so that lack of fit terms could be calculated.
The percentage of ASB and MSW compost and the interaction between biosolids and MSW compost were
included as significant terms in all models. Most models, other than the Block 1 plot and field models,
included either a rate and biosolids percentage term or a quadratic rate term (Table 4). The greenhouse
models generally had better fits, as determined by the coefficient of determination (R2) and root mean
squared error (RMSE) than field plot models when created on an individual block basis. However, the models
that were fit when grouping all of the substrate blocks in either the greenhouse or field plot data were
comparable with respect to the model fit parameters. Where lack of fit statistics were calculable, none of the
models tested were likely to be improved by the inclusion of another unknown parameter.
All of the models created for the greenhouse and field-plot data were optimised based on a blend of biosolids
and MSW compost versus the application of either amendment separately. In general, biosolids were
included at a greater proportion in the optimum amendment blend, ranging from 52% – 63% of the mixture,
and MSW compost was optimum at lower proportions, ranging from 36% – 48% of the mixture (Table 4).
Wood shavings were included in the mixture in five of the nine models generated and ranged from 1% – 3%
of the mixture where they were included. The optimised rate of amendment application was generally
comparable within the greenhouse data and within the field-plot data, but the field-plot data generally had
8 to 44 Mg/ha greater application rates than the comparable greenhouse model. The desirability of the
optimised amendment ratios was comparable among the models, ranging from 0.48 to 0.86. The desirability
can be interpreted as the proportion of the maximum observed value that was achieved by the experimental
units by the predicted estimate. Thus, the predicted biomass under the optimised amendment conditions
was generally 48 to 86% of the maximum observed yield.
The utility of optimisation-centric RSM experimental designs for determining soil amendment application
rates in mine soil reclamation programs was evaluated in this study. Specific objectives were to evaluate
whether standard MA designs versus higher-order designs are suitable for optimisation, and to compare
greenhouse versus field experiments on the optimised amendment mixture and application rates. The RSM
framework, specifically MA designs, was demonstrated to be suitable for optimising aboveground growth of
the selected vegetation species under both greenhouse and field conditions in this study.
The MA subset of RSM was used in this study because the independence of each organic amendment on the
effect of aboveground biomass could not justified and the logistics of amendment acquisition, transport and
application at the operating quarry better support a blending approach. Thus, the optimum proportion and
application rate of the blended amendment was desired. Multiple instances of blended amendments in
practice exist in the literature, and most report that a blend of amendments yields improved plant
performance over singular amendments (Van Ham et al., 2007; Larney et al., 2011). However, these studies
tend to select the proportions of amendment by treatment-comparison experiments, soil physicochemical
calculations, or scientific judgement rather than optimisation-centric experimentation models (Brown et al.,
2005; van Ham et al., 2007; Larney and Angers, 2012). In this study, MA designs rooted in process

optimisation statistics were found to be effective for blended amendment optimisation. Such experiments
offer multiple benefits to amendment determination:
 Fewer experimental units are required in comparison to factorial designs.
 Multiple amendments can be compare individually and in combinations to test their interaction.
 Prediction models generated are better suited for optimising plant performance than the
treatment/mean comparison experiments commonly used in reclamation practice.
This study demonstrated that a blend of amendments, specifically ASB and MSW compost, resulted in greater
plant performance that either amendment applied singularly. Although both amendments contain organic
matter and nutrients, they have unique properties that act synergistically to promote plant growth. Alkaline-
stabilisation processes require the addition of lime and other alkaline by-products, such as kiln dust or coal
combustion fly ash, and pasteurisation at temperatures exceeding 52°C. In addition to organic matter and
nutrients beneficial for plant growth, ASB may enhance physicochemical soil properties by adding soil fines
that can improve soil microporosity, increase cation-exchange capacity and water retention, and provide a
source for micronutrients (Ram and Masto, 2010). However, due to the combined pH elevation and
pasteurisation process, ASB is relatively devoid of microorganisms (Bean et al., 2007). MSW compost, on the
other hand, has reported benefits for stimulating soil microbial communities; the high content of humified
organic matter adds organic matter and nutrients, improves water and nutrient retention and promotes soil
aggregation (Rate et al., 2004; Larney and Angers, 2012; Nair and Ngouajio, 2012; Ohsowski et al., 2012). The
combined effect of both amendments introduced to the mine substrate appears to result in a synergistic
effect on plant growth.
Despite differences in soil texture and nutrient content among the substrates used in this study, the
optimised amendment blend did not vary substantially between the collection locations or between the
greenhouse and field settings. Consequently, it appears that the particular blend of ASB and MSW has a more
dominant effect on plant growth than location-specific soil or meteorological factors. None of the substrates
in the quarry was suitable for sustaining plant growth, as all amended soils had substantially greater plant
growth than unamended controls. Using a range of substrate types as replicates yielded a statistical model
that could generate an optimised amendment mixture with site-wide applicability that covers the range of
soil conditions possible on the site. In addition, a site-wide amendment is logistically easier to implement as
separate blends and rates would not be required in practice. Thus, a combined model encompassing all
replicates, here being material types, is preferred as it covers the range of responses available for the
amendment combinations (Gilmour and Trinca, 2006).
Although the proportions of amendments in the mixture were relatively consistent between greenhouse and
field studies, the optimised application rate for the compiled data models differed by 22 Mg/ha, with more
amendment being required in the field setting to provide optimum growth. This observation could be due to
a variety of circumstances arising in the field, including the effect of fluctuating temperatures and moisture
contents on nutrient cycling and uptake, nutrient leaching from the root zone or fluctuations in plant
available water (Chintala et al., 2012; Sevilla-Perea et al., 2014). Higher field application rates of organic
amendments, in turn, are expected to reduce soil water and temperature extremes, provide additional
nutrients to compensate for those lost due to leaching and have a more stimulating effect on soil microbial
communities in the root zone (Gardner et al., 2010; Nair and Ngouajio, 2012; Larney et al., 2012). The results
of this study suggest that amendment application rates should ideally be determined in a field setting, but
conducting a full scale RSM experiment in the field is more onerous than in a greenhouse setting. Since the
composition of amendments is stable among the greenhouse and field studies, the optimum amendment
mixtures can be determined in a controlled greenhouse environment, and the mixture can then be tested at
various rates in field plots to optimise the field application rates.

Case Studies
Response surface methodologies offer a promising experimental alternative for optimising mine soil
amendments to improve plant growth performance in mine soil reclamation programs. When multiple
amendments offering similar benefits to plant performance are being evaluated, mixture-amount (MA)
designs are suitable to account for the logistical and theoretical assumptions made in the investigation. Here,
an MA experiment was performed to optimise a blend of municipal organic waste by-products for enhancing
soil reclamation at an operating aggregate quarry. Based on the results of this investigation, the following
recommendations for applying MA experiments to soil reclamation programs can be made:
 The MA design should not target polynomial or interaction terms extending beyond the second
order or degree due to observed model overfitting when including higher-order terms.
 Randomised complete blocks defined by multiple substrate types should be used as replicates to
inform a site-wide amendment model rather than performing substrate-specific optimisations.
 Greenhouse studies are adequate for determining the optimum mixture of amendments but do
not yield optimum application rates when compared to the field settings.
 Greenhouse-optimised blends can be trialled at various rates in a field setting to determine
optimum application rates.
Ascough, J.C., Fathelrahman, E.M. and Hoag, D.L. (2013) Using response surface methodology for economic and environmental trade-
offs at the farm level, Air, Soil and Water Research, Vol. 6, pp. 73–89.
Bean, C.L., Hansen, J.J., Margolin, A.B., Balkin, H., Batzer, G. and Widmer, G. (2007) Class B alkaline stabilization to achieve pathogen
inactivation, International Journal of Environmental Research and Public Health, Vol. 4(1), pp. 53–60.
Bolan, N., Kunhikrishnan, A., Thangarajan, R., Kumpiene, J., Park, J., Makino, T., Kirkham, M.B. and Scheckel, K. (2014) Remediation
of heavy metal(loid)s contaminated soils – to mobilize or to immobilize? Journal of Hazardous Materials, Vol. 266, pp. 141–
166.
Brown, S.L., Henry, C.L., Chaney, R., Compton, H. and DeVolder, P.S. (2003) Using municipal biosolids in combination with other
residuals to restore metal-contaminated mining areas, Plant and Soil, Vol. 249, pp. 203–215.
Brown, S., Christensen, B., Lombi, E., McLaughlin, M., McGrath, S., Colpaert, J. and Vangronsveld, J. (2005) An inter-laboratory study
to test the ability of amendments to reduce the availability of Cd, Pb, and Zn in situ, Environmental Pollution, Vol. 138, pp.
34–45.
Brown, S., DeVolder, P., Compton, H. and Henry, C. (2007) Effect of amendment C:N ratio on plant richness, cover and metal content
for acidic Pb and Zn mine tailings in Leadville, Colorado, Environmental Pollution, Vol. 149, pp. 165–172.
Chintala, R., McDonald, L.M. and Bryan, W.B. (2012) Effect of soil water and nutrients on productivity of Kentucky bluegrass system
in acidic soils, Journal of Plant Nutrition, Vol. 35, pp. 288–303.
long-term liabilities: a guidance document, Prepared for the National Orphaned/Abandoned Mines Initiative (NAOMI) by
Cowan Minerals Ltd, Sudbury, ON.
Drozdowski, B.L., Naeth, M.A., and Wilkinson, S.R. (2014) Evaluation of substrate and amendment materials for soil reclamation at a
diamond mine in the Northwest Territories, Canada, Canadian Journal of Soil Science, Vol. 92, pp. 77–88.
Environmental Protection Agency (EPA) (2007) The use of soil amendments for remediation, revitalization and reuse, Publication No.
EPA 542-R-07-013, Solid Waste and Emergency Response (5203P), National Service Center for Environmental Publications,
Cincinnati, OH, USA. 59 p.
Fan, J., He, Z., Ma, L.Q., Nogueira, T.A.R., Wang, Y., Liang, Z. and Stoffella, P.J. (2012) Calcium water treatment residue reduces copper
phytotoxicity in contaminated sandy soils, Journal of Hazardous Materials, Vols. 199-200, pp. 375–382.
Gardner, W.C., Broersma, K., Naeth, A., Chanasysk, D. and Jobson, A. (2010) Influence of biosolids and fertilizer amendments on
physical, chemical and microbiological properties of copper mine tailings, Canadian Journal of Soil Science, Vol. 90, pp. 571–
583.
Gilmour, S.G. and Trinca, L.A. (2006) Response surface experiments on processes with high variation, in A.I. Khuri (ed), Response
surface methodology and related topics, World Scientific Publishing Co., Hackensack, NJ., pp. 19–46.
Houben, D., Pircar, J., and Sonnet, P. (2012) Heavy metal immobilization by cost-effective amendments in a contaminated soil: Effects
on metal leaching and phytoavailability, Journal of Geochemical Exploration, Vol. 123, pp. 87–94.
Kalali, A., Ebadi, T., Rabbani, A., and Sadri Moghaddam, S. (2011) Response surface methodology approach to the optimization of oil
hydrocarbon polluted soil remediation using enhanced soil washing, International Journal of Environmental Science and
Technology, Vol. 8(2), pp. 389–400.

Koo, N., Jo, H-J., Lee, S-H. and Kim, J-G. (2011) Using response surface methodology to assess the effects of iron and spent mushroom
substrate on arsenic phytotoxicity in lettuce (Lactuca sative L.), Journal of Hazardous Materials, Vol. 192(1), pp. 381–387.
Kumpiene, J., Lagerkvist, A. and Maurice, C. (2008) Stabilization of As, Cr, Cu, Pb and Zn in soil using amendments – a review, Waste
Management, Vol. 28, pp. 215–225.
Larney, F.J., Janzen, H.H. and Olson, A.F. (2011) Residual effects of one-time manure, crop residue and fertilizer amendments on a
desurfaced soil, Canadian Journal of Soil Science, Vol. 91, pp. 1029–1043.
Larney, F.J. and Angers, D.A. (2012) The role of organic amendments in soil reclamation: a review, Canadian Journal of Soil Science,
Vol. 92, pp. 19–38.
Nair, A. and Ngouajio, M. (2012) Soil microbial biomass, functional microbial diversity, and nematode community structure as
affected by cover crops and compost in an organic vegetable production system, Applied Soil Ecology, Vol. 58, pp. 45–55.
Naseri, E., Reyhanitabar, A., Oustan, S., Heydari, A.A. and Alidokht, L. (2014) Optimization arsenic immobilization in a sandy loam soil
using iron-based amendments by response surface methodology, Geoderma, Vols. 232-234, pp. 547–555.
Ohsowski, B.M., Klironomos, J.N., Dunfield, K.E. and Hart, M.M. (2012) The potential of soil amendments for restoring severely
disturbed grasslands, Applied Soil Ecology, Vol. 60, pp. 77–83.
Ram, L.C. and Masto, R.E. (2010) An appraisal of the potential use of fly ash for reclaiming coal mine spoil, Journal of Environmental
Rate, A.W., Lee, K.M. and French, P.A. (2004) Application of biosolids in mineral sands mine rehabilitation: use of stockpiled topsoil
decreases trace element uptake by plants, Bioresource Technology, Vol. 91, pp. 223–231.
Schrevens, E. and Cornell, J. (1993) Design and analysis of mixture systems: applications in hydroponic, plant nutrition research, Plant
and Soil, Vol. 154, pp. 45–52.
Sevilla-Perea, A., Hernandez-Soriano, M.C. and Mingorance, M.D. (2013) Carbon mineralization from sewage sludge-amended mine
dump by response surface methodology, International Journal of Environmental Science, Vol. 12, pp. 125–138.
Sevilla-Perea, A., Almendros, G. and Mingorance, M.D. (2014) Quadratic response models for N and P mineralization in domestic
sewage sludge for mining dump reclamation, Applied Soil Ecology, Vol. 75, pp. 106–115.
Solis-Dominguez, F.A., White, S.A., Borillo Hutter, T., Amistadi, M.K., Root, R.A., Chorover, J. and Maier, R.M. (2012) Response of key
soil parameters during compost-assisted phytostabilization in extremely acidic tailings: effect of plant species, Environmental
Science and Technology, Vol. 46, pp. 1019–1027.
Van Ham, M.D., Teshima, M.A. and Dampier, L.M.A. (2007) Biosolids use in gravel pit reclamation, in Proceedings – Wastewater
Biosolids Sustainability: Technical, Managerial and Public Synergy, International Water Association (IWA) Specialist
Conferences, 24–27 June 2007, Moncton, NB.
Zanuzzi, A., Arocena, J.M., van Mourik, J.M. and Faz Cano, A. (2009) Amendments with organic and industrial wastes stimulate soil
formation in mine tailings as revealed by micromorphology, Geoderma, Vol. 154, pp. 69–75.

M. Dey SRK Consulting (UK) Ltd., United Kingdom

C. Williams SRK Consulting (UK) Ltd., United Kingdom
C. Brough SRK Consulting (UK) Ltd., United Kingdom
R. Bowell SRK Consulting (UK) Ltd., United Kingdom
D. Runciman Wastes Ecotech SRL, Romania
V. Turea Wastes Ecotech SRL, Romania
The former Bicapa–Tarnaveni sodium dichromate production plant in Romania was closed in 2001 after
almost 50 years of operation, leaving behind a legacy of hexavalent chromium-contaminated groundwater
and an exposed waste storage facility (WSF). The WSF contains chromite ore processing residues (COPR)
consisting of residual hexavalent chromium, partially processed chromite ore and other waste products.
During operation, the COPR was pumped as slurry into the WSF and dewatered. The reclaimed water was
then returned to the processing plant. The current estimate for the closure of the WSF is approximately
US$ 31 million for a geotextile cover. This would encapsulate the WSF to minimise meteoric flushing of the
WSF, thereby significantly reducing further mobilisation of hexavalent chromium. In addition to the cover, a
pump and treat scheme would also be needed to remediate the contaminated groundwater within the vicinity
of the operation. It is likely that this would operate in perpetuity due to the potential for further release of
hexavalent chromium from the WSF. Whilst this standard approach would significantly reduce the risk to the
environment, it would not remove it.
As an alternative solution, Wastes Ecotech is currently assessing the viability of reprocessing the waste,
thereby not only recovering the chromium from the contaminated site but also turning it into a commodity.
In addition to the chromium, both magnesium and calcium may also potentially be recovered as saleable
commodities. The reprocessing option would also utilise the contaminated groundwater within the processing
circuit and eliminate the requirement for a further treatment scheme. If proven to be financially viable, this
solution would not only remediate the site but would also permanently remove any risk associated with
chromium from the site.
This paper provides background information on the site, an assessment of the reprocessing technologies and
a provisional assessment as to the viability of the alternative solution.
When Wastes Ecotech purchased the former Bicapa–Tarnaveni sodium dichromate plant site in Romania,
the company knew that it would require remediation, as the former site owners’ closure plan was to close
the site gate and walk away. The site is heavily contaminated with chromium, and the options for the site are
to either contain or potentially reprocess the waste.
The biggest environmental concern is the mobilisation of hexavalent chromium from the former works
(Figure 1). Hexavalent chromium is observed as a pollutant within the groundwater and as solid residue on
former buildings and rubble surfaces.
An independent assessment of the site was undertaken prior to the purchase. The recommendation was that
the WSF should be capped with a geomembrane to minimise infiltration of precipitation into the WSF and
thereby minimise mobilisation of hexavalent chromium from the WSF. In addition, a groundwater pumping

The rehabilitation of the Bicapa–Tarnaveni waste storage facility M. Dey, C. Brough, R. Bowell, D. Runciman and V. Turea
and treatment solution should be installed to remediate the already contaminated groundwater. The
estimated cost for these remedial works was approximately US$ 30 million.
15 cm
However, after their purchase of the site, Wastes Ecotech srl contracted the local university to look at the
potential of extracting value from the wastes. In addition to the chromium, both magnesium and calcium
were identified as recoverable commodities. This reprocessing option means that the metals contained in
the WSF could potentially be extracted economically, and thereby the WSF could be considered as a mineral
resource rather than a liability.
The former Bicapa–Tarnaveni chemical works is located near the town of Tarnaveni in Transylvania, Romania,
within the valley floodplain. It is approximately 37 km southeast of the main city of Targu Mures. The Bicapa
chemical site was a former state-owned processing facility that opened in 1916 and produced a number of
products until 2001, when it was closed.
With its long history, the site has seen multiple chemical works on site: from 1916 to 1922, the chemical
works produced gases and explosives; from 1922 to 1948, it manufactured nitrates and fertilisers. Within the
same time period, it also produced ceramics; gold, silver, copper and lead were processed here between
1939 and 1948. From 1950 to 1955, a factory was built specifically for the production of sodium dichromate,
and a second dichromate plant was built between 1971 and 1975. This second plant required the redirection
of the river to provide additional space, the building of a new river dam facility, and the construction of three
storage pits (two solely for dichromate waste and one for carbide process waste) (see Figure 2). Between
1968 and 1986, the chemical plant produced zinc oxides, barium salts, formic acid, hydrofluoric acid, creolite,

Case Studies
Freon-11 and Freon-12. The site as a whole ceased production and closed in 2001. To date, no remediation
of the site has been undertaken.
LEGEND 3
Dichromate Process
Plant
2
1
Waste Storage
Facility
Rough Location of
Former River Valley
Conceptual Cross Y
Section
SCALE
0 100m 200m
Chromite ore was transported to the site via railway from mines in Kazakhstan. Between 1955 and 2001,
dichromate was the primary product produced. All waste products produced by the dichromate factories
between 1970 and 2001 were stored within the WSF, which currently contains an estimated 3.5 million
tonnes (wet) of material. The WSF is a facility adjacent to the former processing plant and is contained by

external concrete retaining walls (Figures 2 and 3). The WSF is underlain by thick layers of alluvial clays and
muds that have minimised the leaching of chromium to groundwater and the nearby river.
Chromium is the 22nd most abundant element in Earth's crust with an average concentration of 100 ppm
(Mason, 1966). Naturally occurring chromium exists predominantly in its chromic (chromium III) form, whilst
the chromate (chromium VI) state is rarely found in nature and is produced mainly from industrial processes.
The processing of chromite ore (FeCr2O4) to produce dichromate salts (Na2Cr2O7) was classically undertaken
using the high lime process whereby the chromite was calcined with sodium carbonate (Na2CO3) in air. Under
these conditions, the chromite is oxidised to chromium (VI), and iron is released as an iron (III) oxide by a
process such as that shown in Equation (1). Adding hot water to the resultant product extracted the
chromium (VI) (forming sodium chromates), leaving behind the relatively insoluble components such as iron,
aluminium, calcium, magnesium and silicon oxides or hydroxides. Typically, the chromate solution is
processed further to produce a purified sodium dichromate salt (Anger et al., 2005). The solution is acidified
with carbon dioxide (Equation 2) or sulphuric acid (Equation 3) to produce the dichromate. This is
subsequently isolated as the dihydrate by crystallisation (Freemann, 2004).
4FeCr2O4 + 8Na2CO3 + 7O2 → 8Na2CrO4 + 2Fe2O3 + 8CO2 (1)
2Na2CrO4 + 2CO2 + H2O → Na2Cr2O7 + 2NaHCO3 (2)
2Na2CrO4 + H2SO4 + 2O2 → Na2Cr2O7 + Na2SO4 + H2O (3)
However, in the case of the Bicapa plant, a local dolomite was used as a lime replacement. Ultimately, the
calcined dolomite residues report to the WSF.
Globally, the production of sodium dichromate, particularly by the high lime process, produced a legacy of
millions of tonnes of chromite ore processing residue (COPR) distributed near and around urban areas (Burke
et al., 1991). The COPR is typically highly enriched in chromium (VI), which is highly toxic, and strict
environmental controls now govern the acceptable concentration of chromium (VI) in groundwater and
surface water that may be affected by COPR. The legacy associated with worldwide COPR has resulted in
detailed research into numerous potential methods to stabilise chromium (VI), typically through reduction
to chromium (III). These methods use nanoscale iron particles (Cao and Zhang, 2006), iron sulphate
(Geelhoed et al., 2003), calcium polysulfide (Graham et al., 2006; Wazne et al., 2007; Moon et al., 2008),
asphalt in combination with iron sulphate (Moon et al., 2009), pyrolysis process with sewage sludge (Zhang
et al., 2009), starch (Wang et al., 2007; Xu et al., 2011), geopolymers (Sun et al., 2014) and bioremediation
(Dhal et al., 2013).
The processing of waste from the historical Bicapa–Tarneveni processing plant is a good case study of a COPR
site with ongoing land management concerns. To minimise further environmental risk and maximise the
potential land use, the chromium (VI) contamination on the site must be appropriately remediated. The
premise of this paper is that remediation of the site can be achieved by reprocessing the waste rather than
by stabilising chromium (VI) to chromium (III), with the added benefit of recycling products. Reprocessing of
COPR has only been partially covered in research to date, which has focused on the recovery of chromium
(Sreeram and Ramasami, 2001; Meegoda et al., 2008; Meegoda and Kamolpornwijit, 2011).
The different oxidation states of chromium exhibit different properties, which are important for assessing
the risk of potential harm to human health. Chromium (III) is an essential trace element involved in the
metabolism of glucose and fats. In foodstuffs, chromium is generally present as chromium (III), and the body
absorbs about 0.5% to 1% of the chromium present in our diet. Yet chromium (VI) compounds can cause
allergic responses (e.g., asthma-like symptoms and allergic dermatitis) in short-term exposure of sensitised

Case Studies
individuals. Long-term exposure of the skin to chromium (VI) compounds can cause sores and ulcers, which
may penetrate deep into the skin if left untreated. Chromium (VI) compounds are classified by the
International Agency for Research on Cancer (IARC) as causing cancer in humans if they are breathed in,
whereas chromium (III) compounds could not be classified and are therefore not considered to cause cancer
(Public Health England, 2012).
Due to the low solubility of chromium (III), its toxicity is limited. In contrast, chromium (VI) is very mobile in
the aquatic environment and highly toxic (Dayan and Paine, 2001). Strict environmental controls on this form
of chromium minimise the risk to human health (UK DEFRA, 2005; European Chemicals Bureau, 2005).
The potential proposed reprocessing solution involves a multistage process that starts with full hydration of
the solids and partial extraction of the highly soluble chromium salts (Figure 4). For the hydration/initial
dichromate mobilisation stage, both thermally pressurised and atmospheric mixing methods have been
evaluated. Simple atmospheric hydration was determined to be the favoured process option. In addition to
hydration, the waste sodium carbonate is added to the solids to promote the mobilisation of the chromium
salts whilst also promoting the carbonation of both the magnesium and calcium. Liquids and solids are then
separated by pressure filtration, and the solutions from this preliminary extractive step are set aside for
chromium recovery through ion exchange; in addition, the filter cake is also backwashed.
The hydrated filter cake then undergoes several sequential carbonation steps to mobilise the magnesium.
This starts with the addition of fresh water and pressurisation with carbon dioxide. The solution is then
recovered before the solids are recarbonated to produce a second magnesium-rich solution. The solutions
obtained are then passed through an ion-exchange stage to recover further chromium, and then they are
decarbonated. Next, the solutions are degassed and heated to 50°C to 60°C to promote precipitation of
magnesium carbonate. The magnesium carbonate solids are then recovered and are thermally decomposed
at 850°C to produce the magnesium oxide product. The preliminary results for this magnesium oxide product
give about a 60% recovery of the magnesium at a product grade of about 96%. Further efficiency
improvements and final product refinements are currently under investigation.
The solid residue from the carbonation steps then undergoes a calcination process at 950°C, and this is
followed by an ammonium chloride (NH4Cl) leaching step to recover calcium and residual magnesium.
Ammonium monophosphate (NH4H2PO4) is then added to the leachate to promote the formation of struvite
(NH4MgPO4) and precipitate the residual magnesium. Calcium is then recovered as a calcium chloride
solution. Further optimisation of this stage in the process is being investigated; possible improvements
include recovery and recycling of the ammonia and refining of the final products. The residue from this
process is then leached with sulphuric acid to mobilise the remaining dichromates and generate a chromium
salt-free waste.
Finally, at all stages in the process, all the dichromate-containing solutions are subjected to ion exchange to
recover the chromium. These various ion-exchange units are periodically backwashed to regenerate the ion-
exchange resins and release the chromium. Current results suggest that the ion exchange is very efficient;
over 99% of the mobilised chromium is recovered, but only 90% of the chromium is mobilised, as the
remaining chromium is probably still present as the original chromite ore. The use of a fixed-bed ion-
exchange unit yields a very pure sodium dichromate product.

There are two principal expected waste components: a solid waste containing silica, sodium carbonate and
sodium sulphate; and a small waste effluent that will also generate a solid waste. Both of these will require
long-term storage. The pilot plant is currently being tested to verify field production and performance
economics; the initial assessment indicates that 60% of magnesium is recovered as oxide with a high purity,
90% of chromium is recovered with very high purity (>99%), and 70% of calcium is recovered as a 40% w/v
chloride solution. These are early stage results, and further optimisation work is still required to improve
recovery rates and remove the impurities present in the magnesium and calcium products.

Case Studies
Two principal excavation methods could be undertaken to reprocess the waste, namely excavator and truck
or hydraulic mining.
Excavator and truck mining has the advantage of requiring less capital outlay, as it can be done through
subcontracting, and it provides greater flexibility. But there are potential traffic and dust issues. Dusts could
be supressed with water mists, but these would need to be supplied from clean water to minimise the risk
that contaminated mists could be dispersed in the presence of the excavator and truck operators. Traffic
issues could be resolved with the use of geogrids combined with geomembranes to disperse the vehicle
loading on the soft ground.
Hydraulic mining would require a significant volume of water, and this could potentially make use of the
contaminated groundwater as the operators are remotely located during the hydraulic mining method.
Whilst hydraulic mining has the potential to further contaminate the groundwater if it is not fully contained,
it produces no dusts and does not require vehicular access on the waste.
At present, the hydraulic mining method is the preferred option for several reasons:
 It could be potentially cheaper, as no access roadways will need to be developed.
 The risk to human health can be contained.
 The risk of greater environmental impact is reduced.
Regardless of the mining option selected, the existing contaminated groundwater will need to be pumped
and treated. This water could potentially be used within the processing plant, thus extracting the chromium
(VI) contained within, rather than requiring a separate water-treatment scheme.
To assess the financial benefits of reprocessing the wastes, the WSF was drilled and assayed to determine
the potential resource. As the WSF was constructed on an historical floodplain, the underlying clays and muds
have formed a natural attenuation barrier for the waste. This can be seen in Figure 5, where the mud and
clay layer has a distinctive brown colour, whereas the chromite wastes have a distinctive green basal layer.
Due to the variation in deformation of these sediments, this change in colour was used to define the base of
the WSF when calculating the volume of the material contained. Because the thickness of the clay base layer
was unknown, the drilling penetration onto this basal layer was minimised to ensure that the clay barrier
remained intact.
cm

The distribution of chromium is fairly even across the WSF, but the notable change in colour at the basal
sediments also shows an accumulation of chromium products within the layer. This is illustrated in the
downhole assays depicted in Figure 6. Whilst the x-ray fluorescence (XRF) analytical method reports major
elements as oxides due to the preparation methodology, this does not mean that chromium is present as a
chromium III oxide (Cr2O3) within the waste.
Cr2O3 (wt%)
0 5 10 15
0%
F01
10% F02
F03
20% F04
F05
F06
30% F07
F08
40% F09
Depth (%)
F10
50%
60%
70%
80%
90%
100%
Based on the current assessment, it is only possible to classify the entire WSF based on the inferred mineral
resources (Table 1). As magnesium is potentially the most valuable commodity to be recovered, the grade
results are reported as magnesium equivalents. The magnesium equivalent conversions for chromium and
calcium were 10.49 and 0.61, respectively.
Category Dry tonnage (Kt) Mg equivalent (%) Cr (%) Mg (%) Ca (%)

Inferred 2,000 42 1.7 14.5 16.7
*Mineral resources are not mineral reserves and do not have demonstrated economic viability. All figures
are rounded to reflect the relative accuracy of the estimate.
For the initial assessment of the economic evaluation, it is assumed that operational costs would be
approximately US$ 168 per tonne of material; this includes mining and processing costs and associated
overheads. Based on these preliminary cost estimates, the current cutoff grade is 5.6% magnesium
equivalents. As the estimated grade of the WSF is 42% magnesium equivalent, there is a significant potential
for economic gain.

Case Studies
Currently, the former Bicapa–Tarnaveni sodium dichromate plant site is a significant environmental risk and
could pose a substantial liability to its owners. Two potential solutions are currently under consideration:
 containing the WSF, with ongoing treatment of seepages to the groundwater; or
 reprocessing the wastes.
In terms of containing the waste, the site investigation/resource drilling has confirmed that the underlying
existing containment of the wastes by the clays and muds is very effective. Therefore, the engineered
containment of the wastes would result in a minimal disturbance of the site. However, the engineered
containment would only be effective for the life of the cover. The proposed engineered cover composed of
geotextiles has a finite life, and as the chromium contamination would remain at site, any failure of the cover
would mean that chromium could be readily mobilised again. In addition, there will always be some seepage
generation from the WSF, mainly as a result of infiltration through the engineered cover, as no engineered
cover is 100% effective; therefore there will be a need for ongoing water treatment. The initial cost estimate
of US$ 30 million for the engineered cover does not include ongoing water treatment costs, so an additional
liability cost will be incurred.
In contrast, reprocessing the wastes recovers the chromium as a saleable product and thereby removes it
from the site. Early financial assessment suggests that this will also provide a financial benefit. If selected,
this solution could potentially generate financial gains rather than consume finances. Ultimately, this will
significantly reduce, if not remove, the liability associated with the site. In addition, careful management of
the groundwater could potentially allow this to be utilised as a resource. Employing this water within the
reprocessing of the waste will both treat the water and recover further chromium. This could then remove
the need for any ongoing water treatment and remove any associated ongoing remediation costs.
However, reprocessing the wastes requires that they be disturbed, and this must be carefully managed in
order to minimise the risk of further contamination of the site. Good stewardship of the site will be required,
and an operating permit will be necessary in order undertake the reprocessing.
This project is still at an early stage, but the assessment undertaken to date suggests that there is significant
benefit — both financial and environmental — to be gained from reprocessing the wastes.
Anger, G., Halstenberg J., Hochgeschwender K., Scherhag C., Korallus U., Knopf H., Schmidt P. and Ohlinger, M. (2005) Chromium
compounds, Ullmann's encyclopedia of industrial chemistry, Wiley-VCH, Weinheim, Germany, pp. 637–640.
Burke, T., Fagliano, J., Goldoft, M., Hazen, R.E., Iglewicz, R. and McKee, T. (1991) Chromite ore processing residue in Hudson Country,
New Jersey, Environmental Health Perspectives, Vol. 92, pp. 131–137.
Cao, J. and Zhang, W-X. (2006) Stabilization of chromium ore processing residue (COPR) with nanoscale iron particles, Journal of
Hazardous Materials, Vol. B132, pp. 213–219.
Dayan, A.D. and Paine, A.J. (2001) Mechanisms of chromium toxicity, carcinogenicity and allergenicity: Review of the literature from
1985 to 2000, Human and Experimental Toxicology, Vol. 20(9), pp. 439–451.
Dhal, B., Thatoi, H.N., Das, N.N. and Pandey, B.D. (2013) Chemical and microbial remediation of hexavalent chromium from
contaminated soil and mining/metallurgical solid waste: a review, Journal of Hazardous Materials, Vols. 250–251, pp. 272–
291.
European Chemicals Bureau (2005) European Union risk assessment report: chromium trioxide, sodium chromate, sodium
dichromate, ammonium dichromate, potassium dichromate, European Communities, Luxembourg.
Freeman, F. (2004) Sodium dichromate, Encyclopedia of reagents for organic synthesis, L. Paquette (ed), J. Wiley & Sons, New York.
Geelhoed, J.S., Meeussen, C.L., Roe, M.J., Hillier, S., Thomas, R.P., Farmer, J.G. and Paterson, E. (2003) Chromium remediation or
release? Effect of iron (II) sulfate addition on chromium (VI) leaching from columns of chromite or processing residue,
Environmental Sciences Technologies, Vol. 37, pp. 3206–3213.
Google Earth 7.1. (2014) Bicapa-Tarnaveni chemical plant, waste storage facility and surrounding area 46°19'02.85"N,
24°16'03.08"W, elevation 295M, Viewed 04 December 2014.
Graham, M.C., Farmer, J.G., Anderson, P., Paterson, E., Hillier, S., Lumsdon, D.G. and Bewley, J.F. (2006) Calcium polysulfide
remediation of hexavalent chromium contamination from chromite ore processing residue, Science of the Total Environment,
Vol. 364(1–3), pp. 32–44.

Mason, B. (1966) Principals of geochemistry, 3rd edition, John Wiley & Sons, New York, 329 p.
Meegoda, J.N. and Kamolpornwijit, W. (2011) Chromium steel from chromite ore processing residue, Frontiers of Environmental
Science and Engineering in China, Vol. 5(2), pp. 159–166.
Meegoda, J.N., Kamolpornwijit, W., Hu, Z., Mueller, R. and Patraju, R. (2008) Conversion of chromium ore processing residue to
chrome steel, in Proceedings of GeoCongress 2008: Geotechnics of Waste Management and Remediation, 9–12 March 2008,
Louisiana, M.V. Khire, A.N. Alshawabkeh and K.R. Reddy (eds), Geotechnical Special Publications (GSP), pp. 383–390.
Moon, D.H., Wazne, M., Jagupilla, S.C., Christodoulatos, C., Kim, M.G. and Koutsospyros, A. (2008) Particle size and pH effects on
remediation of chromite ore processing residue using calcium polysulfide (CaS 5), Science of the Total Environment, Vol.
399(1–3), pp. 2–10.
Moon, D.H., Wazne, M., Koutsospyros, A., Christodoulatos, C., Gevgilili, H., Malik, M. and Kalyon, D.M. (2009) Evaluation of the
treatment of chromite ore processing residue by ferrous sulfate and asphalt, Journal of Hazardous Materials, Vol. 166(1), pp.
27–32.
Public Health England (2012) Chromium: general information, viewed 2 May 2015, https://www.gov.uk/government/publications/
chromium-general-information-incident-management-and-toxicology.
Sreeram, K.J. and Ramasami, T. (2001) Speciation and recovery of chromium from chromite ore processing residues, Journal of
Environmental Monitoring, Vol. 3, pp. 526–530.
Sun, T., Chen, J., Lei, X. and Zhou, C. (2014) Detoxification and immobilization of chromite ore processing residue with metakaolin-
based geopolymer, Journal of Environmental Chemical Engineering, Vol. 2(1), pp. 304–309.
United Kingdom, Department for Environment, Food and Rural Affairs (DEFRA) (2005) Environmental risk reduction strategy and
analysis of the advantages and drawbacks for hexavalent chromium, Risk & Policy Analysts, London.
Wang, T., He, M. and Pan, Q. (2007) A new method for the treatment of chromite ore processing residues, Journal of Hazardous
Materials, Vol. 149, pp. 440–444.
Wazne, M., Jagupilla, S.C., Moon, D.H., Jagupilla, S.C., Christodoulatos, C. and Kim, M.G. (2007) Assessment of calcium polysulfide for
the remediation of hexavalent chromium in chromite ore processing residue (COPR), Site Remediation Technologies, Vol.
143(3), pp. 620–628.
Xu, W., Li, X., Zhou, Q., Peng, Z., Liu, G. and Qi, T. (2011) Remediation of chromite ore processing residue by hydrothermal process
with starch, Process Safety and Environmental Protection, Vol. 89, pp. 179–185.
Zhang, D., Kong, H., Wu, D., He, S., Hu, Z. and Hu, X. (2009) Remediation of chromite ore processing residue by pyrolysis process with
sewage sludge, Bioresource Technology, Vol. 100, pp. 2874–2877.

C. Wells Aboriginal Affairs and Northern Development Canada, Canada
Following the discovery of gold in the Yellowknife, Northwest Territories, Canada, area in the 1930s, Giant
Mine officially opened in 1948. Gold at Giant Mine was found locked in minerals, which needed to be roasted
at extremely high temperatures. Unfortunately, this roasting process also released gases with a highly toxic
by-product, arsenic trioxide. Throughout the 1950s, controls were put in place that minimised emissions to
the air; however, this also resulted in the collection of 237,000 tonnes of highly toxic arsenic trioxide dust. At
the time, scientists and government agencies agreed that storing the waste in underground stopes and
chambers was an appropriate long-term management alternative.
When ore processing ceased in 1999, the care and control of the mine fell to the Department of Aboriginal
Affairs and Northern Development Canada, and attention was focused on the environmental issues left
behind, including the arsenic trioxide stored in underground chambers. The Giant Mine remediation project
was created in 2005 with the overall goal to protect human health and safety, and the environment. To do so
requires the long-term containment and management of the arsenic trioxide waste, ongoing water treatment
and clean-up of the surface elements of the site. The main objectives of the Giant Mine remediation project
are to minimise risks to public and worker health and safety, minimise the release of contaminants from the
site to the surrounding environment, remediate the site in a manner that encourages public confidence, and
implement an approach that is cost-effective and robust over the long term.
The project has recently completed an environmental assessment process under the Mackenzie Valley
Resource Management Act, the governing legislation in the Northwest Territories for projects with the
potential to have an impact on land or water. The project team is now proceeding with a clearly defined list
of requirements established through the process of the project, but faces many challenges going forward,
including technical considerations, regulatory and jurisdictional constraints, consultation and engagement
requirements and resource pressures. It will require a great deal of ingenuity, planning and collaboration to
address these challenges and deliver a successful project for the remediation of the Giant Mine site.
Giant Mine was one of the longest operating and richest gold mines in Canadian history, producing more
than 220,000 kg of gold between 1948 and 2004. While prospectors en route to the Klondike first discovered
gold in Canada’s Northwest Territories (NWT) in 1896, accessibility limited exploration until the mid-1930s.
Gold was first discovered at what is now the Giant Mine site in Yellowknife, NWT (see Figure 1), in 1935, by
prospectors working for Burwash Yellowknife Mines Ltd., whose assets were acquired by Yellowknife Gold
Mines Ltd. in 1937.

Giant Mine remediation project C. Wells
The mine operated under the subsidiary Giant Yellowknife Gold Mines Ltd. until 1940, when the company
could no longer support operations. The mine was stagnant until 1944, when Frobisher Explorations took
over the site based on promising exploration data. Giant Mine officially opened shortly after the end of World
War II, and production began in earnest, with the first gold brick poured on June 3, 1948. From May to
December of that year, the mine produced 8,152 ounces of gold from 49,985 tonnes of processed ore. The
mine continued to produce gold under various operators until 1999, when it ceased operations after Royal
Oak Mines Ltd. went into receivership. At that time, custodianship of the 950 ha site was transferred to the
Department of Aboriginal Affairs and Northern Development (AANDC) on behalf of the Government of
Canada. Miramar Mining Corporation continued to remove ore from the site and process it at the nearby
Con Mine until 2004. The company terminated its obligations under the reclamation security agreement in
2005 and Giant Mine officially became an abandoned mine site, with responsibility for its closure and clean-
up resting with the Government of Canada.
Owing to the length and period of operation, and the refractory nature of the ore at the site, the potential
environmental impact of Giant Mine could be much greater than that of many other mining sites without
proper management, care and maintenance. As with any mining operation, Giant Mine generated a
substantial amount of contaminated waste during the extraction, handling and processing of ore at the site.
There are four tailing ponds with an estimated 16 million tonnes of tailings, as well as an estimated 325,000
m³ of contaminated soils to be remediated or risk managed. The mine water requires treatment before it
reaches the receiving environment of Great Slave Lake. There are numerous buildings in varying states of
disrepair that pose a risk to human health and safety and must be dealt with as part of the remediation plan.
A problematic by-product of mining at Giant Mine is the significant quantity of arsenic trioxide dust. Gold at
the site is found in the refractory mineral arsenopyrite; the ore must be roasted at extremely high
temperatures to extract the gold. This converts the arsenic in the mineral to arsenic trioxide, which is highly
toxic.
At the Giant Mine site, this arsenic trioxide gas by-product and other contaminants were released directly
from the roaster stack, resulting in approximately 7,400 kg of arsenic emissions per day until 1951, when a
Cold Cottrell Electrostatic Precipitator (ESP) was installed. This had the effect of removing a significant portion
of the arsenic trioxide from the roaster emissions, reducing average daily arsenic releases to 2,900 kg per day
by 1956. In 1958, a Dracco baghouse was added to great effect, reducing airborne arsenic emissions to 52 kg
per day by 1959. Over the years, arsenic emissions would fluctuate based on production levels (SENES and
SRK, 2010).
The dust produced at Giant Mine contains approximately 60% arsenic by weight and can be harmful even in
very small amounts. It is water soluble, and can easily make its way into the environment if not managed

Case Studies
properly. Over the operating period of the mine, it is estimated that approximately 237,000 tonnes of arsenic
trioxide dust was collected and stored underground. When collection of the dust started in the 1950s, the
mine operators, scientists and government officials had to decide what to do with the arsenic trioxide dust
produced by the ESPs. They determined that placing it in mined-out stopes and chambers within the mine
itself would provide the best overall storage solution. When the mine eventually closed, it was believed that
the permafrost would eventually re-establish around these stopes and freeze the dust in place.
When AANDC was given care and control of the site in 1999, the department took steps to address the issue
of the arsenic trioxide waste and other hazards at the site. The focus was on addressing the highest risks to
human health and the environment first, with efforts directed at selecting the long-term management
strategy for the dust stored underground, as well as remediation to mitigate immediate risks posed by tailings
deposited along the shore of Back Bay. Over the next several years, AANDC retained specialists to conduct
the necessary studies and options analysis, and presented the findings to both independent experts and the
public for their input. After starting with an initial 54 possible remediation options, AANDC announced in
2004 its intention to proceed with the frozen block method to stabilise the arsenic trioxide in place as a way
to effectively mitigate the risk to the environment while ensuring the protection of human health and safety.
The site lies within the Mackenzie Valley watershed, and is regulated by the Mackenzie Valley Resource
Management Act (MVRMA). The MVRMA is federal legislation aimed at protecting the lands and waters
within the Mackenzie Valley watershed. Since the site is under the care and custodianship of AANDC, it is
also subject to other federal acts, such as the Canadian Environmental Protection Act, the Fisheries Act and
the Migratory Birds Convention Act, among others. It is also situated within the municipal boundaries of the
City of Yellowknife, and so is impacted by the City’s bylaws and permitting requirements.
Between 2005 and 2007, AANDC worked with experts to develop a remediation plan for Giant Mine that
included a proposed approach to deal with the underground arsenic trioxide as well as the various surface
features, such as the pits, tailings ponds, contaminated soils and surface water. In accordance with the
regulatory requirements in the NWT, the project submitted a water licence application to the Mackenzie
Valley Land and Water Board (MVLWB) in 2007 to seek approval for the remediation plan and secure a water
licence to implement the project.
The MVLWB conducted a preliminary screening of the application and indicated that the project could
proceed to the regulatory licencing process. However, the City of Yellowknife intervened under its own
authority and referred the project to an environmental assessment (EA), a process intended to ensure it
would not result in adverse environmental effects. The Mackenzie Valley Environmental Impact Review
Board (MVEIRB) began the assessment process in April 2008. It took until the end of 2008 to finalise the scope
of the assessment and until May 2009 to establish a terms of reference and work plan. At that point, the
project commissioned the developer’s assessment report (DAR), which essentially formed the remediation
plan that would undergo assessment, and then submitted the document to MVEIRB in October 2010.
Information requests, responses and technical sessions took place throughout 2011 and into early 2012,
resulting in a series of technical reports from interested parties later on that year. Public hearings were also
held in 2012, and the public record was closed in October of that year. It was re-opened in February 2013 to
address an additional information request from the MVEIRB, after which it was closed again. The MVEIRB
released its EA report in June 2013 (MVEIRB, 2013). A number of conditions (measures as defined by the Act)
contained in the report were reviewed, and alternate wording proposed. The MVEIRB responded with
suggested modifications to the measures in February of 2014, and in August of 2014, the federal ministers
responsible for the EA issued their final decision on the report of environmental assessment agreeing to 26
legally binding measures, eight of which were revised from those originally proposed.

A number of the measures will need to be completed, or substantially completed, prior to submitting an
updated water licence application for the project. Specific components of the project require stakeholder
input before being included in the updated consolidated project description, and the measures require the
creation of an independent oversight body to provide input into options analysis as well as oversee research
into a long-term solution for the arsenic trioxide dust.
In addition to the requirements from the EA, the project will require a land use permit from the MVLWB for
specific project activities, and various development and demolition permits from the City of Yellowknife. Each
process has its own consultation and engagement requirements that the project must follow in order to
obtain the necessary permits. Consultation, while necessary, important and beneficial, will have an impact
on the overall project schedule and could have cost implications depending on the final remediation plan
that is approved.
It is also worth noting that the various regulatory regimes, while intended to be complementary, can work at
cross-purposes. The MVRMA is intended to mitigate the environmental impacts of land use activities, while
the city bylaws relate to ensuring land use activities occur in an orderly and planned manner. Section 98 of
the MVRMA states that the MVLWB does not have jurisdiction “in respect of the use of land within the
boundaries of a local government to the extent that the local government regulates that use” (Government
of Canada, 2014). In 2011, the MVLWB and the Government of the Northwest Territories, represented by the
Minister of Municipal and Community Affairs (MACA) made a joint determination that allows both the City
of Yellowknife and the MVLWB to regulate the same activity under different processes, as they felt the city
bylaws did not adequately cover the types of environmental considerations that would be addressed by a
land use permit issued by the MVLWB. The only way to harmonise the two processes would be an
amendment to the city bylaws, which could have a significant impact on smaller development projects.
Although the mine is no longer in operation, it is also subject to the metal mining effluent regulations (MMER)
under the Fisheries Act because it continues to discharge mine water effluent. Even though the effluent is
treated prior to release, it is a release of a potentially deleterious substance that could trigger the “harmful
alteration, disruption, or destruction (HADD) of habitat” (Government of Canada, 2002). The effluent can be
discharged as long as the mine “operator,” in this case, AANDC, conducts the required environmental effects
monitoring (EEM). This includes the regular monitoring of effluent and water quality, and biological
monitoring of the receiving environment.
Various other permits or authorisations may be required during the life of the project, such as research
permits (wildlife, scientific, medical), archaeology permits and migratory bird permits.
The project has several major components that will form the overall remediation plan for the site, which
are described in more detail in the following sections. The specific scope for each component will be
determined through implementation of the EA measures and consultation with stakeholders.
As mentioned previously, an estimated 237,000 tonnes of arsenic trioxide dust are stored in stopes and
chambers within the former mine workings. Considerable effort has gone into determining an appropriate
method to manage the risk this material poses, resulting in the selection of the frozen block method. This
approach is a means of freezing the rock around the chambers containing the arsenic trioxide dust and then
maintaining this frozen state through the use of thermosyphons designed for either active or passive heat
transfer. Thermosyphons are a type of heat pump designed to remove heat from the rock surrounding the
arsenic stopes to create an impermeable frozen zone that will completely encapsulate the arsenic dust
indefinitely.
To determine the design parameters for the final freeze system, a freeze optimisation study (FOS) was
conducted on one of the arsenic chambers (Chamber 10). This study involved a full-scale installation of 38
drill holes and freeze pipes, as well as more than 30 instrumented drillholes to collect data during operation

Case Studies
of the system. The FOS was constructed in 2009 and 2010, and testing was initiated in early 2011. The system
was operated continuously during that time, testing active, passive and hybrid operation of the system, as
well as various configurations for the system itself. The objectives of the study included:
 demonstrating large-scale ground freezing;
 estimating parameters needed for engineering design;
 testing implementation methods;
 developing monitoring and data handling methods;
 identifying constraints and opportunities related to procurement and project delivery;
 examining “unknown unknowns,” that is, project uncertainty and unexpected design issues.
The advantages and disadvantages of various design options identified through the FOS, as well as the EA
process, were assessed through a series of trade-off studies. The following studies and variants were
examined:
 wet vs. dry frozen blocks — freezing the dust with or without wetting;
 surface (vertical) and underground (horizontal) piping, or surface only;
 temperature of freezing coolant;
 conversion of system from active to passive
 active and/or passive freezing;
 application of surface amendments;
 requirements for bulkhead design.
The results of the FOS are being assessed at the time of this writing, and will inform the final design of the
freeze system going forward in 2015.
Baker Creek is a 2,520 m channel that traverses the site from the north to the south and discharges into Back
Bay, which is part of Great Slave Lake. The channel has been altered significantly over the course of the mine’s
operation to accommodate mining, processing and road construction. A description of each of the main
reaches of Baker Creek is provided in Figure 2.

The alteration of Baker Creek has created a number of issues that need to be addressed by the project. Water
and sediment quality are significantly impacted by historical spills, the discharge of treated mine water and
offsite inputs upstream, which also impact habitat development and quality. As well, certain reaches of the
creek pass close to pits, and there is a risk of flooding the pits, which could result in uncontrolled flooding of
the mine during severe storm events (greater than one in 500 years). This could result in the uncontrolled
release of arsenic into the environment. As well, the stability of the C1 pit wall adjacent to Baker Creek is
deteriorating, the failure of which would increase the risk of uncontrolled flooding of the mine workings.
A number of options for Baker Creek exist, and it is a requirement of the EA to examine the options in
consultation with regulatory authorities and stakeholders, and specifically with the input from an
independent oversight body prior to making the final decision. The options span from re-routing the entire
creek to a new route that avoids the mine site entirely, to making modifications to certain reaches that
minimise the risk of uncontrolled flooding of the mine and maximise the quality of water and sediment that
ultimately discharge into Great Slave Lake.
Mine records indicate that arsenic from the tailings effluent was controlled by treatment starting in the
1950s, but the details of the treatment and its effectiveness were not described. There was also a new water
treatment circuit commissioned in 1967, using lime to precipitate arsenic from the mill tailings stream before
discharging it into the tailings pond, thus reducing the concentration of dissolved arsenic in the tailings
effluent. The precipitated arsenic was disposed of with the mill tailings.
Water from the mine was pumped directly to Baker Creek without treatment until 1981, when a new water
licence required mine water to be treated to improve the quality of effluent released to the environment.
Pilot scale testing completed in collaboration with Environment Canada resulted in a new effluent treatment
plant starting operation in 1981. Ferric iron and lime were used to precipitate arsenic and other heavy metals.
This was preceded by the destruction of cyanide using first alkaline chlorination, then hydrogen peroxide
oxidation starting in 1990.
The need to regulate flows led to storing mine water in the tailings ponds prior to treatment, which
significantly reduced the available tailings storage capacity. In 1997, a mine water treatment circuit was
installed in the mill, which resulted in the treated water being used in the mill process, reducing the mine’s
consumption of fresh water. Since 1999, when the processing of ore at the site was discontinued, mine water
has been pumped to the south, north and northwest ponds for storage, and then treated in the existing water
treatment plant prior to discharge to Baker Creek during the summer months.

Case Studies
The decision on the final configuration of Baker Creek will be impacted by, and will impact, decisions about
other water management issues at the site. As indicated, surface water running onto the site can have
moderate to high levels of arsenic that contribute to the overall loading of arsenic in water that eventually
discharges from the site. It is also expected that mine water will remain impacted above acceptable levels
that will prevent discharge without prior treatment. This is directly affected by the EA measures, which
require mine effluent to be treated to Canadian drinking water standards prior to discharge into Yellowknife
Bay. A new effluent treatment plant will be designed and built to meet the treatment standard. However, to
minimise the costs of storing, handling and treating water, it will be necessary to limit the input of clean
surface water into the mine. While runoff from tailings ponds and other areas may require treatment, the
intent will be to divert as much water as possible from entering the mine and thus having to be treated.
Capping systems for the tailings designed to maximise runoff efficiency and minimise contact with
contaminated soils and tailings will also contribute to reducing the amount of water requiring treatment.
Storage of water prior to treatment is another decision point for the project, and is dependent on other
factors such as the progress of the freeze. If the mine were to be used for water storage versus surface ponds,
it would require allowing the water level in the mine to rise from its current level of 10 m below the 750 level
(at C-Shaft). The decision to treat water year-round or seasonally is an important design consideration,
because large fluctuations in the water level in the mine that would result from seasonal treatment could
result in an increase in arsenic being discharged from the mine, and could impact the structural integrity of
the backfilled chambers, bulkheads and crown pillars supporting the mine.
Approximately 9.5 million dry tonnes of tailings were originally deposited in the north, central and south
ponds, of which about 7 million tonnes of tailings remain. Between 1988 and 1990, 2.5 million tonnes of
these tailings were reprocessed in the Tailings Retreatment Plant (TRP) and transferred to the Northwest
Pond, to which another 6 million tonnes were added through conventional mill production. The tailings in
these ponds cover a combined surface area of almost 100 hectares, with a maximum tailings thickness of
about 22 m in the central pond.
The tailings are impounded by several dams, meant to contain the tailings and minimise seepage of pore
water. Seepage collection systems were constructed at the perimeter dams where seepage did occur, and
the dams are inspected annually. The performance and safety of these dams was formally reviewed in
September 2004, and no immediate safety concerns were identified. Recommendations to assess dam
performance in more detail and improve operating, maintenance and surveillance procedures have been
implemented as part of the ongoing care and maintenance of the site.
The ultimate remedial concept for the tailings is a cover to prevent human or wildlife contact, and to
encourage vegetation growth and shed runoff/prevent infiltration of surface water. The final design will be
dependent on several factors, and will be determined once the final footprint of the tailings areas is
identified. Tailings are currently being used in the production of paste used to stabilise the underground
workings (further described in section 2.3.5), which also serves to reduce the final volume and aerial extent
of tailings that will require covering.
The mining activities at Giant Mine were meant to be temporary while extracting the gold-bearing ore, with
the mine to be closed using standard practices of the day. However, the last owner went into receivership
and left the site in an operable state, and AANDC took on care and control of the mine and eventual closure.
The underground workings were safe while the mine was in operation, but since active mining ceased in
1999, the underground voids have started to weaken and have the potential to collapse, which has led to the
appearance of sinkholes and subsidence on the surface. As discussed, the collapse of bulkheads or the surface
into the underground chambers, stopes or tunnels could result in injury to workers or visitors to the site, loss
of infrastructure, uncontrolled flooding of the mine and/or a release of arsenic trioxide directly to the
surrounding environment.

To increase the stability of the underground mine workings, the project has started to fill the voids using a
paste backfill delivered through surface drillholes. This process has been used by operating mines for the last
two decades, and as a component of mine reclamation for more than 10 years. The paste is a mixture of
water, cement and tailings in a specifically designed ratio to provide underground structural support that,
once hardened in place, will serve to stabilise the site (Caulfied, 2014).
Owing to the lack of detailed plans of the mine workings, and without the ability to physically verify the paste
delivery for safety reasons, the process is being carefully controlled and monitored using remote video
monitors. The stabilisation of the voids is necessary to secure the site as well as prepare the stopes containing
arsenic trioxide for eventual freezing, and other areas for the potential controlled flooding of the mine.
There are eight open pits and 35 openings to the underground mine, all of which present safety hazards. The
overall mined volume of the pits is more than 2 million m³. Prior to addressing the pits, the openings to the
surface will be sealed in accordance with the Northwest Territories Mine Health and Safety Act and
regulations. The selection of methods must meet the regulations and achieve the objectives for strength and
durability, and is primarily based on the physical characteristics of the opening (size, geometry, location,
inclination) and the quality of the rock surrounding the opening. When the openings are no longer required
for access or ventilation, they will be sealed with structures requiring minimal maintenance to remain stable
and effective in the long term.
A number of the underground stopes are located directly below the pits, and so their long-term stability once
backfilled is a primary consideration when determining how to address the risks the pits themselves pose as
a site hazard. These include hazards related to the pit walls themselves, as well as risks related to water
management: uncontrolled water from below (the mine) could form contaminated pit lakes, and from above
(from Baker Creek) could enter the mine through the pits.
Waste rock will be used to seal up most of the lateral openings, including the portals located in the open pits.
This may also be suitable for sub-vertical openings, such as the raises and shafts. In some cases a concrete
cap may be placed over or inside the opening. The ground conditions would need to be inspected and
assessed for competency as part of the cap design process. In openings where the bedrock is not competent,
a reinforced concrete bulkhead may be constructed deeper inside the opening into sound bedrock.
There are considerable quantities of arsenic-impacted soil throughout the site, though the volumes that
exceed the relevant industrial standard have not yet been fully delineated. The EA requires the completion
of a human health risk assessment and quantitative risk assessment, which will help determine the final
strategy for the various areas of impacted soil based on land use expectations. If soils require remediation,
one potential solution will be to use the impacted material to backfill one or more of the existing pits.
AANDC’s responsibility for the site includes financial liability for the care and maintenance and mitigation of
environmental risks at the site. As the project has evolved, so have the estimated costs to address the issues
at the site. The most recent indicative cost estimates for the project were developed in 2012, and indicated
the total full-up project cost from 2000 to completion in the order of C$ 903 million. This number is subject
to change as a result of the EA and the finalisation of the remediation plan, but is considered reasonable
relative to the magnitude of the problems at the site. The primary challenge is to determine what funding
mechanism(s) will be used to deliver the project.
In response to the issue of major environmental liabilities held by the Government of Canada, due in large
part to abandoned mines sites like Giant Mine, the Federal Government announced the federal contaminated
sites action plan (FCSAP) program in 2004. This 15-year, multi-billion-dollar program has the overall goal of

Case Studies
reducing the risk to human health and the environment, and the associated financial liability associated with
these sites. To date, the project has received 100% of its funding from this program.
The limitation with this funding model is that Canada is a parliamentary democracy, and the sitting
government cannot encumber future governments, so the program is approved in five-year phases, with
funding approved on an annual basis. This makes it very difficult to secure long-term contracts for work and
obtain better value for money. It also means that projects are competing for resources as priorities change,
since there is no long-term commitment of stable funding.
As noted, the current funding program ends in 2020, which is when the Giant Mine project is expected to
start remediation, so a new, long-term, stable source of funds will need to be secured prior to starting active
remediation. This will be a major focus for the project team over the next few years, to ensure the overall
plan to manage and mitigate risks at the site can be accomplished in a reasonable timeframe.
The Giant Mine remediation project is a complex technical undertaking in mine closure. It is further
complicated by the variety of regulatory regimes under which it falls, the high profile it has in the public and
the constraints within the Government of Canada project approval process.
The project components for both the surface and underground remediation are interconnected in various
ways, and decisions made for one aspect of the project will impact the options available for other elements.
The project team will follow the regulatory process to update the remediation plan, obtain a water licence
and other required permits, and seek the necessary approvals to implement the project and reduce the risks
to health, safety and the environment for the long term.
Caulfield, P. (2014) Engineers use waste to support closed Giant Mine, Journal of Commerce, viewed 30 November 2014,
http://journalofcommerce.com/Projects/News/2014/11/Engineers-use-waste-to-support-closed-Giant-Mine-1004166W/.
Government of Canada (2002) Metal Mining Effluent Regulations (SOR/2002-222), viewed 31 January 2015, http://laws-
lois.justice.gc.ca/eng/regulations/SOR-2002-222/FullText.html.
Government of Canada (2014) Mackenzie Valley Resource Management Act (S.C. 1998, c. 25), viewed 31 January 2015, http://laws-
lois.justice.gc.ca/eng/acts/M-0.2/FullText.html.
Mackenzie Valley Environmental Impact Review Board (MVERB) (2013) Report of environmental assessment and reasons for decision,
Giant Mine remediation project, 20 June 2013 (EA 0809-001), http://www.reviewboard.ca/upload/project_
document/EA0809-001_Giant_Report_of_Environmental_Assessment_June_20_2013.PDF.
SENES Consultants Limited and SRK Consulting Engineers and Scientists (SENES and SRK) (2010) Indian and Northern Affairs Canada,
Giant Mine remediation project developer’s assessment report (DAR), viewed http://www.reviewboard.ca/upload/
project_document/EA0809-001_Giant_DAR_1288220431.PDF.


D. Kennard Golder Associates Ltd., Canada

C. MacInnis Aboriginal Affairs and Northern Development Canada, Canada
The former Giant Mine, now under the control of Aboriginal Affairs and Northern Development Canada
(AANDC) on behalf of the Federal Government, is one of Canada’s largest contaminated sites. A key
component of mine closure is to backfill underground voids to prevent collapse of portions of the mine that
could pose immediate risks to public and worker safety and lead to failure elsewhere. In the case of the Giant
Mine site (Site), there is the additional risk that underground collapse could allow surface water to flood the
mine eventually. Severe environmental consequences could result from such a flooding event due to the
presence of arsenic trioxide dust in the underground stopes. To mitigate risks related to underground stability
at the mine site, an innovative short-term advanced remediation backfilling program is underway, while the
long-term overall remedial plan for the site continues to be developed.
AANDC engineers working with Public Works and Government Services Canada, along with industry leaders
in mine closure, developed an innovative and efficient approach to stabilising the underground voids using
tailings paste backfill. Paste backfill is commonly used in operating mines to maximise production, but its use
in mine mitigation and remediation is limited. The Giant Mine project team utilised thousands of tonnes of
tailings that had been deposited on surface as a waste by-product during historical production mining as the
primary component of the backfill required to stabilise the underground voids. It is cost-effective to use on-
site tailings material for underground backfilling, but its use also reduces the future effort required to
remediate surface tailings pond areas.
The project was initiated by carrying out field-scale tests to determine first whether the on-site tailings could
be used to provide suitable backfill material and, second, what amount of binder (cement) would be required.
The early testing showed that suitable backfill material could be produced using the on-site tailings and a
small amount of cement.
Tailings were extracted from the surface ponds and paste was produced by adding water and cement to the
tailings in a mobile mixing system. The paste was delivered to the targeted underground voids via boreholes
drilled from surface. Backfill was contained within the targeted voids by constructing underground barricades.
Few examples of using frozen tailings material to make backfill in large volumes under extremely cold
temperature conditions exist.
Giant Mine (the Site) is an inactive gold mine located approximately 5 km north of the centre of
Yellowknife, Northwest Territories (see Figure 1).
Historically, the mine produced gold from 1948 until 1999, after which stewardship was transferred to the
Department of Indian and Northern Affairs Canada (INAC), now Aboriginal Affairs and Northern Development
Canada (AANDC). INAC immediately transferred management to Miramar Giant Mine Ltd., which ceased all
ore processing activities at the Site but continued to mine and transport ore to the neighbouring Con Mine
for processing until 2004. All mining activities ceased in July 2004, after which INAC reassumed stewardship
and the Deton’Cho/Nuna Joint Venture was retained to operate and maintain the Site in compliance with
current regulations. This group is termed the care and maintenance contractor.

Innovative process for stabilising the subsurface at the Giant Mine Site D. Kennard and C. MacInnis
Approximately 237,000 tonnes of arsenic trioxide dust (dust) produced as waste from roasting gold ore
extracted from Giant Mine was stored underground in mined-out voids, termed stopes in mining
nomenclature, and purpose-built chambers between 1951 and 1999. This dust is hazardous to both people
and the environment, and the long-term remediation plan is to freeze the dust.
Evolving conditions and events over the past few years at the Site have highlighted evidence of
deteriorating/weakening conditions in the underground workings. The stability of certain underground
mining excavations at the Site is of concern to AANDC as their failure would lead to physical surface hazards,
compromise the ongoing and planned remediation work, and release arsenic trioxide into the mine and/or
the environment. Specific arsenic-containing and non-arsenic-containing stopes have been prioritised as
mitigation targets, and this paper describes the general approach to stabilising these openings.
The location of near surface non-arsenic stopes and arsenic-containing stopes and chambers that might pose
risks to public and worker health and safety, the environment, and the Site remediation plan are dispersed
over the Site over a length of approximately 3 km, as shown in Figure 2. The near surface stopes that require
filling are shaded in Figure 2. Mine openings used to access the stopes are also shown in Figure 2.
The level of backfill or arsenic dust in the mined-out stopes and purpose-built chambers is variable, and some
crown pillar stability issues have been identified. Preliminary Site remediation plans concluded that
backfilling voids was the best risk mitigation and remediation approach. Approximately 600,000 m3 of backfill
is required to stabilise all near surface stopes and chambers. The stopes will be backfilled on a priority
sequence based on risk assessments that include an assessment of the probability of failure of the crown
pillars.

Case Studies
The B1-18 non-arsenic stope complex, which was a shallow dipping portion of the orebody mined using cut
and fill mining methods, was chosen as the first location to backfill. The stope complex is an interconnected
series of voids with a complex geometry (see Figure 3). Some of the voids are under a surface watercourse
(Baker Creek) and the risk of mine flooding due to crown pillar failure was a driver in choosing to backfill this
stope first. It was also shallow and relatively close to underground mine access.
Figure 4 shows an example of underground workings that were targeted to be backfilled.

The objective of the work was to tightly fill the stope voids under Baker Creek using a minimal amount of
material readily sourced from the Site and using readily available small-scale equipment that could be
operated by local contractors. Consultant reports describing the work were used to compile the information
in this paper (Golder Associates Ltd., 2014a, 2014b, and 2015).

The chosen backfill material was lightly cemented tailings paste (paste). Tailings, which is finely crushed rock
and is the main by-product of milling gold ore, was proposed as the primary backfilling material. It is abundant
on surface in the existing tailings ponds and is also present in large quantities in the underground as it was
used during historical mining to stabilise voids.
Paste backfill, like hydraulic fill, typically contains sand to silt sized soil particles, water, and possibly binder.
Sources of material can be naturally occurring, such as overburden, or man-made, such as tailings. Compared
to hydraulic fill, paste contains a larger fraction in the silt-sized particle range (less than 0.06 mm). The solids
content, or pulp density, is generally between 70% to 85%, but certainly can be higher or lower. The lower
water content and finer particle size compared to hydraulic fill give paste backfill the range of appearance
from a thin milkshake to toothpaste. A true paste backfill will not segregate, has a measurable yield stress,
and produces very little bleed water (the water available to form ponds). A minimum amount of fines (passing
20 micron) is required in the paste backfill to have reasonable pressure losses in any pipeline that carries it.
The velocity of paste backfill in the pipe under full column flow conditions ranges from 1 to 3 m/s but can be
higher or lower. Binder can be added so the paste attains strength after the binder has cured.
Geotechnical investigations and process design studies along with laboratory testing programs were
completed to analyse the tailings at the Site for suitability as a feed material for paste backfill. The paste was
generally made from a mixture of Site tailings, water, type GU cement (normal Portland cement), fly-ash, and
some rock aggregate. Figure 5 shows a photograph of paste backfill produced at the Site.

Case Studies
The paste was to be delivered to the underground via boreholes drilled from surface into the targeted voids,
either by gravity feed or by pumping under pressure.
The delivery boreholes were drilled over a relatively dispersed area, and therefore mobile delivery of paste
to the boreholes and flexibility in scheduling equipment cycle times was required. Fixed paste plants and
delivery pipelines were not deemed suitable for the Site.
Containment of the flowable paste in the complicated and interconnected voids required the installation of
fill barricades, and some areas of the B1-18 stope complex were not safely accessible due to the deterioration
of the ground support in localised areas.
While the paste is being delivered it is not safe to be in underground workings below or adjacent to the
containment barricades. A remote camera monitoring system was installed to monitor the performance of
the fill barricades and to monitor when spill ports in the fill barricades were leaking paste.
The list of main preparatory activities required prior to backfilling a void included:
 Drilling paste delivery and monitoring boreholes.
 Excavating and stockpiling tailings.
 Rehabilitation of underground workings and construction of underground fill barricades.
 Installation of barricade and stope void camera monitoring system.
Drilling of 150 mm (6″) diameter paste delivery boreholes was accomplished using track-mounted down-the-
hole hammer reverse circulation drilling equipment. Additional boreholes were also required to monitor the
paste backfilling progress and confirm that the voids were full to the specified level were drilled using HQ
sized diamond drilling equipment. In some cases very accurate drilling was required with a tolerable deviation
of only 1%. Figure 6 shows some drilling activities near the B1-18 stope complex area.
The paste delivery boreholes were lined with an inner sleeve/casing to enable pumping of paste at the
maximum pump pressure and not lose paste to cracks and fissures in the bedrock or the overburden.

Excavation and stockpiling of tailings was carried out using tracked equipment (see Figure 7). Limited
preparations for using wheeled trucks on some areas of the tailings were made and trucks were loaded in
the tailings basing and moved the material to the paste production sites. Wheeled equipment that took the
tailings to the paste production areas (see Figure 7).
Underground rehabilitation of the ground support systems to enable construction of fill barricades and
installation of the camera monitoring system was carried out prior to backfilling.
The fill barricade construction locations were chosen at natural choke points in connections between voids
that were targeted to be backfilled and those that were not. Also, some openings adjacent to the voids were

Case Studies
required for future access to lower portions of the mine. The fill barricades were not constructed to hold
pressure; rather, they provided formwork for the paste to be poured in lifts. Figure 8 shows an example fill
barricade constructed with a combination of run-of-mine rockfill and timbers. The figure shows ports cut in
the timbers to allow paste to escape so that excessive fluid pressures did not come to bear on the barricade.
Figure 8 also shows a paste delivery borehole intersecting the working.
A camera monitoring system was installed so that remote observation of the fill barricades and larger voids
could be executed safely. Infrared security type cameras were powered and signal cables routed to a
common observation point. Figure 9 shows a typical camera installation and the image viewing software that
allowed multiple cameras to be viewed at once.
Once all these components were in place, the backfilling of the stope voids could proceed.
The key components of the paste production and pumping system were open-ended mobile mixer trucks,
cement trucks for delivery of batched paste, and concrete pumps, all of which are common sights on

construction projects (see Figure 10). Multiple paste backfill set-ups were running at any one time, at local
target void areas.
The tailings and other components were screened to an appropriate size for the paste delivery holes as well
as the mobile mixing equipment to ensure large pieces did not enter the mixer trucks. This was typically in
the minus 2″ range assuming the paste delivery holes are 8″ diameter. Once the material was sufficiently
mixed in the mixer truck and consistency was within target, the material was transferred to the boreholes
via pump or, in some cases depending on the distance, it could be transferred via gravity flow at delivery
rates of 25 to 75 m3/hour per set-up. A preliminary workflow sheet, from excavation of the tailings in the
basin to placement in the voids, is shown in Figure 11.

Case Studies
The first stage of the backfilling program was a short-term field test conducted in October and December
2013 to confirm paste backfill mix designs and drift barricade construction methodology. During the field
test, several paste backfill mix designs were tested which included different binder contents, different solids
contents, with and without aggregate, and different binder types. The purpose was to test their suitability
for usage underground and to test the equipment and system performance. In addition, remote drift
barricade mix designs and the construction methodology were also tested. Expanding foam material was also
tested as a part of the field trial as an alternative to paste backfill remote barricades.
The field test proved successful in that suitable mix designs were identified and subsequently used in the
production program meeting the design criteria required. The remote drift barricade paste backfill mix
designs and construction methodology were also utilised successfully in the production program based on
the data obtained during the trial.
The field test results determined that mixing Site tailings, treated mine water, and normal Portland cement
binder at concentrations of between 1% and 3% (by weight) would produce paste backfill material suitable
for delivery to the underground voids through boreholes.

Paste mix designs had to vary to deal with variability in the grain size of the excavated tailings, the water
content of the excavated tailings, and the required slump for a particular application. High slump paste was
used for the majority of the bulk fill as it will flow far from borehole insertion points. The mobility of the paste
tailings limits the amount of drilling required. Low slump paste is required for those areas where greater
placement control is required or where remote fill barricades are required.
Figure 11 shows field paste batching trials for low cement content, high slump bulk fills (left) and high cement
content lower slump (right) remote barricade and leak plugging mixes that were tested in a steel shipping
container to simulate backfilling an underground tunnel.
Generally, a minimal amount of binder (cement and/or fly-ash) will be added to create the bulk high slump
(8” to 10”) mix – this would typically be in the 2% to 5% range depending on the particular stope and pour
circumstances. The low slump (4” to 7”) material was produced using an increased solids content, increased
binder content (cement and/or fly-ash) of typically 5% to 15%, and possibly aggregate up to 25% of the mix
as required.
In some specific cases, small volumes of additive chemicals were added to the paste to enhance performance
over limited applications. The additives include:
 cement curing accelerator;
 water reducer/superplasticiser; and
 cement delay admixture.
The accelerator allows the paste to set up and gain strength quickly and was used for multiple applications,
including when underground re-entry soon after placing paste is required. Superplasticiser enhances flow of
the paste through boreholes and underground pipelines/slicklines. Delay admixture delays the curing of the
paste for multiple applications, including long travel times for batched paste to the delivery point.
After all the required drilling and underground preparations and field trials of the paste mix designs were
completed, paste backfilling of the voids was initiated on 21 October, 2013.
Remote fill barricades were some of the earliest paste pours, and Figure 13 shows a partially complete
remote fill barricade in place.

Case Studies
Flexible adaptive management of the paste delivery system was required as multiple fill barricade pours that
needed to cure were happening at the same time and leakage from the stopes required different mixes.
Paste was produced from 21 October until 12 December, 2013. Most of the targeted voids were backfilled
with bulk fill prior to eventually stopping the work due to extremely cold weather. Figure 14 shows the
entrance to a targeted underground void choked off with cemented paste.
Once average daily temperatures fell below about −5°C, ice lenses began to form in the tailings. Figure 15
shows the formation of ice lenses in the tailings feedstock material that had to be thawed with warm water
prior to adding the cement.

Tailings were then moved into a storage building, but as colder weather began the ice lenses began to form
again. Paste production cased when the energy required to thaw frozen tailings and keep process water
warm became excessive due to the cold temperatures.
Paste production began to ramp up during the first few weeks of production, but once cold temperatures
were experienced the production rates fell. Figure 16 shows the average daily paste flowrates for the project
duration.
Figure 17 shows the distribution of the various paste slump mixes (left), and the distribution of cement
content for various paste slumps.

Case Studies
Quality control of the paste mixes included measurement of water content, slump, and uniaxial compressive
strength to test that the backfill adhered to design specification. UCS tests were run on cast cylinders at 1, 3,
7, and 28 days after batching. Figure 18 shows an example paste cylinder before and after UCS testing (left)
and the development of backfill strength for paste pumped into one particular delivery borehole over several
weeks (right). The diagram on the right shows the development of paste UCS in two parts: 1) the first a
representation of the relative volume (totalling 100%) of the paste poured or pumped into each borehole by
date and recipe of the pour; and 2) the laboratory UCS strength of the cast cylinders 1, 3, 7, and 28 days after
batching.

The main lesson learned is that backfilling targeted voids in an abandoned mine is challenging. Areas that
were expected to leak paste did not, and areas that were assumed to be constrained voids leaked extensively.
Notable achievements of the project were:
• Boreholes were drilled with a deviation tolerance of less than 1% to intersect paste backfill was pumped
underground workings 2.0m wide at a depth of 45 m.
• Over 1,500 cylinders were casted for strength confirmation.
• Five different paste backfill recipes were utilised for stope specific pour requirements.
• Remotely placed fill barricades were successfully constructed.
• Various conventional fill barricade construction techniques were developed with available materials and
non-specialised underground mining staff.
• Mobile paste mixing equipment provided the flexibility required to deal with backfilling a complex
underground mine void.
• 11,000 m3 of paste backfill system ran in 2.5 months, at times during very cold temperatures, including
pouring paste backfill at −40°C (see Figure 19).
• Over 1,500 cylinders were cast for strength confirmation.
Producing paste backfill from a mobile system using frozen arsenic-laden tailings at these temperatures had
not been done before (to our knowledge). An adaptive approach to delivering paste that met the
specifications required in the permits and licensing was required. This included recording temperatures of
feed materials and paste going into the ground, as well as tracking via camera and cavity monitoring pour
volumes within the stopes to guarantee tight filling to the back.
Filling the required voids and keeping other parts of the mine open and unhindered by paste called for
ingenuity and flexibility with not only paste recipes, but also the drilling, monitoring, and underground
distribution system logistics and planning. Much of the underground was inaccessible and mine plans were
not complete, so in some cases limiting the flow of paste into unknown areas of the mine required multiple
“if-then-else” strategies during operation.
Mobile paste backfill systems are feasible alternatives to fixed plants and can provide owners with flexibility
and adaptability to site conditions, while still maintaining the quality and reproducibility of a traditional paste

Case Studies
plant. The key is to understand the underground environment as well as the surface implications, for
example: set-up, moves, paste recipe development, and infrastructure or material constraints; and to
develop a specific plan to address these items in a timely and cost-effective manner. This is especially true if
the owner is local, regional, or federal government, which typically do not have the resources – technical or
otherwise – to engage in these activities themselves.
The authors would like to thank AANDC for its permission to present this information.
Golder Associates Ltd., 2014a, Paste production report, Giant Mine, B1-18 Stope complex, Report prepared for Public Works and
Government Services Canada, Rtp. 1314260010-113-R-Rev0-5000, Feb. 26, 2014.
Golder Associates Ltd., 2014b, Void backfilling work summary – Giant Mine, B1-18 Stope complex, Report prepared for Public Works
and Government Services Canada, Rtp. 1314260010-120-R-RevC-1000, Mar. 31, 2014.
Golder Associates Ltd., 2015, Quality assurance report: interim underground stability and geotechnical stability drilling, Report
prepared for Public Works and Government Services Canada, Rtp. 1314260010-166-R-Rev1-4000, Jan. 14, 2015.


M. Larivière Aboriginal Affairs and Northern Development Canada, Canada

D. Hango Public Works and Government Services Canada, Canada
C. Corrigan AECOM Canada Ltd., Canada
Giant Mine is located within the city limits of Yellowknife, Northwest Territories, Canada. It operated from
1948 to 1999, producing more than seven million ounces of gold during its lifespan. This orebody is such that
it required an oxidation process to extract the gold. Chemical roasting was the only efficient oxidation process
available when Giant Mine was developed. Roasting operations began in 1949, converting raw ore into
calcine, which was further processed, and sulphur dioxide and arsenic vapour, which were vented directly to
the atmosphere. The first air emissions controls were introduced in 1951 and captured arsenic in the form of
arsenic trioxide. The complex expanded over time as more efficient technologies became available and were
incorporated into the process train.
When control of the Giant Mine property was transferred to the federal department of Indian Affairs and
Northern Development in 1999, the roaster complex included some of the most contaminated structures in
Canada. It contained free and poorly contained arsenic trioxide dust, asbestos, cyanide and other hazardous
substances. As the complex progressively deteriorated over time, the risks it posed to neighbouring
communities, onsite workers and environment increased with the rising likelihood of building collapse and of
releases of arsenic, asbestos and cyanide. The decontamination and deconstruction of the roaster complex
was undertaken in response to these unacceptable risks to human health and the environment. The technical
aspects of demolishing such a highly contaminated complex were a significant source of difficulties, and
multiple measures, programs, plans and procedures were put into place to protect workers, the public and
the environment. The urgent nature of the work also introduced its own challenges to the planning,
contracting, community engagement, regulatory and execution requirements for the project. A functioning
project team, consisting of individuals from across governmental departments, communications officers,
health and safety officers, managers, consultants and contractors, was critical to successfully meeting the
multifaceted demands of this project. In the end, over a two-year period, the project team successfully
dismantled the 10 structures that comprised the roaster complex; segregated the non-hazardous and
hazardous wastes; and removed and containerised more than 3,000 tonnes of arsenic-contaminated
materials from the complex without adverse effect to the onsite workers, neighbouring communities or the
environment.
Giant Mine is located within the city limits of Yellowknife, Northwest Territories, Canada (Figure 1). It
operated from 1948 to 1999, producing more than seven million ounces of gold during its lifespan. This
orebody is such that it required an oxidation process to extract the gold from arsenopyrite, a mineral
containing arsenic, iron and sulphur. Chemical roasting was the only efficient oxidation process available
when Giant Mine was developed. Roasting operations began in 1949, converting raw ore into calcine, which
was further processed, generating sulphur dioxide and arsenic vapour that were vented directly to the
atmosphere. The first air emissions controls were introduced in 1951 and began capturing arsenic in the form
of arsenic trioxide. Arsenic trioxide is highly toxic: as little as 1 to 2.5 mg/kg of arsenic trioxide is a potentially
fatal dose (Agency for Toxic Substances and Disease Registry, 2014).

Decontamination and deconstruction of the Giant Mine roaster complex M. Larivière, D. Hango and C. Corrigan
The roaster complex was the group of buildings and associated infrastructure whose purpose was to separate
the gold from the Giant Mine ore and treat the air emissions to reduce the release of arsenic trioxide into
the atmosphere. The complex expanded over time as more efficient technologies became available and were
incorporated into the process train. According to some estimates, the roaster complex released as much as
20,000 tonnes of arsenic trioxide into the atmosphere (Canadian Broadcasting Corporation, 2014), whereas
237,000 tonnes were collected by the emissions controls measures and are currently stored underground,
inside mined out rock chambers (SENES and SRK, 2010).
The complex footprint was approximately 7,800 m2 and comprised the following 10 primary structures, as
depicted in Figure 1:
 the Allis-Chalmers (AC) roaster building;
 the calcine plant;
 the Dorrco roaster building;
 the Cottrell precipitator building;
 the Baghouse building;
 the Silo;
 the weight scale house;
 the stack fan building;
 the stack;
 the exterior flues and other ancillary infrastructure.
The silo and weight scale building were not part of the treatment process; they were added to the complex
at a time when the mine was attempting to sell arsenic trioxide commercially.
In 1999, the mine operator declared bankruptcy, and care, custody and liability of the site were transferred
to the federal department of Indian Affairs and Northern Development (now known as Aboriginal Affairs and
Northern Development Canada [AANDC]) as land administrator by default. When this mine site first came
into operation, the bonding requirements for operation to ensure mine clean-up were minimal, and mine
operators were not obligated to set funds aside for future mine reclamation. The legal requirements to this
effect in Canada have since changed, and mining companies are now legally bound to set aside bonding to
cover site reclamation.
The decontamination and deconstruction of the roaster complex was undertaken during a two-year period
starting in 2013. This paper discusses the assessment, regulatory, contractual, technical and climate risks and
challenges associated with the implementation of this project, including lessons learned.

Case Studies
When ore processing and maintenance of the roaster complex ceased in 1999 and control of the Giant Mine
property was transferred to AANDC, the roaster complex included some of the most contaminated structures
in Canada. It contained free and poorly contained arsenic-containing dust, asbestos, cyanide and other
hazardous substances. The buildings had been locked up and not maintained since 1999, and the original
structures (the AC roaster building, the Dorrco roaster building, and the calcine plant) were already 50 years

old. Structural assessments completed by AANDC indicated that the likelihood of building and associated
infrastructure failure was increasing with time, with increasing risks to neighbouring communities, onsite
workers and the environment of releases of arsenic, asbestos and cyanide in the event of failure (AECOM,
2010; SENES and SRK, 2010). At that time, the project team was already seeking regulatory approval to
implement the overall Giant Mine remediation plan. The risks of structural failure and of contaminant release
were deemed serious enough that a separate regulatory review process and approval was sought under
emergency circumstances.
Initial hazardous waste assessments in 1996 for the mine operator (Arctic Environmental Services Ltd., 1996)
and in 2003 for AANDC (North West Consulting Limited, 2003) identified the complex as containing large
amounts of arsenic-containing dust and asbestos-containing materials to differing levels, and as containing
other hazardous materials.
A number of structural evaluations of the complex were undertaken between 1997 and 2010 (Ferguson
Simek Clark, 1997; PWGSC, 2007; AECOM, 2010). These evaluations noted that gaps were present in the roof
and walls of several of the buildings due to asbestos-containing cladding falling off or breaking apart; interior
steel catwalks in the Dorrco roaster building were corroded and becoming structurally unsound; erosion was
visible on exterior and interior brick masonry of the stack, and portions of the steel collar at the top had fallen
off; and the East Cottrell flue – the flue connecting the Cottrell precipitator building to the stack – was in
extremely poor condition, with delaminating external spray-applied asbestos, and cracking and sagging of
the flue itself, which visibly contained a significant amount of residual arsenic trioxide waste. In September
2011, the structural observations, evaluations and concerns were compiled and formalised in an engineer’s
letter report. The report identified noticeable structural degradation in the roaster complex and
recommended that deconstruction of the flues and stack be planned within one year, and that the other
structures be deconstructed within two to three years (AECOM, 2011).
In response to the recommendations from structural evaluations, a detailed waste audit of the roaster
complex was completed immediately, with the goal of preparing a tender package to complete the
decontamination and deconstruction of the complex. The audit estimated that:
 approximately 1,400 m3 (in situ) of arsenic waste was contained in the kilns, tanks, reactors and
air-handling infrastructure and present on building internal surfaces;
 approximately 1,300 m3 (in situ) of arsenic-dust-impacted asbestos-containing material was
present within the complex and on internal building surfaces;
 an additional 850 m3 of porous structural materials were impregnated with hazardous levels of
arsenic;
 cyanide compounds were present in some buildings, mixed with arsenic in some process vessels
and asbestos in surface dust;
 the possibility existed for residual sodium cyanide or cyanide-containing liquids in cyanide circuits
in some buildings;
 approximately 60 m3 of other hazardous waste items were present in the complex (AECOM,
2012a).
The timing of the waste audit and the circumstances under which the work had to be completed were a
challenge. The work was completed in March 2012, when the average temperature was approximately
−20°C). This limited the time that the assessment workers could be safely inside the buildings in the required
respiratory and chemical-resistant clothing to complete the work on a daily basis. Furthermore, there was
limited lighting because there was no power supply to the complex, the buildings were sealed and could not
be opened up owing to the risk of release of asbestos and arsenic-containing dust, and setting up inside
lighting would have required a considerable effort and cost in terms of additional worker training to enter

Case Studies
the complex to set up power supply and equipment, and equipment purchasing (equipment would have to
be left inside because it would become contaminated). The level of debris strewn about the buildings was
such that even running power lines inside would have required a considerable amount of clearing for access.
As such, the assessment work was completed using flashlights and headlamps. Some areas could not be
safely accessed owing to the previously identified structural issues. Finally, in some buildings, particularly the
Cottrell precipitator and the Baghouse buildings, the internal structures and equipment containing arsenic
waste were sealed and coated with asbestos insulation impregnated with arsenic dust, which did not allow
for them to be safely opened up owing to the risk of air quality issues beyond the capacity of the provided
respiratory protection for workers (for example, arsine gas generation from heat generated by cutting, or
large airborne releases of asbestos and arsenic with no capacity for dust control). The high level of debris
present throughout the buildings precluded the use of airlines to complete the work under supplied air.
The climate, logistical and health and safety challenges under which the waste audit were completed
therefore resulted in some significant limitations to the ability to accurately quantify and assess the
conditions of waste, particularly arsenic waste, within the complex. The trade-off between project delay to
enable thorough contractor logistical support to complete a more thorough assessment versus proceeding
to design and tender with a limited assessment and waste quantity uncertainty is discussed further below.
In December 2012, following the detailed waste audit, the urgency to complete the work was re-iterated in
another professionally endorsed report attesting to the conditions and serious risks posed to the
environment and human health and safety (AECOM, 2012b). Prior to initiating the decontamination and
deconstruction work, the roaster complex was located in very close proximity to Highway 4 – the Ingraham
Trail – as shown in Figure 1. The public, including runners and cyclists, heavily used this highway. A partial
structural failure and release of arsenic and asbestos-containing dust was identified as a significant risk to
the public travelling along the highway, but also a risk to the community of Yellowknife and to the overall
environment. The potential for a fire at the AC roaster building, which was the largest of the buildings and
constructed of wood, was also identified as a risk for a significant release of arsenic and asbestos. On a more
localised scale, falling cladding from building or bricks from the stack, the dispersing of arsenic and asbestos-
containing dust originating from inside buildings because of openings, and delaminating of asbestos from the
exterior of the flues were identified as risks to onsite Giant Mine workers. The roaster complex needed to
come down; however, the planning and approvals for the implementation of Giant Mine remediation and
closure were several years away.
Work at Giant Mine is subject to review and approval by the Mackenzie Valley Land and Water Board
(MVLWB). Any work within the jurisdiction of the MVLWB that requires either the drawing of water or the
deposition of waste requires a water licence. This water licence is the instrument by which the MVLWB
authorises projects. For certain projects, the federal requirement for a comprehensive environmental
assessment under the Canadian Environmental Assessment Act (CEA, 2012) is also required prior to issuing a
water licence. This environmental assessment is designed to identify potential impacts of the project, as well
as measures to mitigate these impacts, the latter of which then become conditions of the water licence.
All remediation work at Giant Mine, including the roaster complex decontamination and deconstruction
work, was subject to an ongoing CEA environmental assessment by the Mackenzie Valley Review Board.
However, the Mackenzie Valley Review Board concurred that proceeding with the decontamination and
deconstruction of the roaster complex was necessary to protect human health and the environment, and the
decontamination and deconstruction of the roaster complex could be excluded from the scope undergoing
environmental assessment.
An application was submitted to the MVLWB to proceed to the licensing phase for the work without
preliminary screening or environmental impact assessment review pursuant to paragraph 119(b) (response
to an emergency circumstance) of the Mackenzie Valley Resource Management Act (MVRMA). The

application was accepted, and the project went through a concentrated and rigorous licensing process to
ensure that the deconstruction work could start as soon as possible.
Obtaining regulatory permits under an emergency provision of the MVRMA contributed to heightened
awareness and scrutiny during stakeholder engagements and licensing reviews. There was understandable
public concern that the project just wanted to circumvent the environmental assessment in order to avoid
stakeholder consultations and subsequent delays resulting from those consultations. Collaboration between
the federal government, territorial government, design consultants and successful contractor was essential
to address numerous stakeholder questions/concerns and to ensure that commitments and measures were
integrated into the decontamination and deconstruction design specifications and contractor work plans
implemented during the deconstruction and decontamination. While the regulatory permits were obtained
to complete the work, the local history and high profile of Giant Mine contributed to heightened public and
media awareness of the project, which was a challenge during project planning and particularly during
implementation. Beyond the day-to-day duties during intense onsite activities, the team had to address
numerous media enquiries (many at the same time), reporting for monthly newsletters, monthly MVLWB
reports, weekly air monitoring reports, monthly stakeholder engagement, and interviews and news reports.
A functioning project team consisting of individuals from across governmental departments, communications
officers, health and safety officers, managers, consultants and contractors, was critical to successfully
meeting this one aspect of the multifaceted demands of this project. Recognising the human resource
limitations, the project team brought in additional personnel to assist with these demands and with day-to-
day activities. However, the reality of the situation was such that the senior staff most involved with oversight
of day-to-day work onsite – the key onsite consultant and contractor staff – were required to take time away
from their onsite tasks to address media and stakeholder information requests, which detracted from their
primary objective: to ensure the work was being completed safely, efficiently and according to specifications.
A primary lesson learned from this project was the need to have more senior onsite technical staff to assist
with such requests, including noting this requirement in applicable contractor work specifications for future
work at the Giant Mine site.
A two-phase contracting process for the decontamination and deconstruction work was used to ensure that
only a qualified contractor undertook this work. As a health and safety measure, the two-phase system also
ensured that only qualified companies and personnel participated in the site tour within the high-risk areas
of the roaster complex, as contractors had to go through the first phase of contracting in order to be invited
to the bidders’ tour.
The first phase was to pre-qualify decontamination and deconstruction contractors through a letter of
interest and qualification (LOIQ). The LOIQ briefly outlined the scope of work and risks and provided
evaluation criteria. An evaluation was completed based on the deconstruction/decontamination contractor’s
qualifications, and only firms that met the criteria in the first phase were invited to the request for proposal
(RFP) phase (Phase Two). During the second phase, multiple bidders’ conferences and tours were held to
enable contractors to obtain a better understanding of the project and subsequently produce higher quality
proposals. Proposals were reviewed by Government of Canada project team members, and then work was
awarded. The construction proposals were evaluated on the basis of the “assessed best value” of the highest
combined rating of technical merit (70%) and price (30%). The proposal evaluation included the submission
of an Aboriginal opportunities consideration, as required by AANDC (claims implementation and Aboriginal
territorial relations), owing to the project location within an Aboriginal land claim area. Qualified firms were
also given an honorarium to complete the bidders’ site tour and prepare proposals and plans during the
second phase of the contracting process.
Given the risk the roaster complex posed to humans and the environment, the contract for completing the
work stipulated that it be done within two construction seasons. In addition, the nature of the work was such
that a certain level of work had to be completed by the end of a construction season to ensure that the
buildings were left in a safe configuration at the end of the season. However, the challenges associated with

Case Studies
completing the detailed waste audit assessment of the roaster complex resulted in some significant contract
scope changes during the project’s implementation, particularly at the Cottrell precipitator building. The
sealed ducts, hoppers and precipitator units had significant amounts of arsenic-dust-impregnated asbestos-
containing insulation on their exterior that had not allowed for safe assessment of the interior contents
before preparing the contract specifications. During the course of decontamination work, after the exterior
arsenic and asbestos removal had been completed for the internal ducts, hoppers and precipitator units, the
internal structures were opened up to commence abatement within. At this time, it was identified that the
ducts were completely full of arsenic-containing waste, and that some of it was in a hardened, semi-
cemented state. Upon further assessment, the hoppers located under the electrostatic precipitators were
also identified to contain a large quantity of hardened arsenic waste. In total, it was estimated that the
internal structures of the Cottrell precipitator building contained approximately 700 m3 of arsenic-containing
waste, which was 10 times more than the original design quantities carried in the contract.
Other unexpected conditions were also discovered in other buildings related to both quantity and form of
waste due to restricted visibility and access at the time of the assessment. All of these increases to the scope
of work resulted in it being a challenge for the contractor to complete the work in the contractually allowed
timeframe, and also contractually in terms of securing funding. To ensure that buildings were left in a safe
configuration over the winter in 2013, and to complete the work in 2014, work had to continue until the late
fall and early winter during both the 2013 and 2014 seasons.
Knowing that the assessment of the site conditions was imperfect, the contractor, design and resident
engineering consultants, and federal government representatives worked collaboratively as a team to
identify and quantify the changes in conditions and the required changes to the contract to ensure the project
progressed even with these unexpected conditions. The key was to foster regular, open conversation and
ensure that the additional services were rendered at a fair and reasonable price.
A lesson learned during the course of the project was to sufficiently account for any limitations encountered
during assessment. Incorporating additional contingency into scheduling, accounting for potentially
significant cost increases and having an effective contract change processes are all important components
for a project of this nature. Time permitting, having a formal requirement for the contractor to validate and
confirm conditions early on in the project may also allow the opportunity to plan and negotiate changes
during inactivity (i.e. winter), freeing up resources during the field season.
While the decontamination and deconstruction of the roaster complex was undertaken to address the risks
to humans and the environment, completing the work itself posed a high level of risk to the workers
undertaking the activities, as well as to the public.
While the work control methods, abatement completion standards and medical monitoring required for high-
risk asbestos abatement are well established and regulated, developing similarly appropriate controls for
multiple contaminants was a new endeavour, particularly in a cold climate. Furthermore, the containerisation
and storage of this particular unique hazardous waste in proximity to a city required significant consideration.
Both of these elements required a significant amount of input and acceptance from a multifaceted team of
expert consultants and contractors, regulators, health and safety officers, stakeholders, government
departments and government communication officers.
Prior to commencing work in any structure, an evaluation by a structural engineer was required to identify
and mitigate any risks of structural collapse during work. Detailed hazard assessments were also completed
to address other physical hazards.
The decontamination, or abatement work, posed significant challenges in that it required implementing
controls to address potential worker exposure to multiple contaminants: arsenic, asbestos and cyanide. This
was further complicated by the fact that the level of control for each of these contaminants varied depending
on the actual building being abated within the complex. For example, the concentration and type of arsenic
present within dust and process residuals increased according to where the building was in the process chain:

in the AC roaster, Dorrco roaster, and calcine plant buildings, arsenic was not necessarily in the form of
arsenic trioxide and not present in as high a concentration, whereas in the Cottrell precipitator, Baghouse,
silo and weight scale buildings the form of arsenic was primarily highly concentrated arsenic trioxide dust.
Asbestos-containing materials were also present in different forms in different buildings: the calcine plant,
Dorrco roaster and Cottrell precipitator buildings were constructed of transite with spray-applied insulation
on the building interiors, necessitating complete enclosure of these buildings to complete the abatement
work; within the Cottrell precipitator building, the type of asbestos insulation present included both
chrysotile and crocidolite, the latter of which has much higher health risks for exposure, thereby requiring
more control during abatement work. Completing abatement of arsenic- and asbestos-containing dusts was
further complicated in locations where sodium cyanide was present, given that the typical form of work
control implemented to limit airborne release of these contaminants is water, but using water where sodium
cyanide is present can generate cyanide gas, for which the respiratory personal protective equipment (PPE)
controls used for arsenic and asbestos as particulates would have been inadequate. The selection of cutting
methods for opening up internal structures to remove arsenic also required consideration, in that arsine gas
could be produced where heat was used in the presence of arsenic trioxide, in which case the respiratory PPE
controls used for particulates would have been inadequate.
Appropriate air quality controls and monitoring for multiple contaminants for the protection of abatement
workers and humans outside of the work area needed to be developed and accepted by health and safety
regulators. To this end, the project essentially adopted the same occupational standards for arsenic and
cyanide as for high-risk asbestos abatement work, but increased the level of monitoring, particularly outside
of enclosures, to account for the higher risk and more dispersible nature of arsenic-containing dust. To limit
abatement workers’ exposure to contaminants, buildings undergoing abatement work were wrapped in a
plastic film enclosure and placed under negative air pressure, consistent with high-risk asbestos abatement
practices. Inside the enclosure, air samples were collected to measure asbestos and arsenic air
concentrations, and monitors were installed to measure hydrogen cyanide concentrations. Occupational
samples were collected in each work area, at the frequency of one sample for every 10 workers, to measure
air quality in worker breathing spaces for asbestos and arsenic; personal hydrogen cyanide monitors were
worn with the same frequency. Additional personal monitoring for arsine gas was also worn as required.
Outside enclosure air samples were collected for arsenic, asbestos and hydrogen cyanide at each entry and
exit point to the building enclosures, with additional sampling points placed for every 450 m 2 of building
space (AECOM, 2013). Consistent with asbestos abatement requirements, the contractor was required to
submit abatement exposure risk assessments and proposed mitigative and control measures to complete the
work to the Northwest Territories Workers Safety and Compensation Commission (WSCC), which accepted
the plans.
A measurable standard needed to be developed to document when abatement work for arsenic, asbestos
and cyanide had been completed. In particular, a standard needed to be adopted for arsenic. While there are
some limited surface wipe standards for decontamination of arsenic, none of these were appropriate in the
case of the roaster complex, given the elevated level of total arsenic concentrations present throughout the
Giant Mine site and the Yellowknife area in general. Arsenic-containing materials present within the roaster
complex were identified as hazardous on the basis of leachable arsenic concentrations, consistent with
legislation for hazardous waste characterisation. This resulted in the classification of porous structural
elements, such as wood, being arsenic-hazardous waste. Collecting surface wipe samples of such elements
after internal building decontamination would not have been appropriate, as this material was to be
packaged as arsenic-containing hazardous waste following deconstruction anyway. The project therefore
adopted the same procedure as for high-risk asbestos abatement, using aggressive air clearance sampling for
arsenic as well: surfaces were agitated with a leaf blower; samples were run to confirm that concentrations
of arsenic and asbestos were less than 10% of the occupational exposure criteria for these contaminants;
and cyanide monitors were run to confirm that no hydrogen cyanide was detected (AECOM, 2013). This,
however, was not without its challenges: the soil surrounding the roaster complex is highly contaminated
with arsenic, and perimeter air quality monitoring during abatement work often had outside concentrations
of arsenic higher than inside on windy days or when equipment was trafficking in the vicinity of air monitors.

Case Studies
The buildings therefore needed to be completely sealed and negative air units turned off during air clearance
sampling to avoid measuring impacts being drawn inside from outside contamination.
Once the challenges associated with outside contaminant levels and air quality impacts were addressed, the
use of aggressive air quality air clearance sampling did prove to be an effective and achievable means to
assess the completion of decontamination work within the structures.
It was recognised, however, that despite achieving the air clearance sampling criteria, there was still the risk
of release of inaccessible dust during building demolition. As with the abatement air-sampling program, it
was also identified that the soil surrounding the structures needed strict dust control during the
deconstruction work. Buildings were therefore carefully deconstructed in a step-wise manner under rigorous
dust control measures. Deconstruction was halted if any visible dust releases were observed, and additional
controls were put in place to limit any further releases. As a further precaution, real-time ambient air quality
sampling was conducted for particulate matter of 10 microns or less (PM10) at the broader roaster complex
area to document any dust releases, and every three days, 24-hour samples were collected for PM10, total
suspended particulate (TSP) and total arsenic. Ambient air quality monitoring equipment was set up both
upwind and downwind of the roaster complex area, with four real-time monitors in place at all times, and
sample collection equipment in place at two locations (AECOM, 2013). The real-time PM10 data and the
results of the 24-hour samples were compared to risk-based action level criteria.
The ambient air monitoring program was not without challenges. The ambient air monitoring equipment was
not designed to work effectively at temperatures lower than −10°C. The primary concern initially was related
to battery power, necessitating the use of portable generators to continue powering the equipment.
However, electrical problems proved to be the main issue, and there were a few days on which the real-time
monitors did not work. This was determined to be an issue at −15°C or colder. For these days, the onsite
project team had to rely solely on visual observations. This only occurred on a few occasions, during which,
outside containment, potentially dust-generating work was restricted.
Ice fog, common in Yellowknife early in the winter season, also proved an issue, as it registered as particulate
matter on the real-time monitors. Again, the project has had to resort to rigorous visual observations related
to dust emissions.
The 2014 summer was a particular challenge because of widespread wildfires in the Northwest Territories,
resulting in excessively smoky conditions throughout the region. For several weeks, the PM 10 real-time
monitors were recording constant exceedances of the criteria, making the monitors unable to document
exceedances based on roaster complex activities. Again, this required rigorous visual monitoring of dust
emissions.
The roaster complex ambient air quality monitoring program requirements from the specifications and the
corresponding dust management plan developed by the contractor were the first line of defence against the
dispersion of airborne contaminants from this source area. However, the roaster complex decontamination
and deconstruction work was one of a number of activities taking place at Giant Mine. As a consequence, the
roaster complex program was supplemented by a fence line air monitoring program (at the mine’s
boundaries) and a community air monitoring program developed for the mine activities at large through
collaboration and engagement with stakeholders and input from the Government of the Northwest
Territories, the City of Yellowknife and consultants. The use of this tiered approach was successful in ensuring
that the public was protected, as well as pinpointing other sources of dust beyond the roaster complex
activities.
The abatement air monitoring requirements of the remedial controls program carried in the specifications
established the respiratory protection requirements for the decontamination workers. All workers wore
powered air-purifying respirators, quantitatively fit-tested to provide the highest level of respiratory
protection the respirator could supply, with supplied air respirators worn for specific work requirements
where there was a higher risk of arsine gas generation or hydrogen cyanide. To protect decontamination

workers from dermal contact, two layers of chemical-resistant clothing were worn: one to guard against
particulate exposure and another to guard against chemical exposure.
Completing abatement work during cold temperatures – as was required to meet the contract timeline –
required a significant amount of heating of buildings to protect abatement workers. Cold temperatures
caused the workers’ respirators to fog up, limiting visibility, especially when completing physically demanding
work. The PPE requirements of the abatement were such that workers wore little to no clothing under their
protective suits; they were therefore at risk of exposure to cold if the buildings were not heated. Finally,
abatement work required a significant amount of water application for dust control; the buildings and water
lines needed to be heated to allow for continued dust suppression. Conversely, the summer working
conditions required the same level of PPE as during the winter despite the higher temperatures. Heat
exhaustion and dehydration were daily concerns.
As an added precaution beyond the dust and exposure control measures, air monitoring and PPE
requirements, all workers involved with the roaster complex decontamination and deconstruction work
underwent medical monitoring; to check for occupational exposure to arsenic, weekly urine samples were
collected and analysed for concentrations of inorganic arsenic and methylated metabolites, which is the
analysis recommended by the American Conference of Governmental Industrial Hygienists (ACGIH). To
conform to the regulatory medical monitoring requirements for asbestos, abatement workers were also
required to undergo chest x-rays and pulmonary function tests on a regular basis.
One of the biggest challenges to the project was related to measuring worker occupational exposure to
arsenic through the medical monitoring program. The contractor initiated a medical monitoring program at
the start of the project, collecting urine samples from certain workers according to a set schedule, and
analysing the urine for total arsenic content using an action level of 100 µg/L, in accordance with the protocol
in place before the mine ceased operations. The mine care and maintenance contractor responsible for
overall site health and safety had continued this protocol. However, the extent of monitoring (i.e. who was
monitored at what frequency) was the subject of much discussion among government departments and
regulators to ensure worker protection at the start of the project, with the end result being that all workers
involved with the roaster complex submitted a weekly urine sample for arsenic testing. The urine
concentration at which workers were assessed as being potentially occupationally exposed with work
restrictions imposed was also reassessed by the WSCC during the course of the project. The ACGIH
recommends a biological exposure index (BEI) for inorganic arsenic plus methylated metabolites
concentrations in urine of 35 µg /L (ACGIH, 2001). A concentration above this level suggests that there has
been a risk of arsenic exposure, and is a level that should not be exceeded routinely on a long-term, chronic
basis. However, the measurement of this concentration is fraught with interpretation challenges in that
inorganic arsenic is present in foods; in particular, there has been recent media reporting about the high level
of inorganic arsenic present in rice products (Consumer Reports Magazine, 2012). Inorganic arsenic has also
been identified in many other food sources (Schoof et al., 1999; Cottingham et al., 2013). Indeed, the ACGIH
cautions against the use of the BEI as a strict measure of occupational exposure. There is no way of
distinguishing between occupational exposure to arsenic and ingestion of food containing inorganic arsenic.
Furthermore, such lifestyle choices as smoking or chewing tobacco also contribute to inorganic arsenic
concentrations in urine. Late in the first season of the roaster complex work, the WSCC began to require that
an employer report of injury be submitted for any worker whose urine sample results exceeded the BEI. A
sample result greater than the BEI was also a point at which a worker received individual counselling, a
thorough review and evaluation of the individual’s work practices and removal from high-risk abatement
work areas. As a consequence, documentation of diet in the 48 hours preceding the submission of a urine
sample became a standard part of the medical monitoring to allow a proper investigation of whether there
had been an occupational exposure. Nonetheless, assessment of whether there had been occupational
exposure remained a challenge.
A serious challenge with the medical monitoring program was the lack of laboratories capable of completing
the required analysis in a timely manner. While there were laboratories in Canada that could complete the
testing, they were not equipped to handle the large volume of samples being received on a weekly basis from

Case Studies
the project, which was in the order of 60 samples a week. The laboratory initially contracted to complete the
work was reporting analytical results sometimes as late as three weeks post sampling date. This was a
significant concern to both the project and the WSCC, as it did not allow for workers who had been potentially
exposed to be removed from high-risk abatement work in a timely manner. Three separate labs were used
during the 2013 season to try to rectify this issue, but the turnaround time for analytical results remained a
challenge. Furthermore, upon review of all the data after the 2013 season, it was apparent that different
analytical techniques were used by the various labs, some of which were not capturing the appropriate
arsenic forms for proper comparison to the BEI. This was thankfully rectified in 2014 by a commercial lab
purchasing the required equipment and setting itself up to handle the large volume of samples to complete
the appropriate analysis and report results within seven days. Some samples could also be run high priority,
with a 48-hour turnaround time, for employees whose previous analyses had exceeded the BEI. The
drastically improved turnaround time in 2014 was a significant relief to project management and the WSCC.
Investigations were conducted for each instance of worker urine sample exceedance to try to identify the
source of the exposure. While the project controls during abatement activities provided a thorough measure
of worker exposure during this activity to allow for mitigation, the main lessons learned related to arsenic
exposure arising from the investigations were related to personal hygiene and the need for strict quality
assurance during urine sample collection. The soil contaminant levels for arsenic are elevated throughout
the mine but particularly high surrounding the roaster complex. In 2013, urine samples were being collected
from workers at the roaster complex area in the washroom trailer located outside of the exclusion (high-risk)
zone. However, surface wipe sampling inside all the site trailers outside of the exclusion zone at the roaster
complex often indicated high levels of arsenic. Dust on worker clothing was also identified as a potential
source. Therefore, in 2014, all workers were required to provide samples at the main mine site office building
by the site entrance, to change out of work clothing prior to providing their sample, and to wash hands
thoroughly prior to providing samples to minimise the risk of sample contamination during collection.
Furthermore, in 2014, all onsite trailers at the roaster complex were equipped with boot washes at their
entrances, and trailers were cleaned daily (twice daily for the lunch trailer). Workers were coached constantly
on the need to wash hands prior to eating or smoking and to avoid having work clothes contact table or desk
surfaces. Dirt tracked within site vehicles and equipment cabs was also identified as a source of exposure,
particularly during cold weather, when heaters were in use, necessitating rigorous housekeeping within.
In summary, the medical monitoring program was one of the biggest challenges to the project and the source
of some of the primary lessons learned for the project: when it comes to worker exposure to a workplace
contaminant, it is beneficial to engage the regulators early and to maintain an open dialogue. Furthermore,
it highlighted the importance of thorough assessments related to exposure to arsenic at the site, particularly
related to arsenic levels in soil and the importance of good housekeeping, worker personal hygiene and
sample collection quality assurance.
In the end, over a two-year period, the project team successfully dismantled the 10 structures that comprised
the roaster complex, segregated the non-hazardous and hazardous wastes and removed and containerised
more than 3,000 tonnes of arsenic-contaminated materials from the complex without adverse effect to the
onsite workers, neighbouring communities or the environment.
The next step for the project team will be to design the long-term solution for the arsenic waste temporarily
stored on the purpose-built “material storage area.” The challenge will be to transfer this material to an
underground location and freeze it in a manner that will encapsulate it, along with the other 237,000 tonnes
of arsenic trioxide dust already located underground.
American Council of Government Industrial Hygienists (ACGIH) (2001) Arsenic and soluble inorganic compounds: BEI, 7th edition.
Documentation. Publication # 7 DOC-665, Cincinnati.

Agency for Toxic Substances and Disease Registry (2014) Medical management guidelines for arsenic trioxide, viewed 7 December
2014, http://www.atsdr.cdc.gov/MMG/MMG.asp?id=1200&tid=279.
Canadian Broadcasting Corporation (2014) High arsenic levels found in some Yellowknife-area lakes, viewed 7 December 2014,
http://www.cbc.ca/news/canada/north/high-arsenic-levels-found-in-some-yellowknife-area-lakes-1.2853096?cmp=rss, last
updated 28 November 2014.
Canadian Environmental Assessment Agency (CEA) (2012) Canadian Environmental Assessment Act http://laws-
lois.justice.gc.ca/eng/acts/c-15.21/index.html.
Consumer Reports Magazine (2012) Arsenic in your food, 2012, http://www.consumerreports.org/cro/magazine/2012/11/arsenic-
in-your-food/index.htm.
Cottingham, K.L., Karimi, R., Gruber, J.F., Zens, M.S., Sayarath, V., Folt, C.L., Punshon, T., Morris, J.S. and Karagas, M.R. (2013) Diet
and toenail arsenic concentrations in a New Hampshire population with arsenic-containing water, Nutrition Journal, Vol. 12,
p. 149.
Schoof, R.A., Yost, L.J., Eickhoff, J., Crecelius, E.A., Cragin, D.W., Meacher, D.M. and Menzel, D.B. (1999) A market basket survey of
inorganic arsenic in food, Food and Chemical Technology, Vol. 37, pp. 839–846.
AECOM (2010) Giant Mine flue structural assessment.

AECOM (2011) Giant Mine, structural condition of roaster complex.
AECOM (2012a) Giant Mine roaster complex deconstruction waste audit report.
AECOM (2012b) Recommendation to proceed with demolition of the roaster complex at the Giant Mine.
AECOM (2013) Specifications for roaster complex deconstruction, Giant Mine, Northwest Territories.
Arctic Environmental Services Ltd. (1996) Asbestos assessment at the Giant Yellowknife Mine for Royal Oak Mines Inc.
Ferguson Simek Clark (1997) Site Inspection of roaster exhaust stack for Royal Oak, Giant Mine, Yellowknife, NT.
North West Consulting Limited (2003) An examination of arsenic contamination in the roaster and gas handling complex at the Giant
Mill.
Public Works and Government Services Canada (PWGSC) (2007) Structural inspections, Giant Mine – various buildings, Yellowknife,
NT.
SENES Consultants Limited and SRK Consulting Engineers and Scientists (SENES and SRK) (2010) Giant Mine remediation project
developer’s assessment report, May 2010, p. 5-2.

D.A. Bright Royal Roads University/Hemmera, Canada

N. Sandys Hemmera, Canada
The ability to predict and manage the interactions between mine wastes and water, with the associated
implications for aquatic ecosystem impairment, has evolved immensely over the last two decades and
continues to rapidly change. The major focus so far has been on drainage chemistry from sulphidic ores and
wastes, since oxidation of pyrite, pyrrhotite and similar sulphide minerals has been the root cause of serious
water-quality issues at many mine sites worldwide. However, there are also many examples of compromised
surface water and groundwater quality as a result of geochemical processes involving neutral to alkaline
dissolution and aqueous transport. Neutral mine drainage (NMD) is often taken to mean down-gradient
runoff from sulphidic source materials that are undergoing oxidation following the subsequent reaction with
neutralising minerals such as carbonates; this results in water with circumneutral pH, high hardness and high
sulphate levels. There is no standardised definition of NMD, and this term is increasingly used by researchers
and managers to describe a variety of geochemical processes and issues. The predictive models and tools for
managing the environmental effects of NMD are more poorly developed than for acidic rock drainage (ARD),
especially since predictions depend more on complex interpretations of mineralogy and geochemistry. The
flux of various trace major elements and materials from mine spoils to the hydrosphere is generally far greater
for ARD than NMD; however, the potential for ecosystem-scale impacts from NMD is expected to increase in
proportion with an increase in the spatial scale of mining projects in general. This paper provides a simplified
classification of the various types of NMD that have been encountered as a starting point for developing new
predictive and management approaches for NMD. Several of the tools for preventing NMD-related
environmental impacts, applied during mine development and closure planning, are very similar to the tools
that have been developed for ARD; however, the tools and approaches for NMD place greater emphasis on
management actions at the watershed scale and on interactions between groundwater and surface water.
The effects of mine wastes on water quality are of paramount concern in approving new mines and managing
ongoing operations and mine closures, since the influence of mine drainage on vegetation, fish, wildlife and
humans can be very large relative to the actual mine footprint. While the spatial scale of influence of mine
drainage can be large, past experience has shown that the temporal scale of influence can also be large
relative to the period over which direct economic benefits of mining are accrued. From the 1970s through
the first decade of the new millennium, it has become apparent that metal leaching and acidic rock drainage
(ML/ARD) associated with oxidation of sulphide minerals is among the most environmentally damaging issues
associated with contemporary mining operations (Price, 2005). In recognition of the environmental, societal
and financial costs of ML/ARD, scientists’ work over the last three decades has led to significant progress in
the knowledge about factors that enhance or inhibit acidic drainage, diagnostic tests and interpretations for
prediction and management approaches to reduce the environmental impacts.
ML/ARD is caused by oxidation of sulphide minerals such as pyrite.
𝐹𝑒𝑆2 + 1
2
𝐻2 𝑂 + 15
4
𝑂2 ↔ 𝐹𝑒 3+ + 2𝑆𝑂42− + 𝐻 + (1)
The acidity associated with pyritic mineral oxidation, per the simplified equation (1), is the basis of the term
“acidic rock drainage” (ARD). Additional acidification results from the further oxidation of pyrites by ferric

Beyond ML/ARD: the many faces of neutral mine drainage in the context of mine closure D.A. Bright and N. Sandys
iron (equation 2) and the secondary precipitation of iron oxides along water flow paths (equation 3) (Price,
2009).
𝐹𝑒𝑆2 + 1
2
𝐻2 𝑂 + 14𝐹𝑒 3+ + 15
4
𝑂2 ↔ 15𝐹𝑒 3+ + 2𝑆𝑂42− + 𝐻 + (2)
𝐹𝑒 2+ + 1
4
𝑂2 + 5
2
𝐻2 𝑂 ↔ 𝐹𝑒(𝑂𝐻)3(𝑠) + 2𝐻 + (3)
The tendency of various metals to leach from minerals at higher rates under low pH conditions (metal
leaching, or ML) is an obvious concern for ecological and human health risks associated with mine-affected
surface waters and groundwater.
While ML/ARD issues are typically at the forefront of management concerns for proposed or operating mines
and mine closure, scientists have long recognised that other chemical weathering and dissolution reactions
can be an important influence on water quality, based on both a theoretical understanding of geochemical
processes and real world examples. Nicholson’s (2004) review of water-quality issues associated with neutral
mine drainage (NMD) included a working definition of NMD that involves two factors: (i) runoff from sulphidic
source materials that are undergoing oxidation following the subsequent reaction with neutralising minerals
such as carbonates, which results in water with circumneutral pH, high hardness and high sulphate levels;
and (ii) rapid and accelerated chemical weathering of mineral types other than sulphides. Nicholson’s review
focusses on the second of these.
An extensive search of peer-reviewed and other published literature for the terms “neutral mine drainage”
or “neutral drainage” reveals that research efforts to the present time have overwhelmingly focussed on
mine drainage chemistry from oxidation of sulphidic deposits and the associated secondary and tertiary
geochemical reactions following neutralisation, either in situ or along the water flow path. Nicholson’s (2004)
review on NMD was intended above all as “a starting point for discussion and awareness of an environmental
issue that has had less attention to date than acidic drainage, and one that may become important at
operations without acid drainage issues or where acid drainage has been mitigated.” In the last decade,
however, this issue has not received much attention.
The theoretical, conceptual and practical approaches for detecting, predicting and managing NMD are
arguably already available; these approaches are discussed in recent guidance documents such as Prediction
Manual for Drainage Chemistry from Sulphidic Geological Materials (Price, 2009) and Global Acid Rock
Drainage Guide (INAP, 2009). The names of these documents suggest, incorrectly, that they focus exclusively
on sulphidic mineral dissolution and drainage as opposed to NMD associated with chemical weathering of
other mineral types. Furthermore, the very success of recent international efforts to refine pragmatic
approaches and tools for predicting and managing ML/ARD may be blinding some practitioners at times to
potential environmental issues associated with NMD, as discussed below.
The objective of this paper is to catalyse discussions among scientists and managers about NMD based on
broader definitions as proposed by Nicholson (2004). We briefly discuss types of NMD, adequacy of the tools
currently available to assess mine drainage chemistry and water quality and especially the need to combine
scientific thinking about mine drainage with research on chemical weathering in general. This field offers a
larger focus on pedogenesis and evolution of the geosphere at many scales: from alterations at the mineral
surface to local- and regional-scale weathering, catchment-scale erosion and solute transport, and
continental denudation.
NMD includes two major categories of mineral types and in situ weathering reactions. The first category is
associated with sulphide mineral oxidation accompanied by subsequent neutralisation reactions such that
the drainage is circumneutral or “near neutral” as defined by Price (2009) (e.g., pH in a range of 6 to 8). The
second category is based on accelerated dissolution at the mineral face of any nonsulphide mineral. The
influence of weather based on either of these categories become increasingly important at large geographic
scales, where there is significant mass of geological materials of interest relative to the size of both

Case Studies
catchments that support ecologically productive surface waters and of the recharge areas that support local
and regional aquifers or interact with surface flows.
A generalised discussion of mine drainage chemistry and trace element dissolution potential for all
economically viable mineral deposits is a daunting task. The economic geology “checkerboard” classification
by Dill (2010) or similar classifications can assist us by providing information on the lithological regimes that
may host economically relevant amounts of specific elements. The lithological regimes used by Dill include
the following:
 igneous rocks (ultrabasic, basic, intermediate, felsic, alkaline and carbonatites);
 structures (pipes, faults);
 sedimentary rocks (duricrusts-regolith-veinlike deposits, coarse-grained clastics, fine-grained
clastics and massive rocks, limestones, evaporates, special sedimentary rocks); and
 special facies, including carbon bearing hosts (coal-oil shales-hydrocarbons).
What such schemes point out is that many elements of interest for mining occur in ore bodies and minerals
that are not predominantly sulphidic. Chromite (FeCr2O4), for example, is the predominant mineral source
for chromium, occurring especially in coarse-grained clastic rocks. Recoverable rare earth element (REE)
deposits in Canada and elsewhere occur in bastnasite [(Ce,La)(CO3)F], monazite [(Ce,La,Nd,Th)PO4] and ionic
clays (Parliament of Canadian, 2014).
Perhaps a more simple mineral taxonomy for framing discussions about NMD is the older classification of the
more than 4,700 known minerals into the major groupings shown in Figures 1 and 2 (Gaines et al., 1997). The
Dana classification was developed in 1848 by Professor James Dana of Yale University and includes nine basic
classes: native or pure elements, sulphides, oxides/hydroxides, halides, carbonates/nitrates/borates,
sulphates/chromates/selenates, phosphates/arsenates/vanadates, organic minerals and silicates. Pure
elements are not included in Figures 1 and 2, since water soluble forms at concentrations suitable for
extraction are rare.
It is important to understand the potential for dissolution of various trace elements from minerals in the ore
body of interest both prior to and following excavation and extraction. The mineralised zone almost
invariably contains complex mixtures of other minerals (including gangue minerals) that can interact in highly
complex ways to influence chemical weathering rates and secondary interactions. It is challenging, therefore,
to predict the drainage chemistry of large bulk waste deposits and physically disturbed areas from first
principles. In consideration of NMD, accessory sulphide minerals often occur in association with nonsulphide
minerals of economic interest and can contribute disproportionately to aqueous concentrations of various
metals and metalloids such as arsenic. Mineralised zones and mine wastes are also typically highly
heterogeneous at both microscopic and macroscopic scales, which leads to variations in reactivity and
dissolution potential that are challenging to fully appreciate. The level of effort applied to detailed
mineralogical analysis of ore bodies, host rock, waste rock and tailings, therefore, is generally commensurate
with the accuracy of mine drainage predictions, as discussed in Section 3 of this paper.
As equations (1) and (2) illustrate, sulphate is a major product of sulphide mineral oxidation; it substantially
remains in solution after a change to circumneutral pH in affected waters (e.g., following reaction with
carbonate minerals). Sulphate is often the most effective tracer in water of sulphide mineral oxidation, even
in systems and mine wastes that are non-acid generating.

Old Dana Classification New Dana Classification Common Forms

(Gaines et al., 1997)
Type I NMD: Unprecipated sulfide oxidation products and neutralization byproduct dissolution
Sulfides Sulfides, inc. selenides and tellurides pyrite (FeS2 ),

pyrrhotite (Fe7 S8 )
Sulfosalts
Type II NMD: Accelerated weathering of all other non-sulfide minerals
Oxides/hydroxides Simple oxides magnetite (Fe 3O 4 ),

Oxides containing uranium and thorium hematite (a Fe2O3 )
Forms containing hydroxyl gibbsite (Al(OH) 3 )
Multiple oxides
Multiple oxides with Nb, Ta, and Ti
Halides Anhydrous and hydrated halides halite (NaCl)

Oxyhalides/hydroxyhalides flourite(CaF2 )
Halide complexes, alumino-flourides
Compound halides
Carbonates/nitrates/
borates Acid carbonates calcite (CaCO3 )
Anhydrous carbonates
Hydrated carbonates
Carbonates - hydroxyl or halogen
Compound carbonates
Simple nitrates
Nitrates - hydroxyl or halogen
Compound nitrates
Iodates - anydrous and hydrated
Iodates - hydroxyl or halogen
Compound iodates
Borates - anydrous
Anydrous borates with hydroxyl or halogen
Hydrated borates containing hydroxyl or halogen
Compound borates
Beyond immediate toxicological concerns for human health (based on drinking water ingestion) and aquatic
life in affected areas, elevated sulphate in freshwater systems can profoundly alter wetland and sediment
diagenetic reactions. In particular, organic matter mineralisation in freshwater sediments occurs
predominantly via methanogenesis, as opposed to dissimilatory sulphate reduction, since natural sulphate
levels are limiting (Herlihy and Mills, 1985; Brandl et al., 1990; Bright et al., 1994).

Case Studies
Old Dana Classification New Dana Classification Common Forms

(Gaines et al., 1997)
Type II NMD: Accelerated weathering of all other non-sulfide minerals (continued)
Sulfates/chromates/
selenates Anydrous acid and sulfates gypsum (CaSO4 H2 0 )
Hydrated acid and sulfates
Anhydrous sulfates containing hydroxyl or halogen
Hydrated sulfates containing hydroxyl or halogen
Compound sulfates
Selenates and tellurates
Selenites - tellurites - sulfites
Anydrous chromates
Compound chromates
Phospates/arsenates/ Anydrous acid phosphates apatite (Ca5 (F,Cl,OH) (PO 4)3 )

vanadates
Anydrous phosphates
Hydrated acid phosphates
Hydrated phosphates
Anydrous phosphates containing hydroxyl or halogen
Hydrated phosphates containing hydroxyl or halogen
Compound phosphates
Antimonates
Acid and normal antimonites, arsenites, phosphites
Basic or halogen-containing antimonites, arsenites, phosphites
Vanadium oxysalts
Anydrous molybdates and tungstates
Basic and hydrated molybdates and tungstates
Organic minerals Salts of organic acids and hydrocarbons
Silicates Numerous: see Gaines et al., 1997
Methylation of mercury in lentic and lotic sediments is expected to occur under anoxic conditions and
especially in association with microbial sulphate reduction (Gilmour et al., 1992). Thus, mining-related inputs
of sulphate into freshwater can significantly increase the rates of mercury methylation and biomagnification
potential.
Even in the absence of successive reactions, trace element sulphide mineral oxidation can result in enhanced
dissolution of various metals/metalloids, the continued presence of which in the dissolved state is insensitive
to further neutralising reactions. For example, oxidation of arsenopyrite (FeAsS) results in production of
arsenate oxyanions (equation 4), which will tend to remain in solution across a relatively large pH range in
the absence of a strong sorptive phase such as precipitated iron oxyhydroxides.
𝐹𝑒𝐴𝑠𝑆 + 72𝑂2 + 4𝐻2 𝑂 → 𝐻𝐴𝑠𝑂4 2− + 𝐹𝑒(𝑂𝐻)3 + 𝑆𝑂4 2− + 4𝐻 + (4)
Thus, the aqueous environmental concentrations of arsenate in association with mine drainage are
predominantly controlled by two factors: (i) the rate of arsenopyrite oxidation independent of the amount
of co-occurring neutralisation potential (and rates of neutralisation); and (ii) the availability of iron
oxyhydroxide precipitates (e.g., resulting from oxidation and precipitation of ferrous iron, as shown in

equation 3) as a potential strong sorptive phase. A lack of appreciation of the expected fate of arsenicals
under near-neutral conditions is behind many of the interpretative errors we have encountered in our
reviews of applied mine site hydrogeochemical interpretations.
As discussed by Price (2009), antimony, arsenic, molybdenum, selenium and sulphur tend to occur as
oxyanions in oxidised environments. These oxyanions typically remain in solution at circumneutral pH in the
absence of a strong sorptive phase. Price (2009) also points out that many trace elements that may be
solubilised during sulphide oxidation are relatively soluble under near-neutral pH; these include cobalt,
molybdenum, nickel and zinc.
Demers et al. (2013) provide a compelling example of the ongoing aqueous mobilisation of nickel from older
waste rock deposits at the Tio mine site in eastern Quebec, Canada. Hemo-ilmenite has been mined here for
over 50 years, and waste rock deposits contain mostly plagioclase (labradorite), ilmenite, hematite and some
pyrite, micas, spinel and chlorite. Drainage water from waste rock deposits has remained near neutral, while
nickel concentrations have increased in recent years, and the discharge limit of 0.5 mg/L in effluent has been
exceeded with increasing frequency (albeit still sporadically). Mineralogical analysis of the waste rock
demonstrated that the major source of nickel in the waste rock is millerite (NiS), present as inclusions in
grains of pyrite. Temporal trends in effluent concentrations of nickel, under near-neutral drainage conditions,
were determined to be associated with secondary sorption. In particular, an increasing trend in effluent-
phase dissolved nickel concentrations was determined to be the result of a trend towards saturation of
sorption sites over time, following 25 years in which nickel concentrations in waste rock drainage were not
identified as an issue of concern.
Type I NMD is defined here as runoff from sulphidic source materials that are undergoing oxidation after
subsequent reaction with neutralising minerals such as carbonates, resulting in water with circumneutral pH,
high hardness and high sulphate levels. Equations (5) and (6) provide simplified reactions for neutralisation
of acidity arising from sulphide mineral oxidation through reaction with calcium carbonate.
2𝐻 + + 𝐶𝑎𝐶𝑂3 ↔ 𝐶𝑎2+ + 𝐻2 𝐶𝑂3 (5)
2𝐻 + + 𝐶𝑎𝐶𝑂3 ↔ 𝐶𝑎2+ + 𝐶𝑂2(𝑎𝑞) + 𝐻2 𝑂 (6)
Type I NMD will generally occur (as opposed to ML/ARD) when alkalinity released from minerals such as
carbonates at or down-gradient along an aqueous flow path occurs at a sufficient rate to neutralise the acid
generated from sulphide mineral oxidation. Carbonates, hydroxides and silicates may exhibit neutralising
potential to varying degrees. In general, neutralisation reactions result in the dissolution of base cations such
as Ca2+ and Mg2+ at the site of neutralisation, along with co-occurring elements such as iron and manganese.
Especially for calcium and magnesium in carbonate minerals, rates of dissolution or precipitation are very
rapid relative to the surface or groundwater velocities or other types of reactions; therefore,
dissolution/precipitation reactions are said to be controlled thermodynamically (or to be equilibrium-
controlled) as opposed to kinetically (Price, 2009). Rates of dissolution or precipitation are highly sensitive to
solution pH or dissolved CO2 pressure (pCO2).
Aquatic animals and plants are generally insensitive to increased water hardness associated with higher
dissolved concentrations of base cations, and it is well known that increased hardness is associated with a
decrease in the bioavailability and toxicity to aquatic life of metals such as cadmium, copper, lead and zinc.
Under some circumstances, however, secondary precipitation of carbonates from NMD within the aquatic
receiving environment (i.e., partial-to-complete concretion) can have significant results: reductions in the
permeability of and flow rates through aquatic substrates such as spawning gravels, profound alterations in
benthic habitat, smothering and loss of benthic invertebrates and occlusion of respiratory epithelia of
substrate associated biota.
Calcite deposition has been studied largely as a problem for industrial processes and in the context of natural
precipitation reactions associated with geothermal springs or cave deposits, especially in karst settings. As
reported in Windward Environmental et al. (2014), calcite precipitation has been observed in the Elk River
watershed, British Columbia, in reaches influenced by coal mining discharge. Calcite deposition has also been

Case Studies
observed in association with coal mining within the McLeod River watershed on the eastern slopes of the
Rocky Mountains in Alberta. Calcite deposition is driven by calcium supersaturation in aqueous solutions,
especially in high ionic strength groundwater emanating from coal waste deposits, when carbonic acid
concentrations decline and pH increases as result of a reduction in pCO2. A reduction in pCO2 may occur
especially as a result of increased air-water interactions and off-gassing (e.g., when groundwater emerges at
the ground surface) and as result of photosynthesis within the aquatic receiving environment via the localised
draw-down through uptake and fixation of CO2 by primary producers during photosynthesis. When calcium
becomes oversaturated, it tends to precipitate out onto available solid surfaces as various carbonate
minerals, including calcite.
As shown in Table 1, a large number of different minerals may be present in mineralised areas and waste
deposits at a given site. Our primary interest is in primary and secondary dissolution reactions of nonsulphide
minerals in association with natural or accelerated weathering rates.
Price (2009) defines weathering as “the process by which geological materials are altered on exposure to
atmospheric conditions and agents.” Type II NMD focusses predominantly on chemical weathering rates and
consequences — those based on the dissolution and precipitation potential of substances and
sorption/desorption kinetics in their existing state (e.g., based on anion and cation exchange); hydration;
hydrolysis; oxidation/reduction; and dissolution and precipitation potential of minerals based on chemical
reactions (hydration, hydrolysis and oxidation/reduction). Physical weathering and biologically mediated
weathering may play a role in mine drainage chemistry, but there is insufficient knowledge about these to
develop pragmatic predictive models, and an improved understanding of chemical weathering is likely to be
the priority over the foreseeable future.
Nicholson (2004) notes that several trace elements can exist in a substantially dissolved form under near-
neutral pH conditions; these elements include antimony, arsenic, cadmium, chromium, cobalt, copper, iron,
manganese, mercury, molybdenum, nickel, selenium, sulphate, uranium and zinc. Studies on dissolution
rates and water-borne concentrations of REE (Medas et al., 2013a, 2013b) under near-neutral conditions are
still in their infancy; however, this is likely to be a topic of increasing interest over the next decade. According
to Medas et al. (2013a), the solubility of REE generally decreases at higher pH; however, solubility can be
enhanced through REE complexation with carbonate ligands or humic substances. Complexation with
phosphate ligands, however, decreases solubility.
Selenium flux to the Elk River watershed in British Columbia, Canada, provides a good example of Type II
NMD (Lussier et al., 2003). Since the mid-1990s, it has become apparent that selenium concentrations in coal
mine-affected areas of the Elk River watershed have been increasing. Coal has been extensively mined in the
Elk Valley since 1897, and it is currently mined from five large open pits. Though sulphides are recognised as
important Se-bearing minerals, Lussier et al. (2003) did not observe a significant correlation between
selenium and sulphide concentrations in representative samples collected from the pit walls of the five coal
mines. Detailed studies on 16 samples that were generally representative of the Mist Mountain Formation
overall revealed that from 5% to 21.3% of the total Se was water soluble, 60% to 84% was associated with
sulphides and organic matter and 5.9% to 24.7% is in the silicate matrix. The extent to which watershed-level
effects was associated with direct dissolution (i.e., Type II NMD) rather than sulphide oxidation (i.e., Type I
NMD) in various source areas remains uncertain.
Selenium water-quality issues associated with coal deposits occur worldwide. Ziemkiewicz et al. (2011)
provide results for Se dissolution in coal mine drainage in West Virginia, U.S.A. Water-quality issues
associated with selenium are more commonly associated with discharge from low-sulphur waste deposits
originating from southern West Virginia coal seams than from pyrite-rich seams found in the northern portion
of the state. While the study results were not definitive, they suggested that selenium in the West Virginia
coal deposits exists separately from pyrite, its dissolution rate is independent of pyrite reactions and it exists
primarily as Se0 as opposed to selenide impurities in sulphide minerals.

Advances in our understanding of ML/ARD have resulted in the development and wide-scale use of simplified
diagnostic tests, including acid-base accounting (static tests) and kinetic tests such as humidity cell tests.
Excellent descriptions of mine drainage prediction methods are provided by Price (2009), INAP (2009), Morin
and Hutt (2001) and others. Figure 3 provides a simplified schematic of mine drainage prediction methods,
which is similar to the “wheel approach” devised by Morin and Hutt (2001).
Geochemical
modelling at
Catchment scale source
hydrogeological Site-based
modelling kinetic tests
Laboratory
On-site monitoring kinetic tests
and trend analysis (e.g. column tests;
Mine humidity cell tests)
Drainage
Chemistry
Prediction
Site conditions : Laboratory-based
meteorology & static tests:
climate, shake-flask tests/
hydrology, leachate tests,
hydrogeology acid-base
accounting,
NAG test
Bulk Chemistry: Detailed
Ore Mineralogical
Host Rocks Analysis
Waste Rock
Tailings
The nine major approaches illustrated in Figure 3, which each involve multiple tests and data needs, are not
mutually exclusive. In fact, all nine approaches are needed to varying degrees to develop an adequate
understanding of the influence of mine drainage on the surrounding environment.
Some of these tools have been developed to a high level of detail, such that their application and
interpretations have become relatively prescriptive. For example, standardised methods exist for acid-base
accounting (ABA) methods as do specific variations to account especially for the characteristics of sulphur in
tested geological samples. Conversely, preferred approaches for other important tools such as mineralogical
analysis or geochemical modelling remain relatively abstract and subject to substantial variations between
researchers.
The tools currently used in predicting mine drainage chemistry are too reliant on static laboratory tests on
limited size geological samples for practical application. Often, the effort applied to the other lines of
evidence is limited and subject to outcomes from ABA analysis. For example, the time and cost invested in
humidity cell tests and field kinetic testing is often quite limited, in our experience, if there is no appreciable
evidence from static tests of potential acid generating (PAG) materials. Such a simplification of the
investigative process may be appropriate in cases where the only truly plausible environmental issues arising
from mine drainage chemistry will be in association with ML/ARD; however, over-reliance on static test
results will make accurate predictions of environmental issues associated with NMD highly likely to fail.

Case Studies
The relative value of the nine different approaches for developing accurate predictions of mine drainage
chemistry and its associated environmental impacts is different for geological materials prone to NMD as
opposed to ARD. In particular, NMD predictions require a greater level of effort directed especially to detailed
mineralogical analysis, kinetic testing, geochemical modelling (including reactive transport modelling at
catchment scales) and validation of predictions through follow-up monitoring.
The scale of mining projects worldwide has increased with each passing decade. Contemporary studies of
chemical weathering processes and rates are increasingly relevant as a core theme in geomorphological
research and soil forming processes (pedogenesis), especially for dissolution and aqueous flux associated
with NMD. Such research has largely advanced independently of geochemical evaluations of mining
resources and wastes; since the early 1990s, it has focussed on alterations at mineral surfaces, weathering
in pedons, catchment-scale erosion and continental denudation (see Schott et al., 2012; Moore, 2008;
Phillips, 2005; White and Brantley, 1995).
In particular, many of the geochemical models used in mine drainage chemistry prediction assume dynamic
equilibrium and, in more complex cases, consider reaction kinetics. A number of recent studies of chemical
weathering in the context of pedogenesis, geomorphic change, large-scale biogeochemical cycles and climate
change influences have used models based on chemical dissolution and precipitation kinetics for various
minerals under either far-from-equilibrium or near-equilibrium conditions. Schott et al. (2012) discuss recent
advances in the use of mechanistic numerical models of weathering to predict biogeochemical changes over
scales of a kilometre to thousands of kilometres. Such numerical models generally rely on the Surface
Complexation approach (SC) combined with Transition State Theory (TST) to describe mineral dissolution
kinetics over wide ranges of solution composition. We are not aware that such models have been applied
specifically to mining-related problems; however, such approaches may be particularly promising for some
of the more intractable issues associated with contemporary mines, such as selenium flux into large
watersheds from large-scale coal mining.
Neutral mine drainage can arise under a variety of conditions and can result in aquatic environmental impacts
as a result of the dissolution, entrainment in mine drainage and persistence in surface runoff into water
bodies that support aquatic life. Depending on the minerals present, chemical accessibility, solution
chemistry of meteoric water introduced to the geological materials, sorption capacity along the flow path
and other factors, NMD can have several results: elevated dissolved-phase concentrations of sulphate, base
cations, metal/metalloid oxyanions and metal ions, and precipitation of calcite. Issues associated with NMD
arising from nonsulphide minerals have been given little attention in the context of mining. However,
relevant information on mineral dissolution rates has emerged based on chemical weathering studies driven
by an interest in pedogenesis, geomorphic change, large-scale biogeochemical cycles and climate change
implications. The mechanistic numeric weathering models that have been refined through these interests
may be of value in a mine-management context.
Overall, it is our hope that this paper advances the objectives of Nicholson (2004) in catalysing discussions
about NMD.
This paper was shaped by early discussions with Bill Price, to whom we are grateful.

Brandl, H., Hanselmann, K.W., and Bachenhof, R. (1990) In situ stimulation of bacterial sulfate reduction in sulfate-limited freshwater
sediments, FEMS Microbiology Ecology, Vol. 74, pp. 21–32.
Bright, D.A., Coedy, B., Dushenko W.T. and Reimer K.J. (1994) Arsenic transport in a watershed receiving gold mine effluent near
Yellowknife, Northwest Territories, Canada, Science of the Total Environment, Vol. 155, pp. 237–252.
Clow, D.W. and Drever, J.I. (1996). Weathering rates as a function of flow through an alpine soil. Chemical Geology. Vol. 132, pp.
131–141.
Demers, I., Molson, J., Bussière. B. and Laflamme, D. (2013) Numerical modeling of contaminated neutral drainage from a waste-rock
field test cell, Applied Geochemistry, Vol. 33, pp. 346–356.
Dill, H.G. (2010). The “chessboard” classification scheme of mineral deposits: Mineralogy and geology fro aluminium to zirconium.
Earth Science Reviews, Vol. 100, pp. 1–420.
Gaines, R.V., Skinner, H.C., Foord, E.E., Mason B. and Rosenzwieg, A. (1997) Dana’s New Mineralogy: The system of mineralogy of
James Dwight Dana and Edward Salisbury Dana, John Wiley and Sons Inc., New York.
Gilmour, C.C., Henry, E.A. and Mitchell, R. (1992) Sulfate stimulation of mercury methylation in freshwater sediments, Environmental
Science and Technology, Vol. 26(11), pp. 2281–2287.
Herlihy, A.T. and Mills, A.L. (1985) Sulfate reduction in freshwater sediments receiving acid mine drainage, Applied and Environmental
Microbiology, Vol. 49(1), pp. 179–186.
International Network for Acid Prevention (INAP) (2009) Global acid rock drainage guide (GARDGuide), International Network for
Acid Prevention, viewed 12 May 2015, http://www.gardguide.com.
Lussier, C., Veiga V. and Baldwin, S. (2003) The geochemistry of selenium associated with coal waste in the Elk River Valley, Canada,
Environmental Geology, Vol. 44, pp. 905–913.
Medas, D., Cidu, R., de Giudici, G. and Podda, F. (2013a) Geochemistry of rare earth elements in water and solid materials at
abandoned mines in SW Sardinia (Italy), Journal of Geochemical Exploration, Vol. 133, pp. 149–159.
Medas, D., Cidu, R., de Giudici, G. and Podda, F. (2013b) Geochemical behaviour of rare earth elements in mining environments under
non-acidic conditions, Procedia Earth and Planetary Sciences, Vol. 7, pp. 578–581.
Moore, J. (2008) Biogeochemistry of granitic weathering, PhD dissertation, Pennsylvania State University, State College, PA.
Morin, K. and Hutt, N.H. (2001) Environmental geochemistry of minesite drainage: practical theory and case studies, MDAG
Publishing, Vancouver, Canada, 19 pp.
Nicholson, R.V. (2004) Review of water quality issues in neutral pH drainage: examples and emerging priorities for the industry in
Canada, MEND Report 10.1, Mine Environment Neutral Drainage (MEND), viewed 12 May 2015, http://mend-
nedem.org/mend-report/review-of-water-quality-issues-in-neutral-ph-drainage-examples-and-emerging-priorities-for-the-
mining-industry-in-canada/.
Parliament of Canada (2014) The rare earth elements industry in Canada – summary of evidence, Canadian House of Commons
Standing Committee on Natural Resources, http://www.parl.gc.ca/HousePublications/Publication.aspx?DocId= 6669744&
Language=E&Mode=1&Parl=41&Ses=2 ).
Phillips, J.D. (2005) Weathering instability and landscape evolution, Geomorphology, Vol. 67, pp. 255–272.
Price, W.A. (2009) Prediction manual for drainage chemistry from sulphidic geological materials, MEND Report 1.20.1, Mine
Environment Neutral Drainage (MEND), viewed 12 May 2015, http://www.abandoned-mines.org/pdfs/ MEND Prediction
Manual-Jan05.pdf, 579 pp.
Price, W.A. (2005) List of potential information requirements in metal leaching and acidic rock drainage assessment and mitigation
work, MEND Report 5.1E, Mine Environment Neutral Drainage (MEND), viewed 12 May 2015, http://pebblescience.
org/pdfs/MEND_5_10E_Price_%20Final_Report.pdf.
Schott, J. Oelkers, E.H., Benezeth, P., Godderis, Y. and Francois, L. (2012) Can accurate kinetic laws be created to describe chemical
weathering? Comptes Rendu Geoscience, Vol. 344, pp. 568–585.
White, A.F. and Brantley, S.F. (eds) (1995) Chemical weathering rates of silicate minerals, Reviews in Mineralogy, Vol. 31.
Windward Environmental, Minnow Environmental Inc. and CH2M Hill Ltd. (2014) Elk River Watershed and Lake Koocanusa, British
Columbia, Environmental Synthesis Report, 2014, Report to Teck Coal Ltd., Sparwood, BC, http://www2.gov.bc.ca/
gov/DownloadAsset?assetId=52D09453DEB446228DF1838D65F2FFD5&filename=elk_river_aquatic_env_synthesis_report_
oct_2014.pdf, 223 pp.
Ziemkiewicz, P.F., O’Neal, M. and Lovett, R.J. (2011) Selenium leaching kinetics and in situ control, Mine Water Environment, Vol. 30,
pp.141–150.

T. Sulatycky Reclamation Operations Business Unit, Goldcorp Inc., Canada

D. Bucar Reclamation Operations Business Unit, Goldcorp Inc., Canada
P. Buttineau Porcupine Gold Mines, Goldcorp Canada Ltd., Canada
The Hollinger Mine, first discovered in 1909 by Benny Hollinger, was the largest gold producer in the prolific
Porcupine gold camp located in northern Ontario, Canada, producing more than 20 million ounces of gold
between 1910 and 1989. Between the mid 1980s and the 1990s, the mine site was left in a state of partial
abandonment by a previous owner. Because of the mining methods of the day, the upper levels of the mine
were often mined to surface or left voids with thin crown pillars, creating areas of potential instability, in
addition to the dozens of shafts and raises. Around the turn of the century, rising water levels in the mine
caused the sand backfill used to support underground voids during operation to mobilise deeper in the mine.
The loss of backfill caused large subsidences to occur on the surface around the property, putting many areas
at high risk. One measure implemented immediately to protect public safety was to surround the property
with chain-link fencing and conduct routine monitoring. This locked up 96 hectares of land in the middle of
the City of Timmins. Between 2000 and 2006, the prior mining companies worked with the Ministry of
Northern Development and Mines to rehabilitate mine hazards that were public safety concerns outside of
the core Hollinger property.
In 2007, the property was acquired by Goldcorp Inc. Given the proximity of the mine to the community,
Goldcorp could not in good faith accept a perimeter fence as an acceptable method of closure and began
investigating other options, such as selectively backfilling areas, constructing concrete caps over voids,
removing unstable crowns through blasting and fencing. Technically, several options were considered higher
risk given the complexity and amount of uncertainty. The preferred option consistently was removing crown
pillars and allowing the voids to flood thereby preventing a fall hazard.
In 2012, a closure plan amendment to develop an open pit mine that would centre over the historic mine site
was submitted to and approved by the Ministry of Northern Development and Mines. In 2014, following a
period of extensive community consultation, and with all necessary permits in hand, Goldcorp began the
construction phase, which included stripping overburden, blasting surface bedrock to create flat benches, and
building the perimeter environmental control berm (ECB). All activities required removal of residual gold and
historical mine hazards while working toward final reclamation of the land.
Goldcorp continues to work with the City of Timmins on a subsequent land-use plan such that at the end of
the five-year project, the property can become an area usable by the community for future generations.
Additionally, working with the local Aboriginal community on the application of traditional knowledge
combined with practices of modern rehabilitation techniques, Goldcorp’s end goal is to remove public hazards
and allow an appropriate post-mining land use of the property.
By the early twentieth century, rumours of gold in the Porcupine area, otherwise known as Timmins, located
in northern Ontario, Canada, had been circulating for some time. All prospecting expeditions at the time had
resulted in poor returns, though the efforts were sustained. In October 1909, Benny Hollinger and his partner,
Alex Gillies, started prospecting in the area, following the allure of gold that had inflicted so many others.
The two were exploring the site when Hollinger let out a roar of excitement. He had stripped moss away to
examine the bedrock outcrop beneath it, and found quartz that looked as though someone had dripped a

Developing the Hollinger open pit mine – leaving a positive legacy T. Sulatycky, D. Bucar and P. Buttineau
candle along it – but instead of wax, it was gold. The following year, aggressive development took place, and
supporting infrastructure was constructed at a feverish pace. It was an exciting time in the Porcupine camp.
April of 1968 marked an end to a brilliant chapter in the annals of Canadian mining, as mining operations
ceased at the Hollinger Mine. Producing just over 19 million ounces, the Hollinger Mine was the largest gold
producer in Canada for many years and, at one point, was the largest in the entire British Commonwealth.
With a history of experimentation, innovation and generosity with its acquired knowledge, the Hollinger
Mine contributed a great deal to the progress of mining in Canada and abroad. Within Timmins, the legendary
Hollinger Mine would forever be remembered for its riches, opportunities and community development.
The decision to terminate mining reduced the workforce to a skeleton crew as the mine moved to a state of
care and maintenance. Production shafts were capped with concrete. Equipment, infrastructure, and other
assets that were no longer necessary were sold off. Of particular interest, the company severed parcels of
land adjacent to the mine to support commercial development. The mine continued to maintain a depressed
water level as required to support operations at the neighbouring McIntyre Mine. In the years that followed,
the property would be transferred from owner to owner in search of any residual value. Several near-surface
extraction programs (micro-pits) were mined between 1976 and 1988, after which it was once again left in a
state of partial abandonment. Following the closure of the neighbouring McIntyre Mine in 1988, the Hollinger
Mine was allowed to flood. The legendary mine that once shone so brightly was reduced to a shadow of its
former self.

Case Studies
In the 1990s, as the water level in the mine continued to rise, there was growing concern about the impacts
at surface, particularly owing to the mine’s proximity to the City of Timmins. The mining methods of the day
included excavating from underground right through to surface, leaving shallow crown pillars. Imported sand
from local borrow sources was used to backfill stopes underground to provide structural support to the crown
pillar or support overlying soil. Once saturated, sand loses strength and results in backfill migrating deeper
into the mine as water rises, removing any structural support within the stope. The owner at the time, Royal
Oak Mines, was in a difficult position, with a struggling operation in the wake of deflated gold prices. As
anticipated, the loss of sand backfill in stopes caused crown pillars to fail, which resulted in dramatic
subsidences extending to surface at a number of locations on site and on former mine properties that had
since been redeveloped. The Ministry of Northern Development and Mines issued an order for Royal Oak
Mines to remediate properties of concern around the time the company declared bankruptcy in 1999.
In December 1999, Kinross Gold Corporation announced its intent to purchase Royal Oak Mine’s interest in
the Porcupine camp. As a condition of the purchase, Kinross Gold Corporation entered into an agreement
with the Ministry of Northern Development and Mines (MNDM) to remediate properties of concern. The
initial work consisted of erecting a perimeter fence around the entire mine site, locking up 96 hectares of
land from public access. A comprehensive geotechnical investigation on mine workings outside of the
perimeter fence was conducted, which led to the rehabilitation of dozens of shafts, raises and crown pillars
over a six-year period.
In 2006, the Porcupine Joint Venture (PJV), a joint venture between Kinross Gold Corporation and Placer
Dome, submitted a Closure Plan to the MNDM for the Hollinger Mine property. The proposed rehabilitation
plan for all mine hazards within the mine proper was to leave the area fenced in perpetuity.
Around this time, excitement was growing within PJV as a result of initial diamond drilling on the Hollinger
property.

The concept of mining the remaining gold left behind by underground mines through an open pit scenario
was common in the Porcupine Camp. The Dome open pit and Pamour open pit are both examples of
companies going back to a historic mine property and mining the remnant gold left in the walls of the stopes,
which had been uneconomical to mine using earlier underground extraction techniques.
Over the years, a number of companies had looked at Hollinger, the biggest gold producer in the Porcupine
camp, as an obvious target for this type of scenario. The exploration team at PJV decided to drill a handful of
holes to test the concept in 2004, with positive results. Each hole drilled returned mineable grades and widths
for a potential open pit scenario. As drilling increased, rumours of an open pit in the middle of the City of
Timmins began. From the public’s perspective, the thought of an open pit in downtown Timmins led to all
sorts of wild conclusions. A number of residents had visions of entire portions of the community being
relocated to accommodate this potential new mine. The work was classified within the company to help
prevent negative attention early on and nicknamed “Project Sandy” after Sandy McIntyre, founder of the
McIntyre Mine.
As momentum grew with the exciting results, the next few years brought an increased focus on exploration
drilling to prove up the concept. With eth mine now under the ownership of Goldcorp, the company was
developing plans for an open pit mine at the same time as the closure plan amendments were being drafted
to meet obligations under the Mining Act. A pre-feasibility study was launched to focus on a mining scenario
as well as the social, economic and environmental aspects of the project. Prior to engaging the community,
Goldcorp needed to determine whether there was a viable open pit project.
The first scenario was the same as the one presented the previous year – fence the property in perpetuity.
The second scenario was to rehabilitate areas close to the fence line and adjacent properties and allow for
some minor usage of the property. The third scenario was the option of rehabilitation through mining the
hazards. This solution would benefit the mining company, the workforce and the community.
Through numerous open houses held between 2007 and 2009, these three scenarios were presented to the
community as updated closure scenarios. The preferred mining plan was one of an open pit mined inside the
current fence line, removing the majority of mine hazards by mining through them without impacting the

Case Studies
surrounding property owners, displacing the former “super pit” rumour. Given the proximity to adjacent
landowners and to meet Ministry of the Environment (MOE) requirements, a large separation berm would
need to be constructed on the inside of the perimeter fence to a height of 20 m in order to protect the public
from noise and dust. All ore would be processed at the Dome Mine site and trucked via a newly constructed
haul road. Despite overall public optimism, public consultation became a large focus of the project as it took
shape and the permitting process got underway.
The closer a mine is to a residential area, the greater the impact felt by local residents and the greater the
need to build and maintain trust through transparency and dialogue. Throughout the planning and
development of the Hollinger project, community engagement was critical in understanding community
concerns and finding solutions to address those concerns. Dozens of public meetings, open houses,
presentations and community initiatives have taken place over the past seven years. Goldcorp also looked to
new channels of communication to ensure all stakeholders were kept informed and given several options to
provide their feedback on the operation. The following sections provide a glimpse of the channels proposed.
In 2010, the Hollinger Project Community Advisory Committee (HPCAC) was developed specifically for the
Hollinger open pit mine. The members consist of a diverse group of local stakeholders who critically examine
the issues associated with the project and work with Goldcorp to develop strategies that meet the mutual
needs of the stakeholders, community and company. The HPCAC ensures that stakeholders have the
opportunity to provide comments on key issues relating to the Hollinger open pit and provides a forum for
stakeholders to present their points of view on issues that fall within the mandate of the committee.
One of the key methods by which the project interfaces with the local community and other interested
parties is through the Hollinger Project Information Centre. The centre is a key focal point for channelling
information about the project to the public and is staffed by a full time community liaison coordinator. A
number of displays and information on the history of the Hollinger Mine and the future of the operation are
available for viewing at the centre.
The community feedback protocol, first developed to address the unique challenges associated with the
Hollinger project, now applies to all Timmins operation sites, inactive sites and exploration activities. It
provides another channel of communication between Goldcorp and its stakeholders and encourages
members of the community to register positive comments and observations as well as issues in reference to
the mines’ operations.
The objective of the protocol is to properly address feedback and comments from all stakeholders in a timely
manner and promote understanding of what the community is expecting of Goldcorp.
Stakeholders have a number of options through which to register their feedback, including a dedicated e-
mail address, by phone, in person or through an online form available on the Goldcorp website. Those
providing feedback can also choose to register their feedback anonymously.
Standard procedures ensure that everyone who wishes to be contacted will receive a personal reply within
24 hours. For urgent issues, the appropriate department is required to respond and take action to address
the problem. After the situation has been assessed and either resolved or mitigated, the complainant is
contacted again and informed of the action taken.
All feedback from the community, along with any resulting actions taken by Goldcorp, is captured in a
comprehensive database. The database is also capable of geo-tagging the location of an issue to easily

identify clusters. This information assists Goldcorp in addressing areas of concern through mitigation
measures and other programs.
During the public consultation process, the most common concerns raised by residents included noise from
operations, fugitive dust and vibrations from blasting. While these concerns are addressed through the
implementation of a variety of mitigation measures, Goldcorp wanted to ensure transparency with the
community throughout operations.
In 2012, the company launched a multi-discipline monitoring system that provides continuous live read-outs
of noise, dust and vibration levels from operations. A series of 17 monitoring stations was set up in strategic
locations outside of the environmental control berm (ECB) in the vicinity of sensitive receptors (e.g. local
residents and other land users) and within pit operations. The monitoring data are managed by a third party
to ensure objectivity and confidence in the results. Members of the public can access this data live from a
link on the Goldcorp website 24 hours a day, seven days a week.
The system is also closely monitored by Goldcorp, which allows the company to quickly respond to increased
levels. Internal warning notifications are sent out when sensors are approaching noise or dust limitations,
which allows the Goldcorp dispatchers to move or shut down equipment to lower noise in specific areas of
the pit or dispatch water trucks to reduce dust.
Because of the close proximity of nearby residential, commercial and industrial land users, Goldcorp
developed a best management plan (BMP) to act as a practical manual to address aspects relating to the
management and monitoring of blasting and vibration, noise, fugitive dust, site reclamation and contingency
plans for the proposed project. The BMP is intended to be a working document that may undergo periodic
revisions to accommodate changes in the mine plan and site permitting while ensuring that the objectives of
the BMP are consistently met.
The BMP incorporates best practices currently used in the mining industry in order to identify preventative
or mitigation measures that can be implemented to minimise potential impacts to nearby residents to
acceptable levels as defined by the City of Timmins, regulatory agencies and Goldcorp’s corporate
environmental and CSR standards.

Case Studies
Specific sections of the BMP provide descriptions of the strategies utilised to mitigate and monitor
environmental noise, fugitive dust emissions and ground vibrations, as well as outlining proposed
contingency plans to address potential unforeseen significant events. All of these factors are considered to
be the primary potential issues of concern for residential and commercial land users and for the City of
Timmins.
It was important to Goldcorp to include the community in the development of a subsequent land use plan.
To this end, a series of public consultation meetings and surveys was used to gather input from the
community on its vision of what the mine site could become at closure and post reclamation.
Initially, a working group of 15–20 stakeholders, including representatives from Goldcorp, the HPCAC and
the City of Timmins, was formed and a project charter was developed and became the guiding principles in
the development of the subsequent land use plan.
The objective of stakeholder communications is always to build and maintain trusted, two-way relationships
with communities, government agencies, NGOs and other stakeholders. At Goldcorp – with the Hollinger
operation located in the midst of a bustling urban centre – the need was more pressing, and the experiment
in communication has proven to be equally rewarding.
Following extensive baseline study and community engagement, a closure plan amendment was submitted
and subsequently filed for the project in June 2011, a considerable milestone in the project’s development.
Extensive work continued on the preparation and submission of supporting environmental permits, but this
would be the beginning of a series of challenging setbacks for the project.
The Ministry of the Environment required three permits:
1. a permit to take water (PTTW) for mine dewatering;
2. an environmental compliance approval (ECA) for industrial discharge to manage mine water
pumped from underground; and
3. an ECA for air and noise pollution generated during the operation.
Though the project had received early approval for mine water pumping, subsequent redesign of the project
required an amendment to the permit. The permit was appealed by an opposition group, which delayed
approval for about a year, until it was green-lighted in June of 2012.
The ECA for air and noise became the critical path to project success. First filed in 2011, the application
received project opposition, which Goldcorp addressed by working to revise and clarify the application to
respond to stakeholder concerns. In February 2013, a draft permit was issued, which was immediately
appealed on technical merit.
Following an impressive run in the gold price from 2009 to 2012, 2013 will be remembered as a year of cost
cutting, shelved capital projects and, in some cases, layoffs splattering the headlines as the price of gold
slowly but steadily declined to about $US 1,600/oz in April, at which point it began a precipitous fall to $US
1,200/oz. Investors retreated, and corporations were forced to tighten belts and look seriously at their
futures in light of a lower gold price.
As it was for many projects, 2013 was a turning point for the Hollinger project. The essential air and noise
ECA was in the appeal process, and, based on the economics of the day, the project in its existing form would
be challenged to meet targets. Engineers and geologists consistently revised, evaluated and optimised pit
designs and costs to ensure a robust project. However, not all projects can be evaluated on economics alone.
The Hollinger project is complex, and, in some ways, the social impacts would weigh just as heavily.

The possibility that a permanent fence line would be required in perpetuity if the project could not be
adequately justified raised some concerns. In order to further justify the project, Goldcorp conceptually
evaluated several remediation options internally. These options were developed under the assumption that
zero ore would be recovered.
The concept of remediating a mine hazard to protect the public is simple; either prevent access or remove
the hazard. Each option was evaluated against the following criteria as a measure of preference:
 post-operation available land for public use;
 social acceptance;
 capital and long-term costs.
Ontario Regulation 240/00 under the Mining Act provided specific criteria for and guidance on the
rehabilitation of mine workings and was assumed to be the minimum standard.
Access to a hazard can be prevented using a broad range of possible methods, of which fencing is likely the
least effective. That said, the base case was the cost to erect and maintain a perimeter fence, with
appropriate signage, around the entire mine site, which would protect the public by preventing access to
hazard areas. In practice, this is difficult to achieve, as a fence is often an invitation to curious folk. Timmins
has a long-standing history of public trespassing on old mine sites.

Case Studies
From an economic standpoint, fencing maintenance is not cheap, especially in perpetuity. The capital cost
would be small because there is an existing perimeter fence that meets Ontario Regulation 240/00. Costs to
maintain the 6 km fence were estimated at approximately $CDN 3 million when discounted into perpetuity.
From a social perspective, there would be no available land for public use and therefore it was anticipated
that this option would be unacceptable by the public. The base case was also least preferred by Goldcorp
because it was not the responsible thing to do wither corporately or socially. The liability would remain
behind a chain link fence and require constant monitoring and maintenance in perpetuity.
The first option for comparison against the base case was complete rehabilitation of all unstable crown
pillars, stopes open to surface, shafts and service raises. As described above, there are more than 150 mine
hazards within the perimeter fence that would require remediation for public access. Mine hazards were
assumed to be remediated using a combination of monolithic concrete caps, engineered backfill and plugging
voids with appropriately designed engineered concrete.
The complexity of the mine workings, their size and their density would require prolific engineering
resources. It was also believed that, even if viable, many of these workings would require long-term
geotechnical monitoring. From an economic standpoint, the capital costs would be extraordinarily high and
the project could take years to complete. Estimates ranged in excess of $CDN 150 million with a low level of
confidence. However if the project could meet with success, then the long-term fencing maintenance costs
would be removed.
Socially, this option was anticipated to be widely acceptable, as it allowed full access by the public to the 96-
hectare space. Though this option would make sense from a social and technical perspective, there was
limited confidence that this was a viable option. Equally important, this could potentially complicate or even
sterilise future mineral development.

The second option used for comparison was a combination of the first option and the base case. The concept
was to leave the dense, complex and central portion of the mine fenced off from access and rehabilitate the
isolated or perimeter mine workings in order to tighten up the overall fenced-in area and provide some
usable space. Upon review, three areas in particular were considered viable areas to remediate.
Voids around the periphery would be remediated using a combination of engineered fill and plugging the
void with an appropriate engineered concrete mixture. Instrumentation would likely be installed to monitor
the backfill, but the work could provide some usable space.
Though there was more comfort in the possibility of successfully remediating mine hazards within these
areas, the economics were not much more attractive. The void remediation work alone tipped $CDN 30
million, and fencing maintenance would actually increase to $CDN 3.5 million. Socially, this option was
anticipated to be more acceptable; however, from a liability perspective, the hazard would remain and allow
the public closer access to the central and more dangerous areas.
The best way to prevent access to a hazard is to remove it all together. The concept in option 3 was to remove
as many unstable crown pillars, raises, shafts and stopes open to surface as possible by blasting and hauling
rock to an elevation that would allow the area to flood. Once covered with water, any openings at that point
would no longer be considered a fall hazard. Areas that could not be captured using this method would
remain fenced, particularly peripheral areas or areas close to adjacent landowners. Essentially, the option
becomes a shallow open pit.
From an economic perspective, the capital costs would be high; however, the confidence in success would
be much higher than for options 2 or 3, which rely on highly technical engineering capabilities to remediate
complex voids. Fencing would remain in some areas, but the overall length would be shorter than for options
1 and 3. Further justification was that this option allowed the recovery of ore and access to potential ore
deeper in the mine from, which could generate cash flow that could be used to offset the rehabilitation costs.

Case Studies
Also of importance, this option most closely represented the scenario presented in community consultations
for the previous five years. Voids within the pit could not be accessed, unlike in options 1 and 3, and there
would be much less instrumentation monitoring compared to option 2. The mine would provide employment
for community members and a reasonable quantity of usable real estate for post-closure land use.
In September 2013, the project was given the green light by the Goldcorp board of directors under a revised
operating plan with the justification that the project would be considered a rehabilitation project. Shortly
thereafter, the appeal against the ECA air and noise was denied and the MOE issued the final permit to begin
construction.
In 2014, a construction crew was contracted to assist Goldcorp in the stripping of overburden and
construction of the ECB. Overburden was initially hauled back to the Dome Mine site to be stockpiled. In turn,
waste rock was picked up and hauled back to the Hollinger property to construct the berm. In February 2014,
after much anticipation, the first blast was completed. The blasted waste rock was added to berm
construction. Additionally, the stripped overburden would be placed on the exterior slope of the berm to
form a vegetative medium for future growth.
In January 2014, Goldcorp retained a consultant to assist in the development of a subsequent land use plan
for the Hollinger open pit mine site. The plan outlines the vision for the site after closure based on key
elements included in an agreement with the City of Timmins. Elements under consideration include trails,
aesthetically appealing fencing, benches, a public park, final configuration and landscaping of the berm, and
safe access to the flooded pit, possibly in the form of a beach area. Based on these key elements, the site is
to be rehabilitated into a recreational area that the public can access safely. The plan will be presented to
Timmins City Council for approval in 2015.

Over the last 10 years, the project has met with many challenges and successes. Given the complexity of the
project, it is not surprising that it took this long to get where it is today, even in a mining-friendly town and
the safe jurisdiction of Ontario, Canada. The following highlights some of the lessons learned from the project
to date.
In hindsight, excitement about the project in the early stages of development coupled with a soaring gold
price led to some barriers that could have been prevented. From a project management perspective, the
project moved from pre-feasibility to execution in one step, which led to lower confidence estimates and
designs that added costs during construction.
Initially, the void remediation work, particularly where voids were located underneath the ECB, was not
envisioned to pose a significant risk to berm construction. This proved to be false following a detailed
geotechnical investigation. To meet standards under the Mining Act such that areas could be safe for public
access would require a well thought-out and executed design.
At the time, it took nearly a week to collect readings from the dozens of geotechnical instruments scattered
across the mine site and adjacent properties. During operation, it was anticipated that the monitoring
frequency would increase and more instruments would be required, making it a full-time job.
Project sensitivity and permitting are often difficult to predict, even by the most experienced consultants and
miners. In the case of the Hollinger, it would have been a stretch to accurately plan for the two to three years
of permit review while the project sat in a quasi-dormant state.

Case Studies
The limitations under the BMP for noise fell under two states: (1) construction, which applied up until the
ECB was complete, and (2) mining, which applied thereafter. The limitations have proven to be challenging
to meet in the early stages of construction. The operations team has adjusted procedures and daily activities
frequently as noise-triggered alerts halt activities in a certain area.
Safe production is the highest priority at Goldcorp. Preparing the site for mining activities as safely as possible
was not anticipated to be as challenging, but procedures used at previous open pit mines were not easily
transferrable and tasks took much longer than originally envisioned.
In 2008, no one anticipated the collapse of the U.S. economy. That said, planning at $US 1,800/oz gold versus
$US 1,200/oz gold changed perspectives in a hurry. Had a contingency plan been in place early on, there may
not have been as much of a struggle to gain project confidence when the price of gold did fall.
Community engagement early on in the process was largely considered to play a critical role in project
development. Given the sensitive nature of the project, Goldcorp continues to communicate that it will not
leave the project in the same state as its predecessors, but instead leave a positive legacy that the community
can be proud of. Allowing residents to weigh in early on in the development of the project, raise concerns
and have their voices heard won the community over, ensured a transparent process and provided some
comfort to the regulatory authorities.
As the mine continues to advance, the community feedback protocol ensures that issues are resolved as
quickly as possible. Feedback is heard, responses are prompt and there is open dialogue. Live operations
monitoring also gives the community or other watchdog groups full disclosure of day-to-day activities such
that the project’s operations remain visible and transparent.
Though the community has a century’s worth of experience with mining, poor management and corporate
ethics by historical mining operations in recent memory have left a sour taste for many Timmins residents.
The effort by Goldcorp to grow community support and operate transparently at Hollinger has provided the
necessary assurance to the community, and that has set the bar for future mine development.


B.H. Cranston Paragon Soil and Environmental Consulting Inc., Canada

L.A. Leskiw Paragon Soil and Environmental Consulting Inc., Canada
In areas where soil conditions are poor, successful revegetation and reclamation depends directly on the
ability of the soil amendments to ameliorate growing conditions. The primary objective of this study was to
evaluate the efficacy of various amendments to improve soil conditions, supporting growth and survival of
planted boreal forest species. A secondary objective was to evaluate the success of forest floor transplants
for moss and lichen establishment.
This study was conducted at a reclaimed site at the Cameco Key Lake uranium mine in Saskatchewan, Canada.
Soil amendments applied included commercial peat, lake sediments, flax straw, LFH-mulch, NPK fertiliser and
manure pellets, as well as a control (no amendment). The lowest survivorship of planted vegetation (including
forest floor transplants) was observed in plots treated with fertiliser (69% survival), relative to all other
amendments (81% survival); survivorship was not directly facilitated by any soil amendment. Growth of
planted species (2013–2014) was greatest in the manure pellet (11.4 cm), LFH-mulch (11.4 cm), and peat plots
(9.7 cm). Manure pellet and LFH-mulch treatments also encouraged the natural establishment of the most
species (5 species per plot). Moreover, natural moss establishment (Polytrichum juniperinum) in the manure
pellet plots in 2014 was so extensive (~63% cover) relative to all other treatment plots (typically <5% cover)
that manure pellet plots are now easily identifiable using 2014 satellite imagery. Forest floor transplants were
also confirmed to be a successful method of introducing lichen and moss species to a reclaimed site.
The poor growing conditions associated with mine spoil can directly inhibit the establishment and survival of
all but the most tolerant plants. The results suggest that the addition of organic soil amendments, particularly
manure pellets, can improve growing conditions for planted seedlings as well as for naturally established
vegetation, facilitating the development of self-sustaining, locally common vegetation communities.
Reclamation of mined sites may be achieved using a range of strategies, but the long-term objectives of
rehabilitation can best be reached through the use of vegetation as a basis for stabilisation, aesthetics and
pollution control (Bradshaw and Johnson, 1992; Ye et al., 1999; Tardoff et al., 2000; Haynes, 2009). One
successful approach to revegetation is the addition of soil amendments to improve growing conditions for
vegetation. Which amendment or amendments to apply depends on the starting soil conditions, including
soil chemistry, water-holding capacity, texture and biota (Kumpiene et al., 2008).
Mine spoil is often low in organic matter, nutrients and buffering capacity, and is generally coarse textured,
with low water-holding capacity (Voeller et al., 1998). The addition of organic amendments is a common
technique used to alleviate poor soil conditions (Petersen et al., 2004; Zink and Allen, 1998; Biederman and
Whisenant, 2009) and to increase soil nutrient levels (Blumenthal et al. 2003). The improvements to physical
soil characteristics (including reduced compaction and increased water-holding capacity) may be most
important (Beare et al., 1992; Biederman and Whisenant, 2009) where nutrient levels are not limiting.
Amendments are considered to be a low-cost method of enhancing soil conditions and thus facilitating the
establishment of target vegetation. Organic amendments such as mulch, dewatered biosolids, domestic
refuse, peat and topsoil have been successfully applied to improve the physical properties of the rooting
medium, reduce phytotoxicity and act as a source of slow-release nutrients (Voeller et al., 1998; Tardoff et
al., 2000; Banerjee et al., 2006). Inorganic amendments such as lime and NPK fertilisers have also been used
to buffer pH levels and increase bioavailable soil nutrients (Ye et al., 1999; Andersen and Ostler, 2002;

Using applied research to reclaim a northern Canadian uranium mine B.H. Cranston and L.A. Leskiw
Banerjee et al., 2006), but they do not add organic matter. The growth limitations of target species will help
to determine whether organic or inorganic amendments are most appropriate.
Successful revegetation depends directly on the ability of the amended soil to limit environmental stress on
plants, improve the overall soil conditions and reduce competition with undesirable species (Biederman and
Whisenant, 2009); this is especially important during the early stages of plant development. In Canada,
regulators require the use of native species in most cases. Selecting genotypes tolerant of specific adverse
conditions (e.g. salt, metal, drought, and cold) can increase the survival rates of planted or seeded plants
(Bleeker et al., 2002). Some species, regardless of stress tolerance, are difficult to re-establish in reclaimed
areas; moss and lichens are notoriously so. Ground cover species are critical components of some boreal
forest community types, yet little is known about the requirements for establishing them at reclaimed sites.
Forest floor transplants, including surface soil as well as any native groundcover, graminoid, herbaceous or
woody species, have been used successfully to restore diverse plant communities (Conlin and Ebersole, 2001;
Krautzer and Wittmann, 2006; Aradottir, 2012). The feasibility of using forest floor transplants to re-introduce
moss and lichens to reclaimed areas has not been tested on a practical scale.
The primary objective of this study was to evaluate the efficacy of various soil amendments to improve soil
conditions, thus supporting vegetative growth and survival of planted boreal forest species at a site
undergoing reclamation: the Cameco Key Lake uranium mine. A secondary objective was to evaluate the
success of forest floor transplants for moss and lichen establishment. Incorporating soil amendments was
expected to facilitate the growth and survival of planted vegetation, with those amendments that introduce
the largest amount of organic matter (commercial peat, mulch, manure pellets and straw) having the
strongest effect. These amendments would not only improve the physical conditions of the soil, but also
directly increase the levels of potentially limiting nutrients. A simultaneous companion study — focusing
specifically on soil conditions at the study site — used three criteria (soil chemistry/toxicity, nutrient supply
and water storage capacity) to evaluate the effects of soil amendments on vegetation performance (L.
Olatuyi, personal communication, 2014). The results of the companion study were used to help explain
patterns observed in the establishment, survival and growth of the vegetation.
The study site was located at Cameco’s Key Lake mine in northern Saskatchewan at 57°12'55.14"N,
105°38'05.99"W. Test plots were established on the Deilmann North waste rock pile on 24–27 September
2010. The site is an 81 ha out-of-pit waste rock pile constructed between 1984 and 1997. The rock pile is
composed of approximately 23% sand/till, 47% sandstone and 30% basement rock by volume as excavated
from the Deilmann pit. Basement rock is the deepest geologic formation in this area, overlain by sandstone
and glacial till and outwash. Glacial till materials were excavated from a nearby drumlin and hauled to the
site to create a cover with a nominal thickness of 1 m over the waste rock, which had been graded and roller-
compacted to a depth of about 20 cm prior to cover placement.
The Key Lake region has a continental sub-arctic climate, with warm summers and cold winters. Thirty-year
climate normals (1981–2010) from Environment Canada climate data show an annual cumulative
precipitation of 483 mm, with 318 mm occurring as rainfall. Of the 318 mm rainfall, approximately 69% (217
mm) occurs during the frost-free period between June and August. Natural vegetation in the area is
dominated by boreal species including jack pine (Pinus banksiana Lamb.), white birch (Betula papyrifera
Marshall), blueberry (Vaccinium myrtilloides Michx.), cranberry (Vaccinium vitis-idaea L.), green alder (Alnus
viridis [Chaix] DC), and a variety of moss and lichen species.
Soil and vegetation monitoring plots were established on the study site in 2010. The test plots were grouped
into three replicate blocks: A, B and C (Figure 1). Each block was divided into seven random soil amendment

Case Studies
plots and one control plot, each measuring 9 m × 9 m in size. Each plot was further subdivided into nine 2 m
× 2 m vegetation subplots. Buffer zones (one metre in width) were established between each subplot as well
as between each test plot. Soil amendments applied included no amendment (control), commercial peat
(peat), lake sediments (sediment), flax straw (straw), LFH-mulch (mulch), NPK fertiliser (fertiliser) and manure
pellets (pellets) (Table 1).
Amendment Placement depth Incorporated* Additional notes

Peat Top 10 cm Yes Sunshine PeatTM applied at 6–12 cm
depths
Sediment Top 6–12 cm Yes Blended with subsoil at approximately
1:1
Flax straw 40–50 cm below No Used as a capillary break to increase
surface water-holding capacity
Mulch Top 20 cm No LFH/mulch materials previously stripped
by Cameco and re-distributed over study
plots in mounds
Fertiliser Surface No 18-24-11 (N-P-K) at 400 kg/ha
Pellets Top 10 cm Yes 20 t/ha, sourced from EarthRenewTM.
Manure processed to remove weeds and
pathogens
*Amendments were incorporated into the upper subsoil material using a roto-tiller
Species planted in the vegetation subplots include polytrichum moss transplants, forest floor transplants
(ecoslices), jack pine, bearberry (Arctostaphylos uva-ursi [L.] Spreng.), blueberry, paper birch, sandbar willow
(Salix exigua Nutt.), green alder and trembling aspen (Populus tremuloides Michx.). Seedlings for tree and
shrub species were sourced from Eastern Slopes Rangelands Seeds in Calgary, Alberta, and from Coast to
Coast Reforestation’s Smoky Lake, Alberta, nursery. Seedlings were kept cool and moist during transportation
to the site. All seedlings were wetted with approximately 1–2 litres of water per plant following planting.
In September 2010, all vegetation subplots were randomly assigned one of the seven target tree/shrub
species and planted with four seedlings of that species (three birch seedlings were planted in each plot based
on availability). Owing to plant dieback following the first winter (2011) and poor performance of vegetative
growth during the first spring (2011), all treatment plots were replanted with four additional seedlings (n = 8
seedlings per subplot) in fall 2011 (three for birch; n = 6 seedlings per birch subplot). Seedlings for the
replanting were obtained from the same sources as in 2010. Moss was transplanted in “plugs.” Four plugs
were transplanted to each of the moss subplots in 2011. Four ecoslices of LFH from the forest floor were
transplanted to each of the LFH subplots in 2012. Ecoslices were dug with a spade and were approximately
15–20 cm in diameter and approximately 5 cm thick.
Plant performance was assessed during late summer from 2011 through 2014. Here we present the results
from the 2014 monitoring year. Plant performance was measured in terms of survival and growth of planted
species. The diversity of naturally established species was also determined for each treatment plot and
vegetation subplot.

Vegetation inspection data for 2014 were analysed for plant survival, height and growth across species and
among treatment plots using linear mixed models (LMMs) with plant species × soil treatment + year of
planting as fixed effects, and set (A – C) as a random effect. The number of species that had naturally
established in each treatment subplot was also tested using these fixed and random effects. Significant
differences among treatment means were determined at p < 0.05 using the Tukey HSD post hoc test.
All analyses were completed using R version 2.9.2. All values are reported as mean ± standard error (SE).

Case Studies
Per cent survival of vegetation planted in 2010 and 2011 was examined among treatment plots for each
species in September 2014 (Figure 2). Survival was significantly reduced in fertiliser plots (68.8 ± 1.9%)
relative to all other treatments (81.2 ± 1.6%; p < 0.01, for all contrasts) with the exception of mulch (74.7 ±
1.6%; p = 0.16). The control (82.8 ± 1.5%) and straw plots (86.0 ± 1.4%) had the greatest survivorship. Survival
of planted species in the control plots was significantly greater than that in both fertiliser (p < 0.001) and
mulch (p = 0.01) plots.
In terms of species planted, survivorship of bearberry (46.9 ± 1.9%) and willow (70.5 ± 1.8%) was low in 2014,
although willow did spread significantly via suckering. Alder, aspen, birch, blueberry, ecoslices, moss, pine
and willow all had significantly greater per cent survival than bearberry, regardless of the soil treatment (p <
0.001 for all contrasts). Birch (91.3 ± 1.3%), ecoslices (97.7 ± 0.8%), moss (93.2 ± 2.3%) and pine (88.1 ± 1.0%)
had the highest per cent survival of the planted species (p < 0.05 for all contrasts with alder, aspen, bearberry,
blueberry and willow) (Figure 3).

By September 2014, the average height of planted vegetation was significantly greater in the plots treated
with manure pellets (73.8 ± 7.3 cm) than in all other treatment plots (55.3 ± 7.5 cm; p < 0.001 for all contrasts).
The average heights of all planted species are presented in Figure 4. From 2013 to 2014, growth of woody
species (alder, aspen, pine, birch, blueberry, bearberry) was greater in plots treated with manure pellets (11.4
± 2.7 cm), mulch (11.4 ± 2.2 cm) and peat (9.7 ± 2.3 cm) than in all other treatment plots (Figure 5), but owing
to the amount of standard error, these differences are not significant (p > 0.3 for all contrasts). The year of
planting (2010 or 2011) had no effect on the growth of tree species (p = 0.64).

Case Studies
The number of species that naturally established (ingress) in plots was dependent on the soil treatment
applied (p = 0.001; Figure 6), but not on the species planted in each plot (p = 0.95). Post hoc analysis indicated
that pellets (5.4 ± 0.3) and mulch (5.2 ± 0.4) supported a greater number of naturally established species than
the sediment, straw, fertiliser and control plots (2.7 ± 0.3; p < 0.01 for all contrasts). The peat (4.3 ± 0.4) plots
also supported greater numbers of naturally establishing species than both the sediment (1.6 ± 0.1; p < 0.001)
and control (2.7 ± 0.3; p = 0.02) plots.

During 2013 monitoring, moss cores were beginning to show signs of spread from their planting locations. In
September 2014, significant establishment of the moss Polytrichum juniperinum was observed. The vast
majority of the moss cover was observed in plots treated with manure pellets. These had significantly greater
moss cover (58 ± 5%) than all other treatment plots, which had an average of 3 ± 1% cover (p < 0.0001 for all
contrasts). It is not clear that the establishment of moss in the pellet plots is the result of spread from the
transplants.
Reclaimed mine sites are manmade habitats and come with a range of inherent challenges for vegetation
establishment (Wong, 2003). These sites often lack topsoil and a continuous source of nutrients, and are
subject to sheet erosion, drought, surface mobility, compaction and wider temperature fluctuations than
natural sites (Wong et al., 1999; Wong, 2003). Because of these limitations, it can often be difficult to recreate
the ecological functions of natural sites; adding organic amendments to poor-quality reclaimed soils is
essential for the efficient and successful development of vegetation communities (Wong, 2003).
The results of this study support the expectation that soil amendments can increase the performance of
planted species at a reclaimed uranium mine site. Furthermore, some of the organic amendments (manure
pellets and LFH-mulch) had stronger positive effects on vegetation performance than did the mostly
inorganic amendments (fertiliser, sediments). Of the organic treatments, manure pellets performed the best
by not only increasing the performance of planted vegetation, but also encouraging native species
establishment. In terms of the performance of specific species, bearberry and willow had the lowest survival
rates, although willow did show significant spread through suckering. Moss and lichen establishment through
transplanting was determined to be a viable method of groundcover revegetation. These results could have
broad implications for revegetation practices in mine reclamation and have been used in the development
of a revegetation plan for the area.
Results of the companion soil study (L. Olatuyi, personal communication, 2014) were used to help explain
patterns observed in the establishment, survival and growth of the vegetation. Relevant observations are
presented in the sections below.

Case Studies
Survivorship of planted species was not directly facilitated by any soil amendment; the peat, pellet and straw
amendments did not differ significantly from the control. Survivorship was in fact significantly reduced in the
fertiliser amendment plots relative to the control. Low pH (average pH = 5.9 in upper subsoil) may have been
a driver of the low vegetation performance observed in the fertiliser amendment plots. The acidic soils in the
fertiliser treatment would inhibit plant growth by reducing the amounts of available water and nutrients via
increased nutrient leaching and the inhibition of root growth (Marschner, 1991). For newly planted seedlings,
this could result in a failure to establish and thrive. NPK fertiliser had little or no treatment effect on salinity
and fertility levels relative to the control plots, with the exception of phosphorous, which was two times
higher than background levels.
Unlike vegetation survival, vegetation growth showed a significant positive response to some soil treatments.
This is a common observation, as the limitations on growth are often more subtle than the limitations on
survivorship. In the present study, the average growth of planted species in the pellet treatment was
significantly greater than in all other soil treatments, including the control, where no amendment was used.
The manure pellets appear to have provided a slow-release source of soil nutrients, particularly calcium,
magnesium and phosphorus, which were greater than the levels observed in both the control and fertiliser
plots. Additionally, the average pH level for the pellet treatment plots was 7.4; this level maximises nutrient
availability to plants, as well as reducing levels of metals (Wong, 2003).
In terms of the survival of specific planted species, bearberry fared the worst regardless of the soil
amendment added. This may be because bearberry is an ericaceous species that often relies on ericoid
mycorrhizae to survive harsh environments (Linderman and Call, 1977). Ericoid mycorrhizae produce a range
of hydrolytic and oxidative enzymes important in mobilising nutrients from organic complexes in the soil
(Cairney and Meharg, 2003). The lack of both organic complexes and mycorrhizae in mine spoil may explain
bearberry’s poor survivorship. Without inoculation of mycorrhizae, ericoid species may be difficult to
establish and inappropriate for early stages of revegetation.
Survival of the moss and ecoslice transplants was greatest among all planted species. This suggests that moss
plugs and forest floor transplants are an effective means of introducing moss and lichen propagules to a
newly reclaimed site. Owing to the slow-growing nature and dispersal mechanisms of groundcover species,
natural establishment of mosses and lichens from the periphery of the study site can be prohibitively slow.
Considering the importance of establishing vegetation cover early for biotic (exclusion of weeds and
undesirable species, creation of microsites and facilitation of target species) and abiotic (erosion control, soil
fertility and moisture retention) benefits, introducing these species through transplants is critical.
In addition to an amendment’s ability to facilitate the survival and growth of planted vegetation, it is also
critical for it to facilitate the natural establishment of native vegetation in order to promote a self-sustaining,
locally common habitat. In the present study, the pellet and mulch treatments supported more naturally
established species than did the control. As for growth rates, the neutral pH and availability of calcium,
magnesium and particularly phosphorus facilitated the natural establishment of vegetation in the pellet
treatment plots. The peat plots were also found to have higher volumetric water contents than the control
plots. Moreover, plots treated with manure pellets also showed an obvious and significant increase in moss
cover between 2013 and 2014. At this stage, the driver(s) behind the increased rate of moss establishment
in the pellet plots is not clear. It does not appear to be a result of spread from the moss plugs alone but may
be due to more supportive soil chemistry and fertility. These results add further support to our prediction
that organically based amendments would provide the best growing conditions for planted vegetation on
uranium mine spoil. A combination of organic amendments that includes manure pellets and LFH-mulch is
expected to provide the most suitable growing conditions for vegetation. In areas adjacent to the study plots
(Figure 1, plots A8, B8 and C8) amendments left over from the study were combined and incorporated into

the soil, which was then mounded. Though not monitored for this study, they were observed anecdotally to
have higher species diversity and growth rates than any of the single-amendment study plots.
The high degree of soil disturbance in mine spoil directly inhibits establishment and survival of all but the
most tolerant plants (Wong, 2003). The degree of disturbance, including the fact that soils may be excavated
from great depth, also affects the number, diversity and activity of soil organisms, inhibiting soil organic
matter decomposition and nitrogen mineralization processes in an already challenging environment. The
addition of organic soil amendments, particularly manure pellets, can improve growing conditions for planted
seedlings and mosses as well as for naturally established vegetation, facilitating the development of self-
sustaining, locally common vegetation communities.
The authors would like to thank Lekan Olatuyi, Konstantin Dlusskiy and Kurtis Trefry for assistance in the
field, as well as the Paragon team that initially set up the experiment. Thank you also to Cameco Key Lake
uranium mine, particularly Dana Fenske, for providing the study site, funding and invaluable logistical and
field assistance. Paragon Soil and Environmental Consulting was contracted by Cameco to complete this
work.
Andersen, D.C. and Ostler, W.K. (2002) Revegetation and degraded lands at U.S. Department of Energy and U.S. Department of
Defense installations: Strategies and successes, Arid Land Research and Management, Vol. 16, pp. 197–212.
Aradottir, A.L. (2012) Turf transplants for restoration of alpine vegetation: Does size matter?, Journal of Applied Ecology, Vol. 49, pp.
439–446.
Banerjee, M.J., Gerhart, V.J. and Glenn, E.P. (2006) Native plant regeneration on abandoned desert farmland: Effects of irrigation,
soil preparation, and amendments on seedling establishment, Restoration Ecology, Vol. 14, pp. 339–348.
Beare, M.H., Parmelee, R.W., Hendrix, P.F., Cheng, W., Coleman, D.C. and Crossley, D.A. (1992) Microbial and faunal interactions and
effects on litter nitrogen and decomposition in agroecosystems, Ecological Monographs, Vol. 62, pp. 569–591.
Biederman, L.A. and Whisenant, S.W. (2009) Organic amendments direct grass population dynamics in a landfill prairie restoration,
Ecological Engineering, Vol. 35, pp. 678–686.
Bleeker, P.M., Assunção, A.J.L., Teiga, P.M., de Koe, T. and Verkleij, J.A.C. (2002) Revegetation of the acidic, As contaminated Jales
mine spoil tips using a combination of spoil amendments and tolerant grasses, The Science of the Total Environment, Vol.
300, pp. 1–13.
Blumenthal, D.M., Jordan, N.R. and Russelle, M.P. (2003) Soil carbon addition controls weeds and facilitates prairie restoration,
Ecological Applications, Vol. 13, pp. 605–615.
Bradshaw, A.D. and Johnson, M. (1992) Revegetation of metalliferous mine waste: The range of practical techniques used in Western
Europe, Elsevier, Manchester, UK.
Cairney, J.W.G. and Meharg, A.A. (2003) Ericoid mycorrhiza: A partnership that exploits harsh edaphic conditions, European Journal
of Soil Science, Vol. 54, pp. 735–740.
Conlin, C.B. and Ebersole, J.J. (2001) Restoration of an alpine disturbance: Differential success of species in turf transplants, Colorado,
USA, Arctic, Antarctic, and Alpine Research, Vol. 33, pp. 340–347.
Haynes, R.J. (2009) Reclamation and revegetation of fly ash disposal sites — Challenges and research needs, Journal of Environmental
Krautzer, B. and Wittmann, H. (2006) Restoration of alpine ecosystems, in Restoration ecology: The new frontier, J. van Andel and J.
Aronson (eds), Wiley-Blackwell, Malden, USA, pp. 208–220.
Kumpiene, J., Lagerkvist, A. and Maurice, C. (2008) Stabilization of As, Cr, Cu, Pb, and Zn in soil using amendments — A review, Waste
Linderman, R.G. and Call, C.A. (1977) Enhanced rooting of woody plant cuttings by mycorrhizal fungi, Journal of the American Society
for Horticultural Science, Vol. 102, pp. 629–632.
Marschner, H. (1991) Mechanisms of adaptation of plants to acid soils, Developments in Plant and Soil Sciences, Vol. 45, pp. 683–
702.
Petersen, S.L., Roundy, B.A. and Bryant, R.M. (2004) Revegetation methods for high-elevation roadsides at Bryce Canyon National
Park, Utah, Restoration Ecology, Vol. 12, pp. 248–257.
Tardoff, G.M., Baker, A.J.M. and Willis, A.J. (2000) Current approaches to the revegetation and reclamation of metalliferous mine
wastes, Chemosphere, Vol. 41, pp. 219–228.
Voeller, P.J., Zamora, B.A. and Harsh, J. (1998) Growth response of native shrubs to acid mine spoil and to proposed soil amendments,
Plant and Soil, Vol. 198, pp. 209–217.

Case Studies
Wong, M.H. (2003) Ecological restoration of mine degraded soils, with emphasis on metal contaminated soils, Chemosphere, Vol. 50,
pp. 775–780.
Wong, M.H., Lan, C.Y., Gao, L. and Chen, H.M. (1999) Current approaches to managing and remediating metal contaminated soils in
China, in Proceedings of the 5th International Conference on Biogeochemistry Trace Elements, Vienna, Austria, July 1999.
Ye, Z.H., Wong, J.W.C., Wong, M.H., Lan, C.Y. and Baker, A.J.M. (1999) Lime and pig manure as ameliorants for revegetation lead/zinc
mine tailings: A greenhouse study, Bioresource Technology, Vol. 69, pp. 35–43.
Zink, T.A. and Allen, M.F. (1998) The effects of organic amendments on the restoration of a disturbed coastal sage scrub habitat,
Restoration Ecology, Vol. 6, pp. 52–58.
Olatuyi, L. (2014) Paragon Soil and Environmental Consulting, Edmonton, Canada, personal communication.


S.E. Ames AMES Mine Closure and Reclamation International, Canada
Most early mine operations did not include closure planning. When closing an abandoned mine, operators
are faced with all of the problems that have been left, for example, old equipment and structures,
contamination, unstable landforms and a lack of soil to reclaim the disturbed areas. Closure planning is now
a requirement for both new and operating mines around the world. Its focus is on meeting the end land use
objectives and regulatory requirements. Early closure planning at the mine development stage provides an
opportunity to design landforms for successful closure and more easily meet the end land use objectives of
the site while complying with regulatory requirements. These opportunities may be lost if closure planning is
delayed. For example, the options for reclaiming waste rock piles may be reduced if planning is carried out
later in a project. Early planning can reduce future closure costs by minimising rehandling of mine wastes.
This also affects bonding costs. Innovative approaches for closure that provide long-term site and
environmental sustainability at the post-closure stage increase the opportunity for regulatory and
stakeholder approval, allowing a project to be permitted sooner. Early closure planning also enables
coordination of all aspects of mine development that will allow the closure of the project to occur as planned.
There are challenges in coordinating engineers, wildlife biologists, soil scientists and water specialists to
develop a cost-effective and sensible plan for closure. Inputs and considerations from these different
disciplines in mine planning result in a closure plan that is generally based on current conditions of the mine
site and ensures there is more opportunity to return the site to its former condition, as much as practical.
Sometimes, temporary or early closure is required. For example, some early infrastructure may be buried by
later infrastructure so if a project is stopped, the closure requirements will be different than if the project
proceeded. Early planning for these unexpected situations may also affect closure and bonding costs.
This paper discusses opportunities that were realised when mine planning occurred early in a project and the
challenges and approaches that were encountered when closure planning occurred during operations and
when a site had been abandoned. These approaches can be used in mines in Canada and all over the world.
Closure planning allows the closed mine site to partially disappear into the surrounding environment, as mines
are only borrowing the site for a period of time.
Mine closure planning was not considered over the last several hundred years of mining and therefore, there
are abandoned mines all over the world. Mine closure planning has become mandated recently in many
jurisdictions. The concept of closure and reclamation planning, however, is broad and variable as there are
no international standards. Closure planning often includes planning for reclamation, which is generally
considered to mean revegetation. Closure and reclamation plans can range from a single paragraph to a
detailed plan where species are identified and seeding rates are included. The inclusion of a statement that
the operator will close and reclaim a site is often all that is required to get a project permitted. Therefore,
there is inconsistency regarding the meaning of mine closure and reclamation planning and the requirements
for obtaining a permit. Often the benefits of a mine in a jurisdiction are so critical that environmental
considerations are overlooked to some extent. Often there are fewer requirements for closure planning in a
location where mining is a new industry and the experience of the regulators is more limited than in a
jurisdiction where several mines have operated. In some regions of the world, revegetation is less challenging
than in other areas and therefore, reclamation is less of an issue. The concept of a closure and reclamation
plan is not firm, and the stage of a project when the closure and reclamation plan should be developed is
also not clear. In British Columbia (BC), Canada, there are clear and documented guidelines for mine closure

Opportunities to incorporate mine closure into mine planning S.E. Ames
and reclamation and when this should occur. Closure and reclamation planning is required for new projects
and the expansion of existing projects. However, familiarity and comfort with these guidelines are limited as
new mines and expansions of existing mines are infrequent.
This paper describes the regulatory framework that drives closure and reclamation planning in BC and the
opportunities that can be achieved in reaching the goals of closure and reclamation when mine planning
occurs early in the project. Planning for closure and reclamation is less creative when this planning occurs
during operations or when a site had been abandoned. The principles underlying the various approaches
discussed can be adopted as best management practices across the world with or without a regulatory
framework in place.
In BC, mine closure and reclamation planning is mandated under the BC Mines Act (1996) and the Health,
Safety and Reclamation Code (the Code) for Mines in British Columbia (British Columbia, 2008). Closure and
reclamation requirements are described in Section 10 of the Code. Closure planning is required for all
infrastructures on a mine site, including waste dumps, major impoundments, water structures, openings,
processing facilities and site roads.
The major drivers of reclamation in the Code include the following:
10.7.4. The land surface shall be reclaimed to an end land use approved by the chief inspector that
considers previous and potential uses.
10.7.5. Excluding lands that are not to be reclaimed, the average land capability to be achieved on
the remaining lands shall not be less than the average that existed prior to mining, unless the land
capability is not consistent with the approved end land use.
…
10.7.8. … All surficial soil materials removed for mining purposes shall be saved for use in
reclamation programs unless these objectives can be otherwise achieved. (British Columbia, 2008)
In BC, waste dumps as well as other disturbed areas must be reclaimed.
An estimate of the total expected costs of closure and reclamation over the planned life of the mine, including
the costs of long-term monitoring and maintenance, are also required as part of the development of the
closure and reclamation plan.
The standards set out in the Code rely heavily on baseline studies and the requirement to develop the end
land use objectives. Operators who fail to follow the Code in BC risk not receiving a permit. Other Canadian
jurisdictions, such as Ontario, also have strict closure and reclamation requirements, as mandated in
Regulation O. 240/00, Mine Development and Closure under Part VII of the Mining Act (Ontario, 1990).
Although the regulations vary from country to country and jurisdiction to jurisdiction, good planning will
stand the test of time. It is important to note that regulations often become more stringent as government
agencies revise their closure requirements due to increased experience and knowledge sharing. Many
abandoned mines around the world have left great environmental impacts, which are noticed by the general
public, who now scrutinise these kinds of projects and the decisions of the regulators more carefully.
Approaches to closure and reclamation may become more uniform with the development of international
standards.
As closure occurs at the end of operations, despite the requirement for closure and reclamation planning,
detailed planning for closure and reclamation is often considered a low priority until the end of operations,
when the end of the life of mine (LOM) is imminent. However, BC regulations require a detailed closure and
reclamation plan to be developed as part of the Environmental Application, one of two major documents

Case Studies
required for a new mine project in BC. The second required document, the Mines Act Permit, must include
more detail, including a detailed five-year plan.
There are also two earlier stages prior to the Environmental Application where closure and reclamation
planning can be included in mine planning:
 Preliminary feasibility (pre-feasibility) study
 Bankable feasibility (feasibility) study
The pre-feasibility study, according to National Instrument 43-101 (NI 43-101), is a comprehensive study of
the viability of a mineral project based on a range of components, including mining method; mineral
processing; technical, legal, operating, economic and environmental factors; and financial analysis. The
feasibility study includes similar considerations and serves as the basis for a final decision by a financial
institution to finance a project. Geologists find and delineate the ore body; these specialists are often the
owners of companies, particularly smaller mining companies. During the pre-feasibility and feasibility stages,
company owners generally rely on engineering consultants for the design of the mine and the project in
general. Several speciality engineering companies are often brought in to design different components of a
project such as the waste rock dumps, the tailings storage facilities and the pit design. These specialists work
on their specific tasks and may or may not come together. Frequently, as closure and reclamation planning
are not at the forefront, mine planning in the pre-feasibility and feasibility stages may proceed in a direction
that will not meet the Code. This can result in a large expenditures of effort and money that may require
reversal and redesign of the mine plan, as described below.
This case study is related to the design of waste rock dumps. This project is located in an isolated,
mountainous area with very narrow valleys and limited space for constructing the dumps. During the early
mine planning stages, the waste rock dumps were therefore designed to occupy the smallest footprint
possible because of the limited space. This required the dumps to be constructed by end dumping, which
resulted in long, steep slopes at the angle of repose.
The proponent was not familiar with the Code, and planning proceeded without knowledge of the Code. The
proponent and the waste rock dump engineers were not aware that waste rock storage dumps require
reclamation in BC. The slopes of the proposed design were not only too steep for reclamation but their
proposed location did not allow for resloping so that reclamation could be carried out later ( at the closure
stage). The risk was that a permit would not be given with this plan. Therefore, the waste rock dumps had to
be relocated and redesigned. This change was costly for the proponent. However, the cost to redesign the
mine plan would have been greater if the mine plan was further advanced and other planned infrastructure
had also required redesign and relocation.
The waste dumps were relocated and redesigned in the mine plan to incorporate the appropriate slope for
reclamation. This reduced the requirement to reslope at the closure phase. The cost of resloping is very high,
as resloping of long, high dumps requires rehandling of mine waste. Further, the cost for resloping would be
incurred at the end of LOM, when there are no proceeds from ore sales. Further, the choice to delay resloping
until closure can result in the company carrying large bonds throughout the LOM. Resloping at the end of
mine life also delays closing of the site and presents an additional risk of increased inflation. By incorporating
closure and reclamation planning in the early mine design, proponents can substantially reduce the size of
the bond carried during operations and the interest on this bond, if it is borrowed.
The ability to reclaim the waste rock dumps is a requirement of the Code. However, the inclusion of early
mine closure and reclamation planning in the development of the mine plan can also provide an opportunity
to design unique end land uses. For example, for the case study described above, early input and planning
meant that the design of the waste dumps was further manipulated to include multiple slope angles and

engineered covers in selected areas that, coupled with unique revegetation prescriptions, could result in the
development of a complex of various, unique wildlife habitats.
For this case study, baseline wildlife surveys indicated that this particular project location has important
mountain goat, hoary marmot and grizzly bear habitat. These species are considered valued ecosystem
components (VECs) in BC. It was recognised that some project components, including the waste rock dumps,
would bury critical wildlife habitat. The early design of the waste rock dumps provided an opportunity to
incorporate the potential for these wildlife habitats into the waste rock dump designs.
The closure and reclamation specialist worked with wildlife, terrestrial ecology and surface water specialists
and waste dump engineers to develop a waste rock design for the mine plan that would accommodate these
species. This involved incorporating escape terrain for mountain goats, which required the top of the waste
dumps to be left steep at the angle of repose and absent of vegetation; including suitable vegetation such as
grasses for mountain goat and hoary marmots at lower elevations; and developing grizzly bear habitat in the
lower part of the dump (Figure 1). Mountain goat habitat includes both suitable vegetation for browsing and
escape terrain, which is necessary because mountain goats are at risk of predation by grizzly bears. If the
design included reclaimed slopes over the whole waste rock dump, mountain goat escape habitat would not
have been possible. Not including the design for escape habitat for mountain goats could have represented
a permit risk because of the importance of mountain goats at the site.
The mine plan required a contact water storage facility to collect metal leaching/acid rock drainage (ML/ARD)
from the waste rock dumps and the overall site. The most appropriate location for this pond is at the base of
the waste rock dumps. Because the contact water will not be suitable for wildlife consumption, the presence
of this water storage facility presented a risk to wildlife and, therefore, a risk to obtaining a permit for the
project. Solving this issue required the closure and reclamation specialist to work with the engineers to design
the base of the slope above the water pond to reduce potential wildlife access to the contact water storage
facility. To achieve this, the base of the waste rock dumps was designed to be very steep. This slope could
not be reclaimed, and it would therefore not have any attractant vegetation. As a precautionary measure,
attractant vegetation was planted upslope of the water storage pond to draw the wildlife away from the
water storage facility. This reduced the wildlife habitat attraction of that portion of the dump, including the
contact water storage facility.
The waste rock dump material is very coarse, and this property reduced the potential for natural revegetation
due to the lack of fines, plant nutrients and water storage capacity for plant growth. The team planned for
vegetation patterns that would provide for the unique end land use objective by including forest cover
species above the steep, unvegetated slope that will provide less palatable vegetation and a physical
impediment to animal movement (Forest B, in Figure 1). This design will direct the wildlife across the waste
dumps toward better habitat (Forest, in Figure 1). The Forest habitat was designed to be suited to grizzly
bears. The band of vegetation species above the Forest area provided marmot and mountain goat habitat,
which was connected to the adjacent native habitat such that the mountain goats could travel across the
high-elevation landscape on similar terrain and vegetation types. Similarly, the forest cover for grizzly bear
habitat connected to the adjacent forest such that grizzly bears could also travel across the landscape at a
similar elevation in a similar habitat.

Case Studies
The attention paid to reclamation planning very early in the mine design increased the potential for the
mining company to obtain a permit. Key critical wildlife species vary from country to country and climate to
climate. With good early planning, the appropriate habitat for site-specific species can be achieved. Early
closure planning also allows all aspects of mine development to be coordinated, which will allow the closure
of the project to occur as planned. This type of planning makes it more likely that a project will be permitted
by the regulators and supported by local governments and stakeholders.
The requirement to salvage soil for reclamation is a practice that results in successful reclamation in all parts
of the world. Before the Code was established in BC, soils were generally buried by mine infrastructure such
as waste rock dumps. The Code now ensures that soil salvage considerations are included in the development
of the closure and reclamation plan included in the BC Environmental Assessment Application.
The assessment of the soils in the footprint area for reclamation purposes requires a review of the soil
baseline information, including surficial materials, soil physical and chemical properties, and soil depth.
Topography should also be considered, as suitable soils may not be accessible by heavy equipment such as
dozers. The volume of suitable soil may also not be sufficient as waste rock dumps are mounded, constructed
landforms with a greater surface area than their footprint.
A materials balance calculation is required to assess the amount of soil available for reclamation and the
amount required. The depth of the soil cover is a function of the physical and chemical characteristics of the
soils, climatic factors and the nature of the underlying materials. The decision regarding soil cover depth is
often a compromise. Where there is insufficient soil for reclamation, overburden, if available, may be used
to supplement the soil shortage.
Overburden is often considered a waste material. The requirement to salvage and stockpile soil and
overburden for reclamation can be very costly, and this cost is incurred during the construction phase. The
approximate cost to handle soil and overburden is generally not included in the pre-feasibility and feasibility
studies, although these studies include economic analyses of a project.

Closure and reclamation plans describe the closure and reclamation of a mine site at the end of the LOM.
However, a mine project can be terminated early, and the closure and reclamation plan developed for the
end of the mine project may not be suitable for early closure. The amount of disturbance often increases as
the project progresses. Many jurisdictions do not have a requirement to develop a specific detailed closure
and reclamation plan for the early years of the project and, therefore, they may not be prepared for closure
of the site if this occurs during an early stage.
The BC Mines Act Permit application requires a detailed five-year plan for these early years of a project. In
some mines, such as large open-pit mines, the infrastructure required in the early part of a project, such as
muck pads and local sediment ponds and roads, may be buried by future infrastructure such as waste rock
dumps. A detailed closure and reclamation plan for these early components will not be included in the LOM
closure and reclamation plan as they will no longer be relevant.
The cost of closure and reclamation is required as part of the detailed five-year plan. This cost is also used to
set the size of the financial bond the company must provide to the government. The government uses the
bond requirement as a guarantee to ensure that the mine site can be closed and reclaimed in the event the
proponent defaults or is no longer on the site. The goal of the BC government is to avoid paying for the cost
of closing and reclaiming a mine site regardless of the stage of the project. Thus, the bond will vary over the
LOM.
Closure and reclamation planning for an expanding operating mine can be challenging. In BC, the expansion
of an existing operating mine also requires a Mines Act Permit for the expansion area. At one mine, there
was an opportunity to remine a pit to extract additional ore. When the ore was first extracted, the waste
rock dumps and tailings storage facility were placed on top of the native soil as reclamation was not a
requirement at that time. The plan to reopen the pit includes constructing a new waste rock dump on top of
the drained, former tailings storage facility and a second dump partly on an existing waste rock dump and
partly on undisturbed land. The plan to place new infrastructure on existing mine waste is a good approach
as it reduces the mine footprint. The placement of one of the waste rock dumps on previously undisturbed
land will provide the first opportunity to salvage soil for reclamation for the operating mine.
The soils were assessed for reclamation as required by the Code. Salvaged soils must be stored often for the
LOM, and the location of the stockpiles must be carefully considered to ensure that they are not disturbed
by mining activities. In some instances, new land areas have to be cleared and levelled to store the stockpile.
This increases the mine footprint and the amount of land that will ultimately have to be reclaimed. The
greater the distance between the soil storage and the areas where they will be required for reclamation, the
greater the costs to carry out reclamation. In this case study, the mining company recognised an opportunity
to reduce haulage distance and avoid disturbing more land by stockpiling soils on top of existing, adjacent
waste rock dumps.
The materials balance calculation indicated there was insufficient soil for reclamation of the new waste rock
dumps. So the salvaged soil will be strategically used only on areas of the waste dumps that will result in the
greatest opportunity for successful reclamation. Determining which portions of the waste rock dumps would
be covered in soil involved several factors including adjacent land use. The general end land use objectives
for the site are wildlife habitat, grazing and forest. As the expansion area infrastructure is predominantly
located in a brownfield, it was determined that a strategic location for the soil would be on the parts of the
waste rock dumps adjacent to native wildlife and forested habitat. Therefore, reclamation would increase
the connectivity between the natural wildlife and forested habitat and the area reclaimed to forest and
wildlife habitat.
Aspect was another consideration in maximising the use of the soil. The site is located in a part of BC that has
a dry, warm climate. It was decided that soils would not be spread on south facing slopes because plants on

Case Studies
these slopes are more susceptible to moisture stress than on other aspects. As well, because of the shortage
of soil, soils would not be used on the portions of the new waste rock dumps that occurred adjacent to
existing mine wastes or mining areas.
In this case study, there was an option to reclaim the other parts of the waste rock dumps, though with
poorer quality material. Large amounts of overburden were removed from other parts of the mine site. This
material will be used to cover the portions of the waste rock dumps that will not be soil covered. Although
this is not an ideal approach, the regulators appreciate that a compromise is required when a mine has been
operating without the early benefit of the Code requirements. Compromises such as this can be used in all
locations to reclaim areas where there are insufficient soils for reclamation.
Closure and reclamation plans are required for operating mines in many jurisdictions. This may be related to
the regulatory framework, or it may be required to provide the detailed closure and reclamation cost
required for the development of the asset retirement obligation (ARO) of an operating mine. This is a legal
obligation associated with retiring a tangible long-lived asset, and it requires a detailed mine closure and
reclamation plan. The closure and reclamation plan for an operating mine is similar to a plan for a new mine
in that each component of the mine and future components must be included. Governments also require
closure and reclamation plans when operators close abandoned mines.
Developing a closure and reclamation plan for an operating mine or an abandoned mine does not require
costly baseline studies of wildlife, fisheries, terrestrial ecology, soils and terrain, hydrology and hydrogeology,
such as those that would be required for a proposed mine, because the site has already been disturbed, as
indicated in Figure 2. These mines require a detailed investigation of the mine site, including checks of all
infrastructure such as leach pads, waste rock dumps, open pits, water treatment plants, fuel storage facilities,
pregnant ponds, conveyor systems, tunnels and mill buildings.
The mine waste materials on a site also require laboratory analysis. Essentially, all components of a mine
must be checked and contamination delineated and documented. Laboratory analyses may include acid base
accounting (ABA) and may measure metals and hydrocarbons.

The closure of these sites is challenging and expensive, and there is little opportunity for reclamation. Site
stability is a major consideration. For example, steep, long slopes are subject to erosion (Figure 3) and may
require expensive resloping to achieve stability.
Governments all over the world are concerned about abandoned mines and the cost of clean-up. Safety of
abandoned mines is a major concern, with openings presenting a risk to wildlife and people. Standard
practices such as backfilling are common (Figure 4).
Many abandoned sites have complex histories that require extensive investigation. The long-term exposure
of ML/ARD mine waste materials can result in contaminated water being released to the environment, as

Case Studies
shown in Figure 5. Contaminated water and land adjacent to sources of ML/ARD may require costly
remediation and long-term water treatment.
The opportunities for standard closure and reclamation solutions are generally not available for abandoned
mines. Closure and rehabilitation options often include expensive solutions such as geosynthetic liners and
constructed barriers, ongoing water treatment and engineered wetlands. The lack of reclamation material
may require the need for off-site borrow pits. Material transport is expensive, and the borrow pit areas must
also be reclaimed.
The extent of disturbance and contamination of abandoned mines will continue to be a concern to
governments and jurisdictions, as abandoned mine sites have resulted in extensive off-site contamination to
neighbouring communities, reservoirs, drinking water and fish-bearing streams. The closure of these
abandoned sites will be slow as they are expensive and the costs are often covered by governments.
The future of mine closure and reclamation planning will become more stringent and structured as regulators
share information between jurisdictions and the expertise of technical experts increases. International
standards are being developed that will aid jurisdictions with less stringent standards. Continued research
on reclamation in areas such as in cold climates will result in revisions of the requirements for closure and
reclamation plans and in new standards.
The Mount Polley tailing dam failure in BC in 2014 has resulted in an increased requirement to plan for
infrastructure failures. The new rules state mining firms must consider the possibility of a tailings disaster
and evaluate the environmental, health, social and economic impacts of an accident. This will likely result in
a new section in a closure and reclamation plan that would include approaches for remediating the various
potential failures and the costs associated with this remediation. In BC, because of the Mount Polley tailings
dam failure, there will be a tendency to move away from traditional tailings storage ponds and potentially
toward dry stacking of tailings. Dry stacked tailings will be easier to reclaim and will likely have a smaller
footprint than traditional tailings ponds.

Mine closure and reclamation planning is site specific, and each site requires an approach that will be suited
to the site, the nature of the operation and the regulatory framework. Early planning for mine closure and
reclamation generally results in greater and more cost-effective opportunities for successful closure and
reclamation of a site. Companies are only borrowing the site for a period of time, and early planning allows
much of the site to be returned to a productive state. Good closure planning results in the support of
regulators and stakeholders and can increase the opportunities to obtain a permit. Delays in receiving a
permit can result in financial losses or termination of the mine project. Good planning also reduces bonding
costs and the ARO, releasing more funds to create new opportunities.
Closure planning during the operations phase can result in unsuccessful reclamation if landforms are not
designed for reclamation and if the soil resource is buried under the mine infrastructure. Closure and
reclamation become more costly as waste rock dumps or heap leach pad slopes require resloping at the end
of mine life. This can greatly increase the cost of closure and the costs of bonds. This financial liability is
carried through the operations phase. At this latter stage, large costs are incurred when mining is completed
and the ore has been depleted.
Abandoned mines can be costly to rehabilitate, particularly if environmental contamination has migrated to
the adjacent land and affects water quality, fish, wildlife and human health. Extra costs can be related to
water treatment and the closing of a site without reclamation materials. The visible signs of mining activity
are more difficult to remove as the natural resources are not available for reclamation.
British Columbia (1996) Mines act, RSBC 1996, Chapter 293, Queen’s Printer, Victoria, British Columbia.
British Columbia (2008) Health, safety and reclamation code for mines in British Columbia, British Columbia Ministry of Energy, Mines
and Petroleum Resources, Mining and Minerals Division, Victoria, British Columbia.
Ontario (1990) Ontario regulation 240/00: Mine development and closure under Part VII of the act, Mining Act, RSO 1990, Chapter
M.14.

Mine Closure 2015 – A.B. Fourie, M. Tibbett, Les Sawatsky and D. van Zyl (eds)
A. Schwarz Centre of Water Resources for Agriculture and Mining (CRHIAM), Universidad de Concepción,
Chile
N. Pérez Centre of Water Resources for Agriculture and Mining (CRHIAM), Universidad de Concepción, Chile
G. Chaparro Centre of Water Resources for Agriculture and Mining (CRHIAM), Universidad de Concepción,
Chile
In central Chile, acid mine drainage (AMD) is characterised by high concentrations of copper, which can be
recovered. Passive biochemical reactors are eco-friendly technologies for the treatment of AMD. They use
organic substrate mixtures to drive microbial sulphate reduction and to promote the formation of metal
sulphide precipitate. The performance of conventional passive biochemical reactors, however, is limited by
metal toxicity, and metal recovery is difficult. This investigation works on a new passive biological
treatment, the diffusive exchange bioreactor, which can be tailored to resist metal toxicity. This is achieved
by having separate zones for AMD movement and sulphate reduction, and by allowing the diffusive
exchange of solutes between these zones. This treatment makes higher reaction rates possible because it
can use much finer organic materials as substrates. A key innovation of this research is its use of vertical
tubular screens to convey the AMD through the reactor while simultaneously allowing transverse diffusive
exchange of dissolved species with the substrate. The tubular screens act as reactors for precipitate
formation, settling, and accumulation, and precipitates can be periodically recovered from the screens. This
design promotes higher reaction rates, eliminates clogging, and facilitates the recovery of valuable metals.
Our goal is to optimise this diffusive exchange system. Hence, we evaluated organic substrate source,
method of inoculation, operation with toxic drainage, and three bench-scale prototypes.
We assessed the sulphate reduction potential of several organic substrate mixtures in batch reactors. Then,
we ran three diffusive exchange reactor prototypes using the optimum substrate mixture. During the first
month, the prototypes were fed with sulphate solution only, to determine their sulphate reduction potential.
Then the reactors were fed with strong copper-rich AMD, characteristic of central Chile’s mining operations,
for two additional months. The organic substrate mixture based on cellulose fibres showed the highest
sulphate reduction rates. Adding this organic source into the biological diffusive exchange reactors allows a
reduction in the size of biological passive treatment systems, as it has more reactivity per m 3 than
conventional systems currently in use. During AMD feeding, all prototypes achieved significant metal
removal and acidity neutralisation. The ratio between conductive and reactive volumes and the amount of
interface between these volumes affected removal performance. Importantly, copper precipitates
accumulated at the bottom of the screened tubes and did not affect the hydraulic performance of the
reactors.
Extraction of copper, gold, and silver are the principal mining activities in Chile. During 2014, the country
produced 5.78 million tons of copper, representing 32% of the world’s production (Consejo Minero, 2014).
The extraction of copper occurs in a variety of climates, and several mining operations are located in humid
regions where infiltrating water from rain and melting snow favours the generation of acid mine drainage
(AMD). The oxidation of residual sulphide minerals present in tailings and waste rock from copper mines
can result in AMD with extreme acidity and elevated copper and sulphate concentrations.

Passive biological treatment of acid mine drainage with recovery of copper A. Schwarz, N. Pérez, and G. Chaparro
Because of rising energy and disposal costs of chemical sludges, passive biological solutions are increasingly
being studied as remediation alternatives for AMD. The successful implementation of passive biological
systems in the Chilean copper industry, however, requires mitigation of the toxic effect of copper on the
microbial consortia that degrade the organic matter and reduce sulphate. Suitable reactive mixtures made
up of organic materials and wastes and locally available alkalinity sources (e.g., limestone) must also be
evaluated (ITRC, 2013; Gusek et al., 2004). The composition of the reactive mixture is key for an efficient
sulphate reduction process.
In regard to protecting microorganisms from toxicity, Schwarz and Rittmann (2010) proposed that using
heterogeneous systems instead of homogeneous systems provides bioprotection to the microorganisms. In
the diffusive exchange systems (DES), reactive layers or zones of low hydraulic conductivity are separate
from the conductive layers or zones for transporting wastewater across the system. Sulphate reduction
takes place in the reactive zones, and the generated sulphide migrates laterally to the conductive zones,
where it forms sparingly soluble precipitates with toxic metals. DES reactors can be layered systems or, as
in our present research, beds of reactive material crossed by vertical screened tubes. As AMD flows within
the tubes through the bioreactor downward or upward, its metals are expected to react with the sulphide
within the tubes, forming precipitates that sediment to the bottom. This system avoids direct contact
between microorganisms and the AMD and keeps microorganisms protected from metal toxicity within
regions of high sulphide concentrations in reactive zones.
Pérez et al. (2011) compared the performance in the lab of a homogeneous permeable reactive barrier and
a horizontally layered DES reactor during treatment of AMD with zinc concentrations of up to 200 mg/L.
Although both reactors removed almost all zinc, the DES reactor developed a greater bacterial diversity and
increased sulphate reducing activity. Hence, modelling and experimental evidence indicates that DES
reactors could be useful in remediating toxic AMD with high copper concentrations. Additionally, arranging
layers or zones vertically makes it possible to concentrate precipitates in a reactor zone, from where they
can be extracted and metals can be recovered.
In this work, we first evaluated the sulphate reducing potential of three organic substrate mixtures in batch
systems. The best performing mixture was then utilised in the setup of four bench-scale reactors, three DES
reactors, and a homogeneous reactor. To assess their sulphate reducing capacity, we ran the reactors for
one month, adding only sulphate to the influent. Then we started feeding synthetic AMD characteristic of a
large copper mining operation in central Chile.
Anaerobic cultures with 300 g of different reactive mixtures (Table 1), bacterial inoculum, and synthetic
AMD were set up in 2 L bottles. The microorganisms came from anaerobic sediments from a nearby lagoon
and anaerobic digester content of the local wastewater treatment plant. We included only drainage as a
negative control and ran duplicates of all treatments and the negative control. The bottles were incubated
at ambient temperature for 48 days.
Each mixture was tested with two different drainages. Drainage A was as described by Waybrant et al.
(1998): 3,620 mg/L SO42-, 1,080 mg/L Fe, 0.81 mg/L Zn, 1.15 mg/L Ni, and pH 4–5. Drainage B was as
described by Lindsay et al. (2008): 3,600 mg/L SO42-, 750 mg/L Fe, 100 mg/L Zn, 15 mg/L Ni, 20 mg/L Mn,
10 mg/L Cd, 4.76 mg/L Cu, and pH 4.9.
The bottles were sampled weekly using sterile syringes and agitating vigorously before sampling. The 5 mL
samples were assayed for alkalinity, sulphate, sulphide, pH, redox potential (ORP), and metals. The pH,
ORP, sulphide, and sulphate were measured immediately after sampling. The samples were filtered
(0.45 m). ORP was determined with the electrode Thermo Scientific Orion 9180B and pH with the
electrode Thermo Scientific Orion 3. Sulphate was quantified using method 4500-SO42- E (APHA, 2005) with
Hach SulfaVer 4 reagent. Sulphides (H2S, HS-) were determined using the methylene blue method 4500-S2-D

Case Studies
(APHA, 2005), also with Hach reagents. The metals Fe and Cu were quantified using the Hach methods
Ferrover and Cuver.
Mixture
Material 1 2 3
Cow manure 15% 15% 15%
Leaf compost 30% 30% 30%
Cellulose fibres 45% - -
Sawdust - 45% -
Microcrystalline cellulose - - 45%
Zero-valent iron 10% 10% 10%
Anaerobic sludge (mL) 10 10 10
Cellulolytic bacteria (mL) 2 2 2
Sulphate reducing bacteria 2 2 2
(mL)
The bench-scale reactor prototypes were built using 200 L drums. All reactors were filled with a bottom
layer of sand (5 cm), followed by the reactive mixture layer (62 cm), and capped by a cover of sand (5 cm)
and a layer of water (10 cm). Before filling the reactors, the mixtures (Table 2) were wetted with a sulphate
solution (3,890 mg/L). After construction, the reactors were saturated with the same sulphate solution,
leaving a 10 cm deep layer of solution at the top. The key bacterial inoculum was cow manure; however,
during construction, 750 mL of a sulphate reducing culture and 500 mL of a cellulolytic culture were also
added to the mixture.
Reactors
Components SDES1, SDES2, SDES3 R4
Cellulose fibres 50% -
Cow manure 30% 10%
Alfalfa - 10%
Chips - 50%
Limestone 20% 30%

After construction, when sulphate reduction was evident from sulphate and sulphide measurements within
the DES reactors, we started feeding each reactor 7.5 L/d sulphate solution (3,890 mg/L). After 35 days,
when steady state conditions were reached, we changed the influent per reactor to 3.75 L/d of synthetic
AMD, characteristic of a large copper mine in central Chile, with 3,890 mg/L SO42-, 528 mg/L Cu, 94.8 mg/L
Al, 0.77 mg/L Fe, 84.6 mg/L Zn, 5.46 mg/L K, 6 mg/L Cl, and pH of 4.2. Then, after 54 days, considering the
good performance of all reactors, we doubled the influent flow to 7.5 L/d. We took weekly measurements
of the effluent’s pH, ORP, and concentrations of sulphate and metals (Cu, Zn, and Al, using AAS).
We produced three different DES reactor prototypes by changing the diameter of the screened tubes (4 cm
and 5.5 cm), and the area covered by the tubes (10% and 20%) (Figure 1). The purpose of these
arrangements was to determine transport limitation effects on reactor performance, because DES reactors
rely on diffusive processes.
To compare the performance of the DES reactors with conventional technology, we built a homogeneous
reactor with substrate concentrations as described by Eppley and Gusek (2010), replacing hay with alfalfa.
Table 3 shows the batch results for the three types of cellulosic substrates: sawdust (S), cellulose fibres (F)
and microcrystalline cellulose (C). All the experiments presented negative ORPs of less than −150 mV, in the
range of sulphate reduction (Gibson, 1990). However, the results from one sawdust treatment (S-A)
showed a low sulphate reduction rate (570 mmol/m3/d) and low iron removal (12%), indicating that
metabolic processes were slower due to the lower carbon bioavailability of sawdust compared to pure
cellulose substrates.
We based our selection of the best substrate for the bench-scale reactors on the comparison of volumetric
sulphate reducing rates only, by assuming that changes in sulphate concentrations due to chemical
reactions are minimal compared to biological consumption of sulphate. The microcrystalline cellulose
presented the highest sulphate reducing rates, but this material is too expensive to be considered for large-
scale use. The batch tests with cellulose fibres obtained from a local cellulose industry showed rates almost
twice as high compared to sawdust, a popular substrate used in biological passive reactors. Rose (2010)
considers sulphate reducing rates in the range of 0.1–0.3 moles/m3/d to be good. Since the bottles with
cellulose fibres exhibited sulphate reducing rates significantly higher than this, incorporating this material
in DES reactors could allow for smaller treatment systems or higher metal loads.

Case Studies
Sulphate reduction pH ORP

Iron removal (%)
Treatment rate (mmol/m3/d) (mV)
F-A 1,244 87 6.4 −174
S-A 570 12 6.7 −219
C-A 1,592 86 6.6 −250
F-B 1,274 74 6.6 −197
S-B 700 58 7.3 −185
C-B 1,596 74 7.1 −165
Figure 2 shows the experimental setup of the reactor system. The feed solution was prepared weekly and
stored in a 200 L drum. The reactor effluents were collected in individual settlers connected to a treated
water sump.
SDES3
SDES2 R4
SDES1 Influent
Settler
Sulphate reduction and associated sulphide production was detected from day 20 on in DES reactors
(Figure 3), and appears to have achieved steady state at day 32. Effluent ORP of the DES reactors was less
than −250 mV, and from day 29 on, effluent sulphide was between 20 and 30 mg/L. The homogeneous
reactor, in spite of having a reduced effluent (−200 mV), showed low effluent sulphide concentrations
(< 5 mg/L). Since day 29, all reactors showed sulphate reduction/removal rates in the range of 0.4–
0.7 moles/m3/d. These volumetric rates were calculated based on total bed volume, including reactive and
conductive zones. Considering the low sulphide concentrations generated by the homogeneous reactor R4,
the sulphate must have been immobilised by chemical reactions, such as precipitation with Ca2+ released

from calcite dissolution, and not reduced biologically. The DES reactors allowed reducing conditions to be
established more quickly because reactive zones were not exposed to advection of oxygenated fluid.
30
25
20
Sulfide (mg/l)
SDES-1
15 SDES-2
SDES-3
10
R4
0
5 20 29 32 35
a.
1,2
Sulfate reduc on rate (mol/m3/d)
0,8
SDES-1
0,6 SDES-2
SDES-3
0,4
R4
0,2
0
5 20 29 32 35
b.
Sulphate reduction rates declined by about 50% with synthetic AMD (Figure 4) compared to the sulphate
solution case, varying in range from 0.28–0.33 moles/m3/d.
The homogeneous reactor was very effective in neutralising influent acidity. The effluent pH, regardless of
hydraulic load, fluctuated around a value of 6. On the other hand, effluent pH values of the DES reactors
stabilised above 5 at the lower flow, but when the flow was doubled, effluent pH dropped, tending towards
the influent value of 4 (Figure 5).
In the homogeneous reactor (R4), most copper was removed on the surface due to chemical reactions with
limestone. As the acidity in AMD reacts with the limestone and dissolves it, the pH increases, reducing
copper solubility. In the upper part of this reactor, high copper concentrations probably impede microbial
reactivity, but farther down, as copper concentrations and toxicity are abated, the sulphide produced by
microbial activity also contributes to achieving a higher than 99% removal. In this reactor, the onset of
sulphate reduction was slow. Only low effluent sulphide concentrations were detected during 35 days of
sulphate solution feeding and the first 18 days of synthetic AMD feeding. But after day 18, a constant
increase in effluent sulphide was detected, reaching 42 mg/L on day 68. The flow increase did not affect
effluent ORP nor effluent sulphide in R4 significantly, and the reactor did not reach steady state at day 68.

Case Studies
6,5 2 5 8 11 18 27 36 42 50 56 62 68 76
300
Q= 3,75 l/d Q= 7,5 l/d Q= 7,5 l/d
Q= 3,75 l/d
6 200
100
SDES-1
5,5 SDES-1
SDES-2
0
ORP (mV)
SDES-2
pH
SDES-3
SDES-3
5 R4 -100 R4
Influent
-200
4,5
-300
4
2 5 8 11 27 36 42 50 56 62 68 76 -400
The DES reactors showed high effluent sulphide concentrations and negative ORP during operation at the
low flow of 3.75 L/d, but at the higher flow of 7.5 L/d, no effluent sulphide was detected and ORP increased
accordingly (> 200 mV). The SDES3 reactor presented the highest effluent sulphide concentrations
(maximum value of 73 mg/L. This reactor has a higher interface area between conductive and reactive
zones than SDES1 and a higher reactive volume than SDES2. A higher interface area may improve reactor
performance due to improved mass exchange, and a higher reactive volume augments the amount of
sulphide produced.
All reactors removed metals effectively (Figure 6), with removal rates above 99% for copper (influent
concentration of 528 mg/L) and zinc (influent concentration of 85 mg/L). After the influent flow was

doubled, the high removal rate in the DES reactors persisted for both copper and zinc, but the zinc removal
rate in the homogeneous reactor decreased to 50%. The experiment continues, and it remains to be seen if
these trends will hold. We expect the DES reactors to perform better in the long term because of the
bioprotection of microorganisms within reactive zones.
100,1 120
100
100
99,9
Removal of copper (%)
Removal of zinc (%)

80
99,8 SDES-1 SDES-1
SDES-2 60 SDES-2
99,7
SDES-3 SDES-3
99,6
40
R4 R4
99,5
20
99,4
99,3 0
2 5 8 12 18 27 36 42 50 56 2 5 8 12 18 27 36 42 50 56
Passive biochemical reactors removed high concentrations of metals in synthetic AMD representative of a
large copper mine in central Chile. The conventional homogeneous reactor neutralised acidity and removed
metals effectively. Key is the direct contact of the AMD with the limestone in this reactor, effecting a pH
increase and copper precipitation. Then down-gradient sulphate reduction removed remaining metals.
However, as limestone is consumed from top to bottom, it is expected that a reaction front moves
downward, affecting microbial activity; hence, the long-term performance of this reactor still needs to be
evaluated.
While DES reactors removed metals effectively, they failed to neutralise acidity. It appears that direct
contact between AMD and the limestone is necessary for this purpose. Also, the ratio between conductive
and reactive volumes and the amount of interface between these volumes influenced removal
performance. In particular, the relation between these design parameters and hydraulic detention time
warrants further study with respect to the scale up of this technology.
We would like to acknowledge funding from CONICYT for Doctoral Scholarship 63100133 and projects VIU
120015 and FONDAP/CRHIAM/15130015.
American Public Health Association (APHA) (2005) Standard methods for the examination of water and wastewater, 20th Edition,
American Water Works Association, Water Environment Federation, Washington, DC, USA.
Consejo Minero (2014) Minería en Cifras, viewed 20 January 2015, http://www.consejominero.cl/wp-content/uploads/2014/
12/Miner%C3%ADa-en-Cifras_1214.pdf.
Eppley, R.L. and Gusek, J.J. (2010) Sulfate reducing bioreactor treating acidic coal mine influenced water (MIW) in Western
Pennsylvania, in Proceedings SME Annual Meeting and Exhibit, 29 February – 3 March, 2010, Phoenix, Arizona, Society for
Mining, Metallurgy & Exploration, Pennsylvania, 7 pp.
Gibson, G.R. (1990) Physiology and ecology of the sulphate-reducing bacteria, Journal of Applied Bacteriology, Vol. 69, pp. 769−797.
Gusek, J., Schueck, J. and Blakeslee, C.E. (2004) Bench and pilot scale test results. Passive treatment of acid mine drainage (AMD) at
the Fran Coal Mine, PA., in Proceedings 5th Annual PA Conference on Abandoned Mine Reclamation, Indiana, 39 pp.
Interstate Technology & Regulatory Council (ITRC) (2012) Biochemical reactors for mining-influenced waste, BCR-1, Washington
D.C., Interstate Technology & Regulatory Council, Biochemical Reactors for Mining-Influenced Waste Team,
www.itrcweb.org, pp. 34–49.
Lindsay, M., Ptacek, C.J., Blowes, D.W. and Douglas, D. (2008) Zero-valent iron and organic carbon mixtures for remediation of acid
mine drainage: Batch experiments, Applied Geochemistry, Vol. 23, pp. 2214–2225.

Case Studies
Perez, N.R., Diaz, I.C., Barahona, E., Schwarz, A.O. and Urrutia, H. (2011) Effect of reactive material distribution in the biological
treatment of acid mine drainage, in Proceedings Enviromine 2011, Santiago, 9 pp.
Rose, A.W. (2010) Advances in passive treatment of coal mine drainage 1998–2009, in Proceedings Mining Reclamation
Conference, ASMR, Lexington, KY, p. 847.
Schwarz, A.O., and Rittmann, B.E. (2010) The diffusion–active permeable reactive barrier, Journal of Contaminant Hydrology, Vol.
112, pp. 155–162.
Waybrant, K., Blowes, D. and Ptacek, C. (1998) Selection of reactive mixtures for use in permeable reactive walls for treatment of
mine drainage, Environmental Science and Technology, Vol. 32, pp. 1972–1979.


A.R. Butler NRA Environmental Consultants, Australia

T.R. Anderson NRA Environmental Consultants, Australia
A. Sexton Ernest Henry Mine Pty Ltd – a Glencore Company, Australia
The progressive development of regulatory requirements and controls (regulatory creep) over the life of an
operation can result in environmental monitoring programs with no clear purpose and inefficiencies in data
collection and research efforts. This is particularly the case where rehabilitation completion criteria are ill-
defined or ambiguous, as is common in Queensland Australia but arguably more widely.
In circumstances where environmental monitoring is conducted to achieve compliance, the various
environmental values (i.e. land, water and air) tend to be treated as “silos” rather than integrated
components. Poor understanding of the synergies between components can result in a lack of coordination
between technical disciplines and missed opportunities for efficiencies. Rehabilitation success at a landscape
scale requires that these values be considered collectively. If rehabilitation objectives are well-defined,
operational monitoring and rehabilitation research can be rationalised to meet immediate compliance or
operational needs, and concurrently generate a data-set suitable for validating rehabilitation completion and
for quantifying the residual risk post-closure. This, in essence, is the optimisation approach presented and is
illustrated with examples from a case study at Ernest Henry Mine (EHM), a copper mine in north west
Queensland, Australia.
The optimisation of site information is a conceptually simple approach, strategically staged to manage a
complex risk, further supported by the realisation of associated commercial benefits (direct cash flow savings,
investment returns, and reduced liabilities including bonds). The initial focus at EHM was on amending
licensing and approval conditions to deliver well-defined end land use and measurable rehabilitation
objectives. This was followed by planning to achieve and validate rehabilitation objectives through an
integrated approach, implemented during the operational phase, which rationalises environmental
monitoring, management and research resources devoted across environmental disciplines.
Optimisation aims to achieve a reduced workload and realise cost savings through focusing environmental
and rehabilitation monitoring efforts to serve the dual function of compliance and providing supporting data
and documentation to demonstrate pro-active progression towards closure.
Before “social licence to operate” (MCA, 2005) became mining vernacular, obligatory conditions governing
mine operations existed. These conditions, agreed between the Sovereign owner of the minerals and the
entity which sought to exploit the mineral wealth, were documented in licences prepared under the auspices
of statutory legislation. This legislative regulatory approach to mining operations occurs throughout
Australian jurisdictions.
Community expectations of environmental protection, conservation and enhancement are eventually
promulgated in legislation and ultimately in licence conditions. Over time, the licence conditions have
changed in response to the evolution of community expectations, and subsequent modifications to licence
conditions during mine life have introduced additional complexity to the rehabilitation and closure process.

Optimisation – a rationalised approach to developing mine closure objectives A.R. Butler, T.R. Anderson and A. Sexton
This paper discusses the evolution of regulatory controls (referred to as regulatory creep) during the
operations of a mature mine; the Ernest Henry Mine (EHM), Queensland, Australia and the effect it has had
on environmental monitoring, the development of rehabilitation objectives and the financial bond required.
We describe a strategic review process undertaken to realign rehabilitation objectives and allow a focus on
achieving long-term goals. The approach optimises the value of existing studies and amends research and
monitoring programs to deliver data needed to validate the achievement of desired outcomes at closure.
The approach has application to mature mining operations in similar regulatory environments.
EHM is a large-scale, copper-gold operation located in the Mount Isa mineral district in north-west
Queensland. It is owned and operated by Ernest Henry Mining Pty Ltd (a Glencore company) (EHMPL). The
mine produces copper-gold concentrate in a crush-grind-float-dewater processing plant. The project
commenced construction in December 1995 with an initial anticipated mine life of at least 15 years. Mining
occurred as a single open-cut pit mined via conventional mining methods, before transitioning to
underground mining in December 2011 upon completion of the open pit. Waste rock from the open-pit and
underground operation has been deposited in two dedicated waste rock dumps (WRDs) located immediately
to the north and south of the pit. Tailings are deposited in an upstream constructed tailings storage facility
(TSF). EHM is now mining the ore body below the final open-pit in a two-stage approach, utilising the sub-
level cave method. Block 1 underground production commenced after the open-pit mine life (at the end of
2011) and full underground production will extend the EHM mine life to at least 2026. Acid mine drainage
(AMD) from waste rock was identified as a significant environmental issue within specific stratigraphies.
Waste segregation and handling protocols were developed and potentially acid forming (PAF) waste rock was
encapsulated in clay cells within the dumps. Rehabilitation has occurred progressively on WRDs since 2002.
The mine is approximately 38 km north-east of the rural township of Cloncurry in the Gulf Plains Bioregion
of Queensland. The mine is within the Flinders River catchment, which drains into the Gulf of Carpentaria,
approximately 300 km north (Figure 1). EHM is at the headwaters of Gipsy and Eliza creeks, which are
ephemeral streams that flow from the mine site in a northerly and easterly direction respectively. The nearest
major waterway is the Cloncurry River, approximately 7 km to the west. The region is subject to a monsoonal
climate with the majority of rainfall occurring during the summer months, along with higher temperatures
and evaporation rates. The annual rainfall mean for the EHM area is 515.3 mm (BoM, 2015). EHM is sited on
a broad plain of some 240,000 km2, locally dominated by “black soil plains” (vertisols) with characteristic
Mitchell Grass vegetation. The underlying land tenure of the EHM project area is pastoral lease and the area
has been continuously grazed for the past 150 years. The two primary land uses in the area are mining and
livestock grazing.

Case Studies
The statutory approval process that applies to EHM has evolved since 1995, including comprehensive
changes in the legislative instruments and the regulatory authorities responsible for the administration of
these instruments. During this period of change, the mine has been constructed, commissioned, and the
open cut mine has progressed to completion and through to the currently operating underground operation.
The principal determinants of landform design and rehabilitation outcomes are the mining activities
themselves. However, the regulatory mechanisms and licence conditions that currently apply are
fundamentally different to those that applied when the main waste containing land forms were designed and
the majority of land disturbance occurred. This situation will be common to mature mining operations in
many jurisdictions.
The Impact Assessment Study (IAS) (EHMPL, 1995a) informed the licencing process under the Queensland
Mineral Resources Act 1989, the outcome of which was an instrument of title together with conditions,
commonly referred to as “mine lease conditions” (a process administered by the Department of Minerals
and Energy – DME). The mine lease conditions included requirements to minimise environmental impacts on
and around the mining lease and to undertake rehabilitation of the land within the mining lease, and of any
surrounding land adversely affected.
An amendment to the Mineral Resources Act 1989, promulgated by the Mineral Resources Amendment Act
1995, modified mine lease conditions to include the need to submit an environmental management overview
strategy (EMOS) and a plan of operations (PoO) consistent with the accepted EMOS. The conditions required
the licence holder to conduct mining activities in accordance with the accepted EMOS and PoO. According to
DME and Queensland Minerals Council (QMC) (1994), the intention of the EMOS was to provide a life-of-
mine strategy for the management of environmental effects. It was to contain clear commitments, framed
to enable later assessment of the extent to which the commitments were met; and although mine lease

conditions normally applied, the intent was for the EMOS to replace these conditions (DME and QMC, 1994).
In practice, from 1991 (when the EMOS process became policy) to 2001, the commitments nominated in the
EMOS (which often set targets for water quality and progressive land rehabilitation) were treated as the
principal conditions for compliance reporting and assessment of environmental performance at mine
operations. The EHM EMOS (EHMPL, 1995b), accepted by the regulator, contained 19 commitments.
From 2001, the administration of statutory obligations relating to environmental management on mining
tenure transferred to the Environmental Protection Agency (EPA, now Department of Environment and
Heritage Protection – EHP). Modifications to the Queensland Environmental Protection Act 1994 saw the
EMOS replaced as the key statutory document by the Environmental Management Plan (EM Plan), which
itself no longer needs to be submitted for approvals. Through this period the Environmental Authority (EA)
(as required under the Environmental Protection Act 1994) emerged as the peak regulatory instrument
containing environmental conditions for mining operations in Queensland. Since its introduction, several
versions of the EA, including modifications to conditions (sometimes driven by events unrelated to EHM in
different sectors elsewhere in the State, such as the flooding and pit water discharge at coal mines in Central
Queensland and more recently, coal seam gas), have been issued to EHMPL.
These legislative changes have resulted in an increasing number and greater complexity of licence conditions
during the mine life of EHM. By 1994, eight standard environmental conditions normally applied. Acceptance
of the 1995 EHM EMOS created the equivalent of 19 conditions. The current EA (in effect as of August 2014)
contains in excess of 100 conditions. The increase in the number of conditions has been coupled with changes
to the required standards of environmental performance and post-mine environmental quality, illustrated
by evolving targets for surface water quality. The undertaking given in the 1995 EMOS (accepted by the
regulator) was as follows; “Post mine closure, only run-off water from rehabilitated areas will be allowed to
report to natural drainage systems.”(EHMPL, 1995b). Supporting text nominated performance requirements
for water quality as follows; “The water quality targets for run-off waters at the lease boundary are set by
the livestock watering guidelines criteria, ANZECC 1992, unless background levels of run-off are of a poorer
quality.” (EHMPL, 1995b). The current EA requires the achievement in receiving surface waters of quality
standards sufficient to protect the equivalent of slightly disturbed aquatic ecosystems, though quality must
not exceed stock water standards or the 95th percentile of reference condition whichever is the lower [sic].
In addition, the spatial definition of where performance is measured has shifted from the lease boundary to
within the lease bounds (negating any attenuation affects). Table 1 shows the changes in maximum allowable
electrical conductivity (EC) at EHM. Similar changes have affected groundwater quality limits. The gradual
increase in mandated performance targets has led to the emergence of non-compliance issues despite no
intervening change in groundwater quality.
Licence Details Surface Water Quality Target (EC) (Contaminant Limit))
EMOS (November 1995) 5882 µS/cm*

EA MIN900341305 (31 October 2008) 3000 µS/cm
EA MIN900341305 (17 August 2010) 5970 µS/cm
EA MIN102677911 (14 November 2012) 1000 µS/cm or 95th percentile of reference whichever is lowest
* The 1995 EMOS referred to stock watering guidelines rather than a value. For beef cattle the maximum desirable total dissolved
solids (TDS) concentration was 4000 mg/L. This was converted here to EC using the formula EC (µS/cm) = TDS (mg/L)/0.68 that was
recommended in 1995.
In terms of land rehabilitation conditions, regulatory creep had led to the creation of ambiguous and
incompatible post-mine land use targets. For example, the EA (issued in November 2012) included a
nominated land use of native habitat, but also nominated the achievement of a minimum post-mine

Case Studies
agricultural land suitability class and required the EA holder to nominate, in the post mine land use plan
(PMLUP), vegetation criteria for both native ecosystems (species diversity, abundance and composition and
projective cover) and agricultural productivity (dry matter production and sustainable stocking rates). The
targets in the EA were conflicting, impossible to achieve routinely in all years (such as projective groundcover
of 80%), unnecessarily prescriptive (such as nominating species lists) or ambiguous and lacking clearly defined
end points. As a result, none of the rehabilitation monitoring conducted at EHM between 2007 and 2013 was
able to validate achievement of rehabilitation completion.
Regulatory creep exaggerates an understandable tendency to focus monitoring programs on operational
compliance with monitoring programs being extended to meet changing requirements. In such
circumstances, the various environmental values (defined in the Queensland Environmental Protection Act
1994 as a quality or physical characteristic of the environment that is conducive to ecological health or public
amenity or safety as defined) to be protected (i.e. land, water and air) tend to be treated as “silos” rather
than integrated components. Poor understanding of the synergies between components can result in a lack
of coordination between technical disciplines and missed opportunities for efficiencies. Rehabilitation
success at a landscape scale requires these values to be considered collectively. If rehabilitation objectives
are well-defined, operational monitoring and rehabilitation research can be rationalised to meet immediate
compliance or operational needs, and concurrently generate a data-set suitable for validating rehabilitation
completion and for quantifying the residual risk post-closure. This, in essence, is the optimisation approach
presented.
When the optimisation plan was conceived, the matters highlighted above and illustrated in Figure 2 had
resulted in a situation where rehabilitation objectives were not clearly defined and EHMPL was not in a
position to demonstrate successful progressive rehabilitation (despite good progress being made on WRDs).
This resulted in the regulator not recognising completed progressive rehabilitation and subsequently
requiring EHMPL to provide a proportionate financial bond. Without a well-defined rehabilitation end point,
validation of ongoing rehabilitation performance or an understanding of post-closure residual impacts,
EHMPL had no supporting evidence to secure a reduction in the financial bond in the near term.
The optimisation process encompassed a staged approach to rationalise rehabilitation management,
environmental monitoring and compliance. The key stages undertaken included:
 Stage 1. Initiate changes to the EA conditions to facilitate a modified approach to rehabilitation. A
change in policy by the regulator in 2013 towards less prescriptive and more outcomes-based
licencing with options to adopt standardised (or model) conditions provided the opportunity to
remove the flawed rehabilitation conditions and allowed EHMPL to nominate appropriate
objectives and targets through the submission of supporting statutory documentation (e.g.
PMLUP).
 Stage 2. Focus on rehabilitation objectives and targets considering the whole-of-project and
including the integration of land and water quality targets. This involved a forensic review and
collation of information from historic reports, research programs and monitoring data to assist in
closure planning and to determine the validity of previous or current management practices in
light of the revised objectives and targets. This allowed a focused rehabilitation strategy to be
developed and thus the rationalisation of research and data gathering to not only provide
compliance support, but also data to assist with progressive validation of rehabilitation with a
view to closure.
 Stage 3. Rationalisation across regulatory documents and monitoring programs to identify
information deficiencies and amend research priorities.

Stage 1 was critical, but this paper focuses on stages 2 and 3 of the process. The tasks were as follows.
 Define post-mine land use options and develop rehabilitation objectives that are achievable and
can be readily validated through routine monitoring.
 Examine existing rehabilitation strategy and rehabilitation methods to determine opportunities
and constraints in light of the revised rehabilitation objectives, including:
○ opportunities to improve rehabilitation outcomes in some domains
○ constraints associated with materials balance based on current rehabilitation specifications
and opportunities to optimise available materials to achieve desired outcomes
○ implications for finalising rehabilitation and cover requirements for critical landforms (such as
WRDs and TSF), including a review of the objectives of cover research programs
○ maximise the value of past studies and existing data-sets
○ identify costs savings in terms of required rehabilitation works and ongoing management
(including the financial bond).
 Review environmental monitoring programs in light of the revised objectives, including:
○ ensure programs deliver compliance assessment and provide data to validate progress
towards closure (such as performance of key controls), or achievement of milestones
○ optimise programs across environmental values (land and water) and operations.
 Identify knowledge gaps and develop a program to address them.
The following sections address some of the key tasks undertaken.
To progress rehabilitation planning, it was necessary to cut through the regulatory creep, revert to first
principles and fundamentally redefine the post-mine land use and rehabilitation objectives. Four general
rehabilitation goals have been established in Queensland (EHP, 2014a). These require rehabilitated areas to
be: safe to humans and wildlife; non-polluting; stable, and able to sustain an agreed post-disturbance land
use. These do not require nor do they imply restoration to original condition, but they are a reasonable basis
for setting site-specific rehabilitation objectives. These specific objectives are tailored to a specified end land
use and may differ from domain to domain. Completion criteria are developed for each objective. They are
evidence-based measures that, when achieved, should indicate rehabilitation has been successfully
completed with respect to the agreed objectives. At EHM, the first and third goals in EHP (2014a) were
reasonably well covered (in terms of the need to remove potential infrastructure-related hazards and provide
geotechnical stability at closure) and these are not specifically addressed here. However, there was a need
to address potential post-closure pollution management and define sustainable land uses.
At EHM, two key elements are critical when considering post-mine land use; namely land and water
environmental values. However, land use decisions must also however consider social and economic factors
linked to the beneficial use of these values. Although land and water values may be dealt with separately,
they are interrelated in that the performance of pollution control measures and the final landform and land
use have the potential to influence the quality of water in the catchment.

Case Studies

The concept that no environmental harm (actual or potential) will occur following a significant landscape
disturbance associated with most mining operations is unrealistic. Some level of environmental harm, or at
least the potential for harm, will persist for a long period (>100 years) or, in some cases, in perpetuity
(e.g. Mount Morgan and Mary Kathleen in Queensland, Australia, Iron Mountain in CA, US).
For land (agricultural or natural values), the nature of disturbance (i.e. the resulting altered landform, newly
created materials and growing medium) associated with materially significant land disturbance activities
(such as the creation of WRDs and TSFs) and the complexity of natural and agricultural ecosystems means
that the reinstatement of pre-disturbance conditions is not usually possible. Alternate, and not necessarily
more productive land uses, may need to be accepted but such “losses” might be off-set with “gains”
elsewhere (on or off-lease). Whatever the alternate land use, it must be self-sustaining and not pose a threat
to surrounding and otherwise un-impacted values (i.e. not be dysfunctional).
For water, long-term protection is invariably linked to pollution control. Where AMD is potentially involved,
the initial controls may include waste characterisation, segregation and encapsulation. These approaches
were employed at EHM. However, for closure, the design and implementation of a multi-layer cover system
is invariably adopted as a default control although the specific purpose or cost benefit seems to be rarely
questioned. At EHM, the regulator insisted that the cost of constructing a complex cover system for the WRDs
and TSF, incorporating a double capillary break, be included in the financial bond. One of the tasks of the
optimisation plan was to critically review the need for a complex cover at EHM.
In general, covers are designed to minimise oxygen ingress and acid generation and/or minimise percolation
and seepage generation. The performance of a cover is thus gauged in terms of desired oxygen profiles or
intercepted rainfall percolation rates (in absolute or relative terms). Rarely is performance gauged in terms
of acceptable pollutant loads or target receiving water quality. DITR (2006) states that “Soil covers must
maintain an acceptably low risk of harm to society and the environment over a very long period.” However,
cover materials evolve over time in response to site-specific physical, chemical, and biological processes so
that as-built performance, and performance after several years, does not necessarily represent long-term
performance (INAP, 2003, 2009; Taylor et al., 2003; Rykaart et al., 2006). Limitations of engineered structures
are not necessarily reflected in regulatory documentation or in some notable non-government publications,
but they need to be considered in any serious attempt to examine post-closure residual risks. Often the
regulator’s desired outcome is to establish a cover system, for example, that functions for at least 100 years
or in perpetuity. MEND (2012) recommended cover systems be designed using a minimum 100 year design
life, and ANCOLD (2012) have adopted 1,000 years as a notional post-closure life for the purpose of focusing
design and operational considerations. These requirements are contradictory to reality. No engineered
structure functions in perpetuity (Wilson et al., 2003; Fourie et al., 2012). Typically, geotechnical engineered
structures have a design life of 50-70 years (Fourie et al., 2012) and all engineered structures require post-
construction monitoring and maintenance. Jefferis (1992) suggested several design scenarios for
contaminated land covers. These are instructive for mine waste covers and have been adapted below.
 Design for optimum performance (e.g. percolation of no more than 5% of rainfall) for a
nominated time period after which no defined performance is assured.
 Design for achievable performance with known and quantifiable releases to the environment
(leaking systems do not fall into this scenario if the leakage is greater than design specifications).
This might mean, for example, a design for optimum performance (e.g. percolation of no more
than 5% of rainfall) up to a nominated design event after which a known impact will occur.
 Design for optimum performance with routine monitoring, but build in (and cost) mechanisms for
recovery or treatment of released material (such as seepage), or remedial action should the
system fail or not perform as intended. The most likely failure mechanism(s) should be assessed
during design (see MEND, 2012 summarised in Dobchuk et al., 2012 for an example of a failure
modes and effects analysis used in a regulatory context in Canada).

Case Studies
Intuitively scenario 3 above is preferred as it accepts that structures will have a finite life and that
management measures are needed in case of failure. Scenario 2 provides visibility of design intent and
potentially allows for residual risk to be quantified. This, at least, allows for agreement to be reached on what
is an acceptable residual risk. Arguably, many existing examples of cover systems fall into scenario 1, where
control over containment or seepage generation is effectively reduced once the end of design life is reached
(or sooner in some cases). This is a problem if no effective alternative or additional downstream controls are
in place.
The acceptance of ongoing liabilities associated with closure solutions for waste storages is not attractive for
mining companies wanting a walk-away solution or regulators looking to extinguish long-term liabilities. But
it would appear to be the reality and the sector (regulators and owner/operators) must accept and address
the residual risk if mining impacts are to be managed effectively in the long-term. The regulatory system in
Queensland has a provision for assessing residual risk and for the costs associated with the realisation of that
risk to be retained when the financial bond is returned, following the submission of an accepted final
rehabilitation report (EHP, 2014b). To the authors’ knowledge, this mechanism has not been tested, but it
provides opportunities for developing the rehabilitation strategy at EHM.
Given that even well-conceived complex (and therefore expensive) multi-layer cover systems do not always
perform and that generation of potentially harmful seepage (albeit at a reduced rate) is more likely than not
in the long-term, some have argued that value for money may be best delivered by adopting lower cost
covers and directing savings to long-term maintenance and management (Wilson et al., 2003). In a review of
capping strategies for coal spoil in South Africa, Waygood and Ferreira (2009) acknowledge limitations of
cover performance. They promote consideration of alternatives through an examination of cost benefit by
weighing up the cost effectiveness of covers and the cost effectiveness of seepage water treatment (or by
extension, other downstream controls). The important question Waygood and Ferreira (2009) ask is: “Do the
dollars we spend on the cover produce equivalent or greater cost savings in seepage treatment?”
A key component in designing engineered covers should be to “work back” from the rehabilitation objective
for the receiving environment. The goal of “no water release” (i.e. nil seepage, no run-off) is not possible;
however, focusing effort on controlling the rate and quality of water release, understanding the nature of
the receiving environment (e.g. its assimilative capacity) and likely residual risk is more appropriate.
Obviously there are other risks to consider in designing a cover such as the need for covers to achieve post-
mine land use and this holistic approach has been used in the consideration of achieving rehabilitation
objectives for water. The overall rehabilitation objectives for surface and groundwater are that:
 the releases of seepage and run-off to the surface water receiving environment (i.e. natural
surface water bodies with the spatial definition of the receiving environment clearly defined and
agreed during operations and in post-closure) do not lead to significant (as defined in EHP 2013)
adverse effects on environmental values of concern; and
 the release of seepage to groundwater does not diminish environmental values at locations
beyond the footprint of the rehabilitated mine area.
Whether these objectives are best achieved with complex covers on the WRDs and/or TSF was assessed.
The current EA conditions state only that the void must not cause any serious environmental harm other
than that constituted by the existence of the void itself. No alternative post-mine land use has been
nominated for the void, but it has an important function post-closure. For safety reasons, access to the void
will be restricted thus precluding use as a source of stock water.
At present, water potentially impacted by contact with mine wastes contained in the TSF and WRDs (i.e.
surface water run-off and seepage) is collected in toe drains and directed to surface water storages and
evaporation dams. Post-closure, the site topography at EHM and the close proximity of the WRDs and TSF to

the void will allow such water to be captured in the toe drains and directed to the void to minimise off-site
impacts. The following summary of the void’s behaviour has shown this to be a viable option.
 The final void at EHM is predicted to be an evaporative sink into perpetuity (AGE, 2010).
 Pit water elevation and groundwater levels in the surrounding Mesozoic strata will trend toward
this groundwater elevation.
 The pit will represent a sink in the regional groundwater system and any local groundwater
resources affected or potentially affected by on-site contamination or seepage (associated with
the TSF and WRD) will be captured within the extent of regional drawdown.
 The final void at EHM is assessed not to be a contaminant source to the local or regional
groundwater systems (HCItasca Denver, 2009; AGE, 2010).
These outcomes are critical for assessing potential impacts post-closure and in the development of cover
design and closure strategies for the main post-mine landforms.
The surrounding land at EHM is used for cattle grazing; however, due to the altered nature of the post-mine
landform associated with domains containing the WRDs and TSF, approvals conditions in place at the time of
their construction recognised that rehabilitated land capability would severely limit pastoral use.
A self-sustaining vegetation community suitable for providing native habitat is considered a suitable post-
mine land use for the WRDs and TSF for the following reasons.
 “Natural ecosystems” is the preferred post-mine land use in the EHP (2014a) hierarchy.
 This land use is consistent with the requirements of the EA licence conditions post-2013 that state
rehabilitated land should require no greater management than that required for the land pre-
disturbance. If used for grazing, rehabilitated land on a WRD would require more intensive
management (such as control of stocking rates and erosion management) than for undisturbed
land at the same location, so this would be inconsistent with this requirement.
 As the waste storage structures have been specifically designed and built to minimise
environmental harm from seepage, protecting the integrity of the structure and the protection of
other values of necessity for pastoral use in surrounding areas (such as stock drinking water) is
the primary aim of the rehabilitation, rather than the provision of what, at best, would be low
quality grazing land with severe limitations, making it doubtful whether the inputs required to
achieve and maintain production outweigh the benefits in the long-term.
While it is not possible to completely recreate the black soil (vertisols) plains on elevated waste landforms,
and thus ensure the return of the pre-existing Regional Ecosystem (a unique vegetation community identifier
used in Queensland (Queensland Herbarium, 2014)), there will be similarities in the soil properties and
habitat conditions on the plateaus/tops of structures. It has been demonstrated through monitoring to date
(Gravina et al., 2013; NRA, 2014), that the final vegetation community will have some similarities to that in
the surrounding landscape, therefore providing connectivity with the surrounding landscape. The primary
aim will be to ensure the vegetation community performs critical ecological functions (i.e. is self-sustaining)
and does not diminish the value of surrounding land (e.g. by being a source of significant weeds or eroded
sediment). The development of the specific vegetation objectives and completion criteria are further
described in Butler et al. (2015, in preparation).
Where landforms have undergone little change in elevation or relief and where subsoils remain intact (such
as infrastructure areas), optimisation has identified the possibility of returning the land to pastoral use of
similar value or potentially improved pasture (of increased value), but only where these areas are contiguous

Case Studies
with the surrounding undisturbed land (i.e. they can be managed as one unit with similar management
needs). This now forms the basis for a targeted research program and consultation with the surrounding
landholder to create unrealised economic value in a post-closure environment to off-set losses associated
with the WRDs and TSF.
The main risk associated with waste rock is the generation of AMD and coincident release of metals,
metalloids and salts that have the potential to affect downstream environmental values. EHM has
implemented a range of control measures, since mining commenced in 1995. The design intent mandated in
approvals when the WRDs were designed and first constructed was to protect primary industry (stock water)
receiving water environmental values. These standards have changed, as discussed above, but many of the
design and mitigation measures employed would still be regarded as best practice (see INAP, 2009 GARD
Guide Chapter 6); incorporating:
 long-term waste rock characterisation and selective handling practices
 WRD design and construction controls (waste segregation, waste placement and encapsulation)
 research and monitoring on the effectiveness of control measures such as ongoing kinetic testing
and surface and groundwater monitoring to assess the potential impacts on water quality.
Since the bulk of the WRDs were constructed, surface water affected by seepage and run-off has been
released into the receiving environment. These events included a near 1 in 100 year ARI wet season (2009)
and a 1 in >20 year ARI 24 hour event (2012). They provide models for what might be expected in terms of
residual impacts if the containment system is breached post-closure.
For the optimisation plan, existing information (such as ongoing kinetic testing, water quality monitoring and
biological effects data, rehabilitation monitoring) was reviewed to determine if the design and construction
of the WRDs has resulted in the necessary control of AMD generation and seepage containment as intended
(to progress acceptance of these structures as final landforms). The following main issues were examined
during the validation process.
 Has further research or monitoring substantially validated the design and construction of the
WRD with respect to preventing AMD?
 Has further research or monitoring substantially validated the design and construction of the
WRD with respect to managing other impacts (e.g. neutral mine drainage) associated with
seepage generation?
 Is further development of a barrier cover system necessary to manage seepage and residual risks?
 Considering proposed seepage management commitments at the end of mine life, what is the
residual risk posed by the WRDs post-closure and can this be managed?
The review was able to demonstrate the following points.
 The WRD design and waste placement was appropriate for preventing AMD and for minimising
potential for the generation of neutral mine drainage, though seepage still contained elevated
metals and metalloids (i.e. substantial reworking of the WRDs would unlikely improve outcomes).
 The seepage containment system had functioned well (up to a certain design limit) and that
breaches of the system had not caused significant downstream impacts on environmental values
(i.e. localised, amenable to remediation, effects not exceeding 1 year and no permanent change
to ecosystems or other values in the receiving environment).
 Additional cover specifications (over and above the simple store and release type covers in place
or planned) appear unnecessary to meet the desired rehabilitation outcomes for land and water.

 Potential residual impacts to surface waters could be quantified and their management effectively
costed providing the certainty needed to progress toward an accurate costing of the financial
security to provide for management of any residual risk post-closure.
Cover trials were established on the EHM TSF in 2007 using two store and release cover systems (one with 2
layers at a thickness of 1 m and the other with 3 layers and thickness of 1.9 m). The importance of setting
objectives before finalising the cover design has been discussed previously. At the time of establishment
however, no target net percolation rate had been incorporated into the EHM TSF completion criteria and this
hindered a decision on which cover provided the required performance. The establishment of clear
rehabilitation objectives through the optimisation process provides a sound basis for decisions on which
cover system most cost effectively meets the desired rehabilitation objective for EHM post-closure.
Monitoring data was reviewed to establish whether groundwater values were being impacted or likely to be
impacted by the TSF to further assist in making decisions regarding cover design performance. The review of
groundwater impacts has not revealed, even with an uncapped TSF, that pollution will be sufficient to impact
on groundwater environmental values. As discussed in the section entitled Post-mine land use definition for
the final void at EHM, the void will create a sink in the regional groundwater system and any local
groundwater resources affected or potentially affected by TSF seepage will be captured within the extent of
regional drawdown. Groundwater at EHM does not interact with surface water.
Although it is desirable to minimise infiltration into the tailings as much as is practical to minimise seepage
volumes, there is a point of diminishing return when it comes to cover complexity. The difference between
the two cover systems in percolation measured by the trials or predicted by modelling is relatively small
(especially considering the likely spatial variations in measured percolation rates that can occur in covers that
include a cracking clay growth layer). Both cover systems will be effective in substantially reducing the
potential for seepage generation and, considering the behaviour of the groundwater at EHM, either of the
cover systems would be expected to protect groundwater environmental values post-closure.
The direct rehabilitation (i.e. without the placement of a cover) of the tailings has also been examined, but
research to date has not provided a cost effective solution that provides certainty of outcome. In contrast,
the simple 1 m store and release cover used in the cover trial is similar to that used on the rehabilitated
sections of the WRDs and has shown to be successful in meeting revegetation objectives.
By formalising rehabilitation objectives and using existing data collected across environmental values of
interest, it has been possible to rationally progress decision making regarding the rehabilitation strategy for
the TSF and in the process, potentially save significant costs in terms of cover design and construction.
In comparison to the situation that existed in 2013, the optimisation plan has substantially progressed the
rehabilitation recipe for EHM through the establishment of clear objectives and associated specifications and
performance measures. The application of these rehabilitation objectives to the interpretation of research
and monitoring data available at EHM revealed the following points.
 The current “store and release” cover system installed on large parts of the EHM WRDs coupled
with a seepage collection system directed to the void is considered to be adequate (subject to
further design improvements to the bunds to achieve agreed design intent). EHM can now assert
that existing rehabilitated areas be regarded as complete if they satisfy the completion criteria
developed and that capping and rehabilitation of such areas no longer needs to be costed in the
financial bond.
 A TSF cover design consisting of a nominal 0.50 m base of waste rock material overlain by a
nominal 0.50 m thick growth layer has been shown to be adequate to meet the post-mine land
use objectives with regard to the protection of land and water environmental values. This

Case Studies
provides certainty in terms of materials balance requirements during the operational phase and
allows effective planning and materials management for closure as well as a justifiable alternative
design for reducing the financial bond.
 By retrospectively realigning the rehabilitation objectives with the regulator’s outcomes-based
focus, the opportunity to identify beneficial economic value in a post-closure environment can be
achieved. Provision of improved pasture may only be a single option during this phase, however
the mind shift allows for the investigation of alternative cost saving measures and realisation of
economic value in relation to closure strategies.
 The groundwater review identified that regulatory creep has led to the cumulative monitoring of
many bores for which there is no longer a clear purpose for the data collected. This highlights the
need for periodic program reviews and for every monitoring location to have a specific purpose
clearly articulated in the monitoring program or procedure.
 Surface water programs have been shown to provide data that can be used to determine impacts
from the release of seepage affected waters in terms of scale and longevity. This has allowed for
an assessment of likely residual impacts that may be expected post-closure should the seepage
containment system fail or be exposed to events outside of the design intent. This provides a
starting point for negotiations regarding the provision of a residual risk bond and the
development of a post-closure site-based management plan.
The optimisation process has identified and provided the justification necessary to support a cost reduction
to the rehabilitation recipe used to calculate the financial bond at EHM. Considering the TSF and WRDs alone,
a saving in the order of 20% to 40% relative to the current bond may be possible at EHM. The realisation of
these reductions, and other similar savings not listed, generate forward savings, associated with monitoring,
corrective actions and validation reporting, as well as to the residual risk liability, relative to that which would
prevail if the current nominated recipe is retained. Further cost savings can be made during the operational
phase, where budgets often provide significant allowance for untargeted monitoring with short-term value.
The redirection and focus of key monitoring tasks during the operational phase and, more importantly, their
defined establishment prior to this, allows for the collection of key environmental data than can be compiled
and relied upon to measure the performance of mitigation and rehabilitation measures prior to and beyond
closure. Efficiency and potential operational cost reductions have been inherent in this approach.
The optimisation process utilised a simple approach to a complex risk of significant commercial value. It was
a staged approach, implemented to rationalise rehabilitation management, environmental monitoring and
compliance. The approach is iterative and involves the forensic assessment of: licencing requirements; the
interaction between mine activities and the environment for the purpose of evaluating end land use; and
environmental monitoring data across land, surface water and groundwater disciplines.
The overarching intent was for all rehabilitation management, environmental monitoring and compliance
programs, together with all associated reporting prepared under the auspices of the Licence, to serve the
dual purpose of compliance, ultimate mine closure and final relinquishment.
The resultant productivity improvements (direct cash flow savings, investment returns, and reduced liabilities
including bonds) are compelling and can be measured in tens of millions of dollars during the life of the
project and directly at closure.
This work was funded by Ernest Henry Mine Pty Ltd. The work was commissioned by former Superintendent
Environment Bernadette Williams and we are grateful for her faith in the project.

AGE (Australasian Groundwater and Environmental Consultants Pty Ltd) (2010) Report on the regional groundwater impacts of Ernest
Henry Mine underground mining, report to Ernest Henry Mining Pty Ltd, Project No. G1430, June 2010.
ANCOLD (Australian National Committee on Large Dams/International Committee on Large Dams) (2012) Guidelines on tailings dams
– planning, design, construction, operation and closure, May 2012.
Dobchuk, B., Nahir, M. O’Kane, M. A. and Ayres, B. (2012) Mine waste cover system design – considerations for cold regions, in Mine
Closure 2012, A. Fourie and M. Tibbett (eds), Brisbane, Queensland.
BOM (Bureau of Meteorology) (2015) Climate data online, viewed http://www.bom.gov.au/climate/data/, Bureau of Meteorology.
Butler, A.R., Anderson, T.R. and Sexton A. (2015) The development of rehabilitation objectives and closure criteria for Ernest Henry
Mine, paper to be presented to Australian Mine Rehabilitation Workshop, Adelaide, August 2015 (in preparation).
DITR (Department of Industry Tourism and Resources) (2006) Mine closure and completion, Leading Practices Sustainable
Development Program for the Mining Industry Series, Department of Industry Tourism and Resources, Australia.
DME and QMC (Queensland Department of Minerals and Energy and Queensland Mining Council) (1994) Environmental impact
assessment and management for mining in Queensland, draft, 24 June 1994.
DRI (Department of Resource Industries) (1991) Environmental management for mining in Queensland, Environmental Policy for
Mining, June 1991, prepared by the Queensland Department of Resource Industries.
EHMPL (Ernest Henry Mining Pty Ltd) (1995a) Ernest Henry Project – Draft impact assessment study, January 1995, Vols 1 and 2,
Ernest Henry Mining Pty Ltd, conducted by RUST PPK Environment and Infrastructure Consultants.
EHMPL (Ernest Henry Mining Pty Ltd) (1995b) Ernest Henry Project – Environmental management overview strategy, mining lease
application ML90085, November 1995.
EHP (Queensland Department of Environment and Heritage Protection) (2013) Manual for assessing consequence categories and
hydraulic performance of structures, reference number EM635, Version 4, November 2013, published by Department of
Environment and Heritage Protection, Queensland.
EHP (Queensland Department of Environment and Heritage Protection) (2014a) Rehabilitation requirements for mining projects,
reference number EM1122, Version 2, dated 28 January 2014, published by Department of Environment and Heritage
Protection, Queensland.
EHP (Queensland Department of Environment and Heritage Protection) (2014b) Rehabilitation report, reference number EM872 (Part
1), dated 26 June 2014, published by Department of Environment and Heritage Protection, Queensland.
Fourie, A.B., Tibbett, M. and De Jong, J. (2012) The fallacy of designing for in-perpetuity: Geotechnical assumptions, ecosystem
responses and concepts for managing inevitable change, in Mine Closure 2012, A. Fourie and M. Tibbett (eds), Brisbane,
Queensland.
Gravina, M., McKenna, P. and Glenn, V. (2013) Rehabilitation monitoring at Ernest Henry Mine 2012 assessment, report to EHM by
Centre for Mined Land Rehabilitation, The University of Queensland, April 2013.
HC Itasca (2009) Hydrologic and geochemical analyses in support of the feasibility study of underground mining at Ernest Henry Mine,
report prepared for Xstrata Copper Ernest Henry Mining Pty Ltd, HCI-1862, April 2009 (Appendix three of EHM EMP).
INAP (The International Network for Acid Prevention) (2003) Evaluation of the long-term performance of dry cover systems, final
report, report prepared by O’Kane Consulting, March 2003.
INAP (The International Network for Acid Prevention) (2009) Global acid rock drainage guide (GARD Guide), viewed
http://www.gardguide.com.
Jefferis, S.A. (1992) Remedial barriers and containment, in Contaminated Land Treatment Technologies, J F Rees (ed), published for
SCI by Elsevier Applied Science.
MEND (Mine Environment Neutral Drainage) (2012) Cold regions cover design technical guidance document, Mine Environment
Neutral Drainage, Report 1.61.5c, report prepared for MEND, edited by O’Kane Consulting, July 2012.
MCA (Minerals Council of Australia) (2005) Enduring value – The Australian minerals industry framework for sustainable
development, published by the Minerals Council of Australia.
NRA (NRA Environmental Consultants) (2014) Rehabilitation progress report: 2014 Monitoring Ernest Henry Mine, Queensland,
prepared for Ernest Henry Mine Pty Ltd by NRA Environmental Consultants.
Queensland Herbarium (2014) Regional Ecosystem description database (REDD), version 8.1 (April 2014), Queensland Department
of Science, Information Technology, Innovation and the Arts: Brisbane.
Rykaart, M., Hockley, D. Noel, M. and Paul, M. (2006) Findings of an international review of soil cover design and construction
practices for mine waste closure, in Proceedings 7th International Conference Acid Rock Drainage (ICARD), March, 2006, St.
Louis MO, published by ASMR, Lexington, KY 40502.
Taylor, G., Spain, A., Nefiodovas, A., Timms, G., Kuznetsov, V. and Bennett, J. (2003) Determination of the reasons for deterioration
of the rum jungle waste rock cover, Australian Centre for Mining Environmental Research.
Waygood, C. G. and Ferreira, S. (2009) A review of the current strategy for capping mining spoils, abstracts of the International Mine
Water Conference, 19-23 October 2009, Pretoria, South Africa.
Wilson, G.W., Williams, D.J. and Rykaart, E.M. (2003) The integrity of cover systems – an update, in Proceedings 6th International
Conference of Acid Rock Drainage ICARD, July 2003, Cairns, Queensland, AUSIMM.

Mine Closure 2015 – A.B. Fourie, M. Tibbett, Les Sawatsky and D. van Zyl (eds)
R. Pedlar-Hobbs Denison Environmental Services, Canada

I. Ludgate Denison Environmental Services, Canada
S. Lanteigne Denison Environmental Services, Canada
Mine closure planning and reclamation measures have evolved over the past three decades, as the early 1980s
concept of “planning for closure” has become standard practice. As closure and reclamation considerations
have become routine aspects of mine development from the earliest stages of the mining life cycle, mines that
were previously developed in the absence of closure plan legislation have also designed closure plans for
current operations and planned expansions, and many have initiated progressive reclamation while still in
operation. Canadian mines that developed, operated, and closed prior to the legislated requirement for
closure planning face greater challenges in designing and implementing successful reclamation plans, as they
generally took no measures to reduce the operational footprint and prevent impacts to the environment.
Often, these sites have limited funds available for closure plan design and implementation as they are no
longer generating revenues for the mining company or have been abandoned, and reclamation efforts have
become the responsibility of the government.
Regardless of the stage of the mining cycle during which closure planning is initiated, there are key factors to
consider in designing and implementing a successful plan. These include defining clear reclamation objectives,
assessing available options to meet those objectives, gathering adequate data for evaluating options, having
a flexible and adaptive plan that can be revised during implementation if required, and developing a focused
and integrated monitoring network to track performance of the plan over time. The stage of the mining cycle
during which this process is initiated will determine the reclamation options that are available as well as the
time frame and the cost of achieving success. Measurements of successful reclamation are also different for
mines developed before and after closure plan legislation was adopted. Therefore, there is a need to develop
site-specific performance indicators based on objectives established through collaboration with regulators,
stakeholders, and affected First Nations. This will allow for a balance of expectations regarding what can and
cannot be achieved during the stages of mine closure.
The discovery of uranium in the region of Elliot Lake, Ontario, during the post-World War II era, led to the
development of 12 mines between 1955 and 1958. Three of these – Denison, Stanrock, and Can-Met – are
owned by Denison Mines Inc. By 1996, all three mines had ceased operations and begun the decommissioning
process, which included designing and implementing closure plans for each facility. This paper illustrates how
the key closure planning factors were, and continue to be, essential to the success of the plans designed and
implemented at the Denison Mines Inc. uranium mine sites in Elliot Lake over the past 20 years.
Mine closure planning and implementation of reclamation measures have evolved over the past 20–30 years,
as the early 1980s concept of “planning for closure” (Culver et al., 1982) has become standard practice
(McKenna, 2011). As closure and reclamation considerations have become routine aspects of mine
development from the earliest stages of the mining life cycle, mines in Canada that were previously
developed in the absence of closure planning legislation have also designed closure plans for current

Key factors in developing and implementing a successful mine closure plan – Denison sites R. Pedlar-Hobbs, I. Ludgate, and S. Lanteigne
operations and planned expansions, and many have initiated progressive reclamation while still in operation.
Provinces began adding mine reclamation and closure requirements into mining laws and regulations in the
early 1990s, with specific requirements varying across the country (Cowan et al., 2010). Canadian mines that
developed, operated, and closed or were abandoned prior to Canada’s legislated requirement for closure
planning face greater challenges in designing and implementing successful reclamation plans as they
generally took no measures to reduce the operational footprint and prevent and/or mitigate impacts to
surrounding land, water, and air. Often, these sites have limited funds available for closure plan design and
reclamation activities as they are no longer generating profits for the mining company or have been
abandoned, and reclamation efforts have become the responsibility of the government.
Regardless of the stage of the mining cycle during which the closure planning process is initiated, there are
key factors to consider in designing and implementing a successful reclamation plan. These include defining
clear reclamation objectives, assessing available options to meet those objectives, gathering adequate and
accurate data for evaluating options, having a flexible and adaptive plan that can be revised during
implementation if required, and developing a focused and integrated monitoring network to track
performance and effectiveness of the plan over time. The stage of the mining cycle during which this process
is initiated will determine the reclamation options that are available as well as the time frame and the cost
of achieving success. Measures of reclamation success vary for mines developed before and after reclamation
legislation was adopted. Therefore, there is a need to develop site-specific performance indicators based on
closure plan objectives established through collaboration with regulators, stakeholders, and affected First
Nations. This will allow for a balance of expectations regarding what can and cannot be achieved during the
stages of mine closure.
Measurements of reclamation progress are critical in assessing the effectiveness of mine reclamation
programs over time. These often include indicators such as the diversity and structure of biological
communities, the elimination of contaminants, the capacity of the physical environment to sustain viable
populations, the recolonisation of native species, and integration with the surrounding landscape (Batigelli,
2011; Hendrychová, 2008; Mummy, 2002). Although the closure plan objectives and performance indicators
may be specific to each mine site, the overall goals of mine reclamation are to restore mine sites and affected
areas so that they are self-sustaining, compatible with a healthy environment, and can support human
activities to a degree that is commensurate with the planned end use of the site. This paper examines the
reclamation of the Denison sites to illustrate the benefits of a properly designed and well-executed closure
plan.
The discovery of uranium in the region of Elliot Lake, Ontario, during the post-World War II era led to the
development of 12 mines between 1955 and 1958 (Figure 1). Three of these – Denison, Stanrock, and Can-
Met – are owned by Denison Mines Inc., and the remaining nine sites are owned by Rio Algom Ltd. Most of
the uranium was produced for contracts with the United States Atomic Energy Commission during the late
1950s and early 1960s. Following the cancellation of these contracts, most mines in the area closed, but
Denison Mines continued to operate, supplying uranium to electrical power generating utilities in Japan and
Ontario from the early 1970s to the early 1990s. By 1996, due to diminishing ore grades and high production
costs, all the Elliot Lake uranium mines had ceased operations and begun the decommissioning process,
which included design and implementation of closure plans for each of the facilities. The details of this
process for the Denison properties are described below.

Case Studies
The Denison property, located 16 km north of the city of Elliot Lake, consisted of a conventional mill and
underground mine that operated between 1957 and 1992. Over this time, Denison generated a total of
63 million tonnes (Mt) of uranium tailings that were deposited into two bedrock lined basins, which are now
referred to as tailings management areas (TMA-1 and TMA-2).

The Stanrock property, located 21 km northeast of the city of Elliot Lake, was also developed in the 1950s
and consisted of a mine, mill, and TMA. The underground mine used conventional milling to recover uranium
from 1958 to 1964. From 1964 to 1970, and from the late 1970s to 1983, a mine water recovery operation
was used to extract uranium from water that was allowed to flood the underground workings. During the
latter period, the mine was rehabilitated and a small amount of ore was hoisted to surface and hauled to the
Denison mill for processing.
The Can-Met property is located adjacent to the Stanrock property, and both the mine and mill operated
from 1957 until 1960. Tailings from both the Stanrock and Can-Met facilities were discharged into the natural
basin of a small lake located immediately south of both mines. This became the Stanrock tailings basin (TMA-
3), which contains approximately 5.7 Mt of tailings.
Prior to closure, the mines operated in accordance with stringent licensing requirements of the Atomic
Energy Control Board of Canada (AECB), now known as the Canadian Nuclear Safety Commission (CNSC), in
consultation with other federal and provincial regulatory agencies. When closure of the mines was
announced in the early 1990s, the Canadian regulatory framework governing mine closures was still in the
process of being finalised. The decommissioning process for the uranium mines was the largest privately-
owned, multimine closure program seen up to that time in Canada (Payne et al., 2004). Over the course of
three years, extensive environmental hearings were conducted to evaluate the proposed closure plans and
the potential environmental and socio-economic impacts. The approvals process was facilitated by the
establishment of a joint review group (JRG) comprised of more than 15 federal and provincial regulatory
agencies, with the CNSC as the lead regulatory agency. This collaborative effort allowed for coherent
responses and decisions on all issues during the hearing process.
Both companies involved in the decommissioning process (Denison Mines Inc. and Rio Algom Ltd.) shared
common overarching objectives for the reclamation of all sites, which were to protect public health and
safety, minimise long-term environmental impacts from the decommissioned facilities, and ensure the
decommissioned sites would not prejudice the survival of affected communities after the mine closures.
Fortunately, since the 1970s, the physical, chemical, and radioactive characteristics of the Elliot Lake tailings
have been studied by both companies in conjunction with government agencies such as the Canada Centre
for Mineral and Energy Technology (CANMET) and the Mine Environmental Neutral Drainage Program
(MEND) and by various university research groups. This work provided the companies with a significant
database of information that was essential in planning, designing, and implementing the closure plans (Payne
et al., 2004).
The Elliot Lake uranium deposits are classified as low grade (< 0.1% U3O8). Although the radioactive
characteristics of the tailings were a concern for the JRG, detailed radiological modelling conducted as part
of the National Uranium Tailings Program has confirmed that exposures and risks to the public from the
Denison facilities are very low (Denison Mines Ltd., 1995). However, in addition to radioactive elements, the
tailings contain an average pyrite content of 6%. When pyrite is exposed to air, it reacts with oxygen to
produce sulphuric acid. This in turn mobilises metals and radioactive isotopes and has a detrimental effect
on receiving waters if not treated. All of the water from the TMAs drains into a network of rivers and lakes
that eventually discharge into Lake Huron. Acid generation was deemed to be the most significant issue for
consideration during the development of decommissioning options, although long-term physical stability of
the sites and achieving radiation doses as low as reasonably achievable were also important (Denison Mines
Ltd., 1995).
A number of closure options were considered and evaluated against the established design criteria and
objectives prior to selecting the final reclamation plans for each site (Table 1). Site-specific probabilistic risk
assessments were also completed to examine the likelihood and consequences of failure resulting from
discrete disruptive events such as earthquakes, floods, or droughts having return periods of up to 1 in 10,000

Case Studies
years. The risk assessments also considered the expected costs of environmental releases resulting from
discrete events and the cost of ongoing care and maintenance (Belore et al., 1999; Kam et al., 1999a, 1999b;
Welch et al., 1996). The ability of each option to meet the specified design objectives and criteria for each
TMA was assessed on the basis of the option’s potential environmental, economic, and social impacts. The
evaluation also took into account the guidelines for environmental assessment laid out by the Federal
Environmental Assessment Review Office (FEARO) panel and the recommendations made at public hearings.
Following thorough evaluation of all the options and consideration of the risk assessment results, the panel
concluded that the final approved reclamation plans were designed and constructed with the best available
technology to address the problem of perpetual containment (CEAA, 1996). The decommissioning plans that
were selected and implemented at the Denison and Stanrock sites are described in the following sections.
Option Description
Base case Long-term collection and treatment of contaminated seepage and run-off water
Water cover Use of natural precipitation and run-off water to flood the exposed tailings
Water-retaining perimeter dams required
Soil cover Use of non-acid-generating soil material to cover the exposed tailings and encourage a
rise in the water table elevation
Complete Hydraulic relocation of tailings
removal
Underground disposal
Deep lake (Quirke Lake) disposal
In situ A modification to the base case
(Stanrock
Construction of new, low permeability dams to establish and maintain an elevated
only)
water table near the tailings surface
Similar advantages to a water cover
A water cover was selected as the most suitable option for decommissioning TMA-1 and TMA-2 at the
Denison site (Figure 2).

The water cover was selected based on the following criteria:

 Acid generation is eliminated or nearly eliminated.
 Airborne release (i.e., dusting) is eliminated.
 Geography of the area is well-suited to flooding.
 Risk of structural failure is very remote.
 Flooded tailings would result in no requirement for new lands.
 Burden on future generations is limited.
 Basins are returned to a use very similar to their pre-mine development use.
 Radiation dose to receptors will be much less than anticipated limits.
 Water cover is an effective barrier to intrusion by man.
The decommissioning of the Denison site was completed in 1996 using the water cover option at both TMA-
1 and TMA-2 (Figure 2). However, the process of implementation was unique to each TMA. Constructing large
dams would not have been economical in TMA-2 nor would they have been sustainable within a small
watershed.
As a result, tailings were removed from TMA-2 to lower the elevation and meet the requirements of a water
cover. The tailings were relocated to two areas (TMA-1 and the underground workings) using hydraulic

Case Studies
monitors and slurry pumping. The appearances of TMA-1 and TMA-2 before and after decommissioning are
illustrated by pairs of photographs in Figure 3.
a) b)
c) d)
All surface facilities at the Denison site were demolished, and the site was revegetated, with the exception
of the executive lodge, the two effluent treatment plants, and roadways required for long-term access. Prior
to demolition, equipment was removed for sale when possible, and scrap steel was removed for recycling.
Asbestos was removed and placed into the Denison landfill located at the northeast end of TMA-2, which
was subsequently graded, capped with a till cover, and revegetated. All other hazardous materials, such as
laboratory chemicals, were disposed of by a firm licensed to undertake that work.
An in situ management plan was chosen as the preferred option for decommissioning the Stanrock TMA
(Figure 4).

The reasons for selecting this option included the following:

 The majority of the acid generating tailings would be stored below the water table.
 Airborne dust and radon releases would be reduced.
 Surface water quality would be within accepted loadings.
 No new areas were required.
 The basin would be returned to conditions similar to those of pre-mine development.
 The option was cost effective and addressed all concerns within the existing TMA.
 Future risks of major system failures were minimal.
The decommissioning of the Stanrock and Can-Met sites was completed between 1992 and 2000 (Ludgate et
al., 2005). New containment dams were constructed downstream of the existing structures to raise the water
table. The dams incorporated a water-retaining core of compacted till and a bedrock foundation beneath the
dam core that was grouted to minimise seepage. Filter drains were provided to prevent internal erosion and
a build-up of pore water against the dam. Although there is a small head-pond, water is generally not
impounded in the TMA, but drains from the surface, passes through a spillway in Dam A, and flows, along
with seepage collected from all of the dams, to a treatment plant southwest of the plant. Although
constructing the dams effectively raised the water table, it could not be raised high enough to completely
cover all of the tailings due to fractured bedrock Unoxidised tailings above the water table still react to form
additional acid, which is flushed out along with the acid already present. Modelling conducted as part of the
environmental assessment predicted that it may take up to 50 years to deplete the entire contained acid
inventory (Denison Mines Ltd., 1995).

Case Studies
The demolition of the Stanrock surface facilities was carried out in 1992 and 1993. Prior to demolition,
equipment and material were sold for reuse or recycling when possible, and all hazardous materials were
removed and disposed of appropriately. Facilities that were demolished included the mill complex, head
frames, mine offices, workshops, warehouse, mine hoist rooms, pump houses, and sewage plant. The
underground crushing and grinding plant, located approximately 60 m below the surface, was used as a
disposal site for solid waste and was completely filled with demolition material. Due to the need for long-
term water management and treatment of acidity, radium, and mine-related metals at the Stanrock site,
some infrastructure was retained. Demolition works excluded some power lines, an electrical substation, the
effluent treatment plant, the Dam G pumps, the trestle, and the pipeline, as these services are still required
into the foreseeable future.
Each of the TMAs was upgraded to ensure that they meet current standards for the protection of health,
safety, and the environment. There are a number of programs in place to ensure the continued safe
operation, care, and maintenance of these decommissioned sites. The general programs include site security,
radiation protection programs, health and safety programs, inspection programs, tailings management
operating programs, monitoring programs, reporting programs, and emergency and contingency response
programs.
In addition to routine monitoring of all site infrastructure by care and maintenance staff, annual inspections
are also completed by professional engineers in order to verify that the facility is performing as designed,
confirm that routine care and maintenance activities are being completed as required, and identify any
potential areas of concern that may require remedial action. To date, there have been no issues at any of the
structures, and the TMAs and basins are operating as designed.
Denison has undertaken extensive environmental monitoring programs since the early 1970s. At the time of
closure, each mine had its own environmental monitoring program conducted under an operating license
from the AECB (now CNSC) or a Certificate of Approval (now Environmental Compliance Approval) from the
Ontario Ministry of Environment and Climate Change (MOECC). In 1997, both Denison Mines Inc. and Rio
Algom Ltd. reviewed their existing monitoring requirements in terms of their relevance to current
environmental data and predictions of changing conditions associated with decommissioning, as outlined in
the Environmental Impact Statement (Denison Mines Ltd., 1995). The evaluation resulted in the development
of a focused and integrated monitoring network to track the performance and effectiveness of the
reclamation plans over time. The integrated strategy is subdivided into four programs that supersede the
monitoring requirements listed in the licenses and Certificates of Approval for the TMAs. The monitoring
programs include the following:
 Serpent River Watershed Monitoring Program (SRWMP): This replaced the various mine-specific
receiving environment monitoring programs with one comprehensive program focused on water
and sediment quality, benthic invertebrate and fish communities, and radiation and metal doses
to humans and wildlife.
 In-basin Monitoring Program (IBMP): This companion program to the SRWMP assesses the health
risks to biota potentially feeding at each of the aquatic and vegetated management areas.
 Source Area Monitoring Program (SAMP): This program monitors the nature and quantities of
contaminant releases to the watershed.

 TMA Operational Monitoring Program (TOMP): This program generates the data used to track
TMA performance and supports decisions regarding the management and discharge compliance
of the TMAs.
Each program was designed to directly complement the other three programs in terms of monitoring
locations, parameters, and sampling frequency and thus ensure that the overall monitoring framework is
comprehensive and interpretable. These programs are objective-driven and allow for modifications to be
made over time in response to changes in the conditions at the sites.
Data collected under TOMP and SAMP demonstrate the effectiveness of decommissioning activities at both
the Denison and Stanrock sites. Since decommissioning, there have been substantial reductions in the
quantity of reagents used at the effluent treatment plants or directly within the TMAs. For example, at the
Denison site, the liming of the surficial tailings and the use of water cover has resulted in circumneutral pH
in both TMA-1 and TMA-2. As a result, the need for caustic treatment has been modest and generally only
required to treat acidic rain and meltwater (Figure 5).
Decommissioning activities were also effective in reducing the concentrations of heavy metals and sulphates
in the TMAs, as indicated by the trend of decreasing concentrations present in pretreated water collected
from TMA-1 between 1992 and 2014 (Figure 6). However, with the reduction in sulphate concentrations,
there has been a slight increase in radium concentrations. This is due to the remobilisation of radium, which
substitutes in the structure of barite (Ba,Ra)SO4. Despite the relatively low solubility of barite, this
observation is not surprising based on initial modelling, and additional studies have shown increased mobility
and release of radium in pyritic uranium tailings when sulphate ion mobility control has been depleted for
both on-land and underwater disposal scenarios (Davé, 1999a, 1999b; Davé et al., 2002). The
decommissioning of TMA-1 and TMA-2 has been successful in addressing the primary concern of acid
generation and has resulted in improved water quality. Ongoing and future management of the TMAs is
focused on the possibility of increased radium mobility with the depletion of sulphate over time within the
basins. Recent modifications to water management and treatment have included installation of a filter berm
at the outlet of the settling pond and installation of a siphon system to better control influent flow for
treatment (Ramsay et al., 2013).

Case Studies
At Stanrock TMA-3, there have also been great improvements in the reagent requirements for water
treatment as a result of decommissioning activities. Prior to the construction of the new dams, the TMA
surface run-off and the combined seepages required large quantities of lime to neutralise the water before
it was released. Since dam construction and decommissioning, the consumption of neutralising agents and
barium chloride has decreased substantially due to decreased tailings water acidity and radium
concentrations in the influent water (Figure 7).
The SRWMP is conducted on a five-year cycle, and four cycles of the program have been completed to date
(1999, 2004, 2009, and 2014). Water quality in receiving waters is better than criteria established for the
protection of aquatic life (Table 2) and in most cases, the concentrations of mine indicator parameters (pH,
SO4, Ra, U, Ba, Co, Fe, and Mn) have improved or remained consistent over time (Bailey et al., 2013).

Sediment quality is below upper limits established for the protection of aquatic life, but remains above lower
limits and background concentrations. Benthic invertebrate density and community structure in exposure
lakes are different from those in reference lakes, but the magnitude and number of differences are
decreasing over time, indicating a gradual recovery. Public radiation doses are below Health Canada
Guidelines of 0.3 mSv/y (Davé, 2014).
pH SO Ra(T) U Ba Co Fe Mn
# (mg/L) (Bq/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L)
Location samples 1
Assessment 6.3– 100 1 0.005 0.047 0.0009 0.47 0.098
collected Criteria 8.5
MDL 0.1 0.005 0.0005 0.005 0.0005 0.02 0.002
Reference
D4 4 7.2 4.6 <0.005 <0.0005 0.013 <0.0005
P-22 4 6.9 3.2 <0.005 <0.0005 0.015 <0.0005
SR-05 4 6.9 3.7 <0.005 <0.0005 0.006 <0.0005
SR-14 4 6.9 4.2 <0.005 <0.0005 0.01 <0.0005
SR-18 4 6.9 4.3 <0.005 <0.0005 0.043 <0.0005
SR-19 4 7.0 3.7 <0.005 <0.0005 0.022 <0.0005
Near field
D5 4 7.1 14.9 0.053 0.0017 0.059 <0.0005
D6 4 6.6 33.0 0.006 <0.0005 0.017 0.34 0.215
DS-18 4 6.8 73.5 0.101 0.0013 0.016 1.68
M-01 4 6.9 10.3 0.014 0.0032 0.015 <0.0005 0.82
Q-09 4 7.0 37.0 0.068 0.0021 0.073 <0.0005
Q-20 4 7.0 21.0 <0.005 <0.0005 0.018 <0.0005
SC-01 4 7.0 21.0 0.006 <0.0005 0.01 <0.0005 0.12
SR-06 4 6.8 47.5 0.057 0.0009 0.313 <0.0005
Far field
SR-01 4 6.9 34.0 0.017 0.0013 0.038 <0.0005
SR-08 4 7.0 165.0 0.026 0.0009 0.018 <0.0005
1. These represent the most recent assessment criteria for the protection of aquatic life or the
upper range of background concentrations (BCMOE, 2006; CCME, 2003; Ontario, 1994).
Denison Mines Inc., in association with Rio Algom Ltd., has developed an extensive public information
program to ensure effective communication of site activities with local residents. A more intensive program
was deemed necessary because of the changing demographics of the community. Elliot Lake is no longer a
mining town, but it has become a leading provider of retirement living services.

Case Studies
The public information program includes annual site tours (in conjunction with the annual Elliot Lake Uranium
Heritage Days, held in early July), public presentations, public awareness meetings, development of an
information web site, newsletters, and the release of reports. An Elliot Lake rehabilitation information web
site (www.denisonenvironmental.com) was developed to provide both historical information and current
monitoring results. Biannual presentations are made to the city council, and public meetings are held
periodically to inform the public of important events and receive feedback on site activities. An annual
newsletter is distributed through the local newspaper and is also available on the web site. All environmental
monitoring reports are provided to the city and are available to the public at the local library.
Ongoing consultation and collaboration with Serpent River First Nation (SRFN) is also a key part of long-term
care and maintenance of the Denison and Stanrock sites. Meetings, community presentations, and site tours
are held on an ongoing basis and as requested to address specific questions or areas of interest. SRFN has
also participated in wildlife studies within the Serpent River watershed to incorporate traditional knowledge
into the site monitoring programs.
Despite the significant challenges faced by Denison Mines Inc. when the mines ceased operations in the early
1990s, the decommissioning process in Elliot Lake has been successful. Reclamation objectives were
established through collaboration with regulators, local communities, and affected First Nations. Options
available to meet the objectives were determined by drawing upon an extensive database of information
that was established in conjunction with government and university research groups. Reclamation options
were evaluated against design criteria and objectives, risk assessments were completed, and the final
reclamation plans were approved by the FEARO panel (CEAA, 1996). During implementation of the
reclamation plans, a focused and integrated monitoring network was developed, which allows tracking of the
performance and effectiveness of the plans over time. This monitoring network also allows for adaptive plans
to be created and implemented as required to ensure the objectives of the reclamation plans continue to be
met.
Batigelli, J.P. (2011) Exploring the world beneath your feet – soil mesofauna as potential biological indicators of success in reclaimed
soils, in Proceedings 15th International Conference on Tailings and Mine Waste, 6–9 November 2011, Vancouver, Canada,
Norman B. Keevil Institute of Mining Engineering, Vancouver.
Belore, H., Balins, J., and Payne, R. (1999) Flood and drought risk models for tailings management areas, in Proceedings Canadian
Dam Association Annual Conference, October 1999, Sudbury, Ontario.
British Columbia Ministry of the Environment (BCMOE) (2006) British Columbia approved water quality guidelines,
http://www.env.gov.bc/wat/wq/bcguidelines.html.
Canadian Council of Ministers of the Environment (CCME) (2003) Canadian water quality guidelines for the protection of aquatic life,
in Canadian environmental quality guidelines, Canadian Council of Ministers of the Environment, Winnipeg.
Canadian Environmental Assessment Agency (CEAA) (1996) Decommissioning of uranium mine tailings management areas in the
Elliot Lake area, Report of the Environmental Assessment Panel, Ottawa, Canada.
long-term liabilities: a guidance document, prepared for National Orphaned/Abandoned Mine Initiative (NOAMI), Cowan
Minerals Ltd., Sudbury, Ontario, 123 pp.
Culver, K.B., Chakravatti, J.L., Gorber, D.M., Knapp, R.A., and Davis, J.B. (1982) Close-out concepts for the Elliot Lake uranium mining
operations, in Proceedings International Symposium on Management of Wastes from Uranium Mining and Milling, 10–14
May, 1982, Albuquerque, New Mexico Vol. 14, Iss.18, pp. 157–168.
Davé, N.K. (1999a) Underwater disposal of pyritic uranium tailings, part I – laboratory studies of Ra-226 diffusion from fine and coarse
tailings, Division Report, MMSL 99-077(CR), Mining and Mineral Sciences Laboratories, CANMET, Natural Resources Canada,
Ottawa, Ontario.
Davé, N.K. (1999b) Underwater disposal of pyritic uranium tailings, part II – laboratory studies of Ra-226 diffusion from gypsum
depleted composite tailings, Division Report, MMSL 99-078(CR), Mining and Mineral Sciences Laboratories, CANMET, Natural
Resources Canada, Ottawa, Ontario.
Davé, N.K., Paktunc, A.D., and Lortie, L. (2002) Laboratory column leaching studies of pyritic uranium tailings and waste rock from
Quirke Mine, Rio Algom Limited, Elliot Lake, Ontario, Division Report, MMSL 01-022(CR), Mining and Mineral Sciences
Laboratories, CANMET, Natural Resources Canada, Ottawa, Ontario.

Davé, N.K. (2014) Biological recovery of freshwater and marine habitats containing submerged reactive mine tailings, Draft Report
MMSL 14-0XX(INT), Mining and Mineral Sciences Laboratories, CANMET, Natural Resources Canada, Ottawa, Ontario.
Denison Mines Ltd. (1995) Environmental impact statement – decommissioning of the Denison and Stanrock tailings management
areas, Denison Mines Ltd., Elliot Lake, Ontario.
Hendrychová, M. (2008) Reclamation success in post-mining landscapes in the Czech Republic: a review of pedological and biological
studies, Journal of Landscape Studies, Vol 1, pp. 63–78.
Kam, S.N., Morrell, R. and Ludgate, I.R. (1999a) Containment dam design for decommissioning of the Denison Mines tailings
management areas in Elliot Lake, in Proceedings Canadian Dam Association Annual Conference, October 1999, Sudbury,
Ontario.
Kam, S.N., Payne, R.A., and Knapp, R.A. (1999b) Remediation and management of historical uranium tailings sites in Elliot Lake, an
overview, in Proceedings Canadian Dam Association Annual Conference, October 1999, Sudbury, Ontario.
Ludgate, I., Primeau, P., Laliberte, R., and Pedlar, R. (2005) The decommissioning of the Denison uranium facilities in Elliot Lake, in
Proceedings from Canadian Nuclear Society conference on waste management, decommissioning and environmental
restoration for Canada's nuclear activities: 'Current practices and future needs', 8–11 May 2005, Canadian Nuclear Society,
Ottawa, Ontario (Canada), 22 pp.
McKenna, G. (2011) Mine Closure 2011 conference – lessons learned, Canadian Reclamation, Vol. 12(1), pp. 38–43.
Mummy, D.L. (2002) Microbial biomarkers as an indicator of ecosystem recovery following surface mine reclamation, Applied Soil
Ecology, Vol. 21, pp. 251–259.
Ontario (1994) Policies, guidelines, provincial water quality objectives of the Ministry of Environment and Energy. Ontario Ministry
of the Environment and Energy, Queen’s Printer for Ontario.
Payne, R.A., Balins, J.K., and Kam, S.N. (2004) Elliot Lake – a sustainable transformation, in Proceedings Sustainable Post-Mining Land
Management Conference, 4–6 November 2004, Krakow, Poland.
Welch, D.E., Davis, J.B., Payne, R.A., Becker, D.E., and Balins, J.K. (1996) Criteria and procedures for assessing the closure performance
of uranium tailings facilities in Elliot Lake, Ontario, in Proceedings CDSA/CANCOLD Joint Canadian Dam Safety Conference,
October 1996, Niagara Falls, Ontario, Riley's Reproduction & Printing Ltd., Calgary.
Bailey, J.L., Celis-Sagado, M., Heneberry, J., and Yan, N. (2013) Using zooplankton community structure and bioassay to compare
recovering acidified and contaminated lakes in Northern Ontario, in Proceedings: Sixth Annual Ontario Mine Reclamation
Workshop, 18–19 June 2013, Cobalt, Ontario.
Ramsay, K., Luchinski, J., Conway, A., Morrell, R., Bronson, J., Pedlar-Hobbs, R., and Ludgate, I. (2013) Improvement of water
treatment at a former uranium mine through improved management practices and better effluent discharge control, in
Proceedings Sixth Annual Ontario Mine Reclamation Workshop, 18–19 June 2013, Cobalt, Ontario.

M. Haakensen Contango Strategies, Canada

V. Pittet Contango Strategies, Canada
J. Spencer Capstone Mining Corporation, Canada
J.H. Rodgers Jr. Clemson University, USA
J.W. Castle Clemson University, USA
When designed and executed in a scientifically guided manner, constructed wetland treatment systems
(CWTSs) can treat various contaminants in water. A process-driven treatment plan was developed to convert
contaminants in the water to minerals in the soil, improving water quality and decreasing risk to the
environment. This site-specific CWTS is being developed for Capstone Mining Corporation’s Minto Mine
(Yukon), using a scaled phased approach to allow for improvement, optimisation and flexibility for
modifications along each step. These phases are: (1) site assessment and information gathering, (2)
technology selection and conceptual design, (3) pilot-scale testing and optimisation (controlled environment),
(4) onsite demonstration-scale confirmation and optimisation and (5) full-scale implementation. The first
three phases have been completed successfully, with Phase 4 underway. Performance highlights include
successful application of microbial profiling technologies (genetic and growth based) to guide system design
in a site-specific context. Microbiome technologies applied in a site assessment and throughout pilot-scale
testing allowed for identification of natural copper- and selenium-attenuating microbial communities and
ecosystems at the mine site, correlation with native wetland plant species and confirmation of the
enhancement of beneficial microbial populations through design of the CWTS. The pilot-scale CWTSs
confirmed plant amenability to transplantation, and the design selected for further testing on site achieved
92% removal of copper (mean influent 146 μg/L, outflow 11.3 μg/L) and 41% removal of selenium (mean
influent 10.2 μg/L, outflow 6 μg/L) using synthetic influent designed to mimic the worst-case water chemistry
of a long-term closure scenario. A mass-balance of the pilot-scale systems confirmed that the elements were
sequestered to sediments of the CWTS, with less than 0.5% of the copper and 2% of the selenium transferred
to the plant leaves. The pilot-scale system allowed for selection of the optimal design (from three different
designs) and for different water chemistries to be tested for treatment, mimicking early closure (containing
ammonia and nitrate) and long-term closure (no longer containing ammonia and nitrate). At the conclusion
of testing, the pilot-scale CWTS was converted to a hybrid bioreactor/CWTS to evaluate its potential to
function as a semi-passive treatment system, should the need arise (e.g. change in regulatory requirements
or change of influent water chemistry). Conversion to the hybrid bioreactor/CWTS allowed identification of
critical aspects of the transition period and improvement of selenium removal, having influent of 12 μg/L and
outflow of 3.9 μg/L, with lowest recorded outflow concentration of 1.9 μg/L (84% removal). Data from the
pilot-scale testing were used to calculate system-specific removal rate coefficients, allowing for more accurate
sizing estimates for full-scale and modelling effects of seasonal flow-rate and water chemistry variations. The
optimised design was applied to construct a demonstration-scale CWTS at the Minto Mine during fall 2014 to
refine findings and calculations, including seasonal variations.

Process-driven design and piloting of a site-specific constructed wetland for copper and selenium treatment in the Yukon M. Haakensen et al.
The Minto Mine, operated by Capstone, is located 240 km northwest of Whitehorse, Canada, on the west
side of the Yukon River. The Minto property lies within the eastern part of the Dawson Range, with
elevations from 700 to 1,000 m; the landscape has rounded mountains intersected by broad valleys and
drainages that are part of the Yukon River watershed. The climate in the Minto region is subarctic
continental characterised by long, cold winters and short, cool summers. The area experiences moderate
precipitation in the form of rain and snow and a large range of temperatures on a yearly basis with a mean
annual temperature below 0°C.
The Minto Mine has been in commercial operation since October 2007, and the deposits being mined are
copper sulphide mineralised zones. Surface water and groundwater quality is a key consideration in the
evaluation of potential effects of mining and mineral development projects, and changes to water quality
parameters have the potential to affect aquatic and human use of water resources. A reclamation and
closure plan (RCP) is required under both the Yukon Territorial Waters Act and Waters Regulation in the
form of a water licence and the Quartz Mining Act, under direction of the director of mineral resources, in
the form of a quartz mining licence. The RCP is intended to address the long-term physical and chemical
stability of the site and closure of the proposed features and disturbances associated with the mine. As part
of the RCP, a constructed wetland treatment system (CWTS) is being designed, evaluated and optimised for
water treatment at closure through a phased program. These phases are: (1) site assessment and
information gathering, (2) technology selection and conceptual design, (3) pilot-scale testing and
optimisation (controlled environment), (4) onsite demonstration-scale confirmation and optimisation and
(5) full-scale implementation. The CWTS is being designed with the objective of treating water during free-
flowing months. Low flow volumes in the winter will freeze and will discharge with snow melt during
freshet. The findings from Phase 3 pilot-scale testing and optimisation in a controlled environment are
summarised below.
When designed and executed scientifically, CWTSs can treat various contaminants in water (Hawkins et al.,
1997; Huddleston and Rodgers, 2008; Murray-Gulde et al., 2008; Rodgers and Castle, 2008). Once
established, CWTSs can become self-sustaining ecosystems with the plants providing yearly renewal of
carbon to fuel microbial activity. As such, they possess the desirable potential to remediate contaminated
mine drainage for as long as it is generated. Having a site-specific approach allows for the appropriate set of
operational conditions to be achieved, where complex coupled reactions can take place for treatment of
targeted constituents. CWTSs have been used to mitigate risks to aquatic receiving environments from a
variety of aqueous contaminants, including copper (Cu) and selenium (Se) (Kadlec and Wallace, 2009), which
are of interest to the Minto Mine.
Many biogeochemical reactions take place in a wetland environment, but specific reactions may decrease
the aqueous concentrations and bioavailability of some contaminants (e.g. Cu, Se), thus lowering risk to the
receiving environmental systems by decreasing exposure to contaminants. Removal mechanisms include
filtration and settling, sorption to soils or dead organic material, uptake into plants or biofilms, complexation,
degradation, mineralisation and volatilisation. A process called dissimilatory sulphate reduction can
chemically reduce available aqueous sulphate to sulphide through the metabolism of specific bacteria. Once
dissimilatory sulphate reduction has occurred, the sulphide provides a mechanism to remove dissolved
cations such as Cu from water by complexation to produce copper sulphide, an insoluble precipitate. The
mechanism for selenium treatment in a wetland is more direct, through dissimilatory selenium reduction
that is also microbially mediated, sometimes by some of the same bacteria that catalyse dissimilatory
sulphate reduction. The most common forms of aqueous Se are selenite and selenate, with valence states of
IV and VI, respectively. When these are reduced in a wetland to an insoluble state, Se precipitates as

Case Studies
elemental Se(0); however, at lower reduction states, it can also be reduced to selenide (-II) valence state,
which will volatilise.
The insoluble forms of Cu and Se are then removed from the water through precipitation/settling and
filtration mechanisms inherent in wetlands. These are then sequestered into the sediments over time
through accretive processes. Accretion is the naturally occurring process of wetland sedimentary material
(precipitates, decaying plant material, etc.) accumulation over time. CWTSs can be designed to have different
accretion rates. Once an accreting CWTS is established and mature, targeted constituents are essentially
sealed away under layers of sediment, rendering them less bioavailable and less susceptible to re-entry into
the water column. Because of this design attribute, there is no need to dredge the CWTS or harvest wetland
plants for the system designed for the Minto site; in fact, this type of activity would disrupt treatment
functions and re-expose previously sequestered constituents. This process mimics what occurs in natural
wetlands; therefore, it is often the best option for long-term, low-maintenance, effective treatment.
In order to fully understand the function of biogeochemical processes such as the mineralisation of aqueous
dissolved copper and selenium, it is critical that the microbial populations performing these functions be well
characterised and understood. For example, information about the total quantity and relative abundance of
sulphate-reducing bacteria (SRB) can help gauge the effectiveness of a CWTS design and probable robustness
of the system. Microbiome analyses and growth-based assays can provide insights into microbial
communities in CWTSs, allowing for more effective decisions on design and optimisation while helping to
determine the stability of microbial communities in fluctuating conditions.
In order for CWTSs to be predictable and robust, they must be designed, piloted and optimised through
phases in a site-specific manner to allow for improvement along each step (Haakensen et al., 2015). This
manuscript describes results from pilot-scale testing that involved selecting the optimal CWTS design (from
three different designs) and testing different water chemistries for treatment, mimicking early closure
(containing ammonia and nitrate) and long-term closure (no longer containing ammonia and nitrate). At the
conclusion of testing, the pilot-scale CWTS was converted to a hybrid bioreactor/CWTS to evaluate its
potential to function as a semi-passive treatment system, should the need arise (e.g. change in regulatory
requirements or influent water chemistry). This included the addition of an organic carbon amendment
(alfalfa hay and straw) to stimulate reducing conditions for improved selenium treatment. Alfalfa hay and oat
straw were chosen, as they would provide both short- and long-term carbon sources compared to a metred
dosing of more available electron donors such as ethanol. Results from pilot-scale testing were then used to
calculate system-specific removal rate coefficients, allowing for more accurate sizing estimates of the full-
scale CWTS and modelling for effects of seasonal flow-rate and water chemistry variations.
Two plant species were of interest for the pilot-scale CWTSs based on data collected during site assessments
(reports 2013-0100-256 and 2013-0100-257 on the YESAB registry). Carex aquatilis (sedge) was chosen, as it
is readily available at the Minto site and was one of the first plants to colonise cleared or disturbed areas that
were wet or flooded, indicating that it may be a good candidate for quickly establishing the CWTS.
Furthermore, data from the site assessment suggested that Carex hosted a range of beneficial microbes and
was conducive to reducing conditions, presumably with low radial oxygen loss (ROL). Moss was used in the
second and third pilot-scale CWTS designs, as Carex and moss coexisted at Minto locations having shallower
water. As such, we wanted to assess whether moss contributes positively or negatively to treatment. It is
well known that moss has a high uptake rate of cations (such as copper and selenium) and is a relatively
benign sink for these elements (i.e. is not a food source for invertebrates or higher animals, and as such does
not contribute greatly to bioaccumulation [Haines and Renwick, 2009; Longton, 1997; Suren and
Winterbourn, 1991]). This was confirmed with the site assessment in 2013, which produced data indicating
that moss at the Minto Mine has a high uptake of elements such as copper, and could therefore be an
effective means of removing elements from seepage through uptake.

Key considerations for plant selection for CWTSs include physiology (e.g. ROL, water depth tolerance),
biomass production per year versus decomposition rate (to allow for accretion), bioconcentration
tendencies, plant density in the wetland cell (to aid in flow distribution), flow rate tolerance, structural
capacity and prevention of sediment re-suspension, evapotranspiration rate and provision for microbial
habitat. ROL is a crucial parameter to consider when designing a wetland cell’s targeted conditions (e.g.
oxidising vs. reducing; Taylor, 2009), which in turn directly affects treatment mechanisms (e.g. dissimilatory
reduction). Furthermore, plant selection was based on site-specific information, as plant characteristics vary
not only between plant species, but also between ecotypes, and it was important to plan for a CWTS design
that would not be subjected to a drastic transition of plant species over time.
The pilot-scale CWTSs were designed using a process-driven approach based on information gathered
through a site assessment specific to the purpose of guiding CWTS development or other applicable passive
treatment technologies (reports 2013-0100-256 and 2013-0100-257 on the YESAB registry).
Pilot-scale surface-flow CWTSs were constructed in dedicated indoor greenhouse pilot facilities in Saskatoon,
Saskatchewan, Canada. The ambient temperature of the indoor facilities was kept at 20°C (+/- 6°C) with
minimum 12-hour photo periods. The CWTSs were planted with either Carex aquatilis (sedge), or Carex and
aquatic moss (with and without 5% v/v pine biochar added to the soils) for a total of three types of CWTSs
being tested in duplicate, resulting in six pilot-scale treatment systems in this study. Each system had two
cells in a series in order to better define removal rate coefficients through a range of concentrations. Each
cell had a 41 cm diameter and a height of 57 cm (Figure 1). Pine biochar was chosen as a soil amendment, as
previous studies have demonstrated that addition of biochar to the hydrosoil can result in faster
establishment and growth of newly planted wetland plants (Haakensen et al., 2013). Additionally, it was
hypothesised that the biochar may provide greater surface area for beneficial microbial communities to
establish.
The hydrosoil mixture was the same for all pilots: sand amended with 2% woodchips (v/v) and 5% peat,
except that the third system design also included a 5% pine biochar amendment. The sand and amendments
were tested for metal and nutrient contents. The selected biochar was low in soluble salts and had a high
surface area, low metals content and moderate available organic carbon and phosphorous. Moss and Carex
were harvested from the Minto site and used to vegetate all pilot-scale CWTSs. To establish the plants, soil
was filled to a depth of 20 cm at the bottom of each cell, and submerged under 10 cm of tap water. Ten to
12 Carex were planted per cell, with an approximately 250 mL volume of moss (compressed), as appropriate
for the system design. The water depth was changed throughout the pilot-scale testing to evaluate effects
on performance and develop strategies for optimisation.

Case Studies
A thorough water characterisation was performed to determine the range of the following constituents:
acidity, alkalinity, biological oxygen demand, carbonate, chemical oxygen demand, chloride, conductivity,
hardness, ion balance, iron, manganese, metals, metalloids, pH, sulphate, total Kjeldahl nitrogen, total
organic carbon and total suspended solids. This is particularly important when considering complex systems
such as mining-influenced waters, as some constituents (e.g. chloride) can interfere with wetland function,
while others can affect the removal of targeted constituents in beneficial or detrimental ways (e.g. iron,
sulphate, ammonia, nitrate, alkalinity). zz
The CWTSs were designed for treatment of the water chemistry predicted to be present at the Minto site
after closure. As such, this water does not yet exist and had to be synthesised as an approximation based on
water quality modelling. The worst-case post-closure water chemistry for the month of October was used,
as it had the highest concentrations of key elements for any month expected to have free-flowing water. The
simulated water recipe underwent five iterations of design, testing and optimisation to ensure the
composition and method of assembling resulted in analytical chemistry that matched the predicted water
quality. The recipe was then assembled with both high-nitrogen and low-nitrogen additions to mimic
concentrations of blasting residues (ammonia and nitrate) in short-term and long-term closure scenarios.
The final recipe and resultant water chemistry are provided in Table 1. Before performance testing began,
the pilot-scale systems were established with tap water for approximately six weeks. This was followed by a
10-week acclimation period (i.e. no testing) with the high-nitrogen synthetic influent to avoid misconstrued
data regarding treatment due to sorption, as well as to establish plants.
The pilot-scale systems were operated for more than 30 weeks, with testing in three phases to determine
the effect of ammonia and nitrate concentrations and added organic carbon on constituent removal. The
first phase of testing was over a six-week period to mimic short-term closure with a high nitrogen
concentration (high N period) based on the predicted maximum worst-case ammonia and nitrate
concentrations (Table 1). This was followed by a second phase of testing over approximately six weeks during
which only very low concentrations of nitrogen (low N period) were present in the influent water in order to
simulate a long-term closure scenario. Although the chemistry of the water is expected to improve over time
in long-term closure (e.g. because of covers and establishment of vegetation), the chemistry was kept
consistent for both periods of testing other than for nitrate (and accordingly ammonia, which can be oxidised
to nitrate). This is because nitrate can affect the treatability of compounds, such as selenium, that are
removed from water through coupled biogeochemical reactions and dissimilatory reduction, as nitrate has a
high affinity for available electrons and is preferentially reduced before other compounds. However, it is
known that after blasting activities have ceased, the concentrations of ammonia and nitrate will decrease
over time. If other constituents were modified at the same time, it would not be possible to ascertain the
effects of nitrate on performance. Therefore, the two periods (high and low N) were tested to gain
information about the expected performance of the system under these two scenarios.

Average influent concentration by testing period

Targeted
Element (total) Unit Acclimation High N Low N Hybrid bioreactor-
concentration1
CWTS
Aluminium (Al) mg/L 0.100 0.089 0.085 0.080 0.098
Arsenic (As) mg/L 0.0030 0.0031 0.0033 0.0033 0.0037
Cadmium (Cd) μg/L 0.334 0.011 0.236 0.336 0.392
Chromium (Cr) mg/L 0.0065 0.0058 0.0064 0.0071 0.0075
Cobalt (Co) mg/L 0.0039 0.0039 0.0042 0.0043 0.0044
Copper (Cu) mg/L 0.13 0.140 0.146 0.146 0.176
Iron (Fe) mg/L 3.1 3.1 3.1 3.2 3.5
Lead (Pb) μg/L 0.70 0.44 0.67 0.59 0.71
Magnesium (Mg) mg/L 50 49 49 51 53
Manganese (Mn) mg/L 2.6 2.5 2.5 2.6 2.7
Molybdenum (Mo) mg/L 0.012 0.013 0.015 0.013 0.013
Nickel (Ni) mg/L 0.0072 0.0077 0.0087 0.0076 0.0080
Selenium (Se) mg/L 0.010 0.010 0.011 0.010 0.011
Sodium (Na) mg/L 44 40 41 42 43
Strontium (Sr) mg/L 2.9 2.5 2.5 2.5 2.6
Sulphur (S) mg/L 85 85 84 86 88
Uranium (U) mg/L 0.0051 0.0012 0.0012 0.0012 0.0014
Vanadium (V) mg/L 0.016 0.015 0.013 0.014 0.018
Zinc (Zn) mg/L 0.032 0.052 0.037 0.040 0.042
Other Parameters
Total ammonia (N) mg/L 0.2-0.5 2 0.61 0.50 0.29 0.05
Dissolved nitrate (N) mg/L 0.8-28 2 29 30 0.86 0.84
pH pH 7.5-8.0 7.89 8.16 7.96 8.01
Alkalinity (Total as mg/L - 107 107 100 100
CaCO3)
Bicarbonate (HCO3) mg/L - 133 130 122 130
10
Water depth cm - 10 17 30
17 3
Hydraulic retention 37± 3
hr - 37± 3 53 ± 2.5 87 ± 3
time 53 ± 3 3
1 Targeted concentration based on modelled worst-case post-closure predicted water quality.
2 Ammonia and nitrate were not added to the synthetic water for the low N and hybrid bioreactor-CWTS periods.
3 Water depth was changed from 10 cm to 17 cm during the high N period to test plant tolerance to deeper water.
The third and final stage of pilot-scale testing included the addition of an organic carbon amendment (alfalfa
hay and straw) to stimulate reducing conditions for improved selenium treatment. This hybrid bioreactor-
CWTS period was conducted for approximately seven weeks to test how the CWTS functions once converted
from conventional CWTS operation to the hybrid configuration. During the hybrid bioreactor-CWTS testing
period, 1.4 g of alfalfa pellets were added per litre of water in each of the first cells in a series. Straw was
added at 10 cm3 per L to the first cells, and 5 cm3 per L to the second cells in a series. The alfalfa pellets and

Case Studies
oat straw were subjected to analytical and leachability testing prior to use to ensure suitability for this
application.
Constant flow metering pumps were used to set the flow rate of water into the system. Flow rates were
confirmed regularly (at minimum weekly) and adjusted as necessary to within +/−10% of the desired flow, to
achieve a known hydraulic retention time (HRT). Outflow rates were also measured to confirm that they were
not significantly different from inflow rates, indicating that evaporation was not a major influence on water
quality (i.e. through concentration).
Dissolved oxygen (DO; mg/L), temperature (°C), pH, specific conductivity (SPC; μS/cm), and oxidation-
reduction potential (ORP; mV) were measured on a routine schedule with an YSI Professional Plus handheld
unit. The relative oxidation-reduction potential of the soil (referred to in this document as ‘relative redox
potential’ to avoid confusion with water ORP measurement) was measured using inert electrodes (copper
wire probes with platinum tips) that were permanently installed in the soil of the cells and remained there
for the duration of the project to ensure accurate readings (Faulkner et al., 1989). To take a reading of the
relative redox potential between the soil and water, a reference electrode (Accumet Calomel) was suspended
in the water column above the inert electrode and measured in millivolts by a voltmeter.
Water, sediment and plant samples were collected throughout the pilot-scale study, with analytical testing
performed by a third-party accredited laboratory. Outflow concentrations for pilot-scale system designs
and/or testing periods were compared, with a p-value calculated using the non-parametric Kruskal-Wallis
test. Sediment samples were also collected for the microbial analyses outlined in Section 2.5. In addition to
routine analysis for metals in plants, a replicate of each plant sample (Carex and moss) collected at the end
of the low N period was used in a trial to determine whether the plants leach constituents of concern, such
as Cu and Se, after freezing and thawing of surrounding water. In this trial, 200 mL of water from the
corresponding CWTS cell was added to the plant sample to completely cover the plant material. The samples
were frozen at −20°C for five days, and then thawed at room temperature (~22°C) overnight. The plant
samples were removed and lightly squeezed to return residual water to the container, and the plant tissue
was sent for metals analysis to compare with pre-freeze concentrations.
A removal rate coefficient (k) can be calculated according to Equation (1).

−ln⁡(𝐶𝑓⁄𝐶𝑖)
𝑘= 𝑡
(1)
Cf = final concentration
Ci = initial concentration
t = HRT
Once k has been calculated from known data (such as the data used in this study), the equation can be
rearranged to solve for the HRT (t) and, hence, estimated full size needed for the CWTS to achieve outflow
objectives, given a known influent concentration and flow rate. Likewise, it can be used to solve for an
outflow concentration, given a known retention time and influent concentration, thereby being a necessary
tool for use in any CWTS design. That said, removal rate coefficients are highly specific and must be developed
in a site-specific manner for each element of interest. While they may sometimes be applied in a conceptual
manner to other situations/sites, caution should be taken in applying a removal rate coefficient developed
for one design and water chemistry to a very different chemistry or design basis. It is often the case (as it is
here for copper) that k must be calculated and applied for different ranges of certain constituents. For this
reason, the pilot-scale CWTSs were constructed with two cells in a series to test different ranges of element
concentrations as they are treated through the system.

Most-probable number (MPN) assays were performed for sulphate-reducing (active standard ASTM D4412),
selenite-reducing (Siddique et al., 2006) and nitrate-reducing (nitrate reduction test, supplied by Sigma-
Aldrich) organisms. Total heterotrophs (grown with YTS250 medium; Lefrançois et al., 2010) were also
quantified in aerobic and anaerobic conditions. In brief, sediment samples were diluted 1:100 with a 0.1%
peptone solution. This starting dilution was then diluted serially from 1/400 to 1/419,430,400. At minimum,
all tests were conducted in duplicate. Wells were incubated at +30°C without light and assessed for visible
growth (formation of a bacterial pellet) and/or colour change specific to the type of media according to the
publication or manufacturer’s protocol after 27 days. The MPN of organisms capable of each metabolism was
then calculated as previously described (Blodgett, 2010). The MPN of organisms in each pilot-scale system
was compared over time using the non-parametric Kruskal-Wallis test.
DNA was extracted from all sediment samples using the MO BIO Powersoil Powerlyzer DNA extraction kit,
with the addition of phenol during cell lysis per the manufacturer’s alternative protocol. Targeted DNA
sequencing was used to identify bacteria present in each sample via polymerase chain reaction (PCR)
amplification of the v3/v4 region of the 16S ribosomal RNA gene (Klindworth et al., 2013). Library preparation
and sequencing were performed per the manufacturer’s instructions for MiSeq v3 paired-end 300 bp
sequencing (Illumina) for all samples and positive and negative controls. All raw sequences were filtered to
remove low-quality reads based on the following criteria: average quality less than Q30, shorter than 350 bp,
not having an exact match to the forward primer and having any base called N (unknown). Bioinformatics
pipelines consisting of internally developed scripts and selected QIIME scripts (Caporaso et al., 2010; Edgar,
2010) were used to process the reads. Similar sequences were clustered into groups called operational
taxonomic units (OTUs) using a 97% identity threshold and the pick_de_novo_otus.py script. All OTUs with
less than 10 representative sequences across all samples were discarded as a quality-filtering step to remove
OTUs that may have arisen owing to sequencing errors. Taxonomic classification of the OTUs was performed
using the Greengenes database version 13_8 (DeSantis et al., 2006; McDonald et al., 2012). The percentage
of the microbial community classified as organisms of interest (e.g. SRB) was also compared across samples.
Weighted UniFrac distance measure (Lozupone et al., 2011) was used for ordination analyses using the
“phyloseq” package (McMurdie and Holmes, 2013) in R. The most abundant OTUs were compared to the Cu,
Se and S concentrations in the sediment to calculate the Spearman correlation coefficient and associated p-
value.
Over the course of the study, a total of 45,000 L of synthetic water (influent) was received by the pilot-scale
CWTSs (in 41 batches), approximately 1,200 L of synthetic water per week, in a pilot-scale CWTS footprint of
1.56 m2. The system operated with a flow rate of 20 mL/min, resulting in an HRT ranging from 34 to 94.3
hours, depending on water depth during the respective period of testing (Table 1).
The performance of each pilot system, based on minimum, maximum and average measurements of copper
and selenium concentrations as well as several parameters that aid in understanding treatment performance,
is summarised in Table 2 for all three testing periods. The Carex and moss systems achieved an average
copper removal of 92% (mean influent 0.146 mg/L, mean outflow 0.0113 mg/L) and average decrease of 41%
for selenium (mean influent 0.0102 mg/L, outflow 0.006 mg/L) using synthetic influent designed to mimic
worst-case water chemistry for a long-term closure scenario (i.e. low N period). While copper treatment was
relatively stable through the high N and low N periods, selenium treatment took longer to establish and
improved though the high N period, and then was stable during the low N period (Figure 2). While biochar

Case Studies
had no significant effects on performance, for both the high N and low N periods, all systems planted with
moss and Carex had significantly better copper treatment than systems with Carex only (p < 0.005). In
contrast, there was no statistically significant difference among the three pilot-scale system designs for
selenium removal in any of the testing periods.
ORP
Copper Selenium DO
Period pH (water,
(mg/L) (mg/L) (mg/L)
mV)
Influent Avg 0.146 0.0102 8.00 - -
Avg 0.018 0.0076 7.49 5.36 157
Carex Min 0.016 0.0058 6.99 4.08 82
Max 0.020 0.0097 7.94 6.69 215
Avg 0.013 0.0073 7.64 5.35 247
High nitrogen
Carex + moss Min 0.011 0.006 7.19 2.59 104
Max 0.016 0.0095 8.29 7.55 215
Avg 0.012 0.0067 7.84 5.75 140
Carex + moss +
Min 0.009 0.0058 7.39 3.50 84
biochar
Max 0.014 0.0085 8.28 8.70 205
Influent Avg 0.146 0.0102 7.96 - -
Avg 0.021 0.0061 7.54 5.61 168
Carex Min 0.017 0.0053 7.16 3.99 126
Max 0.032 0.0067 8.05 7.72 220
Avg 0.011 0.0060 7.71 5.73 159
Low nitrogen
Carex + moss Min 0.010 0.0056 7.26 3.02 113
Max 0.013 0.0065 8.32 9.78 188
Avg 0.012 0.0063 7.6 5.21 182
Carex + moss +
Min 0.011 0.0058 7.15 2.57 106
biochar
Max 0.015 0.0069 8.25 9.04 374
Influent Avg 0.176 0.0119 8.01 - -
Avg 0.021 0.0040 7.16 1.52 -129
Carex Min 0.0075 0.0019 6.91 0.02 -175
Max 0.045 0.0056 7.62 3.23 -79
Hybrid bioreactor- Avg 0.023 0.0040 7.20 1.73 -150
CWTS Carex + moss Min 0.0083 0.0025 6.95 0.11 -212
Max 0.044 0.0054 7.82 5.61 -101
Avg 0.026 0.0040 7.19 1.86 -134
Carex + moss +
Min 0.0083 0.0028 6.96 0.03 -195
biochar
Max 0.064 0.0054 7.71 4.71 -30
* The average outflow for the hybrid bioreactor-CWTS period includes data from 19 June, when there was a spike in concentrations.
If this data point is not included, the average outflow concentrations are 0.015, 0.017 and 0.015 for Cu mg/L, and 0.003, 0.003 and
0.004 for Se mg/L for the Carex, Carex + moss and Carex + moss + biochar systems, respectively.
The hybrid bioreactor-CWTS configuration was tested as a contingency measure in case higher removal rates
are needed than achieved by the CWTS at a given point in time (e.g. influent water quality or outflow
objectives change). There was no significant difference between the treatment performances of the three
system designs for either copper or selenium during the hybrid bioreactor-CWTS period. In the case of
selenium, the reconfiguration from CWTS to the hybrid bioreactor-CWTS resulted in a statistically significant
decrease in outflow concentration compared to that prior to the reconfiguration, with a lowest measured
outflow concentration of 0.0019 mg/L and average of 0.004 mg/L (Figure 2; p < 0.001). This improvement in
selenium treatment was expected based on the design targeting more favourable conditions for selenium
reduction, such as lower DO and ORP resulting from the additional electron donors from microbial

decomposition of the alfalfa and straw. In regards to copper, the lowest outflow concentration of the study
was also measured during the hybrid bioreactor-CWTS phase (0.0075 mg/L); however, treatment
effectiveness decreased briefly in all systems prior to this (Figure 2, 19 June data point). Even during this brief
disruption, 86% of the total copper entering the system was removed from the water. This disruption in
treatment effectiveness coincided with low DO and moderately reducing conditions in the water column
being attained in all cells (DO < 0.5 mg/L, ORP < −80 mV, compared to > 3.5 mg/L and > +142 mV prior to
adding alfalfa and straw). It should be noted that the leach test conducted on the alfalfa and straw prior to
addition determined that there was no discernible release of copper from these organic materials. The cause
of the disruption could not be determined definitively, with the disturbance being registered during a single
sampling point and a maximum potential disturbance period of four weeks.
The tests conducted in this study demonstrate that the CWTS designed for Minto can be repurposed as a
hybrid bioreactor-CWTS resulting in improved selenium and copper treatment within the same footprint
(Figure 2), and without need for added construction. However, operating the system as a hybrid bioreactor-
CWTS would require greater operational maintenance than it would in its original CWTS configuration. While
the hybrid bioreactor-CWTS configuration improved selenium treatment and demonstrated the lowest
outflow concentration of copper of the entire trial, the short upset of treatment performance during the
transition is meaningful in terms of monitoring during site implementation. It should be emphasised that if a
transition from a typical CWTS design to a hybrid bioreactor-CWTS style is performed, careful monitoring of
the CWTS would be required, as some elements may have decreased treatment effectiveness for a brief
period. During this timeframe, if concentrations of elements exceed regulatory guidelines, water could be
recycled through the wetland, or a contingency water treatment plan could be put in place for the outflow
over the period of treatment disturbance.
While per cent removal is often the most widely discussed aspect of system treatment performance, by itself
it is not a valuable measure. There is a great difference between 99% removal with a starting value of 10,000
(leaving 100 units of a constituent), compared to 99% removal for a starting value of 10 (leaving only 0.1 units
of a constituent). In fact, the per cent removal of a constituent in a system provides almost no information
of value without the context of HRT and initial and final concentrations. As such, it is more appropriate to
discuss and evaluate treatment performance in the context of reaction kinetics. The use of two cells in each
series allowed for the opportunity to assess the amount of treatment occurring at different points in the
pilot-scale CWTSs, and thereby for the calculation of removal rate coefficients to be developed for ranges of
concentrations. As can be seen in Figure 3, most of the copper is removed in the first cell, with only a small

Case Studies
amount of treatment occurring in the second cell. Conversely, selenium treatment occurs in a more linear
fashion, through both cells in the system. This information can be used to provide more accurate sizing and
performance estimates of a full-scale CWTS.
Based on the treatment performance measured in this study, removal rate coefficients were calculated for
the long-term closure scenario with low ammonia and nitrate concentrations. The removal rate coefficient
of k = 0.01 for selenium can be applied between the concentrations of 0.002 and 0.012 mg/L, k = 0.055 for
copper between 0.036 and 0.150 mg/L, and k = 0.025 for copper between concentrations of 0.008 and 0.035
mg/L. At concentrations higher than 0.150 mg/L, a greater value of k is likely applicable. It remains to be
experimentally determined through onsite demonstration-scale testing whether it is possible to achieve even
lower outflow concentrations of selenium and copper with this wetland design. However, lower
concentrations of both elements have been achieved with CWTS designed for other sites, suggesting it is
thermodynamically possible (Haakensen et al., 2013). These removal rate coefficients can be used with
confidence at a temperature of 20°C, and could be adjusted conservatively by using the Arrhenius equation.
The Arrhenius equation is considered conservative in this circumstance because some of the rate-affecting
processes for treatment in a CWTS are sorption, filtration and precipitation, which are not impacted by
temperature in the same way as chemical and biological reactions.
While plants are often thought of as a treatment pathway through uptake, this pathway is not sustainable
for metals, as they can be released upon decomposition. However, plants can provide multiple benefits to
the function of a CWTS, and uptake of elements can be minimised through appropriate designs that target
mineralisation and decreased bioavailability. In these systems, copper treatment was highly correlated to
sulphur deposition in soils, and not correlated to iron concentrations.
Mass balances were calculated for each pilot-scale system to assess the fate and distribution of elements
removed from the water during the course of testing. For systems that were planted with both Carex and
moss, 64–75% of the copper removed from the water was sequestered into the top six centimetres of
sediment. In contrast, systems with only Carex planted and no moss had only 33% of the copper ending up
in this top sediment fraction. For systems planted with both Carex and moss, 68–83% of treated selenium
was sequestered to this top portion of the sediment, but less than 52% was in the sediment of the system
planted with only Carex (actual percentage cannot be calculated, as the selenium was below the sediment
detection level; the 52% assumes actual concentrations in the soil are equal to the detection limit). For both
elements, only 1–3% of the total load removed from the water was found in the above water vegetation
(Carex) at the end of pilot-scale testing. It was not possible to determine the concentrations of elements in
moss, as the majority did not survive in the hybrid bioreactor-CWTS conditions. As such, the majority of
copper and selenium treated by the system was sequestered to the top six centimetres of sediment, with
only a small portion being taken up into the Carex.

Although it was expected that moss would perform well in treating water through sorption, it was also
expected to generate high DO concentrations in the water (through photosynthesis), which could negatively
impact the targeted negative oxidation-reduction conditions necessary for sulphate and selenium reduction
in the systems. Furthermore, if the CWTS was built without moss, it was expected to eventually include moss,
as this was the natural state of wetlands found at Minto through the site assessments (reports 2013-0100-
256 and 2013-0100-257 on the YESAB registry). Therefore, it was necessary to test whether it was beneficial
or detrimental to have moss in the CWTS design.
All plants were harvested and transported from the Minto Mine with the above-ground biomass (leaves)
removed before transplanting to ensure that measurements were reflective of actual uptake during this
study. The Carex and moss both had a very high survival rate in the pilot-scale systems, with increase of stem
count and total mass through the entire study (Figure 4). The plants were established for six weeks after
transplantation with tap water running through the pilot-scale CWTSs before synthetic influent was started.
The maximum average stem/leaf height was reached after approximately three months. During these first
three months, the average stem/leaf height increased, while the total number of stems remained the same.
It was only after these plants achieved the maximum height that they began to send up additional stems. The
stem count was still increasing when the study ended, suggesting that maximum plant density had not yet
been achieved and would likely take more than the length of a single growing season to establish (Figure 4).
During system takedown, it was apparent that the Carex roots had spread throughout the cells, providing a
strong foundation for the plants. Moreover, many of the roots were associated with sulphate-reducing zones,
as was evident by black areas in the surrounding sediment.
It is important to the overall treatment system design to determine if plants may release elements during
freezing and thawing of the water, as would occur through natural seasonal progression. A freeze-thaw trial
was conducted in which Carex and moss were collected and sent for metals analysis both before and after
being submerged in water from the wetland and freezing to −20°C followed by thawing to room temperature.
It was found that Carex released 18–38% of accumulated selenium and 0–13% of accumulated copper based
on the decrease in tissue concentrations after freezing and thawing. In contrast, moss was found to have
more than a two-fold increase in the concentration of these elements after being frozen and thawed. It
appears that through the freeze-thaw process, the moss has an enhanced sorption of these compounds,
removing an additional amount from the water. Based on the pilot-scale system design and plant densities,
Carex would theoretically release on average 0.14 mg of copper and 0.27 mg of selenium per square metre
of wetland area, while moss would theoretically bind 511 and 11.8 mg/m2, respectively. This finding is
important, as it demonstrates that despite some release of constituents by Carex through freeze-thaw, the
moss will uptake an excess of what is released by the Carex, thereby assisting with treatment during the
initial spring thaw before biogeochemical processes are at full capacity.

Case Studies
Using genetic- and growth-based methods, bacterial communities during all CWTS pilot-scale testing phases
were compared. The soil used to construct the pilot-scale systems had low bacterial diversity (Simpson’s
reciprocal index), which increased as plants were established in the pilot systems (prior to influent running
through the systems). Similarly, the relative abundance of SRB was very low in the initial soil and increased
as plants were established in the systems. The high N and low N phases had similar bacterial diversity and an
increase in relative abundance of SRB (genetic) and total heterotrophs (growth-based) compared to before
influent was running through the systems. The hybrid bioreactor-CWTS period also showed an increase in
bacterial diversity, and the highest relative abundance of SRB. The low N and hybrid bioreactor-CWTS phases
had overall similar microbial populations, except that the hybrid bioreactor-CWTS had a higher relative
abundance of SRB. This indicates that either: (1) the microbes present in the low N phases were well suited
to the bioreactor-CWTS conditions, or (2) that the community was still in flux at the final sampling, and
treatment may have continued to improve over time in the hybrid system as more anaerobic organisms
flourished.
The results of the growth-based MPN analysis indicate an abundance of organisms capable of denitrification
was present in the high N period. Although in lower abundance, denitrifying bacteria were present through
the low N period and the hybrid bioreactor-CWTS period. The MPN for selenite-reducing bacteria ranged
from hundreds of thousands to tens of millions of organisms per gram of soil and was similar between the
different system designs and testing periods. The number of SRB steadily increased over time in all designs,
starting with a MPN of approximately 1,000 organisms per gram of sediment, which is similar to that found
associated with Carex through the site assessment (reports 2013-0100-256 and 2013-0100-257 on the YESAB
registry), and increased by two orders of magnitude through the study. This increase is consistent with trends
observed in the relative soil redox measurements of the pilot-scale systems becoming more negative over
time, and is indicative of the system becoming more established and achieving the desired reducing
conditions targeted for treating constituents of interest. Both genetic and growth-based methods found the
presence and abundance of SRB were positively correlated with final sulphur concentrations in sediment.
This shift in microbial communities (paired with the shift in population observed for denitrifying organisms)
is indicative of the responsiveness and adaptability of the CWTSs to different influent chemistries. It has
previously been shown that once established, the microbial communities in CWTSs are robust and maintain
similar community compositions even after freezing and thawing (Haakensen et al., 2013).
Several OTUs related to known SRB were found to correlate with sediment concentrations of copper (p <
0.001) and sulphur (p < 0.1), suggesting that their presence contributes to reduction of sulphate to sulphide,
which then binds to copper and precipitates into the sediment. This is supported by the appearance of black
areas surrounding the Carex roots and black precipitates on the sediment surface observed during system
takedown. This copper-sulphide mineralisation mechanism is ideal for treatment, as it produces a stable form
of copper in the sediment. We can tell from the genetic sequencing that these same types of SRB were found
in samples collected during the August 2013 Minto site visit, particularly associated with Carex roots. It is
likely that these beneficial microbes were brought to the pilot-scale system along with the Carex during
transplanting. This indicates that despite differences in sediment, water and environmental conditions
between the harvesting location and the pilot-scale CWTS, these microbes are robust and can be further
encouraged to thrive and perform beneficial functions by being provided the ecological niches that they
prefer in the context of the CWTS (e.g. Carex roots, low ORP, and low DO).
Based on data collected during pilot-phase testing, we conclude that the combination of Carex and moss is
the ideal CWTS design for long-term, sustainable copper and selenium treatment for the Minto Mine. In this
design, Carex provides stability to the wetland soil and sediment with extensive root structure, additional
treatment capacity based on the roots drawing water down into the sediment, sulphate-reducing zones and
renewal of organic carbon in the sediment. The moss provides sorption and uptake of elements even through
freeze-thaw events. Moreover, both Carex and moss are known to be accreting, peat-forming plants, which

is beneficial to soil quality in the CWTS over time. Using two cells in each series permitted removal rate
coefficients to be developed for ranges of concentrations, allowing for more accurate sizing and performance
estimates of a full-scale CWTS.
Phase 4 of the design and optimisation program is underway, with a demonstration-scale CWTS built onsite
at the Minto mine in the fall of 2014. Based on research findings presented in this paper, the demonstration-
scale CWTS is constructed as a free water surface flow CWTS planted with both Carex and moss. Biochar was
not incorporated in the soil, as the findings of the pilot-scale studies showed that it did not result in improved
treatment or greater abundance of beneficial microbes.
We would like to acknowledge Jenny Liang, Qingxiang (Mark) Yan and Emily Charry Tissier, all of Contango
Strategies, for work performed during pilot-scale testing.
Blodgett, R. (2010) US Food and Drug Administration, Bacteriological analytical manual, Appendix 2: Most probable number from
serial dilutions. http://www.fda.gov/Food/FoodScienceResearch/LaboratoryMethods/ucm109656.htm
Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Fierer, N., Peña, A.G., Goodrich, J.K., Gordon,
J.I., Huttley, G.A., Kelley, S.T., Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M.,
Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., Zaneveld, J., and Knight, R. (2010)
QIIME allows analysis of high-throughput community sequencing data, Nature Methods, Vol. 7, pp. 335–336.
DeSantis, T.Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E.L., Keller, K. Huber, T., Dalevi, D., Hu, P., and Andersen, G.L. (2006)
Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB, Applied and Environmental
Microbiology, Vol. 72, pp. 5069–5072.
Edgar, R.C. (2010) Search and clustering orders of magnitude faster than BLAST, Bioinformatics, Vol. 26, pp. 2460–2461.
Faulkner, S.P., Patrick, W.H. Jr. and Gambrell, R.P. (1989) Field techniques for measuring wetland soil parameters, Soil Science Society
of America Journal, Vol. 53, pp. 883–890.
Haakensen, M., Terry, M.W., Pittet, V., Castle, J.W. and Rodgers, J.H. Jr. (2013) Effects of freeze-thaw and biochar on sequestration
and localization of elements within oxidizing and reducing pilot constructed wetland treatment systems, SETAC North
America 34th Annual Meeting, 17–21 November 2013, Nashville, USA, Presentation 593, recording available online at
http://setac.sclivelearningcenter.com/index.aspx?PID=9484&SID=187033
Haakensen, M., Pittet, V., Spacil, M.M., Castle, J.W. and Rodgers, J.H. Jr. (2015) Key aspects for successful design and implementation
of passive water treatment systems, Journal of Environmental Solutions for Oil, Gas, and Mining, Vol. 1, pp. 59–81.
Haines, W.P. and Renwick, J.A.A. (2009) Bryophytes as food: Comparative consumption and utilization of mosses by a generalist insect
herbivore, Entomologia Experimentalis et Applicata, Vol. 133, pp. 296–306.
Hawkins, W.B., Rodgers, J.H. Jr., Gillespie, W.B. Jr., Dunn, A.W., Dorn, P.B. and Cano, M.L. (1997) Design and construction of wetlands
for aqueous transfers and transformations of selected metals, Ecotoxicology and Environmental Safety, Vol. 36, pp. 238–248.
Huddleston, G.M. III, Rodgers, J.H. Jr. (2008) Design of a constructed wetland for treatment of copper-contaminated wastewater,
Environmental Geosciences, Vol. 15, pp. 9-–9.
Kadlec, R.H. and Wallace, S.D. (2009) Treatment wetlands, 2nd edition, CRC Press Taylor & Francis Group, Boca Raton, USA.
Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M. and Glöckner, F.O. (2013) Evaluation of general 16S ribosomal
RNA gene PCR primers for classical and next-generation sequencing-based diversity studies, Nucleic Acids Research, Vol. 41,
pp. e1.
Lefrançois, E., Quoreshi, A., Khasa, D., Fung, M., Whyte, L.G., Roy, S. and Greer, C.W. (2010) Field performance of alder-Frankia
symbionts for the reclamation of oil sands sites, Applied Soil Ecology, Vol. 46, pp. 183–191.
Longton, R.E. (1997) The role of bryophytes and lichens in polar ecosystems, in Ecology of arctic environments: 13th special symposium
of the British Ecological Society, 1st Edition, Blackwell Science Ltd., Oxford.
Lozupone, C., Lladser, M.E., Knights, D., Stombaugh, J. and Knight, R. (2011) UniFrac: An effective distance metric for microbial
community comparison, ISME Journal, Vol. 5, pp. 169–172.
McDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A., Andersen, G.L., Knight, R., and Hugenholtz, P. (2012)
An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea, ISME
Journal, Vol. 6, pp. 610–618.
McMurdie, P.J. and Holmes, S. (2013) phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census
data, PLoS ONE, Vol. 8, p. e61217.
Murray-Gulde, C.L., Bridges, W.C. and Rodgers, J.H. Jr. (2008) Evaluating performance of a constructed wetland treatment system
designed to decrease bioavailable copper in a waste stream, Environmental Geosciences, Vol. 15, pp. 21–38.
Rodgers, J.H. Jr. and Castle, J.W. (2008) Constructed wetland systems for efficient and effective treatment of contaminated waters
for reuse, Environmental Geosciences, Vol. 15, pp. 1–8.

Case Studies
Siddique, T., Zhang, Y., Okeke, B.C. and Frankenberger, W.T. Jr. (2006) Characterization of sediment bacteria involved in selenium
reduction, Bioresource Technology, Vol. 97, pp. 1041–1049.
Suren, A.M. and Winterbourn, M.J. (1991) Consumption of aquatic bryophytes by alpine stream invertebrates in New Zealand, New
Zealand Journal of Marine and Freshwater Research, Vol. 25, pp. 331–343.
Taylor, C.R. (2009) Selecting plant species for optimize wastewater treatment in constructed wetlands, MSc Thesis, Montana State
University, Bozeman, USA.


M. Venhuis EcoMetrix Incorporated, Canada

S. Shaw EcoMetrix Incorporated, Canada
R.V. Nicholson EcoMetrix Incorporated, Canada
S. Bartle Consultant, Mining Engineer, Canada
J. Stirling ArcelorMittal Dofasco, Canada
The Sherman iron mine, in Ontario Canada, began operation in 1968 and produced 22 Mt of iron pellets before
closing in 1990. The banded iron deposit yielded magnetite ore and had associated carbonate and sulphide
facies. At the start of the operation, some sulphide mine rock was used for road fill and some was stockpiled
before the potential risks of acid generation with high-sulphide rock were recognised. After recognising the
acid generation potential in 1977, the mine operators relocated sulphide rock from roads, some stockpiles
and developing pits and deposited it in a 340 ha tailings impoundment. At closure, the five pits were allowed
to fill with water, and most developed outflow within about five years.
The closure plan requires the water leaving the site to meet the provincial water quality objectives (PWQO)
for Ontario for several metals and pH. At closure, the off-site flows were close to but typically exceeded the
PWQO values for metals, but they have always achieved the pH objective between 6.5 and 8.5. Water quality
continues to improve over time, and the off-site flows have met the PWQO objectives since 1995. Acid base
accounting (ABA) results suggest that there is sufficient neutralisation potential in the vast majority of mine
rock to prevent acid generation. However, there are some features such as a road embankment and a small
rock stockpile that exhibit acidic drainage conditions including low pH and elevated iron concentrations. The
local watersheds have, however, assimilated the residual acid drainages and have continued to meet PWQO
values to date in the off-site flows. This paper outlines a detailed investigation undertaken to assess the state
of the environment with a focus on the small areas with residual acid and metal sources and the measurable
effects on the downstream aquatic environment. Mine rock was characterised in detail to verify geochemical
stability of those materials that have remained neutral over the 25 years since closure.
Planning for closure for new mine projects involving sulphide minerals requires successful identification and
mitigation of potentially problematic mine materials that will be permanently stored on-site after mine
operations cease. Much effort is typically taken to predict what materials may be potentially acid generating
(PAG) and/or metal leaching (ML) and therefore may require special management to mitigate the effects of
such processes on water quality over the long term. The science of predicting mine material behaviour and
water quality effects and the management of potentially problematic mine materials has improved over the
past few decades (Price, 1997; Price, 2009; INAP, 2014). Yet there is strong evidence that our collective
predictions regarding PAG materials and mine-influenced water quality have been less than perfect (Kuipers
et al., 2006). One of the weaknesses identified in the review of prediction performance is the lack of follow-
up and performance monitoring to provide valuable feedback for prediction and mine material management
programs; this feedback can be used to modify and refine our hypotheses, methods and interpretations in
this evolving science.

Evaluation of acid generation in mine rock at an iron mine 25 years after closure M. Venhuis, S. Shaw, R.V. Nicholson, S. Bartle and J. Stirling
There are abundant examples of failed predictions or inappropriate selection of management methods for
potentially problematic mine materials. Some of these examples are related to mines that operated at times
when issues of sulphide mineral weathering and acid production were not well understood. Others have
developed for various reasons, ranging from optimistic predictions of material behaviour or mitigation
performance to inappropriate allocation of resources to mine material management (Robertson, 2011).
Some historic mines with acid generation issues represent substantial liabilities for current and future
generations of taxpayers. One example is the Faro zinc-lead mine in Yukon, Canada, where the owner became
insolvent. However, other historic mines with a history of acid issues have been rehabilitated ad hoc by the
mine operators. This was the case for the uranium mines in the Elliot Lake region of Canada (Nicholson et al.,
2012); these mines represent success stories for environmental stewardship. Whether historic mines with
sulphide materials were failures or successes, they provide opportunities to learn and to improve the science
of mine material characterisation and management to improve our potential for environmental protection
with future mining projects.
The magnitude of the acid drainage issue associated with metal mines was not even suspected in Canada
until about 1968 (Hawley, 1977). While Hawley was an early crusader for the control of acid drainage in
Ontario, Canada, his focus was almost entirely limited to mill effluents and tailings because the majority of
observations of acid drainage were associated with underground operations that had little mine rock to
manage on surface.
The former Sherman iron mine near Temagami, Ontario, operated from 1968 to 1990, a period when acid
generation related to sulphide mine materials was emerging for the metal mine industry in Canada. The
Sherman open pits had banded iron deposits that contain three main facies including oxides (hematite and
magnetite), carbonates (calcite and ankerite) and sulphides (pyrite and marcasite). No predictive studies for
acid drainage had been completed for the site prior to closure.
After arriving in 1977, the mine manager recognised the potential effects of the high-sulphide mine rock and
selectively segregated that material for storage in the tailings facility. At closure, a sulphide-rich zone above
the final flood level in one pit was removed by blasting and relocating the material to the pit for permanent
underwater storage. At the time, these mine waste management methods for environmental control were
unprecedented. Nonetheless, there were sulphide mine rock materials that were not segregated prior to
1977 and that either were used for road building or made their way into rock stockpiles, resulting in local
areas of acid drainage on-site. Although the quality of water on and flowing off the site has been either stable
or improving since closure and meets the mine closure permit requirements, the provincial regulator
(Ministry of Northern Development and Mines) requested in 2013 that the mine site be evaluated. The
assessment of the site provided an opportunity to investigate current conditions using up-to-date methods
and interpretations compared against environmental performance since closure in 1990.
ArcelorMittal Dofasco is currently responsible for the former Sherman mine property, which is located near
the town of Temagami, 100 km north of North Bay, Ontario, Canada. The Sherman mine operated between
1968 and 1990, when it was closed. The mine was a source of iron ore, which was extracted from open pits
from a typical Algoma-type banded iron deposit of Precambrian age. The ore consisted of magnetite, which
was formed into pellets and shipped to Hamilton for Dofasco’s steelmaking operation.

Case Studies
The ore was extracted from open pits including the south (the first pit to be developed in 1968), north, west,
east and Turtle pits. The ore was milled on-site, and the tailings were discharged to a tailings facility to the
north of the mill and northeast of O’Conner Lake (Figure 1). The tailings consist of finely-ground chert
containing about 8% to 10% iron and other gangue minerals such as carbonate minerals and trace sulphides,
primarily pyrite and marcasite. The tailings basin covers approximately 340 ha. Most mine rock was deposited
adjacent to the pits. Some rock was used for roadbed construction. An estimated 73.2 Mt of mine rock is
located in 13 distinct stockpiles and as fill in several areas (Figure 2). Rock that was identified after 1977 as
high sulphide and acid generating was segregated, relocated and deposited within saturated tailings or below
water in the flooded south pit.
Although the iron deposits primarily consisted of magnetic iron oxide (magnetite) and carbonate minerals,
there are small zones of sulphide minerals that occurred in bands adjacent the ore. One band on the north
side of the south pit was mined and included with the preproduction stripping of this area to open up the
ore zone and supply the necessary fill for early mine development. At that time, the acid-generating potential
of these sulphides was not recognised. Some of this mixture of high-sulphide and barren rock was used to
construct a service road from the west end of the south pit to the discharge of Lake Tetapaga, where a control
structure was built. In the mid-1970s, it was recognised that the sulphides in this road were degrading the
quality of the water in the Tetapaga River. The road material was reclaimed in 1977 and buried in the tailings
basin. Water quality in the Tetapaga River subsequently improved and returned to previous conditions.

The south pit haulage road at the east end of Lake Tetapaga (east embankment, Figure 2) also received some
of the high-sulphide material. Seepage from the east embankment exhibits low pH along a 100 m length of
shoreline at the toe of the embankment. A stockpile of mine rock from the early mining of the south pit,
located at the south end of the Lake Tetapaga Causeway (SP-NE) also exhibits acidic seepage. After 1978,
sulphide materials from the south and east pits were identified, segregated and hauled to the tailings basin
for burial. The water within the south pit was also recognised to be acidic in the early 1970s, and mitigation
measures were developed to prevent acid generation after the pit had been mined out and flooding to a
natural level had occurred. After the south pit had flooded to overflow levels, it was noted that some sulphide
material remained exposed in the pit walls above the natural water level. The operators blasted the rock
where sulphide was exposed to lower the top level of sulphide material in the pit walls. In addition, a three-
metre-high semipermeable berm was constructed at the end outflow notch of the pit in order to raise the
natural water level in the pit sufficiently to submerge all of the sulphide band in the south pit walls. These
mitigation measures in the early 1990s were unprecedented and were performed in the absence of
regulatory input.
The mine site is located on a major watershed divide, with flow from the west watershed draining through
Tetapaga Lake, Lake Temagami and into Lake Huron in the Great Lakes watershed (Figure 3). The drainage
from the east watershed flows into Net Lake and eventually into the Ottawa River and the St. Lawrence River
system. As part of the closure plan commitments in 1990, provincial water quality objectives (PWQO) were
stipulated as the maximum targeted values for the outlets of Lake Tetapaga (STN#2) and Link Lake (STN#27),
which represent outflows from the site from the east and west watersheds, respectively. In addition, mine
effluent guidelines (MEGs) as listed in the Guidelines for Environmental Control in the Ontario Mineral
Industry (Ontario Ministry of Environment, 1981) were specified as maximum target values at four on-site
stations, including Vermillion Creek (STN#9), the east arm of Lake Tetapaga at the causeway (STN#12) and
the outlet of Turtle Lake (STN#6). The appropriate PWQO and MEG values are provided in Table 1. The
constituents of potential concern (COPCs) for the Sherman mine site include aluminum, copper, iron, nickel
and zinc. Key chemical indicator parameters, such as pH, sulphate and alkalinity have also been analysed.

Case Studies
The water leaving the site has been monitored since mine closure in 1990, and COPC concentrations are
shown to be either stable or declining with time. Metal concentrations leaving the site in both catchments
(monitoring stations #2and #27) have been at or below PWQO values for the last several years, as shown in
Figure 4.
PWQO MEG
Parameter
mg/L mg/L
pH 6.5 – 8.5 5.5 – 10.5
Alkalinity --- ---
Acidity --- ---
Sulphate --- ---
Total aluminum 0.075 1.0
Total copper 0.005 1.0
Total iron 0.3 1.0
Total nickel 0.025 1.0
Total zinc 0.02 1.0
Water quality on-site has shown consistent improvements since the 1990s and currently meets MEG criteria.
Concentrations at the monitoring stations #6, #9 and #12 have generally been at or below PWQO values for
the past several years. The discharges from the west and south pits, specifically, have shown improvements
that are consistent with mitigation of sulphide oxidation by flooding, and these are not negatively affecting
downstream water quality (Figure 5). The improving quality of pit waters also provides a strong indicator of
mine rock behaviour. The limited catchments areas of the west and south pits also include drainages from

several mine rock stockpiles. This means that mine rock drainage has not negatively affected pit water quality
in the past and is not doing so now.
Water quality generally improves within each watershed as it flows downstream on-site and eventually off-
site. For example, concentrations of sulphate decrease from the south pit overflow to Turtle Creek (STN#6)
and to Johnny Creek (STN#27), or from the west pit overflow to Vermillion Creek (STN#9) and to Tetapaga
River (STN#2), as depicted in Figure 6.

Case Studies
The potential residual effects of local acid generation on-site at the east embankment and at the SP-NE
stockpile do not appear to be adversely affecting the off-site water quality. The limited acidic drainage from
the east embankment is reflected in the water quality at STN#12, which represents water in the east arm of
Lake Tetapaga before it flows through the causeway to the west arm and out through the Tetapaga River
(STN#2). While the pH remains neutral, the sulphate concentrations appear to be at steady state
concentrations of several hundreds of mg/L, suggesting ongoing inputs of sulphide mineral oxidation
products with subsequent neutralisation (Figure 7). The copper concentrations are representative of other

metal concentrations at that location that exhibit either stable or decreasing values over time, suggesting
that the ongoing acid generation at the east embankment has not led to deteriorating water quality in the
east arm of Lake Tetapaga.
Some of the acidic drainage from the SP-NE stockpile reports to the east arm immediately east of the
causeway. However, the majority of acidic drainage from that rock appears to report to the west arm
immediately west of the causeway. All of the acidic drainage entering the east and west arms of Lake
Tetapaga is included in the outflow of the west arm at STN#2. As shown in Figure 4, the water quality at
STN#2 meets PWQO values and continues to improve over time. This implies that the acid generation
intensity at both the east embankment and the SP-NE rock pile may be declining with time.
As part of an investigation of acid generation at the mine site in 2014, over 500 individual mine rock samples
were collected from the 13 mine rock stockpiles, the east embankment and the causeway. These samples
were recovered from 48 boreholes and 42 test pits. However, only a subset of 199 samples was submitted
for analysis.
The mine rock contains sulphide minerals, mainly pyrite, with sulphide-sulphur contents ranging from 0.01%
to 5% S and a geometric mean value of 0.24% S. The sulphide-sulphur content represents approximately 88%
of the total sulphur in the mine rock, and 12% of the total sulphur is present as sulphate, which is consistent
with weathering for three decades or more. Carbonate minerals within the mine rock from the west and
north pits were primarily composed of calcite and dolomite, with siderite and ankerite making up a
substantial portion of the carbonates in the waste from the south, east and Turtle pits. Overall, carbonate
contents ranged from 0.06% to 18% CO3 with a geometric mean value of 2.2% CO3. The acid base accounting
(ABA) characterisation included carbonate analysis on all samples and modified Sobek test (Lawrence and
Wang, 1996) with siderite correction (Skousen et al., 1997) on a large subset of rock samples. Comparison
between the calculated carbonate neutralisation potential (Carb-NP) and the Sobek-NP results gave a strong

Case Studies
correlation coefficient (R2 = 0.98) with slightly smaller Carb-NP values than Sobek-NP values below levels of
100 kg CaCO3/t. The geometric mean Carb-NPR values for all mine rock stockpiles ranged from 3.0 to 63
(Table 2); this suggests non-PAG material, with the exception of the EP-NE stockpile, which had a geometric
mean Carb-NPR value of about 1.7, which may be characterised as PAG.
Number of
Mine rock Number of Geometric
Minimum Maximum samples with
stockpile samples mean
Carb-NPR < 1.0
NP-N 3 12.8 4.0 27.8 0
EP-N 6 2.9 0.3 8.2 1
EP-NE 6 1.7 0.5 7.0 1
SP-NE 21 7.8 0.6 56.3 2
SP-SW 15 8.3 0.7 30.6 2
TP-N 2 11.7 11.3 12.2 0
TP-NE 4 20.2 12.8 55.7 0
TP-E 7 10.9 2.5 119.4 0
WP-N 21 7.6 0.02 7.6 2
WP-NE 6 6.0 0.7 18.6 1
WP-S 16 30.3 2.1 733.3 0
WP-SE 8 6.4 0.4 188.8 3
WP-SW 15 62.9 6.7 568.0 0
East embankment 50 7.6 0.1 358.9 7
Causeway 12 4.7 0.3 26.2 1
Total 192 9.8 0.02 733 20
On average, the mine rock comprising all the stockpiles at the site has excess NP compared to AP. Therefore,
as a whole, one would not expect any stockpile to be net acid producing. However, acidic seepage is occurring
from the SP-NE stockpile near the causeway across Lake Tetapaga and from the east embankment. A review
of the historic records suggests that net acid production has occurred at those locations for at least the past
two decades. The seepage is characterised by pH values of less than 3.0, which indicates that the ABA results
for these stockpiles are not providing a complete picture of the conditions for acid generation.
Results of quantitative evaluation of materials by scanning electron microscopy (QEMSCAN) analyses suggest
that material from the south pit is characterised by discreet banded layering of the sulphide and carbonate
phases. This layering was not seen in the mine rock from the northern geological strike zone encompassing
the west and north pits, which had the sulphide and carbonate phases well disseminated through the rock
(Figure 8). The banded layering could result in sulphide oxidation greatly exceeding carbonate neutralisation
as the carbonate minerals would be largely inaccessible because they would be locked into the matrix of the
rock fragments.

To further evaluate the potential for acid generation in the mine rock, the rate of depletion of the available
sulphide and carbonate content was estimated for each of the mine rock stockpiles based on calculated
loading rates and pore water concentrations. The depletion time calculations represent a more realistic
assessment of the potential for the mine rock stockpiles to produce net acidic conditions, as they incorporate
observed field conditions represented by soluble contents of oxidation and neutralisation products rather
than the bulk material properties used for ABA.
Since the mine rock stockpiles at the site have been in place for several decades, the rates of sulphide
oxidation and carbonate consumption (mg/kg/a) are likely to be near or at steady state. Soluble sulphate
concentrations on the rock particle surfaces estimated from shake flask extraction (SFE) tests (mg/kg) were
used to calculate the pore water concentrations with measured moisture contents of the rock samples (Table
3). The SFE tests were performed on test pit samples with no processing in order to preserve the integrity of
the dissolved constituents in the pore water or moisture in the rock. Similarly, calculated calcium and
magnesium concentrations are an indicator of the rate of carbonate dissolution, which results from acid
neutralisation reactions. The sulphate loading rate is the amount of sulphate produced from sulphide
oxidation and transported or loaded to the aqueous phase each year, and it can be calculated per tonne or
per cubic metre of mine rock. A bulk density of 2,000 kg/m3 was assumed. An infiltration rate of 0.51 m
(510 L/m2/a), or 60% of total precipitation, was estimated for each stockpile. The calculated loading rates
(mg/kg/a) were based on the reaction with one cubic metre of rock and are provided for each stockpile in
Table 3.

Case Studies
Geometric mean pore water

Loading rate
Mine rock concentrations
(mg/kg/a)
stockpile (mg/L)
Sulphate Calcium Magnesium Sulphate Calcium Magnesium
East embankment 1,391 491 136 355 125 35
NP-N 446 176 42 114 45 11
EP-N 364 109 89 93 28 23
EP-NE 559 142 135 143 36 34
SP-NE 515 226 74 131 58 19
SP-SW 554 239 58 141 61 15
WP-N 895 279 96 228 71 24
WP-NE 719 267 71 183 68 18
WP-S 358 118 79 91 30 20
WP-SE 528 120 119 135 31 30
WP-SW 193 112 28 49 28 7
TP-N 1,403 303 124 358 77 32
TP-NE 135 78 42 34 20 11
TP-E 843 250 111 215 64 28
The depletion rate of sulphide in mg S/kg/a was then calculated from the sulphate loading rate (mg SO4/kg/a).
Similarly, the depletion of carbonate (as dolomite) (mg CaMg(CO3)2/kg/a) was calculated by converting the
combined calcium (mg Ca/kg/a) and magnesium (mg Mg/kg/a) loading rates. The times to deplete the
available sulphide and carbonate were then determined by dividing the contents in mg/kg by the depletion
rates in mg/kg/a (Table 4).
The total amounts of sulphide and carbonate present in each cubic metre of mine rock was based on the
geometric mean contents. However, a portion of the sulphide and carbonate content will be locked in the
larger matrix of the mine rock and will likely not be available for reaction. Therefore, it was assumed that
only 50% of the total sulphide and carbonate contents were available in each stockpile.
For all stockpiles, the estimated carbonate depletion times (Table 4) are significantly higher than the
corresponding sulphide depletion times when geometric mean pore water and solids content values are
selected. These results correspond to the conclusions drawn from the ABA testing results, which indicated
that, on average, all the mine rock stockpiles had more than sufficient capacity to neutralise all of the acid
produced from sulphide. However, similar to the conclusions reached using the ABA results, the depletion
times calculated using geometric mean solids contents and calculated pore water concentrations do not
account for the observed acidic drainage in the east embankment and SP-NE stockpile.
Additional investigations are underway to determine the differences in stockpile behaviour and the reasons
why some stockpiles are exhibiting acidic conditions even though overall ABA conditions suggest that
sufficient neutralisation potential exists within the stockpile.

Sulphide Carbonate
Depletion rate Sulphide Carbonate
Mine rock depletion depletion
(mg/kg/a) content content
stockpile times times
Sulphide Carbonate (%S) (a) (%CO3) (a)
East embankment 118 839 0.13 6 1.99 36
NP-N 38 289 0.22 29 5.23 278
EP-N 31 299 0.21 34 1.15 59
EP-NE 48 428 0.23 24 0.74 27
SP-NE 44 408 0.11 12 1.56 59
SP-SW 47 394 0.14 15 2.16 84
WP-N 76 514 0.21 14 3.04 91
WP-NE 61 451 0.24 20 2.74 93
WP-S 30 292 0.09 15 5.36 282
WP-SE 45 372 0.12 14 1.50 62
WP-SW 16 185 0.05 15 5.82 484
TP-N 119 595 0.14 6 3.07 79
TP-NE 11 172 0.09 39 3.39 302
TP-E 72 509 0.08 6 1.73 52
Overall, mine rock at the former Sherman mine is considered non-acid generating. Although results of ABA
testing indicate that the mine rock is non-PAG, there are some features such as road embankments and a
small stockpile of mine rock that contain acid-generating rock that exhibits low pH conditions in local
drainages. Although the reasons for this are not fully known at this time, preliminary results suggest that
mineralogy could play a role. Additional studies are being undertaken to further investigate the differences
between stockpile behaviours in relation to acid generation. Nonetheless, both site watersheds have
assimilated the residual acid drainages and have continued to meet PWQO values in the off-site flows. The
lack of significant ARD issues at the mine site is due to the recognition of the potential issues related to the
presence of sulphide-bearing rock and forward thinking mitigation measures, which were both employed at
a time prior to these methods becoming industry standards. These practices, which were unprecedented for
environmental control, have resulted in water quality leaving the site at concentrations generally well below
closure plan objectives and criteria for a number of years.
Hawley, J.R. (1977) The use characteristics and toxicity of mine-mill reagents in the province of Ontario, Ontario Ministry of
Environment, Toronto, Ontario, Canada.
International Network for Acid Prevention (INAP) (2014) Global acid rock drainage guide (GARD Guide), viewed 23 February 2015,
http://www.gardguide.com.
Kuipers, J.R., Maest, A.S., MacHardy, K.A. and Lawson, G. (2006) Comparison of predicted and actual water quality at hardrock mines:
the reliability of predictions in environmental impact statements. Kuipers & Associates and Buka Environmental, Butte, MT,
U.S.A., 228 p.
Lawrence, R.W. and Wang, Y. (1996) Determination of neutralization potential for acid rock drainage prediction, MEND report 1.16.3,
Mine Environment Neutral Drainage (MEND), Ottawa, ON, 149 p.

Case Studies
Nicholson, R.V., Ludgate, I., Clyde, E. and Venhuis, M. (2012) The successful reclamation of acid generating tailings in the Elliot Lake
uranium district of Canada: a good-news case history, in Proceedings 9th International Conference on Acid Rock Drainage
(ICARD 2012), 20–26 May 2012, Ottawa, Ontario, W.A. Price, C. Hogan and G. Tremblay (eds), Curran Associates, Inc., Red
Hook, NY.
Ontario Ministry of the Environment (1981) Guidelines for environmental control in the Ontario mineral industry, Ontario Ministry
of the Environment.
Price, W.A. (1997) Draft guidelines and recommended methods for the prediction of metal leaching and acid rock drainage at
minesites in British Columbia, British Columbia Ministry of Employment and Investment, Energy and Metals Division.
Price, W.A. (2009) Prediction manual for drainage chemistry from sulphidic geologic materials, Canadian MEND report 1.20.1, Natural
Resources Canada.
Robertson, A.M. (2011) Mine waste management in the 21st century: challenges & solutions beyond incremental changes, in
Proceedings Tailings and Mine Waste 2011, 6–9 November 2011, Vancouver, B.C.
Skousen, J.G., Renton, J., Brown, H., Evans, P., Leavitt, B., Brady, K., Cohen, L. and Ziemkiewicz, P. (1997) Neutralization potential of
overburden samples containing siderite, Journal of Environmental Quality, Vol. 26(3), pp. 673–681.


P. Littlejohn BioteQ Environmental Technologies, Canada

D. Kratochvil BioteQ Environmental Technologies, Canada
A. Consigny BioteQ Environmental Technologies, Canada
An important aspect of mine closure is the management of waste residues such as waste rock, tailings, water
treatment by-products and other metallurgical process waste. Issues with waste residues include their mass
and volume, long-term stability and environmental impact, the methods required for safe disposal and the
metal value that may be recovered through reprocessing or offtake. Well-planned management of residues
is required not only at sites that are currently in production but also at legacy sites and for projects that are
in the permitting and planning stages. In order to navigate the challenges presented by management of
residues, many industrial players are employing novel technologies to achieve sustainability. This paper
examines three case studies involving the use of new or unconventional technology to improve sustainability
and reduce the cost of residue management at different stages of mine life. One case study involves recovery
of copper value from low-grade stockpile decades after active mining ceased. The second details a new arsenic
management plan at an operating copper smelter, which focuses on dramatically reducing the volume of
arsenic waste. The third describes a novel approach to environmental control to reduce selenium discharge
to ultralow limits.
A wide variety of waste residues and by-products are generated during the operation of a mine. The
properties of these materials change depending on the mining methods and mineral processing techniques
used, environmental treatment requirements and the stage of the mine life. Residues that result from mining
water treatment also require management. The character of water treatment by-products and the strategies
used to manage them depend on the chemistry of the water treated, the treatment objective and the
technology used to achieve that objective. As water treatment is often required for many years of a project’s
life and even into closure, residue management becomes an important long-term liability for mine owners.
Accurately accounting for the cost of residue management thus becomes important for determining not just
the up-front capital cost of water treatment but also the true life-cycle cost of different treatment methods.
Viewed through this lens, cost-effective residue management involves reducing and, if possible, eliminating
waste volume, improving long-term waste stability and minimising losses of value to waste. This paper
describes three case studies where water treatment residue management is a key driver in the project. In
these projects, novel technology is applied to manage waste sustainably.
As much as 20% of total global copper production comes from active or passive heap leaching of copper from
oxide and low-grade sulphide ores using solvent extraction (SX) combined with electrowinning (EW). The
process economics of SX-EW at most mine sites dictates that the concentration of copper in the leach solution
must be maintained at > 200 mg/L. During the active life of a mine, this is easily accomplished by stacking
fresh ore on top of, or next to, the old ore. However, at the end of the mine life when no additional ore is

Using novel technology for residue management and sustainable mine closure P. Littlejohn, D. Kratochvil and A. Consigny
available, copper concentration drops below the SX-EW plant target. At this point, the plant shuts down and
the mine operators face significant costs associated with decommissioning and closing the heap leach
operation. Although the large majority of copper has been leached and recovered prior to closure, the value
of residual recoverable copper remaining in heaps at the point of SX-EW plant closure is often very significant
relative to the overall heap leaching closure cost.
New copper recovery processes such as BioteQ’s BioSulphide® process can enable mine operators to cost-
effectively recover most of the residual copper value remaining in heaps that have reached the end of their
life. This not only reduces the inventory of potentially mobile copper that may leach into the environment
after closure, but also makes the proceeds from the sale of the recovered copper available to fund closure
costs. The present case study represents an example of this approach.
This case involves a site that was one of America’s largest copper producers for much of the twentieth century
before operations were halted in the 1970s. After active operations concluded, a low-grade stockpile
continued to leach copper, but the grade of the leach solution declined below the level required for SX-EW
plants to operate economically. In 2004, BioteQ installed a copper recovery plant using the BioSulphide
process, which allowed a cost-effective copper recovery to continue for an additional nine years.
The concept of metal recovery that is applied here to the closure of a heap leach operation may apply equally
to sites with acid rock drainage, where the mass of valuable metals present in contaminated wastewater
represents an opportunity to offset the long-term water treatment and/or site closure costs.
The BioSulphide process involves two separate processes: generating hydrogen sulphide gas in a bioreactor
and using biogenic hydrogen sulphide gas to precipitate metals from water. In the bioreactor, bacteria are
fed elemental sulphur and acetic acid under controlled conditions to generate a rich stream of hydrogen
sulphide gas, as per equation (1). The bioreactor where this chemistry takes place is completely isolated from
the stream to be treated, simplifying process control and flexibility.
4S0(s) + CH3COOH(aq) + 2H2O(l) → 4H2S(g) + 2CO2(g) (1)
In the following stage of the process, the biogenic hydrogen sulphide gas is brought into contact with metal-
bearing streams in a gas-liquid contactor. The design of the contactor vessel and the tight automatic control
process conditions inside the contactor allow a selective precipitation of targeted metals of value or toxic
metals that require removal from the metal-bearing stream. In the present case, copper is the metal of
interest. It precipitates according to the reaction chemistry shown in equation (2).
Cu2+(aq) + H2S(g) → CuS2(s) + 2H+(aq) (2)
The resultant copper sulphide slurry is clarified to produce treated water that is discharged from the plant
and a copper sulphide solid that is dewatered in a filter press. A schematic of the BioSulphide process is
shown in Figure 1.

Case Studies
The BioSulphide facility was commissioned in 2004. The plant was designed to produce up to 3.8 tonnes per
day of sulphide (as hydrogen sulphide gas) with a treatment rate of up to 12,000 m3/day of leach solution.
The design rate of copper recovery from the leach solution was 99%, and the treated solution was recycled
to the top of the stockpile.
The following figures show results from the plant over a decade of treatment. During this time, the volume
of water treated and the concentration of copper in the water varied depending on seasonal rainfall and
mine remediation activities. However, as Figure 3 shows, the process proved to be robust and highly
adaptable to the variations in feed concentration and flow rate. This is partly due to the flexible design of the
plant, but the inherent rapid reaction of copper with sulphide also contributes to the high process reliability.
Throughout its operational history, the plant consistently removed more than 99% of the feed copper and
recovered the metal as 40%–45% metal-grade concentrate appropriate for sale.
Feed to BioSulphide® Plant

August 2004 to August 2013
12,000
10,000
8,000
m3
6,000
4,000
2,000
Copper Concentrations
August 2004 to August 2013
500
400
300
mg/L
200
100
0
Cu in Influent Cu in Effluent
The operating costs of this process consist of both fixed costs for operator salaries and overhead and variable
costs for reagents and power. These variable costs are proportional to the mass load of copper in the stream.
The cost of power and reagents for this process is approximately US$ 1.76/kg of produced copper. When the
flow rate and metal concentration of the runoff were high, fixed costs represented a specific cost of
US$ 0.88/kg of copper. However, as Figures 2 and 3 show, both the flow rate and the metal concentration of
the stream declined. As the metal load diminished, the fixed costs of the plant became a more significant

part of the overall operating cost. At the time of plant closure, the mass of copper recovered dropped to
approximately 204,000 kg per year, driving the total cost of production to US$ 5.07/kg. At this point, the
incremental value of copper recovered was deemed too low compared to the operating cost of the plant.
The operation of the BioSulphide plant at this legacy site in the American Southwest allowed for the recovery
of a significant amount of additional revenue from a closed heap leach pad. This represented not only a
recovery of metal value but also the conversion of an environmental liability to a revenue stream. This
approach contrasts sharply with conventional lime neutralisation. If lime had been used at this site, the
following chemistry would apply:
2H2O(l) + CuSO4(aq) + Ca(OH)2(s) → Cu(OH)2(s) + CaSO4.2H2O(s) (3)
However, this approach would have generated a significant volume of gypsum and copper hydroxide residue,
which would require disposal. This residue material would not only represent a waste of contained metal
value, it would also require long-term management to prevent the precipitated heavy metals from
remobilising. Taking an average flow of 6,000 m3/d at 250 mg/L copper content, lime neutralisation at this
site would have produced approximately 32 tonnes per day of 20 weight % slurry. This would have amounted
to a total of 110,000 m3 of water treatment residue over the life of the plant. Instead, the copper sulphide
produced by the site was sold to a local copper smelter, generating approximately US$ 9 M in revenue over
its decade of operation. This revenue was able to offset the cost of ongoing remediation and water
treatment. Additionally, because the copper sulphide solid was sold to an off-site smelter, there was no
additional cost or associated long-term liability for residue disposal.
Historically, copper producers have avoided deposits with elevated levels of arsenic. However, as copper
deposits with low arsenic content are depleted and the global demand for copper continues to increase,
major copper producers have been forced to start extracting copper from arsenic-rich ore bodies. As a result
of the close association of copper with arsenic in sulphide minerals separated from tailings by flotation, a
significant portion of the arsenic present in these ore bodies ends up reporting to copper smelters.
Consequently, arsenic management and the costs associated with arsenic disposal at copper smelters have
become noteworthy challenges facing copper producers worldwide. When sulphide minerals containing
arsenic are smelted, arsenic is volatilised with the SOx-laden off-gas. This off-gas must be scrubbed prior to
discharge. While some arsenic reports to dusts that are removed in high-temperature particulate scrubbers
such as electrostatic precipitators, a significant portion of the total arsenic entering smelting operations
reports to water-based scrubbers that produce liquid blow-down that is highly acidic and rich in dissolved
arsenic. In South America, this blow-down is referred to as Efluente Planta Acido Sulfurico (EPAS). The life-
cycle cost of the treatment of EPAS, which includes the cost of disposal of waste residue of treatment,
represents a major environmental treatment challenge. An example composition of EPAS is shown in Table 1.
Parameter Value
pH 0.3 to 0.8
Free acidity as H2SO4 (mg/L) > 70,000
Copper (mg/L) 300 to 1,500
Arsenic (mg/L) 4,000 to 9,000
Other metals of value present Re, Mo

Case Studies
In the present case study, the project site is a copper smelter in Chile. The site has been in operation for
decades and has seen a gradual upward trend of arsenic concentration in the copper concentrate it treats.
This trend is expected to continue. Currently, arsenic in the EPAS solution is treated through lime
neutralisation of the EPAS liquor with the addition of ferric iron in aerated tanks and a long retention time.
Introducing lime into the stream neutralises the sulphuric acid to produce gypsum; air sparged into the lime
reactor tanks oxidises As(III) to As(V), which readily precipitates with ferric iron, forming a mixture of ferric
arsenate and ferric hydroxide with arsenic adsorbed onto the ferric hydroxide solids. Calcium arsenate is also
produced in this process. The salient reactions are shown in equations (4) through (7).
Ca(OH)2(s) + H2SO4(aq) → CaSO4.2H2O(s) (4)
HAsO2(aq) + 0.5O2(g) + H2O(l) → H3AsO4(aq) (5)
6Ca(OH)2(s) + 2H3AsO4(aq) → 2Ca3(AsO4)2(s) + 6H2O(l) (6)
Fe2(SO4)3(aq) + 2H3AsO4(aq) → 2FeAsO4(s) + 3H2SO4(aq) (7)
Major issues associated with the current method of treatment make it unsustainable in the long term:
 Very high mass and volume of arsenic laden waste: Depending on the stream composition, the
waste solids production rate can vary from 100 to 250 kilograms of dry solids per cubic meter of
EPAS treated. The waste residue contains less than 5% As on a dry-weight basis but is considered
hazardous waste. At the site in question, this material is disposed of off-site, and the transport
and disposal costs associated with this are prohibitive.
 High life-cycle costs: The reagents and power costs are high primarily because the oxidation of
As(III) to As(V) is a kinetically slow process. The consumption of lime and ferric iron reagents is
very high, and the solid residue disposal is expensive.
 The value of acid contained in EPAS and metals that can potentially be recovered is lost by lime
neutralisation and precipitation of all metals into the waste solids residue.
 Low water recovery: Due to the poor dewatering properties of the waste solids, a large amount of
water is lost, and the waste solids further increase the total tonnage and volume of the waste
produced. Water recovery represents a major issue for smelting operations located in regions
with water scarcity.
The approach BioteQ developed for the copper smelter in question involves using the BioSulphide process
to treat the EPAS solution instead of lime neutralisation. This system involves contacting EPAS liquor with
biogenically produced hydrogen sulphide gas. The overall process and flow sheet is the same as that
described in section 2.1. However, the sulphidation of trivalent arsenic occurs via the reaction chemistry in
equation (8):
2HAsO2(aq) + 3H2S(g) → As2S3(s) + 4H2O(l) (8)
Other heavy metals that have an affinity for sulphide, such as copper, zinc and lead, also report to this
precipitate. After clarification in a thickener, the resultant sulphide precipitate is filtered and stored on-site
in a secure landfill.
The benefits of the EPAS treatment using the BioSulphide process include the following:
 The volume of waste residue generated through treatment is significantly reduced. The metal
sulphide residue solids contain over 50% w/w As on a dry-weight basis, offering an order of
magnitude reduction in solids handling and disposal costs.
 Power and chemical costs are reduced based on the following:

○ The reaction in equation (8) is fact, and equilibrium is very favourable for all arsenic to
precipitate at the extremely low pH of typical EPAS without the need for neutralisation.
○ The As(III) present in EPAS is precipitated as As(III) into As2S3 (synthetic orpiment), thus
eliminating the need for oxidation prior to removal from EPAS.
 Water recovery is higher than in the current system.
 Acid contained in EPAS is recovered.
 The recovery of metals such as Cu, Re and Mo from EPAS becomes a possibility.
The current system involves trucking arsenic-bearing waste to a third-party site. As this material is hazardous,
the mine owner bears both the cost and the liability for transport. Part of the proposed upgrade of the
existing EPAS treatment using the BioSulphide process includes constructing a waste residue storage facility
directly at the mine site; the greatly reduced mass and volume of waste means that local waste residue
disposal is possible.
Currently this project is in the final stages of detailed process design. Construction of the facility is anticipated
in 2016, and commissioning and operation are expected to take place in late 2016 to early 2017. The site
owner selected the BioSulphide process over not only the current lime/ferric precipitation EPAS treatment
method but also over the atmospheric scorodite precipitation method. Many in the scientific community
consider scorodite to be probably the most stable arsenic solid. However, scorodite residue typically contains
less than 25% As compared to the more than 50% achieved by the BioSulphide process. Based on life-cycle
cost, the atmospheric scorodite process is so much more expensive than the BioSulphide process that the
savings from using BioSulphide outweigh the incremental improvement in the stability of waste residue
brought about by the scorodite process. Stability is a relative term linked directly to the conditions applied in
the waste solids residue disposal facility; under certain storage conditions, sulphide residue stability is equal
to or greater than the stability of scorodite.
Selenium has recently garnered significant attention as a contaminant of concern in a broad range of mining
operations around North America. Since selenium is in the same group as sulphur on the periodic table,
selenium can often substitute for sulphur in mineral structures such as coal or sulphide minerals. As such, it
can be mobilised by acid mine drainage processes and be released to the environment. While small amounts
of selenium are a necessary micronutrient for many organisms, concentrations even in the single digit parts
per billion (ppb) range are deleterious to fish species, causing birth defects and warping skeletal structures.
For this reason, regulatory guidelines for the maximum selenium concentration in industrial discharges in
North America range from 5 to 20 ppb, with some site-specific requirements as low as 1 ppb.
Historically, the issue of selenium contamination has been understood to relate primarily to coal-fired power
plants. As coal containing selenium is burned, selenium is volatilised. When the combustion off-gas (flue gas)
passes through a wet scrubber, selenium reports to the scrubber wastewater. Selenium treatment must be
used in order for this wastewater to be discharged to the environment safely. Selenium-bearing streams are
typically warm and laden with organic chemicals; they have steady flow rates and metal loads due to the
relative consistency and stability of power plant operations. As such, these flue gas wastewater streams are
ideal feed for the various types of biological treatment systems that make up the majority of existing
selenium treatment plants.
In the past, the industrial focus on selenium treatment in water has been on these flue gas wastewater
streams. However, both regulators and mine operators have more recently recognised the importance of
managing selenium and removing it from a wide range of streams impacted by resource industry activity.
These streams include runoff and seepage from mining overburden, waste rock and tailings storage facilities.

Case Studies
In cases such as these, the character of the selenium-bearing stream differs significantly from that of flue gas
wastewater. In contrast with flue gas wastewater, runoff streams typically have the following traits:
 large seasonal variability in flow and selenium content,
 cold water, often near freezing in northern climates,
 nutrient-starved receiving environment with no tolerance for added chemical oxygen demand
(COD) or biochemical oxygen demand (BOD) nutrients,
 minimal organic content, and
 multiple remote outfalls requiring treatment.
Residue management must be considered in any long-term water treatment system. Biological selenium
treatment systems produce a biological sludge by-product containing dead or inactive biomass and a large
percentage of water. The selenium removed from the feed water is contained in this biological sludge
by-product as tiny particles of elemental selenium. The amount of sludge generated depends on the type of
biological reactor used and the operating strategy, but in modern biological systems, approximately 0.5 L of
wet biological sludge is generated per cubic meter of water treated. As this material is biological in nature, it
will necessarily change over time. Little is known about how this material behaves in the long term.
Managing this material represents a further point of contrast between the treatment of flue gas wastewater
and mine runoff water. Coal-fired power plants typically generate large volumes of fly ash and metal-laden
gypsum waste from flue gas desulphurisation. This material is typically stored in secure landfill facilities
designed to minimise water contact and mobilisation of toxic contaminants. When biological selenium
treatment is used, the amount of biological sludge generated is small relative to the amount of fly ash and
metal-laden gypsum, and as such it can be blended with other waste material. In contrast, selenium water
treatment for mines involves larger volumes of water and less access to secured landfill repositories. The
long-term implications of residue management over the multidecade life of the mine is a key driver for the
owner of the property described in this case study.
The project in this case study is a large copper-gold mine that is currently in the permitting process. As the
ore body and project site are located in a sensitive watershed area, the requirements of water treatment on
the site are being carefully evaluated. Of particular interest is selenium, which is present in certain fractions
of the ore body and associated overburden and waste rock. The owner of the site is keenly aware of both the
importance of selenium and the life-cycle costs of selenium treatment to ultralow discharge targets. In order
to minimise the costs of active water treatment for selenium, the owner has taken two major actions.
First, the owner understands that the risk of selenium contamination in seepage is from a specific type of ore
on the property. Therefore, the mine plan calls for sequestering this type of material in a predetermined
waste rock holding facility. This means that the selenium released through weathering of the ore will be
concentrated in a smaller volume of water. In this way, active water treatment for selenium will be required
on a considerably smaller flow of water than if the selenium-bearing material was spread across the entire
waste rock storage facility.
Second, the owner is investigating alternative treatment methods, with an emphasis on reaching ultralow
discharge targets while minimising waste residue for management. To this end, the owner has investigated
alternatives to conventional biological systems currently on the market.
Selen-IX™ is a selenium treatment system developed by BioteQ to selectively remove selenium from
industrial water in order to reach environmental discharge targets. The first stage of the system uses a strong
base anion-exchange resin in the sulphate form to selectively remove selenium from water to concentrations
of less than 10 ppb. Once the resin is saturated with selenium, the ion-exchange resin is regenerated using a

concentrated sodium sulphate brine solution. During regeneration, selenium loaded on the resin is
exchanged with sulphate from the sodium sulphate brine. This generates a low-volume brine stream that
contains the selenium removed from the influent water.
The spent regenerant rich in selenium is then treated using an electrocell in which an iron anode is dissolved.
This introduces iron into the selenium-rich regenerant. Iron and selenium then precipitate from the solution
as an inorganic iron oxyhydroxide laden with selenium. Following this treatment, the iron-selenium solids are
separated from solution in a clarifier with the aid of flocculant, and the now selenium-depleted solution is
recycled for re-use in the ion-exchange system. The system is shown schematically in Figure 4.
For this project, a containerised Selen-IX pilot unit capable of treating 2 L/min of influent feed water was
mobilised to a pilot site in the British Columbia Lower Mainland. Surface water from the proposed mine site
was shipped to the pilot site and spiked with metal species to match runoff quality predicted by geochemical
modelling. At the time of writing, this pilot is ongoing and expected to be completed at the end of the first
quarter of 2015.
The composition of the feed water of the current project is shown in Table 2:
Parameter Value
pH 4.1–4.5
Dissolved Se (mg/L) 0.103
Nitrate as N (mg/L) 32
Sulphate (mg/L) 1,950
Total dissolved solids (mg/L) 2,910
In the current pilot program, the target effluent selenium concentration is 1 ppb or less. This objective has
been achieved, as shown in Figure 5. These results were generated over a week of locked-cycle pilot
operation.

Case Studies
The character of the solid iron-selenium by-product from treatment is of great interest to the mine owner
sponsoring this pilot program. While full solids characterisation results are not available at the time of writing
due to the ongoing nature of the program, X-ray diffraction (XRD) analysis has been carried out on a sample
of the material. The spectra of the material, shown in Figure 6, shows that the dominant species in the
material are magnetite and elemental selenium. The relatively low base counts per second also show that
the material is largely crystalline rather than amorphous. Toxicity characteristic leaching procedure (TCLP)
testing of material generated during the pilot campaign has shown that the material is nonhazardous.
The amount of solid residue generated through Selen-IX is proportional to the amount of selenium removed
by the system and dependent on the effluent discharge criteria; the amount of iron required increases as
discharge targets become more stringent. In this project, approximately 0.25 L of filtered material is expected
to be generated per cubic meter of water treated, a reduction of half compared to modern biological systems.
Mine owners in North America are increasingly faced with the necessity of actively treating water from mine-
impacted runoff for selenium. Biological treatment systems currently represent the best available technology
for the removal of selenium from waste water. However, mine-impacted water has a number of

characteristics that make implementing biological treatment technically or logistically difficult: ultralow
regulatory limits may not be practical to reach through biological treatment, and there are concerns about
the long-term stability of residue management methods and the ability of treatment systems to respond to
rapid variations in water flow rate and temperature. The receiving environments for treated water are often
nutrient starved and thus intolerant of nutrient and BOD plumes that can result from biological treatment.
All of these factors lead to high operating and capital costs for conventional systems, which leads to interest
in novel systems such as Selen-IX.
Industry challenge Selen-IX

< 1–5 ppb Se regulatory limits Demonstrated Se < 1 ppb from streams
with > 100 ppb feed
Quantity and long-term stability of residue Compact inorganic Fe-Se solid of magnetite
and elemental Se
Variable flow and mass loads Easy to adapt IX cycles on the fly
Cold temperature of water Insensitive to temperature
Intolerance of receiving environment for No organics or phosphate added; no need
nutrients, BOD for nutrient polishing
Multiple, remote points of discharge IX modules lend can be decentralised
High life-cycle cost of ownership Significant savings in life-cycle cost
Demonstrating effective water treatment for selenium is a key hurdle the proposed mine in question must
get over in order to pass permitting. The owner recognises that effective water treatment goes beyond
reaching ultralow discharge targets; the long-term life-cycle cost of residue management must be
considered. The prospect of generating a stable, inorganic iron-selenium by-product is a significant driver for
the owner’s interest in Selen-IX. While the technology is still at the pilot scale, piloting results to date
demonstrate its potential as a sustainable alternative to conventional selenium treatment methods.
When water treatment for mining is evaluated, there is a tendency to focus on the initial capital cost of
treatment installation. While this analysis is important for acquiring financing and investment, when the life-
cycle cost of treatment is considered, the ongoing operation and maintenance costs are often as significant
as or more significant than the capital cost. This is particularly true when treatment involves generating waste
residue or by-products. By-products take a variety of forms, ranging from nonhazardous tailings and metal-
bearing waste with potential for remobilisation to biological sludge or liquid brine. Depending on the nature
of the treatment in place, these residue streams can represent both environmental challenges and
opportunities to recover value. In order to minimise the costs and maximise the benefits of environmental
treatment, these residues and their management strategies must be well understood. The three case studies
highlighted in this paper describe unique situations where novel technologies were used to manage waste in
a cost-effective and sustainable way.

F.A. Maimela University of the Witwatersrand, South Africa

T. Zvarivadza University of the Witwatersrand, South Africa
Surface mining, by its nature, causes considerable environmental challenges, which must be resolved if the
mines are to be correctly closed. The voids left need to be rehabilitated in accordance with the dictates of
environmental regulations so that the land is left in an economically usable state. This paper examines the
effects of waste dumping optimisation for final closure profile at a South African colliery pit. The pit is currently
being mined, and next to it are two previously mined pits, which were left as open voids and not backfilled.
Unlike in the past, rehabilitation is now enforced through regulations, and companies ought to comply; this
means the colliery must rehabilitate both the two opened voids and the pit that is currently being mined. The
material from the current pit is to be used to fill in the voids, which will lead to a material shortage at closure.
This research shows that there will be enough waste to backfill the two voids that were left open and there
will be a shortage of material to backfill the current pit at the end of mine life. Void dimensions will be
minimised by concurrent rehabilitation after the two voids have been backfilled; thus discard from the plant
will be used to compensate for the deficit.
This research was conducted to help companies optimise waste and meet legal requirements by rehabilitating
the disturbed area at the end. The paper highlights the crucial benefits of waste dumping profile optimisation.
Optimising waste dumping helps companies avoid having a big void at the end without enough material to
fill it. This strategy ensures that open voids are closed at competitive costs at the end of mining and the area
is restored to economically usable conditions. Decant elevations were used in order to prevent water from
mixing with the discard. A drainage pattern was established for water flow after rehabilitation, and the
colliery pit will be rehabilitated at the end. Converting the pit to wetland is another plausible option that can
be pursued. This will help bring back bird and plant life. An additional benefit of conversion to wetland is a
more diversified ecosystem with species such as macro-invertebrates, frogs, small mammals and birds. This
strengthens biodiversity, which is the backbone of life on earth.
Surface mining poses a threat to human existence as well as to the environment. This means that areas
disturbed by mining activities should be rehabilitated to achieve their original form. Detailed plans are
essential for closure, and rehabilitation plans should be designed and implemented with eventual closure in
mind (Wells, 1986). If closure is not taken into consideration in the initial stages of planning, the experience
and consequences can be traumatic (Wells, 1986) and can include having a final void at the end of the mine
life with little or no material to fill it and rehabilitate the area. This can invite high penalty duties for a mine.
This research was conducted at a South African colliery located next to two opened voids, which need to be
filled with material and levelled. The waste being stripped off from the box cut is used to fill in the two voids,
which means that no box cut material is being stockpiled. This may result in a material shortage for the final
closure profile. It is therefore necessary to optimise use of the waste to avoid having a big void at the end
without enough material to fill it. This research was conducted in order to help companies avoid such
consequences and instead meet legal requirements by rehabilitating the disturbed area at the end. As Wells
(1986) notes, “a considerable amount of planning is required for rehabilitation of land that has been
disturbed by open cast mining if it is to meet legal constraints and the requirements of Chamber guidelines.”

Waste dumping optimisation for final closure profile: a South African colliery case study F.A. Maimela and T. Zvarivadza
(This statement refers to guidelines discussed by the Chamber of Mines of South Africa, 1981). Planning for
rehabilitation should be incorporated with other mine plans as this will help in reducing rehabilitation costs.
Optimising waste dumping schedules is vital to ensure that open voids are closed at the end of mining and
the area is restored. If done properly, rehabilitation costs can be reduced. Figure 1 illustrates how important
it is to optimise waste dumping for final closure since high cost percentages can be avoided if this is done
efficiently. The expenditure for rehabilitation at the colliery was calculated to be R 11,314,920 (in 2014
values). This expenditure will escalate in the ensuing years in line with the consumer price index.
The study presents an alternative way to fill in the final void should there be any material shortage at closure.
This study investigates how to avoid having a large residual mining void at closure by optimising waste
dumping. The colliery we researched is situated next to two open voids that are currently filled with water:
void 1 and void 2. The colliery is in its developmental stage, where a box cut is being opened with mining
blocks of 100 m × 60 m. Mining will proceed by mining strips with 100 m × 45 m blocks. The waste or
overburden that is being removed is to be used to fill in voids 1 and 2. This means that no box cut material is
being stockpiled, which may lead to a material shortage at closure for backfilling the final void. Waste
optimisation is thus vitally important to eliminate the residual void and minimise rehabilitation costs.
Rehabilitation has to be done at closure to meet the legal requirements. The integrated water management
system, surface drainage system and environmental management plans must be considered. Companies also
need to consider financing options, including a financial provision account for closure and long-term
maintenance. This paper summarises research on the following activities: optimising final void dimensions,
meeting legal requirements at the end of mining and calculating rehabilitation costs for the pit.
Legislation that has been promulgated in order to ensure that mine closure is appropriately considered
includes the National Environmental Management Act (NEMA) and Section 38(1)(d) of the Mineral and
Petroleum Resources Development Amendment Act (MPRDA), which both relate to the closure of mined land.
Kotze (2013) notes that both acts state that “mining operations should as far as it is reasonably practicable,
rehabilitate the environment affected by the prospecting or mining operations to its natural or

Case Studies
predetermined state, or to a land use which conforms to the generally accepted principle of sustainable
development.” However, it is not always easy to restore land to exactly the way it was before disturbance by
mining. Companies are expected to at least restore it to a state where it is still beneficial for the current
generation and generations to come.
Rehabilitation and closure responsibilities are shared among the government, mining companies and other
stakeholders. The government, according to Swart (2003), provides guidance on environmental constitution.
The government carries the responsibility to be accountable to the public and to promote sustainability. It
can do so because it regulates the mining industry. This responsibility means the government inherits
enduring problems and mining legacies. Mining companies and stakeholders work hand in hand with the
government to comply with relevant legislation regarding rehabilitation and closure. Sections 39–42 of the
MPRDA are most relevant to this research.
Sections 39–42 of the MPRDA (2002) deal with requirements of rehabilitation:

 Section 39: The Environmental Impact Assessment (EIA) and the Environmental Management Plan
(EMP) are prerequisites of this section. Companies are required to state the type and extent of
damage to the environment, to identify methods they plan to use to restore the land to a usable
form and to provide costs (Chadwick, 2012).
 Section 41: This section requires financial provision, which is the money set aside in order to
provide for closure. This money is deposited in a rehabilitation account owned by the government
as a way to ensure that there is financial provision for rehabilitation at closure (Swart, 2003).
 Section 42: This section requires management of stockpiles and mine wastes. It works hand in
hand with the National Water Act, 1998. This section helps to ensure that companies manage
disposal of their wastes during mining, decommissioning and closure.
A holistic approach combining best practices from different mines operating in South Africa is desirable. This
approach tells how to avoid poor rehabilitation. Chadwick (2012) argues that assessing the economic viability
of the plan, including funding for post closure care and maintenance, is one of the most important
approaches to mine closure. This ensures that the mine closure plan is viable and funding is readily available
and properly maintained. The land must still be capable of sufficient productivity to sustain life after closure.
This is ensured by audits of the stages of mining up to closure. Chadwick (2012) advises that companies must
not rely on communities to fulfil their rehabilitation duties since ignorance of smaller impacts may grow to
more critical issues in the future if not attended to at early stages. Mining companies need to be innovative
and have an accurate overview of rehabilitation issues in order to adequately address these. This positions
them to address issues well beyond those raised by the communities.
Economic and technical issues, national and political unrest, and small-scale mining are some of the reasons
that lead to mines being abandoned (Rapson, 2004). The prices of mining commodities depend on demand
and supply, and these issues make it difficult for prices to remain stable. In cases where prices fall, the mines
can find it challenging to sustain operations, which can result in premature closure or companies going
bankrupt and abandoning the mine. Technical issues, including the adoption of new methods that have yet
to be tested for effectiveness and efficiency, are another significant factor; the sites on which testing was
being done are frequently not rehabilitated (Rapson,2004). Small-scale mining is mostly not legal and is
therefore unregulated; thus artisanal miners simply walk away and abandon their mines after mining.

Failure to meet the requirements of the legislation with regard to rehabilitation and closure can lead to poor
rehabilitation or abandoned mines. Several social, environmental and economic challenges are presented by
abandoned mines. Potential issues include safety and health problems such as the risk of people falling into
the mine voids, resulting in fatalities. Cartagena et al. (2001) note that land degradation and water pollution
emanating from abandoned mines create damaging economic impacts. Every stage in the mine value chain
poses different environmental, social and economic impacts that need to be appropriately managed. Mine
life cycle closure planning will help avoid a cumulative overwhelming burden at the time of mine closure.
There are different approaches that help in ensuring that rehabilitation is achieved at the end of mining and
legal requirements are met. These are discussed in the following sections.
Mine rehabilitation includes movement of material, and it is essential to optimise this process (Caccetta and
Kelsey, 2001). Caccetta and Kelsey (2001) have described a mathematical model to assist in this optimisation.
This model uses a block model to represent the landform and the desired landscape that must be met at the
end of the process. “The amount of material moved (cut) must be equal to amount of material disposed (fill)
and material movement must be minimum” (Caccetta and Kelsey, 2001). The block model works hand in
hand with a subgradient optimisation algorithm to simulate blocks that will be cut and blocks that will be
filled. The model then generates the shape of the final pit and illustrates how much material can be cut and
filled to obtain a balance and reshape the land successfully. The benefit of the mathematical model is that it
enables a rapid solution for different pit shapes, which can then be evaluated at early stages of mining before
closure. This method was tested using two real mining methods and the optimum reshaping profiles were
achieved.
Knowing the risks involved in specific processes and their severity on the processes is significant. Matrices
can be used to rank the severity of different risks. The ranking assists in indicating whether a risk should drive
the company to look at options that will help to reduce the risk at an early stage. The matrix can also be used
to investigate sustainable development parameters (environmental, social and economic) to ascertain the
risk they carry. Table 1 shows how the risk probability and consequences are ranked.
The matrix, according to Rapson (2004), is used together with the indicated key to rank the severity of the
risk involved. Regarding the risk of having a material shortage at the end of mining at the colliery researched
for this study, the probability of this happening is certain; this gives it a probability ranking of 5 from the
matrix table. The consequences are catastrophic, since this is a threat to the communities affected (people
can fall inside the pit and die), so the risk gets a consequence ranking of 5. To find the total risk ranking, this
consequence ranking is multiplied by the probability ranking:
Total risk ranking = 5 × 5 = 25 (1)

Case Studies
Ranking Probability Ranking Consequences

5 Certain/common 5 Catastrophic
Almost certain/has
4 4 Major
happened
3 Likely/could happen 3 Serious
2 Possible but not likely 2 Minor
Unlikely/practically
1 1 Negligible
impossible
These rankings work in hand in hand with the matrix displayed in Figure 2.
The severity of the risk of this colliery having a material shortage at the end of mining falls within the red
region and thus is very high. Water pollution is another risk to consider. It is certain that water will be polluted
if rocks with acidifying potential are left exposed and untreated, so this risk gets a probability ranking of 5.
The consequence of this risk will be catastrophic as well since the rivers close by will be polluted and thus
aquatic life will be threatened. The water table can also be polluted, leading to shortage of potable water for
drinking. So this risk is given a consequence ranking of 5. When this consequence ranking is multiplied by the
probability ranking, the rank of this risk is also high:
Total risk ranking = 5 × 5 = 25 (2)
This approach helps companies focus on their major risks and thus generate solutions that can reduce their
severity. Using this approach addresses risks to the three spheres of sustainable development and thus
eliminates risks of water pollution and failure to rehabilitate land.
The study considered other adjacent pits as well as stockpiles and focused on the empty voids that were
being filled by the box cut material, potentially leaving large open voids at the end of mining. The research
also looked at the surface drainage pattern at closure. Surveying methods were used to calculate the volume
and area of the land that will be disturbed. The results from the voids adjacent to the pit are presented below.
Volume calculations using surveying methods showed that the total area to be disturbed is 41.9 ha. This

helped to determine how much material will be required. The material shortage was 564,463.10 m3. To
compensate for the shortage, discard (low grade ore) is to be placed in the pit.
Consultations with the planning, environmental, geology, surveying and rehabilitation departments were
conducted throughout the project. The detailed research methodology is as outlined below:
 Desktop studies and physical investigation of the mining lease area were done in order to identify
options that might be suitable to achieve the main objectives.
 The site was visited at the outset in order to see the lease boundary and know the area to be
disturbed.
 The volumes were obtained including estimation of the final void volume, volume of the whole pit
and volumes of the two adjacent opened voids to be filled with box cut material.
 The volumes were used to determine the volume of material available to fill in the two adjacent
voids.
 The material required to fill in the final void and the shortage were calculated.
 During the search for extra material, it was noted that the discard from the plant needed a
storage area. The discard was then considered as an option to compensate for the water
shortage.
 The elevation at which the discard should be placed to avoid contact with groundwater and
surface water was calculated. (This was to avoid damage to groundwater and surface water that
could be caused if the water passed through the discard.)
 The decant elevations (the elevations to which water will rise when the pit has been backfilled)
were obtained from the environmental management department and used to calculate the
discard elevation. (The convention is that discard elevation should be 1 m below decant elevation
to avoid a mixture.)
 Pit floor and decant elevation were used to establish the elevation at which the discard should be
placed.
 Rehabilitation costs were calculated.
 Conclusions and recommendations were drafted based on the results and analysis.
The literature review and the research conducted during the study indicated options for solving the problem.
The best economic choice was chosen from the options examined and discussed below.
Normally, the box cut material from the pit being mined is stockpiled to be used at closure. The colliery has
a dump situated away from the current pit being mined. In this option, the material would be placed at the
dump to be used later on to fill in the final void. As a consequence, the two previously mined voids would
have no material to fill them. This would be a violation of environmental regulations set by the Department
of Mineral Resources (DMR), and this might lead to fines. Therefore material might need to be transported
from other pits nearby, which would be costly.

Case Studies
The mine could do concurrent rehabilitation during mining in order to have a much smaller void at the end
of mining. This would require multiple operations in the pit and would require more careful planning in order
to avoid overlapping processes during mining and rolling over for rehabilitation.
For the third option, we compared dumping distances from the current pit to the dumping site or the voids.
This helped in finding a more cost-effective option as material can be dumped at the nearest destination. It
takes 779 m to haul material from the box cut at the pit to the dumping site and 505.3 m to dump at open
void 1. It would make sense to dump material at the open void since it is closer. The following cost analysis
explains the choice to dump at void 1.
Using year 2014 rates, it costs approximately R 6/m to use dumpers (trucks to dump material from the pit)
in moving waste.
Moving cost to dump = amount for dumper to move material(R/m) × distance travelled (m)
= R 6 × 779 m = R 4,674
Moving cost to void 1 = amount of dumpers to move material (R/m) × distance travelled
= R 6 × 506 m =R 3,036
It is therefore cheaper to move material from the current pit to the void. Financial provision accounting as
required by section 41(1) of MPRDA comes with additional benefits since, for every hectare that is levelled,
R 300,000 can be reclaimed from the lodged financial provision in accordance with the Act. Backfilling of
void 1 had already begun at the colliery, and 3.5 ha has been covered. This equates to R I.05 million, which
can be reclaimed if the area is levelled. This income can be used to supplement funds for the mining processes
going on at the pit.
Table 2 shows the amount of money that can be reclaimed from the provisions account if the voids are
completely levelled. The voids cannot be filled in immediately; however, every hectare being levelled can
account for up to R 300,000.
Void 1 Void 2
Area = 3.05 ha Area = 13.06 ha
Rehab cost = R 300,000/ha × 3.05 ha Rehab cost = R 300,000/ha × 13.06 ha
= R 915,000 = R 3,918,000
The analysis points to the third option as the optimal solution that can be implemented.
The volumes of pit, voids, final void and total overburden were calculated using surveying and planning data,
and the results are as presented in Table 3. These calculations enabled material balances to be done. Knowing
these volumes helps in using volume calculations to determine how much material is needed to fill in the two
open voids and to calculate the volume of the shortfall in material. Since the dimensions of the final void

would have been estimated, this helps in determining whether the waste material will be enough for
backfilling at closure. The volumes are tabulated below.
Total waste Void 1 Void 2 Volume of the final void

30,252,014 m3 (BCM)* 4,701,000 m3 1,464,000 m3 3,149,175 m3
*The total waste is in bank cubic meters (BCM), total in situ. This volume increases by a swell of 25% when
blasted. The new total waste volume is 38.2 ×106 m3.
Once we knew the total waste volume, we could calculate the areas covered by the two voids to determine
how much waste would be needed. Void 2 is already being filled with some of the overburden material being
removed from the pit, and the calculation below explains how much volume is left to be filled.
Void 1: Area covered = 3.05 ha; total volume = 1,464,000 m3
Void 2: Area covered = 13.06 ha; total volume = 4,701,000 m3; fill volume = 29,928.21 m3
Remaining volume = 4,671,071.79 m3
The calculation above shows that void 1 is still not filled, and void 2 has been filled to about 3.5 ha. The total
amount of waste since mining commenced at the pit in April 2013 to December 2013 helped in working out
how much waste has been moved so far and used to fill the voids. This helps in working out the total volume
that will be required to backfill the whole pit after removing some waste material to the voids. The volumes
are given in Table 4, and the calculation prior to these volumes is given after the table.
Waste moved from April 2013 – December 2013

Month April May June July August September October November December Total
Volumes 538,390 976,430 763,450 729,700 643,480 561,910 572,597 945,824 705,694 6,437,475
(m3)
Total waste volumes to fill voids (1 and 2) = 6,165,000.0 m3

Total waste volumes to fill voids 1 and 2 = 6,165,000 m3
Amount of waste left in pit = total pit volume – what has been moved so far
= 38,197,997.93 m3 – 6,437,464 m3 = 31,760,533.93 m3
The total amount of waste moved as of December was used to fill voids 1 and 2. The amount of waste that is
available to fill the pit is calculated as follows:
Amount of waste to fill pit = waste remaining after filling voids + waste left in the pit
= 31,760,533.92 + (6,437,464 − 6,165,000) = 32,032,997.92 m3
The disturbed area excluding the final void is 52.10 ha; therefore, the material required to backfill the
disturbed area is 31,937,030 m3. The amount of waste that remained after filling the disturbed area was
determined to be 96,000 m3. This volume of waste can be stored for final use at closure.
Void 1 = 3,136 m2; void 2 = 1,839 m2; total length = 633 m

Total void volume = volume of void 1 + volume of void 2

Case Studies
= 1,985,088 m3 + 1,164,087 m3 = 3,149,175 m3

The total length was an estimate from the borehole data collected from geological assessments.
Discard dumps are prone to spontaneous combustion when dumped on land (Rapson, 2004); this happens
due to lack of compaction and exposure to oxygen. Should this happen when there are people around, it can
have severe health effects and can even cause death. The fire can also damage the equipment and
infrastructure around the mine. Placing discard in the pit will compact it and reduce its exposure to oxygen,
reducing the opportunities for spontaneous combustion.
According to conceptual design, if discard is to be placed in the final void, decant and daylight elevations
must be established (Barnard, 2012). Note that decant elevation is the elevation where water comes out on
surface. It is the same as daylight elevation. This helps to determine the elevations to which discard can be
placed. In this case, daylight elevation was 1,561.6 m. Discard must be placed 1 m below daylight elevation
to avoid clean water mixing up with discard. This means that any elevation below 1,560.5 m in the pit at
closure can be used to place discard. However, elevations above 1,560.5 m cannot be backfilled using discard.
These elevations helped in determining the positions at which there will be discard placement and thus the
amount of discard volume required and in calculating the overburden to fill in the remaining elevation.
Figure 3 illustrates the placement of discard and overburden relative to the elevations. This was done using
geological data for elevations. Note that overburden is the waste directly below the topsoil up to the top of
first seam while mid burden is the waste between seams.
The discard amount worked out to be 2,526,169.325 m3, as given in Table 5.
Location Discard amount (m3) Depth of discard

Final void 2,526,169.325 1 m below daylight elevation (1,561.6 m)
Total volume of void = 3,149,175 m3

Overburden + midburden = total volume of final void – discard volume

= 3,149,175 m3 – 2,526,169.325 m3
= 623,005.675 m3
There is a material deficit of 527,000 m3 (623,000 m3 − 96,000 m3) from the volume after filling the remaining
part of the pit; therefore, this material will be blasted from the high wall of the final void, creating slopes of
1:14. This means that 527,000 m3 of overburden will be blasted from the final void high wall, thereby reducing
the length of the final void and yielding a slope of 1:14 or preferably 1:10, depending on the material blasted
from the high wall. This is because the material will have been used to cover the remaining part of the pit
excluding the final void.
Table 6 shows the rehabilitation costs of the pit as determined using year 2014 rates.
Task Additional information Rate Measure Units Cost

Conceptual design Overall rehabilitation planning R 650,000 lot 1 R 650,000
Specific design Design of different final landform R 450,000 lot 1 R 450,000
Box cut spoils
placement Hauling spoils to the dump site R 27 m3 R0
Discard placement Compensation for the shortfall of
material; discard should not be placed
last, as this will result in leaching of salts
into the topsoil, resulting in poor
vegetation growth. R8 m3 R0
Final shaping Includes dozing, truck and shovel to
ensure smooth profiles and free-draining
profiles R 175,000 hectare R 9,205,000
Fertilising and
preparation R 9,000 hectare 0 R 473,400
Seeding Application of the seed mixture used R 1,000 hectare 0 R 52,600
Maintenance, year 2 Fertilising, cutting and bailing R 2,300 hectare 0 R 120,980
Rehabilitation Performance assessments on the
assessments vegetation to determine success and
progress R 10,500 each 5 R 52,50
Total R 11,314,920
The discard cost is R 0 because the expenditure is met by the plant. This is an advantage to the mine, since
there will be material to fill in the final void at no expense. The same can be noted for the box cut spoil
placement, which will also cost R 0 because the box cut material is being dumped into the two empty voids

Case Studies
and not stockpiled. The total rehabilitation cost for the pit is R 11,314,920; this cost can be minimised by
concurrent rehabilitation. The final pit void is approximately 0.5 ha; therefore, the total rehabilitation cost is
R 1,249,600. The rehabilitation cost significantly reduces for rehabilitating the final void only as compared to
rehabilitating the whole pit at closure.
Surface mining causes environmental disturbances and therefore it is essential for mines to be conscious of
the need to meet legal requirements. The voids at the end have to be rehabilitated so that the environment
is restored to its original form or a form that can sustain an economic end use. The studied colliery has taken
rehabilitation as a priority. The pit will be rehabilitated at the end of mining while the two empty voids are
filled as mining continues. The final void dimensions can be optimised by concurrent rehabilitation since this
helps in reducing the total disturbed final area to be rehabilitated. This also has financial benefits since up to
R 300,000/ha levelled can be reclaimed. Blasting eight strips with a total area of up to 5 ha will help in claiming
up to R 1.5 million from the environmental provision account.
Rehabilitation costs for the pit were calculated to be R 11,314,920. This cost decreases when concurrent
rehabilitation is considered since the final area is reduced to 0.5 ha, which leads to a rehabilitation cost of
R 1,249,600. This study illustrates how rehabilitation can be optimised, leading to the conversion of land for
a sustainable end use at mine closure. Planning for closure at the conception of any mining project is critical,
as this enables workable solutions encompassing concurrent rehabilitation and final closure profile planning
to be reliably implemented.
The company will have to apply for and obtain a water use licence as outlined in section 21(g) of MPRDA in
order to be able to dump discard in the final void.
The rehabilitation implementation strategy should be adequately incorporated in mine planning right from
the onset. Early considerations ensure that challenges such as slope instabilities are detected early and slopes
are redesigned at an early stage. Rehabilitation costs are measured relative to the area disturbed; therefore
rehabilitating as mining commences eventually reduces the total area disturbed. This will result in lower
rehabilitation costs. Again, concurrent rehabilitation will help subsidise the mining process since there is a
certain amount that can be reclaimed per hectare levelled.
Rehabilitation aims to create a suitable sustainable environment; this includes bringing back the ecosystem.
The final void can be converted into a wetland. Drainage can form from the rehabilitated area through the
valleys formed. Wetlands up to a metre deep can also be built along the free drainage system, and sandstone
rocks from spoils can be used to help decrease the velocity of water running along the drainage during rain
seasons. As the water flow slows and rain continues, the amount of water builds up, leading to the
introduction of bird life, fish and plants in the ecosystem. Wetlands have several benefits including creation
of habitat, erosion control and flood attenuation. Wetland creation is different from just creating grassland:
wetlands help to bring a more diversified ecosystem inclusive of species such as macro-invertebrates, frogs,
small mammals and birds.

Caccetta, L. and Kelsey, P.J. (2001) Models for mine site rehabilitation, viewed 20 November 2014, http://www.mssanz.org.au/
MODSIM01/Vol%204/Caccetta2.pdf.
Cartagena, P., Campusano, R. and Hoskin, W. (2001) Abandoned mine sites: Problems, issues and options: Discussion paper presented
by UNEP, viewed 04 December 2014, http://www.eclac.org/drni/noticias/noticias/9/7029/ProgrammeUNEP.pdf.
Chadwick, P. (2012) Financial provisions for rehabilitation and closure in South African mining: Discussion document on challenges
and recommended improvements (summary), WWF-World Wide Fund For Nature (Formerly World Wildlife Fund), Cape
Town, South Africa, viewed 04 December 2014, http://awsassets.wwf.org.za/downloads/summary_mining_report_8aug.
pdf.
Chamber of Mines of South Africa (1981) Handbook of guidelines for environmental protection, Vol. 3/1981: The rehabilitation of
land disturbed by surface coal mining in South Africa.
Kotze, C. (2013) Importance of end land use and mine closure planning increases in South Africa, Mining Weekly, 22 November 2013,
viewed 03 December 2014, http://m.miningweekly.com/article/importance-of-end-land-use-and-mine-closure-planning-
increases-in-sa-2013-11-22-2.
Rapson, L.A. (2004) The development of a risk-based computer database to prioritise the environmental rehabilitation of defunct and
abandoned collieries in the Witbank area of South Africa, MSc research report, University of the Witwatersrand,
Johannesburg, viewed 03 December 2014, http://www.coaltech.co.za/chamber%20Databases%5Ccoaltech%5CCom_DocMa
n.nsf/0/6C2987FF58B0452642257AD80027B0CC/$File/LINDA%20RAPSON%20MSc-%20Abandoned%20Collieries%20.Pdf .
Swart, E. (2003) The South African legislative framework for mine closure, Journal of The South African Institute of Mining and
Metallurgy, no. 7, pp. 489–492.
Thompson, R.J. (2005) Surface strip coal mining handbook, South African Colliery Managers Association, viewed 09 December 2014,
http://www.sacea.org.za/.%5Cdocs%5CSACMA%20Surface%20Strip%20Coal%20Mining%20Handbook_rev1.pdf.
Wells, J.D. (1986) Long-term planning for the rehabilitation of open cast workings, Journal of The South African Institute of Mining
and Metallurgy, Vol. 86(3), pp. 89–93.
Barnard, P. (2012) BECSA 4 Mines long term rehabilitation planning, Golder Associates Africa (Pty) Ltd., Johannesburg.

B.A. Power Azimuth Consulting Group Partnership, Canada

R. Baker Azimuth Consulting Group Partnership, Canada
Expectations for the environmental performance of post-closure mine landscapes are changing, and risk
assessment is increasingly used to guide closure activities. The environmental risk assessment process
evaluates the likelihood and magnitude of adverse effects to humans and ecological resources as a result of
exposure to chemical and/or physical stressors. In relation to environmental contaminants, this paper
discusses shifts in mine closure goals, describes risk assessment concepts and illustrates how risk assessment
can help meet those goals. From an environmental perspective, mine closure has typically focussed on mine
water release and reclamation. However, over the past decade, the mining industry, regulators and
stakeholders have been communicating more specific closure concerns related to exposure to mine
contamination. As well, First Nations have been more involved in making decisions and driving land-use goals
for the post-reclamation environment. Risk assessment is increasingly used to test whether closure plans will
perform related to environmental goals. This paper describes practices for conducting a problem formulation
in a mine closure scenario and provides a sample risk assessment framework for describing the process of
assessing risks of contaminant exposure to ecological receptors under current and post-closure scenarios.
The paper provides examples of risk assessment used to meet two mine closure goals: (1) protection of wildlife
from mine-related contamination in the post-closure landscape, and (2) suitability of post-closure conditions
for First Nations traditional land use. These examples underline the importance of planning ahead for mine
closure; engaging with regulatory agencies, First Nations and stakeholders; and ensuring that technical work
is defensible and will answer important risk-management questions. This paper is targeted at readers who
are new to risk assessment and its use in mine closure.
To meet regulatory requirements and reduce environmental liabilities, companies and governments
increasingly close mines formally after the resource is exhausted or no longer considered viable. Historically,
mine closure focussed on the physical attributes of a mine site such as sealing mine openings, managing
tailings dams and decommissioning mine buildings and infrastructure. Engineering risk assessment is used to
analyse fatal flaws and failure modes so that closure plans can document risks and mitigation measures.
From an environmental perspective, mine closure has typically focussed on mine water release (passive or
active discharge to receiving waters) and reclamation (restoring mines to equivalent capability). However, in
recent years, the mining industry, regulators and stakeholders have asked more questions about risks from
exposure to mine contamination. For example, where historic mine sites are orphaned and abandoned and
potentially contaminated, responsibility usually goes to the Crown. In that case, the Canadian Institute of
Chartered Accountant’s Public Sector Accounting Handbook’s PS3260 requires that environmental liabilities
be evaluated. Other presenters at this conference (Stewart, 2015; Healey et al., 2015) describe liability
assessment and risk management of contamination at mine sites in more detail.
For obvious reasons, metal mining operations are often situated where natural mineralisation is anomalously
high. Natural weathering of mineralised zones can lead to elevated background metal concentrations. When
this natural weathering is accelerated by years or decades of mining disturbances, persistently elevated metal
concentrations in soils, surface waters and aquatic sediments can result. In addition, production of minerals

Using environmental risk assessment to meet changing expectations for mine closure B.A. Power and R. Baker
for economic use involves a series of physical and chemical processes that may result in environmental
contamination. This may occur at any time, from excavation of the ore that contains the metal in mineral
form through to production of the metal in a marketable form.
Expectations for the environmental performance of post-closure mine landscapes are changing, and
environmental risk assessment is increasingly used to guide closure activities. The risk assessment process
evaluates the likelihood and magnitude of adverse effects to humans and ecological resources (e.g., plants,
invertebrates, fish and wildlife) as a result of exposure to chemical and/or physical stressors. Assessing the
risks of metals is challenging because they are naturally occurring substances and some of them are essential
for physiological processes.
This paper provides an introduction to risk assessment in the context of mine closure with a focus on mine
closure goals and the ways that risk assessment and risk management can help meet those goals. The
intended readers are not specialists in risk assessment but work in complementary fields and have an interest
in how risk assessment can contribute to mine closure.
Depending on when a given mine was developed and operated, it may or may not have a closure plan in
place. Ideally, the goals for mine closure will be clearly stated and relevant to the regulatory regime at the
time of closure; however, unless the mine is currently operating, this is often not the case. A mine may have
been in a care-and-maintenance phase for an extended period of time or it may be orphaned or abandoned.
In both these cases, an important first step is to clarify the goals for mine closure, including expectations
regarding future land use, site access by the public, reclamation and water quality. Determining such goals
should involve the applicable regulatory authorities, Crown agencies if the land is under a lease that is being
returned to the Crown, key stakeholders in the vicinity of the mine or affected by mine closure, and Aboriginal
and First Nations organisations.
Typical mine closure goals related to public safety are increasingly accompanied by goals related to
environmental protection, which can be significant drivers for mine closure. Mines are often in wilderness
settings, and the land is often returned to the Crown or sits within the traditional lands of local First Nations.
First Nations are increasingly influencing land-use goals for the post-reclamation environment, as was the
case for a case study presented herein for an historic mercury mine in the traditional territory of the Takla
Lake First Nation.
Risk assessments of mine sites often overlook links between the risk assessment design and the mine closure
plan and input from affected parties. These links are discussed later in this paper (see Section 3.2). Risk
assessment is a powerful tool for comparing the outcomes of different remediation scenarios and responding
to key questions from the public, regulatory agencies and members of local First Nations communities.
The practice of risk assessment can test whether closure plans will perform according to environmental goals.
The results of risk assessment can be used to support closure planning and to provide guidance on
management decisions about targeting areas for clean-up or remediation where undue risks are posed;
identifying areas to be left alone to continue to recover naturally; and helping agencies, First Nations and
stakeholders determine whether the resources they are concerned about will be protected.
This description of the methodology of environmental risk assessment at mines sites focusses on both
problem formulation, or “front-end” issues (Section 3.2) and the link between risk assessment and risk
management, or “back-end” issues (Section 3.3). Technical guidance on risk assessment for both human
health and ecological concerns varies to some extent across jurisdictions and in terms of level of detail and
complexity. Regarding human health risk assessment (HHRA) guidance, a screening-level risk assessment is

Risk Assessment
often not sufficiently detailed to support risk-management decisions where potential human health risks are
identified (Health Canada, 2010). A preliminary quantitative risk assessment is a simple quantitative risk
assessment that uses conservative assumptions so that risks are more likely to be overestimated than
underestimated. If risks are identified in a preliminary assessment, then more detailed approaches become
necessary to provide more accurate (realistic), defensible and representative estimates of risks. The
International Council on Mining and Metals has produced human health risk assessment guidance for metals
(ICMM, 2007).
Ecological risk assessment (ERA) guidance also varies by jurisdiction and can vary substantially depending on
when it was developed because the practice has evolved substantially since the early 1990s. Screening-level
risk assessment is generally not useful for mine sites; more quantitative and complex approaches tend to be
merited. There is some guidance for ERA of metals (e.g., USEPA, 2007; Fairbrother et al., 2007; ICMM, 2009),
and there is a rich body of literature regarding bioavailability of metals.
Regardless of the site or regulatory regime, problem formulation, which is more widely practiced in ERA than
in HHRA, is the important first step in risk assessment. Problem formulation clarifies the nature of issues
associated with contamination at a site and the method for addressing those issues. According to
Environment Canada (Government of Canada, 2012), problem formulation generally entails the following
elements:
 Description of the site management goal(s) and the specific assessment goal(s) of the risk
assessment: For example, if a site management goal is to reclaim a site as wild lands, the
assessment goal may be to assess whether there are potential effects under current conditions
relative to human land use and protection goals for wildlife that are agreed in consultation with
resource managers and First Nations.
 Review of the regulatory context for the site and the risk assessment, including applicable legal
instruments and policy: During closure, mine sites often fall under the jurisdiction of multiple
government agencies and legislation, and this overlap can be challenging. For example,
contaminated sites regimes, which are often designed for urban settings, are not designed for
application to mine site closure under-mine-related regulations. Similarly, management of
hazardous wastes in a mine setting presents complications.
 Review of existing site information, including ensuring that enough is known about the site to
support the risk assessment process: There is limited guidance on site investigation of
contamination at mine sites, and this often results in gaps that must be filled during the risk
assessment process. For example, site data may not reflect the mine closure plan, cover all
exposure pathways, or capture all the areas of potential concern from a contamination
perspective.
 Selection of contaminants of concern (COCs) and description of their characteristics that are
relevant to the ERA (i.e., transport, fate and effects): COCs are typically identified based on
analytical data for soils, sediment, surface water and groundwater and screened against the
jurisdiction’s applicable screening values (criteria, standards, etc.). Practitioners should consider
collecting reference or background data for groundwater chemistry early in this process, as this
can be useful to place site data in context.
 Consideration of COCs should reflect the activities that took place on site, including extraction,
beneficiation (e.g., milling and leaching) and mineral processing (e.g., smelting and refining): It is
particularly important to identify metals and metalloids that tend to increase in concentration as
they move up through food webs (e.g., mercury and selenium). Specific conditions that can
change the bioavailability of the contaminants (e.g., acid rock drainage affecting metals

speciation) should also be understood. Contaminants such as hydrocarbons from fuel storage and
spills may need to be evaluated in addition to metals.
 Selection of receptors of concern (ROCs) that could be affected by contamination and that will be
evaluated in the risk assessment: For human exposure, it is important to understand future site
use, particularly land use by First Nations at sites on traditional lands. For ecological receptors, it
is usually not feasible or necessary to include every possible species; therefore, a subset of
candidates are selected as surrogate ROCs. Special consideration may need to be given to listed
species if they are present on the mine site.
 Development of conceptual site model(s) that document the exposure pathway by which COCs
may come into contact with the receptors of concern: Examples of exposure pathways include
water and food consumption (for humans and wildlife) and direct contact (for invertebrates). A
conceptual site model can be expressed in a table, matrix, diagram or pictorial format. This should
be supported with text that cross-references the rationale used to select ROCs and exposure.
Because risk assessment is an iterative process, a conceptual site model should be updated as
more information becomes available to refine the problem formulation. The overall complexity of
the conceptual site model should be proportional to the complexity of the site. Of particular note
for mine closure, expectations for understanding mine site hydrogeology and geochemistry have
changed significantly in recent years. Practitioners should anticipate increased emphasis on
understanding the fate of mine water under closure conditions in the context of the conceptual
site model for both current conditions and into the future.
 Clarification of protection goals and associated acceptable effect levels (AELs): These relate to the
mine closure goals; typically, protection goals and AELs vary according to the intended land use
(e.g., human access) and/or by receptor (e.g., species at risk may be afforded a higher level of
protection than common species). Protection goals and AELs should be discussed with decision
makers early in the risk assessment process to build consensus. For mines, it is important to
discuss how background concentrations will be interpreted against site-related contamination.
Land use should be discussed in detail as small refinements can change the outcomes in the mine
closure plan (see Section 4.2).
 For ERA, development of lines of evidence for each assessment endpoint: Lines of evidence
specify how measurement endpoints will be used to evaluate potential risks. They combine
exposure and effects for each receptor group to describe risk. For example, a line of evidence
could include water chemistry (exposure) and toxicity through use of laboratory bioassays
(effects).
 Weight of evidence: The approach to assessing risk across the various lines of evidence should be
described to the extent possible; how will the various pieces of information that are collected for
combinations of receptors, media and pathways be synthesised into risk estimates? The planning
for weight of evidence analysis begins during problem formulation.
Overall, problem formulation should frame the issues, including the goals, context and nature of potential
effects; and it should design and plan an approach to assess risks that specifies the tools to be used and the
method for evaluating results.
An output of problem formulation is clarity about the information that must be collected and analysed to
meet the stated goals (the lines of evidence, as described above). During the analysis phase of risk
assessment, the data are collected and analysed. In many cases, risk assessment follows a tiered approach,
in which screening tools are applied at an early stage to determine whether further work is needed or to
prioritise future investigations. For example, a screening-level risk assessment might use conservative,
simplifying assumptions to assess risks. If risks are acceptable using conservative simplifications, there is

Risk Assessment
probably no need for further work. On the other hand, if potential effects are identified, a more detailed and
accurate risk assessment may be warranted. In the latter case, the return on investment tends to be greater.
For HHRA at mine sites undergoing closure, lines of evidence that may be important include frequency and
nature of site use, information about country foods (gained by sampling food and understanding
consumption by local communities) and bioavailability of metals in the soil (gained through physiologically-
based extraction techniques).
To evaluate risks to ecological receptors at mine sites, the following types of data are often collected to
support lines of evidence: data on concentrations of contaminants in environmental media (sediment, water,
soil/tailings); documentation of ecological status of different areas within the mine site (e.g., habitat
assessment, vegetation surveys); measures of exposure in food items (e.g., vegetation, berries, insects,
benthos, fish); modelling to predict risks to wildlife from exposure to metals from water and food; and
assessment of aquatic ecosystems (e.g., benthic community assessment, toxicity testing). Food chain
modelling is a key line of evidence for mine closure. This often uses site-specific data to estimate the exposure
(dose) of a metal that a receptor such as a mammal or bird might receive via its environment and food (see
Figure 1).
The dose is then compared with toxicity reference values (safe levels where harm is not anticipated, typically
derived from the literature), and a ratio is calculated (hazard quotient or HQ) (see Figure 2). When HQs are
less than one, practitioners can confidently conclude that risks are below harmful levels. When HQs are
greater than one, this usually means that they should revisit assumptions with site-specific data to determine
if those potential risks are real. During mine closure, metal concentrations can be extremely high in soils at
some mine features (e.g., ore concentrate handling areas), but those metals can be bound in the rock and
unavailable to biological organisms (i.e., low bioavailability). In other cases, metals concentrations may be
somewhat elevated but highly leachable; hence they can be very bioavailable and merit management
attention.

The food chain modelling approach can be quite sophisticated and can integrate site-specific factors including
spatial extent and magnitude of contamination, concentrations in environmental media, ingestion rate and
dietary preference, habitat quality and contaminant uptake efficiency. Importantly, where surface soils are
contaminated, post-closure risks can be predicted to evaluate performance of closure designs. An example
is provided later in this paper.
Aquatic risk assessments for mine closure usually focus on the fate and effects of metals, ions and sulphate
in the receiving environment. Toxicity testing, benthic community assessment, fish surveys and habitat
assessment can be important lines of evidence for aquatic ERAs. For mine closure, predictions of future water
quality (e.g., downgradient of mine and/or in pit lakes) can be used to predict risk under future conditions.
Risk assessment is a tool for decision-making but, ultimately, the need for risk management at any site
depends on judgments by risk managers about the acceptability of estimated risks and their associated
uncertainty (Hope, 2007). This means that frequent and clear communication between risk assessors and risk
managers is needed to ensure that results of a risk assessment are understood and useful from a
management perspective. To be effective, the risk assessment process needs to start early enough to
contribute to decisions made during development of the mine closure plan.

Risk Assessment
Conventional approaches to mine closure will typically address the following: geotechnical considerations
(i.e., tailings dam, slope stability); sealing of mine openings; management of mine water; demolition of
infrastructure and removal of land or landfilling; decommissioning of roads; and reclamation (i.e., site
preparation, seeding/planning and monitoring). Requirements to assess contamination on mine properties
vary by jurisdiction; however, expectations that mine closure will address contamination are increasing in
both scope and frequency. Whether to meet regulatory requirements or for due diligence, mining companies
are considering how best to address contamination as part of mine closure. With specific exceptions, it is
neither desirable nor practical to manage mine wastes through disposal (e.g., removal to landfill). Risk
management typically involves isolating mine wastes through the use of covers, which breaks the pathway
from contaminated soil to both ecological and human receptors. In these situations, it is also important to
evaluate leachability of mine wastes and to include this evaluation in the design of the cover (to control
groundwater and/or surface water contamination).
To assist risk managers in evaluating management options, it is helpful to depict the ERA conclusions for each
receptor group and area of the mine site in terms of both risk (i.e., the magnitude of contaminant-related
impacts) and uncertainty (i.e., the degree of confidence in the estimate of magnitude and cause of impact).
These conclusions, which represent common outputs of ERAs, can be overlaid on the range of likely
management options that might be considered under various scenarios for mine closure. Canada does not
have a formal decision-making framework for the acceptability of environmental conditions after mine
closure. However, these steps are common to effective risk-management planning in general:
 Site assessment (assessment of contamination) and risk assessment (ERA, HHRA)
 Site closure and remediation/reclamation planning
 Decision-making framework to identify risk-management options for each mine area based on
risks and uncertainties
 Risk-management plan
Some of the greatest challenges are in defining “acceptable risk” and in balancing competing interests
(National Research Council, 2009). Competing interests may be those traditionally identified with
sustainability (i.e., environment, economy and society) but they may also be technical (e.g., decommissioning
a tailings pond with a flourishing wetland may result in losing the wetland). This perspective informs the work
of the Cumulative Effects Management Association (Brocke and Ferster, 2010), which notes that “if meeting
one goal conflicts with meeting others, then the consequences have to be weighted.”
Ultimately, results of the risk assessment process may lead to four possible management outcomes for each
area of the mine site (e.g., mill site, tailings pond, waste rock):
 No further action: there is no risk and/or risks are acceptable (planned risk controls are sufficient).
 Risk controls and/or source removal: current risks are unacceptable and new/improved risk
controls and/or source removal are needed.
 Other solutions: current risks are unacceptable; however, risk controls and/or source removal are
considered impractical or may have greater impact than residual contamination.
 Further studies: residual uncertainty is considered too large for making sound risk-management
decisions.
The Pinchi Fault region, located in central British Columbia (BC), is naturally enriched in mercury because of
the abundance of cinnabar mineralisation in bedrock along several portions of the fault. The Bralorne-Takla

mine was developed to produce mercury for a short period of time during the Second World War. The mine
was subsequently abandoned, and wastes from mining operations remained on the site. As part of its
mandate to address potential human health and environmental impacts on Crown Land, the Crown
Contaminated Sites Program (CCSP) of the BC Ministry of Forests, Lands and Natural Resource Operations
identified the former Bralorne-Takla mine as one of its priorities. Accordingly, a series of environmental
studies was commissioned by CCSP; the risks to human health and ecological receptors (small mammals and
birds) were found to be unacceptable.
Working with the Takla Lake First Nation, the CCSP evaluated remediation options and chose an approach
that involves removal of the most contaminated (hazardous/leachable) material, covers for excavated areas
and other mine waste and risk management of residual contamination. Figure 3 shows a map of the site and
outlines the area being remediated. The remediation will result in clean surface soils over the entire area.
Risks to terrestrial receptors under post-closure conditions were evaluated using food chain modelling and
effects assessment as described above. Potential risks to most of the 30 wildlife receptors were negligible.
More detailed evaluation of risks using dose-response data (see Figure 2) was warranted for some receptors
(mostly small mammals), and in a few cases, potential risks were estimated to be more than negligible. The
ERA clearly showed that post-remediation/closure, there will be a substantial decrease in risks to wildlife,
particularly small mammals and small home-range birds. In addition, the mine site will be reclaimed to
optimise wildlife habitat. This example demonstrates the use of risk assessment as a tool to test mine closure
plans.
Figure prepared by SNC-Lavalin Inc. for Azimuth (2014); reproduced with client permission.
The Sä Dena Hes mine near Watson Lake, Yukon Territory, is being closed by Teck Resources Ltd. (Teck).This
zinc-lead mine operated for 16 months between 1991 and 1992. In 1993, the site was put into a state of care

Risk Assessment
and maintenance, and since 2000, the mine has been in temporary closure. The Sä Dena Hes mine is one of
the first mines to undergo formal closure in the Yukon with involvement of the mine owner.
The local First Nations, particularly the Liard First Nation, have a long history in the area of the mine. They
expressed a desire to use the site for cultural activities such as gathering berries or plants used in traditional
medicines. Members of the general public may also use the land at the mine site for recreational activities
such as hiking and hunting. During mine closure planning, an engagement process was designed and
implemented to understand how First Nation community members used the land and watershed around the
mine prior to mine development. Members of the community visited the site and worked alongside field
crews to evaluate current conditions. Community elders and resource harvesters were specifically involved
to inform and direct sampling locations. Samples of soil, water, berries and traditional medicines were
collected to measure metal levels. The data were used to estimate how much people could be exposed to
these metals if they spent time at the mine or consumed country foods or traditional medicines from the
mine following closure. The estimates were compared to the maximum levels of human exposure allowed
under environmental regulations.
In this instance, lead was the main concern. Since this metal is more toxic to humans than many other metals,
potential human health risk estimates based on lead exposure became a key determinant of the nature and extent of risk
management required. The results of the HHRA defined areas around the mine site that required remediation
or risk-management measures to eliminate or lower the potential for risks to humans from exposure to
metals. As a result, the mine closure plan included the conventional approaches described in Section 3.4, and it also
detailed where exposure pathways needed to be broken through the use of consolidation/covers and other risk
management measures. The final result will be a formally closed mine site that is both chemically and physically safe for
human use and that also protects the environment.
Expectations for evaluation of mine site contamination during mine closure are evolving, as are requirements
for remediation and visions for land use. Whether to meet regulatory requirements or for due diligence,
mining companies are considering how best to address contamination as part of mine closure. Risk
assessment is a proven approach to support remediation and risk management; it takes into consideration
both human health and the environment. The practice of risk assessment is increasingly being used to test
whether closure plans will perform related to environmental goals.
These two examples of risk assessment underline the importance of planning ahead for mine closure;
engaging with regulatory agencies, First Nations and stakeholders; and ensuring that technical work is
defensible and that it will answer the important risk-management questions.
Derrill Shuttleworth prepared the illustrations in the main body of the paper. Gregg Stewart of Crown
Contaminated Sites Program (BC Ministry of Forests, Lands and Natural Resource Operations) agreed to
provide information from the historic Bralorne-Takla mine site ERA. Teck Resources Ltd. (Michelle Unger) is
acknowledged for sharing the case study of Sä Dena Hes mine.
.
Brocke, L. and Ferster, R. (2010) Guide to the landscape design checklist in the Athabasca oil sands region, Cumulative Environmental
Management Association (CEMA), Fort McMurray, Alberta.
Fairbrother, A., Wentsel, R., Sappington, K. and Wood, W. (2007) Framework for metals risk assessment, Ecotoxicology and
Environmental Safety, Vol. 68(2), pp. 145–227.
Government of Canada (2012) Federal contaminated sites action plan (FCSAP) – ecological risk assessment guidance, Environment
Canada, Gatineau, Quebec.
Health Canada (2010) Federal contaminated site risk assessment in Canada, Part V: guidance on human health detailed quantitative
risk assessment for chemicals (DQRAChem), Ottawa, Ontario.

Healey, N., Power, B. and Kristoff D. (2015) An environmental risk assessment framework for developing financial liability estimates
for closing five abandoned metals mines in northern Saskatchewan, Canada. in Proceedings Mine Closure 2015, 1–3 June
2015, Vancouver, BC, A.B. Fourie, M. Tibbett, L. Sawatsky and D. van Zyl (eds), InfoMine, Canada.
Hope, B.K. (2007) What’s wrong with risk assessment? (editorial), Human and Ecological Risk Assessment, Vol. 13, pp. 1159–1163.
Hull, R.N., and Swanson, S. (2006) Sequential analysis of lines of evidence – an advanced weight of evidence approach for ecological
risk assessment, Integrated Environmental Assessment and Management, Vol. 2(4), pp. 302–311.
International Council on Mining and Metals (ICMM) (2009) Metals environmental risk assessment guidance (MERAG), International
Council on Mining and Metals, London, United Kingdom.
International Council on Mining and Metals (ICMM) (2007) Health risk assessment guidance for metals (HERAG), International Council
on Mining and Metals, London, United Kingdom.
National Research Council (NRC) (2009) Science and decisions: advancing risk assessment, Committee on improving risk analysis
approaches used by U.S. EPA, Board on Environmental Studies and Toxicology, Division on Earth and Life Studies, The National
Academies Press, Washington, D.C.
Stewart, G.S. (2015) British Columbia’s crown contaminated sites program – a ten year review, in Proceedings Mine Closure 2015, 1–
3 June 2015, Vancouver, BC, A.B. Fourie, M. Tibbett, L. Sawatsky and D. van Zyl (eds), InfoMine, Canada.
United States Environmental Protection Agency (USEPA) (2007) Framework for metals risk assessment, EPA 120/R-07/001, United
States Environmental Protection Agency, Washington, D.C., viewed 5 May 2015, http://www2.epa.gov/sites/production/
files/2013-09/documents/metals-risk-assessment-final.pdf.

D.J. Theberge Shell Canada Energy, Canada

E. Ferster Shell Canada Energy, Canada
The social performance standards for Canadian oil sands operators are under constant scrutiny from the
general public, non-governmental organisations (NGOs) and special interest groups, which has resulted in
heightened, intense, and often controversial media coverage both locally and internationally. With multiple
operating mines within close proximity in a regional area, cumulative effects management is a significant
issue. Through the Government of Alberta’s Land Use Framework, the Progressive Reclamation Strategy is
discussed as part of the Lower Athabasca Regional Plan, whereby initiatives are put in place that are intended
to improve security, clarity and environmental performance within the oil sands mining industry (Government
of Alberta, 2012). Included in the strategy, the Mine Financial Security Program is designed to provide
incentives for progressive reclamation, but also contains a mechanism for non-performance in the form of
financial penalties. The strategy also includes clarification of the reclamation certification program, as well
as the introduction of reclamation milestones for increased transparency. A recognised reclamation milestone
– termed “ready-to-reclaim” is described as “areas that are no longer required for mine or plant purposes and
are available for reclamation but where reclamation activities have not yet commenced” (Government of
Alberta, 2014b). Determining which lands within an operating oil sands mine lease meet the definition of
ready-to-reclaim is not always a straightforward exercise – especially in the early stages of the operation and
given the dynamic nature of open pit oil sands mining.
With rapidly evolving mine plans and regulatory policy, selecting an area to reclaim which will not be required
for future mine operations is challenging and difficult to accomplish with an objective and empirical approach.
The decision to reclaim a particular area must balance the risk of completing the reclamation against the risk
that the area is not truly ready-to-reclaim, due to some unforeseen circumstance occurring in the future. As
such, the intensity of progressive reclamation is dependent on either the company’s ability to accept, or
mitigate, risk. The challenge broadens when attempting to quantify risk, when often many of the
consequences are qualitative. At Shell, an internal standard project risk assessment process was modified and
adapted specifically for progressive reclamation planning in oil sands mining. Several iterations of the model
have occurred since it was first adopted as a tool to assist with reclamation planning – initially evaluating the
probability and consequences of the proposed reclamation area becoming re-disturbed. It has since been
expanded to a multi-step process where the proposed reclamation area is first evaluated against restrictive
planning criteria to generate the probability of the area being re-disturbed. Restrictive planning criteria were
introduced to reduce the inherent ambiguity in determining probability. The same area is then evaluated to
measure the consequences of the area being re-disturbed. Other developments include the introduction of
time as a risk mitigator. Development of the risk assessment process is iterative and expected to evolve as
improvements are realised and reclamation certification criteria are further clarified through the Government
of Alberta’s Progressive Reclamation Strategy.

A risk based approach to planning for progressive reclamation in the Canadian oil sands D.J. Theberge and E. Ferster
It is clear that due to the dynamic nature of mine planning in the Canadian oil sands there is a need for
adaptable and dynamic reclamation plans that manage the inherent risk of planning for, and execution of,
progressive reclamation. A systematic process was developed to determine which areas within the mine are
ready to be reclaimed. The process, termed the Reclamation Risk Assessment Process, allows the
organisation to determine its level of risk tolerance and is used as an input into the mine reclamation plan.
The Reclamation Risk Assessment Process – or RRAP – is meant to enable consistent, objectively derived
reclamation plans through introducing scaled restriction criteria which set the probability of a reclamation
area being re-disturbed in the future. The maximum restriction criteria values are then evaluated against the
maximum consequence to generate a risk rating. The adaptation to the RRAP allows for improved assurance
of mine reclamation plans and a more rigorous reclamation planning process. The RRAP is also useful to
identify focus areas to mitigate risk and in turn improve reclamation planning quality. Through continued
development and improvements, the RRAP has proven useful, as demonstrated through increased
transparency of reclamation plans, a more rigorous quality assurance process, and improved clarity in
reclamation targets and target derivation.
Understanding the resource, regulatory policy, physical environment, social pressure – internally, locally and
internationally – and the nature of oil sands mining operations is integral to balance interests in planning for
progressive reclamation. Oil sands operators have to balance the risk involved with all of the above issues,
as well as the internal risk, in planning for, and completing, progressive reclamation.
The oil sands, which are a mixture of sand, water, clay, and bitumen, represent 97% of Canada’s oil reserves,
giving Canada the third largest reserves in the world. There are three main oil sands deposits: Athabasca,
Peace River, and Cold Lake. Only a portion of the Athabasca deposit is within close enough proximity to the
surface to render surface mining practicable. The remainder of the Athabasca deposit, as well as the other
two deposits, must be recovered through in situ methods. The minable portion of oil sands reserves
represents only 20% of the total 167 billion barrels within oil sands reserves, which could impact
approximately 4,800 km2 of land (CAPP, 2014). The characteristics of the oil sands deposit are highly variable,
ranging in ore grade, proportion of sands and clays, thickness of overburden, chemical composition, and so
on. Under Directive 082, the Alberta Energy Regulator (AER) establishes four criteria that are used to
determine the volume of bitumen that must be recovered from the mineable resource – they are as follows
(AER, 2013):
 oil sands cut-off grade,
 minimum mining thickness,
 ratio of Total Volume to Bitumen in Place (TV/BIP), and
 processing plant recovery.
Of the four criteria, the first three may affect surface land use of mining operations and therefore progressive
reclamation. Depending on the deposit, surface real estate availability and flexibility could be limited.
Oil sands mining operations are located in the north-eastern part of Alberta, north of the urban service area
of Fort McMurray, within the Regional Municipality of Wood Buffalo. The area is within the Boreal Forest,
and is characterised by deciduous, mixed wood and coniferous forests amongst expansive wetlands
(Government of Alberta, 2012). There are many lakes and streams, which are part of the Athabasca drainage

Risk Assessment
basin. The relief is generally subdued, with level to gently rolling topography of glacial deposited surficial
geology.
Since the mineable oil sands are restricted to a specific geographic location, cumulative effects management
is a significant issue. There are seven mines operating within the mineable area, with imminent expansions
and new projects. Progressive reclamation enables increased rates of reclamation and a reduction of negative
cumulative effects over the life of the mining operations. Through the Environmental Protection and
Enhancement Act (EPEA) issued by the Government of Alberta, oil sands mine operators are required to
reclaim to “target the establishment of a self-sustaining locally common boreal forest, integrated with the
surrounding area” (Government of Alberta, 2007). Understanding the ecological functions of the locally
common boreal forest, as well as having the ability to reconstruct boreal forest landscape and watersheds to
those functions, are integral to the reduction of risk in reclamation plans.
Progressive reclamation has been built into regulatory policy at various levels, including frameworks, regional
plans and strategies. The Government of Alberta, under the Alberta Land Stewardship Act, introduced the
Land Use Framework (LUF) in 2008, which included a new approach to managing the Province’s land and
natural resources. The LUF includes the development of new land use regions where a regional plan would
be developed for each region. The Lower Athabasca Regional Plan (LARP) was released in 2012, which
includes the mineable oil sands area, north of Fort McMurray. The LARP is a strategic document giving
direction for the region over the period of 2012 to 2022, setting the stage for “robust growth, vibrant
communities and a health environment” over the next 50 years (Government of Alberta, 2012). The LARP is
reviewed every five years to ensure continued relevance. Public consultation is a component of the renewal
process. Some key objectives of the LARP include balancing economic development, environmental and
social considerations through a cumulative effect management approach, determining economic,
environmental and social objectives for the region, as well as strategies, approaches and tools to achieve the
desired objectives (Government of Alberta, 2012). A strategy included in the LARP, which is pertinent to the
discussion of this paper, is the Government of Alberta’s Progressive Reclamation Strategy.
One of the goals of the Progressive Reclamation Strategy is to ensure disturbed lands are reclaimed in a
timely fashion. As will be discussed further in Section 2.5, since the mines are quite large in expanse, with
most operating over many decades, it is not practicable to wait to complete all reclamation upon closure.
Furthermore, with multiple mines operating in the same area, cumulative disturbance is increasingly
significant. The Progressive Reclamation Strategy includes three main components: updating the reclamation
security policy; enhancing reclamation reporting; and clarifying the reclamation certificate program
(Government of Alberta, 2012).
The Mine Financial Security Program (MFSP) was established in 2010 by Alberta Environment and Sustainable
Resource Development (AESRD), but the management of the program now resides with the AER. The MFSP
process ensures security is in place to protect the public from costs associated with closure decommissioning
and reclamation, while balancing industry’s opportunities for responsible resource development (AER, 2014).
The MFSP adopts an approach to managing financial risks that recognises that the resource value associated
with an approved project is an asset in terms of the Net Revenue generated by its operations. Where an
Approval Holder has MFSP Assets at least three times greater than the MFSP Liability, is 15 years or more
from the end of its reserves, and is keeping current with its reclamation plans, additional financial security
above the base amount is not required. Where an Approval Holder has MFSP Assets less than three times its
MFSP Liability, or is nearing the end of the mine productive life, or is not meeting its approved reclamation
plans, additional financial security is required.
There are four types of financial security deposits, focusing on various potential risks in the lifecycle of a
mine, in the MFSP:

 Base Security Deposit (BSD) provides immediate funds for the government, if the Approval Holder
defaults, to maintain security and safety at the site until a new Approval Holder takes over or the
site is closed, plus an additional amount to address some of the residual risks of the program. If
no new Approval Holder takes over the site then the BSD can be used for abandonment,
remediation and surface reclamation of the site. The BSD is based on the sector, and reflects the
amount and complexity of work required to make the site safe.
 Operating Life Deposit (OLD) addresses the risks at the end of the mine’s life. The Approval Holder
is required to post financial security when there are less than 15 years of reserves left so that all
outstanding abandonment, remediation and surface reclamation costs are fully secured by the
time there are less than 6 years of reserves left. The OLD is offset by the BSD (i.e., the OLD
payable is only the amount it exceeds the BSD).
 Asset Safety Factor Deposit (ASFD) addresses the risks that an Approval Holder’s MFSP Assets fall
below a level deemed adequate to ensure that all MFSP Liability can be fully funded. The Approval
Holder posts financial security when the MFSP Asset to MFSP Liability ratio falls below 3.00.
Sufficient financial security is posted to bring the ratio back to 3.00.
 Outstanding Reclamation Deposit (ORD) addresses the risks posed by an Approval Holder
deferring reclamation. The Approval Holder posts financial security when they do not reduce
liability according to a reclamation plan approved by Alberta Environment.
Assets under the MFSP represent the estimated financial capability of an Approval Holder’s project to address
its future environmental obligations.
The three-year reclamation plans in the MFSP submission are aligned with the first three years of the 10-year
Mine Reclamation Plan (MRP), which is a requirement of the Environment Protection and Enhancement Act
(EPEA) Operating Approvals of all oil sands mines operators. The 10-year MRP’s are submitted every three
years. If operators fall short of completing the reclamation as per the approved MRP and MFSP, an
outstanding reclamation deposit may be triggered based on the shortfall of completed reclamation.
Completed reclamation is tracked through the annual Environment Reports, which are also a requirement of
a project’s EPEA Approval. To supplement the annual Environment Reports, a GIS based geodatabase
component is also completed. The information collected from operators for land use and reclamation
milestones is displayed on an interactive government website, which is available for review by the general
public. Operators now report on four broad-based categories under the “landcover class” designation, which
are: natural, cleared, disturbed and reclaimed. Each of the four landcover classes is then further subdivided;
for example, the “disturbed” landcover class includes categories called “ready for reclamation” and “soils
placed.” A particular land area can move from “disturbed – soil placed” into the “reclaimed” landcover class
once vegetation has been planted or ingressed. If there is potential for the area to be disturbed at some point
in the future, it is categorised as “reclaimed – temporary.” If not, and the area has been planted with the
closure prescription, it can then be termed “reclaimed – permanent.” Ultimately, operators must apply to
the Government of Alberta for the reclaimed lands to be certified; whereby the lands are returned to the
Crown once certification is received.
There are currently 104 ha of certified reclaimed lands within the mineable oil sands region – a reclaimed
overburden disposal area now named Gateway Hill (Government of Alberta, 2014a). In order for the area to
have received certification, years of reclamation monitoring are required to demonstrate that the reclaimed
lands met the equivalent capability objective. With changing opinions, new research and different
expectations of what reclaimed lands should look like, defining equivalent capability has become less clear
and more difficult to evaluate success. In July 2013, the Government of Alberta released a document entitled
“Criteria and Indicators Framework for Oil Sands Mine Reclamation Certification” designed to provide
direction and guidance related to oil sands mine reclamation and closure planning (Government of Alberta,
2013). The Criteria and Indicators Framework (CIF) was developed by the Cumulative Environmental
Management Association (CEMA), which is a multi-stakeholder association with representatives from local

Risk Assessment
First Nations, government and oil sands mines operators. With the publication and release of the CIF, the
Government of Alberta website noted that the Framework is not yet complete and requires further
development.
It is not hard to find oil sands mines in the media – either locally or even internationally. Typically, reports
carry a negative connotation. Social performance is one reason why companies strive to complete
progressive reclamation. Oil sands companies want to be recognized as leaders in reclamation and
sustainable developers of Alberta’s resources. Equally important as completing progressive reclamation is
ensuring the areas remain reclaimed and are not subject to re-disturbance – something that is not always an
easy task in the unique scale and dynamic nature of oil sands operations.
Oil sands operations are relatively large and expansive, with single mines occupying up to 200 km2 or more,
and can be in operation for many decades. When oil sands mines start operating they require a large portion
of land to support the operation. In addition to the mining areas, land is required for plant facilities, external
tailings storage facilities, overburden disposal areas, support infrastructure and reclamation material storage
areas. Once a significant enough portion of the pit has been mined and containment dykes have been built,
tailings deposition can move into the mined out pits, reducing the reliance on external tailings facilities.
Reclamation of tailings is a significant issue; once clays inherent to the ore body become suspended in tailings
solution following extraction of the bitumen, it could take decades, or even centuries, for settlement and
consolidation to occur if left on their own. Various new technologies have emerged in order to attempt to
accelerate the time required for fine tailings to consolidate and become ready to reclaim. Many of the
technologies have placed additional pressure on land use during operations. As discussed in Section 2.1,
depending on the geology, real estate may be even more constrained due to ore recovery and tailings
management requirements. Dynamic mine plans and changes in real estate demands often trigger additional
risk to reclamation efforts.
Shell Canada Ltd. (Shell) operates two oil sands mines north of Fort McMurray, Alberta. They are referred to
as Muskeg River Mine (MRM) and Jackpine Mine (JPM), and are operated by Shell on behalf of the Athabasca
Oil sands Project (AOSP), a joint venture between Chevron Canada Ltd., Marathon Oil Canada Corporation
and Shell Canada Ltd. MRM and JPM began operating in 2002 and in 2010, respectively. Of the overall
disturbance footprint, the majority of the land base continues to be utilised for operations at both mines. As
of December 31, 2013, a total of 324.1 ha of reclamation had been completed at MRM, with 158.0 ha as
temporary reclamation and 166.1 ha as permanent reclamation (Shell, 2013b). At JPM, a total of 57.1 ha of
reclamation had been completed, with all as temporary reclamation (Shell, 2013a). In recent years, the
corporate desire to increase reclamation and set performance targets had led to the requirement of a more
thorough and rigorous process for determining which areas should be targeted for reclamation. Furthermore,
reclamation planning at Shell lacked the objectivity required to properly quantify the risk of having to re-
disturb a reclaimed area at some point in the future.
Prior to the RRAP, selection of areas to be included in reclamation plans were mainly based on the physical
state of the landform with regards to construction to a given geotechnical design and future plans for the
landform given the mine and tailings plan. Some qualitative criteria was employed, but was typically only to
the interpretation of the reclamation planner and his/her experience. No formal process was in place to
derive the selection of areas that should be reclaimed in a given year of the plan. It was recognised a more
formal process was required for assurance of the plan. Issues also arose when reclamation plans were being
adapted in order to satisfy a predetermined reclamation target, even if there was some question of the

potential for the given area to be re-disturbed at some point in the future. It is not uncommon in the oil
sands, in some capacity, for previously reclaimed area to be re-disturbed. Re-disturbance may occur for many
reasons, including: change in mine plan and mine progression, geotechnical reasons, changes to dump and
dyke plans, changes to regulatory requirements, accidental disturbance due to close proximity to operations,
and so on. Reclamation plans must be robust enough to adapt to uncertainly in a dynamic planning
environment.
A Shell standard project risk assessment matrix (Figure 1) was adapted to satisfy the requirements, conditions
and constraints of oil sands mine reclamation planning. The basis of the risk assessment matrix is to measure
the consequence and probability of some event or incident occurring and to generate a risk rating.
Consequence types are defined into further categories, whereas probabilities are defined by a percentage.
Risk ratings increase in severity from small, material, severe and critical as seen in Figure 1. Early in the
development of the RRAP, for when determining reclamation areas, it was decided that the event being
measured would be the risk of the reclamation area in question being re-disturbed at some point in the
future. This approach was employed since it was considered an ultimate event (worst-case scenario) when
using the RRAP to determine reclamation risk.
PROBABILITY
0-5% 5-20% 20-50% 50-80% 80-100%
Very Low Low Medium High Very High
Very
CRITICAL
High
High SEVERE
IMPACT OF CONSEQUESCE
Medium MATERIAL
Low SMALL
Very
Low
Following the adaptation of the standard project risk assessment matrix, there were some issues with the
consequence categories. Since the model had been adapted from what was intended for a different use, it
became clear afterwards that the consequence categories had not been sufficiently adapted to reclamation

Risk Assessment
planning. As such, the sensitivities and technical requirements for reclamation planning were not adequate.
Consequence categories in RRAP were later revised, as discussed in Section 3.2.2. Earlier versions of the RRAP
also lacked objectivity and ability to quantitatively determine the probability of a consequence occuring.
Although the probabilities were quantitative, the derivation was not. The probability was mainly determined
through expert opinion or experience of the planners.
Through internal stakeholder consultation, it was identified that earlier versions of the RRAP also lacked the
ability to determine whether there was sufficient information, or if field conditions were such that if the
reclamation was completed in the specified period, it would be to the required standard of quality. To
address the concern, the risk was differentiated by either being a physical risk or functional risk. Physical risk
was defined as re-disturbance required due to some factor external to reclamation planning, such as changes
in mine plans, geotechnical issues, and so on, whereas functional risk was defined as re-disturbance required
due to some factor relating to the quality of reclamation. Functional factors such as insufficient design
consideration for closure drainage, inability to meet reclamation objectives, contamination, and so on, may
lead to re-disturbance for remediation. Examining physical and functional risk separately resulted in nearly
duplicated work and the potential requirement for two separate models. Consequently, a decision was made
that only physical risk would be evaluated until the RRAP could be revised to be more robust by including
functional risk component.
Since the earlier versions and trials, the RRAP was further adapted to satisfy requirements of reclamation
planning since, as described previously, it was found to be insufficient due to the inherent ambiguity in
determining the probability without objective or empirical means. It was also determined that the
consequences would need to be redefined to be more applicable for reclamation. An additional shortcoming
was its failure to consider functional risk. As the criteria within the model evolved and its purpose was better
defined, the concept of using time as a risk mitigator was employed. Figure 2 shows the current version.

Through working with the RRAP, a significant flaw was identified with regards to the determination of
probability along the x-axis of the matrix. A concept was tested where rather than using a probability of re-
disturbance to set the restriction along the x-axis, reclamation planning criteria was developed and
introduced to set the x-axis restriction. Probability was then replaced by what is termed restriction criteria.
After several revisions, a total of six categories of restriction criteria were determined. The development of
restriction criteria also enabled the RRAP with the capacity to consider functional risk. The restriction criteria
are defined below in Sections 3.2.1.1 through 3.2.1.6.
For the purposes of this document, landform is defined as the structure on which reclamation is planned, as
identified in the closure plan. The landform construction status refers to only that portion of the landform
that is being planned for reclamation. Criteria are quantified by determining if that portion of the landform
has been constructed, or when completion is expected. Factors such as if construction is completed to design,
or any outstanding issues such as overbuilds, operational surface drainage, geotechnical concerns, and so on,
on parts of the landform, which could affect the reclamation and therefore may be restricting reclamation,
are all considered. The landform construction criteria is meant to determine the point at which a particular
area is considered physically ready for reclamation. If the landform on which the reclamation area sits is not
constructed to completion, there is much higher potential of re-disturbance of the proposed reclamation
area. These criteria are primarily concerned with construction and landform status.
Operational status criteria is meant to determine whether the landform is required for current or future
operations and which portions of the landform are required for those operations. This criteria is somewhat
linked to the landform construction status criteria; however, is intended to increase or decrease the risk,
depending on how much of the landform is involved in operations. Typically, increased operations leads to
increased risk.
The reclamation criteria restriction is best suited to address functional risk in the overall RRAP, since it
includes broad subjects such as target ecosite and ecosite phases, soil and vegetation prescriptions, closure
drainage, reclamation unit criteria, geomorphic design, coarse woody debris, wildlife habitat, and so on. The
level of reclamation criteria restriction is assessed by the level of knowledge, clarity and understanding of
reclamation criteria and ability to achieve desired reclamation outcomes within the reclamation area in
question. It can be linked to closure plan criteria, so that there is transparency between reclamation
objectives, risks and outcomes. Reclamation criteria likely presents opportunity to develop a more robust
RRAP model since there may be opportunity for improvements with increasing understanding of reclamation
criteria required for certification.
Major reclamation issues are defined as critical flaws of an overall reclamation plan, such as material balance
deficiencies, and tailings capping design. The degree of restriction on the risk assessment is related to the
degree that major reclamation issues are affecting the specific proposed reclamation area. For example, if
the reclamation area is situated on an overburden disposal area, tailings capping design issues are of little
relevance.
Reclamation area size criteria was selected as a quantitative measure to remove ambiguity. While not
consistently the case, smaller reclamation areas are more likely to be re-disturbed. The reclamation area size
criteria is intentionally inversely related to the cost consequence. Therefore, risk associated with a small or a

Risk Assessment
large area will always be higher. This is intentional so that cost and area size will drive behaviour in planning
to avoid small fragmented areas and/or to ensure reclamation criteria is known for large areas.
Understanding that for some large areas, this would force risk to almost always be at an unacceptable rate,
the RRAP does not evaluate the risk involved with not reclaiming a given area. In which case, there is
mechanism for a case specific exemption to the risk rating. The exemption may also be linked with some
mitigation strategies.
The regulatory status criteria is a measure of the level of regulatory submissions and approvals granted for
the particular proposed reclamation area. This might range from closure plans to annual plans or the MFSP.
The selection and definition of consequences of re-disturbing a reclamation area was given careful thought.
Several iterations were completed to ensure that the consequences worked well with the restriction criteria.
If the restriction criteria or consequences were not aligned or improperly selected, the resulting risk rating
may not be appropriately representative. This relationship will likely have opportunity for improvements in
the future once it has been more rigorously tested. A key assumption was introduced where any re-
disturbances, which would theoretically occur would be done so while be compliant with regulations.
Therefore, a regulatory consequence would not be required to be evaluated against. The following
consequences are measured in the RRAP; environmental, cost and reputational.
Since the conception of the RRAP, the “environment” consequence underwent the most change. Having been
built off the standard project risk assessment matrix, the environmental consequence was initially developed
from the more generic conventional sense of the oil and gas business, and it was not sufficient for mining
since it focused on spills and contamination. The consequence is measured by the size of its environmental
effect; therefore, since the consequence of an area being re -disturbed is being measured, defining
environmental consequence was somewhat difficult. The consequences vary depending on the age of the
reclamation, the proximity to the natural off-lease environment, level of use by wildlife, and so on. In order
to increase the ability to quantify the consequence, a “size of re-disturbance” component was assigned,
which is based on the assumption that the larger the re-disturbance, the larger the environmental effect.
This also enables the consequence to be inversely related to the reclamation area size restriction criteria.
However, a smaller area of re-disturbance of wetlands would create higher environment effect than a larger
re-disturbance of upland ecosites. Certain qualitative measures were also included such as loss of wildlife
habitat and risk of contamination from adjacent areas used by operations.
As stated in Section 3.2.1, the cost consequence is inversely related to the reclamation area size restriction
criteria. The cost consequence is based on the costs to salvage the reclamation soils from the reclaimed area
as well as the costs to reclaim the area again in the future. This represents the potential financial loss to the
operation if a reclaimed area is re-disturbed in the future.
Considering the social pressures tied to operations in the oil sands, reputational consequence is a valid
measure of re-disturbance risk for a reclaimed area. It is noteworthy that there were only little changes to
the reputational consequence assessment through the revisions of the RRAP. The consequence is measured
by the level of impact, either internal or external.

Another important component of the RRAP is the concept of using time as a risk mitigator. Since many of the
future reclamation areas are not built, and therefore not considered “ready-for-reclamation,” it is necessary
to assume that when they are built they will then become “ready-for-reclamation” and reclamation activities
can proceed. Reclamation plans beyond five years are considered conceptual, and therefore are considered
a “small” risk. The RRAP is intended for reclamation occurring within the first five years of the current
operations plan. Where the risk was determined to be too high in a certain year, it is assumed that if the
reclamation is deferred to a later point in the plan, the risk level should then be acceptable since the
associated risk was reduced to an acceptable level to be included in the plan. If for some reason the risk had
not been sufficiently reduced, the areas are reassessed in the next annual plan.
Targets are useful to show goals, objectives and business ambition, and to align annual performance goals
with business goals. Robust goals should be developed that are specific, measurable, achievable, relevant
and time-framed. The RRAP may be useful in target setting since it allows the Business to set its level of risk
tolerance as an input to reclamation plans. Output from the reclamation plans will help ensure the
development of robust and objectively derived goals and targets. Opportunity for improvement of
reclamation planning process then comes from the RRAP’s ability to identify the high-risk components of a
proposed area so that appropriate mitigation strategies can be put in place. Once strategies are put in place,
improvements are continually made through multiple iterations. If successful in risk mitigation strategies,
and the business’ risk tolerance remains equal, overall reclaimed area in reclamation plans should increase
with more iterations.
Based on the previous methodology, the RRAP model is utilised as follows:

 Proposed reclamation polygons for the first year of the plan are determined based on as built
relative to final design, previously completed reclamation areas, future mine plans, internal and
regulatory commitments, and so on.
 Proposed reclamation polygons are evaluated against the six restriction criteria categories in
Section 3.2.1. A restriction rating (1 – 5) is assigned and input into the RRAP model.
 The proposed reclamation polygons are then evaluated against the three consequence categories
in Section 3.2.2 and a consequence rating (1 – 5) is assigned and input into the RRAP model.
 Using the restriction criteria rating (1 – 5) and the consequence rating (1 – 5), the RRAP model is
utilised to derive the risk of completing the reclamation in the first year of the plan. The model
will calculate the risk and assign the severity based on the location within the model matrix.
 If the risk is found to be too high for what the Business has approved, appropriate mitigative
measures are used to decrease the risk – likely through deferral of initiating reclamation in the
proposed area until such point the risk is deemed acceptable.
 Reclamation areas are adjusted in the reclamation plan to reflect the outcome of the RRAP model
mitigation requirements.
The RRAP currently includes process flow methodology, disclaimer and intent, definition of terms and a
model worksheet. The model worksheet essentially contains all the data for the reclamation areas within the
reclamation plan. A description of the proposed area is summarised, and based on the process flow and
methodology, the maximum consequence and restriction criteria rating generate a risk rating. Should the
area not meet the minimum risk requirements, a mitigation strategy is assigned and a new rating is

Risk Assessment
determined based on the mitigation strategy. The mitigation strategy may include a deferral date for when
mitigation strategy completion is expected. The prospective reclamation area is then be included in the plan
according to the date identified in the mitigation strategy. There is an opportunity for environmental
performance improvement through commitments to complete the recommendations of the mitigation
strategy. Ultimately, if risks are mitigated, the likelihood or re-disturbance in minimised.
Since the RRAP is being used as a planning tool, it is important that it be reviewed regularly to ensure
continued improvement and applicability. Ideally, a solid foundation has now been developed and changes
will be mainly technical rather than functional. Revisions should also consider impacts to reclamation plans
and regulatory commitments. It is recognised that the RRAP is just one tool, and to be effective, it must be
built into a planning process and framework. Future work will develop any support processes to make
improvements on the RRAP.
Oil sands mining is a complex, dynamic and expensive process, conducted primarily in remote areas under
very challenging geographic and climatic conditions. In addition to these operational challenges, operators
must contend with a complex regulatory regime and address challenging social performance issues from
NGOs, the general public and special interest groups. Reclamation activities at oil sands projects face the
same challenges and scrutiny. When it comes to reclamation, the pressure on oil sands operators to “get it
right – the first time” is intense, leaving little room for error or costly mistakes. The need for reclamation
planners to use every available tool at their disposal to ensure successful reclamation is very important and,
in fact, necessary for its accomplishment.
The RRAP presented in this paper is just one tool to help reclamation planners at Shell to “get it right.”
Experience to date has shown that the lack of a systematic approach to managing risk associated with
planning reclamation was a major deficiency in the planning process. The RRAP, even though fairly new to
Shell’s reclamation planning process, has proven to be a very valuable tool by quantifying risk, identifying
focus areas to mitigate risk and improving the QA/QC and target-setting process. It notably uses time as a
risk mitigator, which provides flexibility when building progressive reclamation plans.
As reclamation techniques become more robust and reclamation planners gain more experience, it will be
important to consider whether the model should be refined or modified to remain instructive to the
process. However, this study demonstrates that maintaining some form of risk-based assessment is likely to
add significant value to the reclamation planning process
AER (2013) Alberta Energy Regulator, Directive 082: Operating criteria: resource recovery requirements for oil sands mine and
processing plant operations, February 13, 2013, viewed 21 January 2015, http://www.aer.ca/documents/directives/
Directive082.pdf.
AER (2014) Alberta Energy Regulator, Mine Financial Security Program (MFSP), Abandonment and Reclamation, MFSP, viewed 20
January 2015 , http://www.aer.ca/abandonment-and-reclamation/liability-management/mfsp.
AESRD (2013) Alberta Environment and Sustainable Resource Development, criteria and indicators framework for oil sands mine
reclamation certification. Prepared by Mike Poscente and Theo Charette of Charette Pell Poscente Environmental
Corporation for the Reclamation Working Group of the Cumulative Environmental Management Association, Fort McMurray,
AB. September 18, 2012.
CAPP (2014) Canadian Association of Petroleum Producers, upstream dialog, the facts on: oil sands Feb 2014, viewed 15 January
2015, http://www.capp.ca/getdoc.aspx?DocId=242473&DT=NTV.
Government of Alberta (2007) Alberta Environment, Environmental Protection and Enhancement Act Approval for the Muskeg River
Mine Expansion, Approval No. 20809-01-00, Construction Operation and Reclamation of the Musket River (Lease 13,30 and
90) Oil Sands Processing Plant And Mine, November 2007, 54 pp.
Government of Alberta (2012) Alberta Environment and Sustainable Resource Development, Lower Athabasca Regional Plan 2012-
2022, August 22, 2012, 26 pp.

Government of Alberta (2013) Alberta Environment and Sustainable Resource Development, criteria and indicators framework for
oil sands mine reclamation certification, July 2013, 1 p.
Government of Alberta (2014a) Alberta Environment and Sustainable Resource Development, Focus, Oil Sands Mining Development
and Reclamation, Reclamation Reporting, Definitions, March 5, 2014, viewed 15 January 2015,
http://www.oilsands.alberta.ca/reclamation.html
Government of Alberta (2014b) Alberta Environment and Sustainable Resource Development, Oil sands, Alberta’s Clean Energy
Future, Reclamation view on January 22, 2015, viewed 22 January 2015, http://esrd.alberta.ca/focus/state-of-the-
environment/land/response-indicators/oil-sands-mining-development-and-reclamation/reclamation-reporting.aspx.
Shell (2013a) Shell Canada Limited, Jackpine Mine Annual Report, submitted to the Energy Resource Conservation Board, Approval:
8512E, 2013, 25 pp.
Shell (2013b) Shell Canada Limited, Muskeg River Mine Annual Report, Submitted to the Energy Resource Conservation Board,
Approval: 8512E, 2013, 29 pp.

N. Lower Palmer Environmental Consulting Group, Canada

M. Quach Palmer Environmental Consulting Group, Canada
R. Marsland Marsland Environmental Associates, Canada
Mine waste can pose a risk to the natural environment, especially if on-site closure conditions change. At
remote locations in particular, such sites can be viewed as an unknown risk by regulators, and mining
companies are often asked to monitor or treat an unknown risk or to develop management options to reduce
them. This paper presents a case study for a historic mine site in northwest British Columbia. Discharges from
the mine portals and waste dumps are entering the river, which flows into southeast Alaska. Aquatic risk
assessment was used to evaluate the potential for contamination of aquatic biota due to changes in water
quality. To complete the risk assessment, scientists studied water quality upstream and downstream from
mine site discharges and completed a detailed review of historic data. The team identified cadmium, copper,
lead and zinc as the contaminants of potential concern and assessed the discharges’ impact on a variety of
fish species including coho salmon, sockeye salmon, Chinook salmon and bull trout. Although significant levels
of copper and zinc were found downstream from the mine, the drainage was assessed to pose a low risk to
fish in the river. The potential risks to fish were measured using a hazard quotient (HQ): if this is greater than
one, it can indicate unacceptable risks to fish and other aquatic life. The study concluded that, regardless of
whether a water treatment plant were to be operating or not, the HQs were less than one for the majority of
the year, including during critical time periods for fish migration and spawning. The study identifies data gaps
on the ecology of the river, for example, the use of the river by fish during the winter and spring thaw; if
needed, additional data collection could be targeted to supplement the aquatic risk assessment. This case
study showcases an approach to evaluating ecological risk at remote locations and guiding focused data
collection and management options. This demonstrates the need to implement treatment and management
strategies that are appropriate to reduce off-site impacts.
Mine sites can pose a risk to the natural environment, particularly if on-site conditions change and discharge
from mine waste materials enters adjacent natural waterbodies. It can be difficult to accurately assess the
degree of risk of effects to the natural environment, particularly in remote and northern sites where
extensive monitoring can be time consuming and data can be difficult to collect. It can also be difficult to
determine a focused and effective mitigation plan if the risks are unknown or too general in nature.
At a remote historical mine site in northwest British Columbia (BC), effluent from the old mine portals and
waste rock piles converges and drains into a river, which ultimately flows through southeast Alaska. In an
attempt to reduce this effluent loading, a temporary water treatment plant (WTP) was operated on-site for
six months. The plant was originally envisaged through an environmental assessment (EA) commitment to
provide an interim treatment facility for the incremental loading that might occur as a result of relocating
potential acid generating (PAG) historical waste rock to facilitate construction as part of a new mine project.
Operating costs were significantly higher than anticipated for this WTP, and design parameters were not met.
As part of the WTP commissioning, a monitoring and surveillance program was developed to monitor the
effects of the WTP on the receiving environment. This monitoring program continued even after the plant
was shut down. An aquatic ecological risk assessment (AERA) was subsequently conducted to specifically

The role of aquatic ecological risk assessment to guide effluent management from mine waste N. Lower, M. Quach and R. Marsland
address the potential for impacts of this historical mine discharge to fish in the receiving waterbody. This
AERA also compared the risks between the period when the WTP was in operation and when it was not.
The AERA combined historic literature with current water-quality data of the mine effluent and the receiving
environment. Surface water-quality data was available on a mainly monthly basis from 2008 to 2013. This
period included both the operational and nonoperational periods of the WTP. Data on sediment and benthic
invertebrates in the receiving waterbody were limited. The hydrologic regime of the receiving waterbody
(seasonal major glacial outbursts) precluded collecting stable baseline datasets on the benthic invertebrate
community. Therefore, the AERA focused on fish, in particular salmonids, which are prevalent in the receiving
water body, considered to be the most sensitive fish receptor in the known fish community and are of
significant cultural and economic importance in the local area.
The AERA followed a tiered approach using the steps below:

 Document relevant receptor information (for example, fish community and life history data).
 Compile available water-quality data.
 Identify water-quality screening levels and determine constituents of potential concern (COPCs).
 Estimate environmental concentration (EEC) for each COPC.
 Determine appropriate toxicity reference values (TRVs) specific to the receptors of concern and
exposure media.
 Characterise potential adverse effects associated with each COPC.
 Characterise risk through estimating hazard quotients (HQs) and considering uncertainties.
The specific approach for the risk assessment was as follows.
Water quality screening levels were derived for the identification of COPCs through consideration of
provincial and federal criteria, and with consideration of background water quality (i.e., water representing
regional levels). Where background levels were greater than regulatory criteria, the 90 th percentile
background value for each relevant parameter was selected as the COPC screening level. This is consistent
with the BC Ministry of Environment guidance for completing risk assessments and was considered a
conservative approach, given that there were generally 47 or more monitoring events available for the
background parameters.
Water quality for the monitoring location nearest to the point of discharge to the river (site exfiltration pond
SE-2) was selected for the identification of COPCs that would be considered in the aquatic receiving
environment with respect to their potential to cause adverse effects in fish.
Summary statistics and trend graphs were generated for the measured COPCs at each of the monitoring
stations located within the river to gain an improved understanding of COPCs within the aquatic receiving
environment and to form the basis for fish exposure levels. The EEC for each COPC at each receiving
environment location was the 90th percentile value for the available data sets.
Project-specific TRVs were derived by considering published toxicity literature specific to freshwater fish.
Salmonid data was prioritised, in keeping with the project scope. The exposure pathway was from direct
surface water contact.
Risk estimates were generated in the form of hazard quotients (HQs) for each COPC at each of the three
monitoring locations in the receiving environment, for both dissolved and total metals. HQ results greater
than one indicate that the potential for unacceptable adverse effects exists, and this suggests that more
refined consideration may be warranted to reduce uncertainty and/or mitigate the risk. The HQ is calculated
by dividing the estimated environment concentration (EEC) by a single point toxicity reference value (TRV).

Risk Assessment
The collective HQ results, particularly those with results greater than one, were further considered prior to
forming a conclusion with respect to mine discharge and its potential to adversely affect fish as a whole in
the receiving waterbody.
In order to come up with a level of potential risk based on the HQ, the Recommended Guidance and Checklist
for Tier 1 Ecological Risk Assessment of Contaminated Sites in British Columbia (Government of British
Columbia, 1999) was used. According to this document, if a receptor is known (or assumed) to be present, it
is categorised as follows:
 Low risk is when HQ < 1.

 Moderate (intermediate) risk is when1 < HQ < 100.
 High risk is when HQ > 100.
Further details on these three levels of risk are provided by the Science Advisory Board for Contaminated
Sites in British Columbia (2008):
 Low risks: Implies that adverse effects are likely not present based on the totality of data available.
Low risk differs from the term negligible risk in that the former designation is more appropriate for
situations where the conclusion is based on the balance of probabilities;
 Moderate (or intermediate) risks: Implies that some degree of adverse effects are likely, based on
the totality of data available. Risk estimates suggest that risk management or remediation is
necessary, unless further refinement of the risk estimate is conducted
 High (or severe) risks: Implies that adverse effects are likely based on the totality of data. Risk
estimates suggest that risk management or remediation is necessary and that this conclusion is
unlikely to change even if further refinement of the risk estimate is conducted.
Aquatic monitoring data in the area was available from 1988, and more intensively from 1994 as part of an
environmental assessment (EA) certificate. Various environmental effects monitoring (EEM) studies were
also conducted as a follow-up to the EA and for preconstruction activities. Data was available not only on
water quality but also on fish health and contaminant loadings.
The receiving waterbody from the mine site is one of the major tributaries to the Taku River. The Taku River
is a transboundary river originating in northwest BC and flowing 266 km before emptying into the Taku Inlet
just south of Juneau, Alaska. The Taku River commercial salmon fishery is vital to the traditional and
subsistence-based lifestyles of the Tlingit people. Throughout much of the year, the local hydrograph is snow
and glacial melt-driven. However, on at least one occasion per year, the river is subject to extreme flood
surges from a glacier-impounded lake that drains quickly (jökulhlaups). During such events, the water levels
can rise over a period of 24 to 48 hours with flows > 2,000 m3/s, compared to typical maximum daily flows
of 300 – 400 m3/s. Near the confluence with the Taku River, the receiving waterbody has a catchment area
of 781 km2, and approximately 42% of this is covered in glacier. The mean annual discharge was estimated
to be 39 m3/s, with the lowest monthly flow in March or April. Within the study area, the receiving waterbody
is considered as one reach, with no obvious break based on gradient or other important hydraulic or habitat
features.
Comprehensive fish and fish habitat data was available for the site from work completed between 1989 and
2007. The site and receiving waterbody was therefore considered to be well-characterised. The receiving
environment supports several nonsalmonid species (e.g., stickleback and sculpin) and up to nine salmonid

species, including all five Pacific salmon species (Oncorhynchus sp.), anadromous and resident Dolly Varden
(Salvelinus malma) and bull trout (S. confluentus), steelhead/rainbow trout (O. mykiss) and cutthroat trout
(O. clarki). Within the receiving environment watershed, juvenile Coho salmon (O. kisutch) and Dolly Varden
were the most common and ubiquitous species captured during monitoring programs.
Fish habitat quality throughout a large portion of the floodplain is limited by elevated turbidity during the
open water season and extreme turbidity and flow during the seasonal glacial outburst floods. The receiving
waterbody is primarily used as a migration corridor and only provides temporary refuge habitat for salmonids
and other fish species. However, there are well-defined, clear-water (groundwater upwelling) side channels
downstream of the mine site, which provide important overwintering, spawning and rearing habitat for both
resident and anadromous fish.
The risk assessment focused on the four most common and abundant species in the receiving waterbody:
Chinook salmon (Oncorhynchus tshawytscha); coho salmon (O. kisutch); sockeye salmon (O. nerka); and Dolly
Varden/bull trout (Salvelinus malma/S.confluentus). Chinook salmon are not ubiquitous in the immediate
receiving environment, but likely use the Taku River confluence and therefore may occur within the zone of
influence. All key receptor species are of economic, cultural and/or recreational importance to marine and
freshwater fisheries. Juveniles from all these species spend at least 18 months in freshwater before migrating
to the ocean. Dolly Varden/bull trout comprise both anadromous and resident forms. Salmonid species are
known to be sensitive to environmental or chemical perturbations and therefore are suitable receptor
species that conservatively measure potential risks from water quality. Figure 1 provides the residence
timings of the selected receptor species.
Acid rock drainage (ARD) has been leaching from the mine site since operations ceased in 1957. The major
metals in the PAG waste rock are aluminium, calcium, iron, magnesium and sodium. Field bin testing showed
leachate from the PAG waste rock is acidic with < pH 3.5 and with elevated zinc and copper concentrations.

Risk Assessment
Since the shutdown of the WTP in 2012, the discharge from the historic portals has been directed to the site
exfiltration pond, along with site runoff, waste rock runoff and neutral mine water discharges.
Available receiving environment water-quality data for samples collected between June 2008 and July 2013
was used in the risk assessment. Specifically, four stations were established: station W10 is 4.5 km upstream
of mine site discharges (background); station W46 is downstream of WTP effluent discharges; station W51 is
325 m downstream of site exfiltration pond drainage; and station W32 is 2.7 km downstream of site
exfiltration pond. Station W10 is representative of water-quality conditions upstream of the mine discharges.
Stations W51 and W32 represent the water quality downstream of the mine discharges. W46 is located
directly downstream of the WTP discharge but upstream of the site exfiltration pond (SE)-2.
As part of a previous EEM study, monthly samples were collected at these stations between June 2008 and
February 2009, and then additional monitoring was conducted as part of the WTP discharge permit. From
August 2012 to July 2013, stations W10 and W32 were sampled biweekly from August to December, and
then weekly from January to June 2013, for a total of 37 samples from each site. Samples were taken from
stations W46 and W51 weekly from September 2012 to July 2013.
Monitoring of the receiving waterbody includes analysing water samples for total and dissolved metals and
specified physical parameters: pH, conductivity, turbidity, total suspended solids, hardness and alkalinity.
Station W10 is representative of water quality conditions upstream of the mine discharges, and stations W51
and W32 represent the water quality downstream of the mine discharges. Concentrations of copper and zinc
remain relatively stable at station W10; however, the concentrations at stations W32 and W51 increase as
flow diminishes over the winter, then peak in late April (due to snowmelt and the first flushing of the waste
rock combined with the portal discharges into SE-2), and then decrease again with increased river flows.
During the ice-free periods, the concentrations of total and dissolved metals are similar between station W10
upstream of the mine and station W32 downstream of the mine. Even station W51, immediately downstream
of the mine, is comparable to stations W10 and W32. There appears to be enough dilution during high flows
that mine effluent concentration in the receiving waterbody is vastly reduced, even without the WTP
operation.
There is good correlation between W32 and W51, which reflects the dilution occurring on average of about
20 to 50-fold. Table 1 summarises the dilution rations from W51 to W32 for copper, cadmium, zinc and lead
during prefreshet (April 20 – May 4, 2013), high flows (September 12, 2012 – January 5, 2013; May 11 – July
20, 2013) and low flows (January 19, 2013 – April 13, 2013). Dilution ratios were based on median
concentrations for each parameter for each station. The dilution ratios are lower during high flows, as
compared to low (including prefreshet) flows because abundant, rapid dilution is available, even between
the mine and W51.

Parameter Prefreshet High flows Low flows

Dissolved copper (Cu) 26 15 30
Total copper (Cu) 67 10 46
Dissolved cadmium (Cd) 21 18 29
Total cadmium (Cd) 20 9 29
Dissolved zinc (Zn) 22 21 34
Total zinc (Zn) 24 11 34
Dissolved lead (Pb) 16 1 5
Total lead (Pb) 198 4 47
Risk potential screening levels used to identify COPCs considered provincial and federal regulatory criteria
and local background water quality concentrations in the receiving environment. The criteria included British
Columbia Water Quality Objectives (BCWQO); Canadian Council of Ministers of the Environment (CCME)
water quality guidelines; and local background concentrations as defined by the 90th percentiles for
measured parameters at upstream monitoring station W10 (Table 2). The screening level selection approach
was as follows: the background water quality took precedence over the BCWQO and CCME values; otherwise,
the lowest of the provincial and federal values was selected as the screening level, except where the
background level exceeded regulatory criteria. In such cases, the 90th percentile for the background data was
carried forward as the screening level; this was the case for total aluminium, chromium, copper, iron,
vanadium and zinc.
The COPCs identified were based on comparison of the project screening levels with the available data
representative of the worst-case water quality being discharged from the mine. The station used for the
identification of COPCs was station SE-2, which is a sampling point located prior to discharge into the aquatic
receiving environment. This approach identified a rather broad range of COPCs that were then carried
forward into the evaluation of risk at the monitoring stations representative of the receiving environment
(stations W32, W46 and W51).

Risk Assessment
Background
Screening
Parameter Units CCME BCWQO 90th
level
percentile
Fluoride (F) mg/L 0.12 2 0.06 0.12
Nitrite (N) mg/L 0.06 0.02 0.03 0.02
Nitrate (N) mg/L 13 3 0.06 3
Dissolved sulphate (SO4) mg/L 218 13.9 218
Dissolved chloride (Cl) mg/L 120 150 1.81 120
pH pH units 6.5 – 9.0 6.5 – 9.0 7.7 6.5 – 9.0
Total aluminium (Al) μg/L 100 4,538 4,538
Total arsenic (As) μg/L 5 5 2.5 5
Total barium (Ba) μg/L 1,000 89 1,000
Total beryllium (Be) μg/L 5.3 0.1 5.3
Total boron (B) μg/L 1,500 25 1,500
Total cadmium (Cd) μg/L 0.09 0.01 – 0.06 0.11 0.16
Total chromium (Cr) μg/L 1 1 7.1 7.1
Total cobalt (Co) μg/L 4 25 4
Total copper (Cu) μg/L 4 2 10.7 13
Total iron (Fe) μg/L 300 1,000 5,428 6,265
Total lead (Pb) μg/L 7 12 3.4 7
Total lithium (Li) μg/L 14 25 14
Total manganese (Mn) μg/L 700 116 700
Total mercury (Hg) μg/L 0.026 0.00125 0.025 0.00125
Total molybdenum (Mo) μg/L 73 1,000 2.7 73
Total nickel (Ni) μg/L 150 25 9.2 25
Total selenium (Se) μg/L 1 2 0.26 1
Total silver (Ag) μg/L 0.1 0.05 0.04 0.05
Total thallium (Tl) μg/L 0.8 0.8 0.00546 0.3
Total titanium (Ti) μg/L 2,000 264.4 2,000
Total uranium (U) μg/L 15 300 15
Total vanadium (V) μg/L 6 11.8 11.8
Total zinc (Zn) μg/L 30 7.5 20.8 31.9
Subsequent to the screening, a more detailed consideration was given to each of the preliminary COPCs. This
process identified that mercury was not detected in the discharge at SE-2 or the receiving environment
stations. Given that mercury has not been detected at the mine, or within the local receiving environment, it
was not carried forward for quantitative evaluation in the risk assessment. Similarly, when the screening was

applied to the remaining list of preliminary COPCs, only seven were found to be present at concentrations
exceeding the project screening levels in the receiving environment: aluminium, cadmium, copper, iron, lead,
nickel and zinc. Further review of these metals resulted in four metals being carried forward for quantitative
evaluation in the risk characterisation because they exceeded the criteria in at least 10% of the samples (and
more than once) at one or more of the monitoring stations. These were cadmium, copper, lead and zinc. The
screening identified total metals as the COPCs; this is a function of the regulatory surface water criteria being
set for total concentrations rather than dissolved. However, to provide a more comprehensive evaluation of
the potential for risk in the aquatic receiving environment, HQs were calculated for both the total and
dissolved concentrations for each COPC.
Exposure to COPCs was characterised using two methods: COPC concentrations in surface water and COPC
concentrations in fish tissues, where data was available. Data was available for juvenile Dolly Varden
collected from upstream and downstream from the mine discharge site. The results found no significant
differences in tissue metal concentrations between sample site locations (Hitselberger, 2012).
Each surface water sample result represents an estimated environmental concentration (EEC) from which an
HQ was calculated. The fish receptors are also known to prefer certain habitat conditions, and Table 3
compares the water-quality parameters from the background river main stem and clear-water side channel
to the guideline criteria.
Range in main stem Range in clear-water side

Parameter CCME guidelines
background concentration channels
Hardness – 23 – 33 mg/L 29 – 57 mg/L
Conductivity – 54.0 – 78.2 µS/cm 68.4 – 132 µS/cm
pH 6.5 – 9.0 7.65 – 7.81 7.64 – 7.90
TSS 49 – 111 mg/L < 3 – 40 mg/L
Turbidity – 44 – 145 NTU 0.48 – 44 NTU
Total
0.005 – 0.1 0.89 – 2.9 mg/L 0.01 – 0.51 mg/L
aluminium
< 0.000070 –
Total cadmium 0.000017 0.000021 – 0.000026 mg/L
0.000070 mg/L
Total copper 0.002 – 0.004 0.0047 – 0.0076 mg/L < 0.0010 – 0.0033 mg/L
Total iron 0.30 0.94 – 2.95 mg/L < 0.03 – 0.69 mg/L
Total lead 0.0010 – 0.0070 0.0015 – 0.0028 mg/L < 0.0005 – 0.0007 mg/L
Total zinc 0.03 0.009 – 0.0160 mg/L < 0.0050 mg/L
The effects of a chemical contaminant on an ecological receptor are characterised by an exposure-response

curve. The toxicity reference values (TRVs) selected for salmonids were considered to be threshold
concentrations or doses/intakes of the COPCs that could cause harm if exceeded. The TRVs considered for
use in this risk assessment were based on chronic toxicity tests carried out under standardised laboratory
conditions (Table 4). All TRVs were based on average hardness measured over the sampling events at the
four sampling stations.

Risk Assessment
Chemical TRV chronic (µg/L) TRV acute (µg/L) Acute to chronic ratio
Cadmium 1.3 1.3 1
Zinc 187 460 2.5
Lead 15 1,170 78.0
Copper 13 19 1.5
HQ box-plot graphs were generated for each of the four COPCs at stations W46, W51 and W32 for WTP
operational and nonoperational scenarios. For cadmium, plots indicate a potential unacceptable hazard to
fish at station W51, even when the WTP was operational. Total cadmium HQ also exceeded one in one sample
at W10, suggesting that total cadmium may be naturally elevated. Plots indicate a potential unacceptable
risk to fish at W51 when the WTP was not operational. The maximum HQ when the WTP was not operational
was approximately eight times higher than the maximum HQ when the WTP was operational.
Copper HQs indicate a potential unacceptable hazard to fish at station W51 when the WTP was operational.
Dissolved copper also marginally exceeded the HQ of one at sample stations W46 and W32; however, total
copper HQs were less than one. Total copper exceeded the HQ of one in one sample at station W10,
suggesting naturally elevated copper. Dissolved and total copper HQs when the WTP was not operational
indicate a potential unacceptable hazard to fish at stations W5 and W32. Almost all fish HQs based on
dissolved and total copper at site W51 were greater than one. Two fish HQs for total copper at stations W10
and W46 marginally exceeded the HQ of one, indicating natural elevations. The selected TRV was equivalent
to the P95 value for copper at W10.
Dissolved and total lead HQs for surface water exposures to fish indicate a negligible risk at stations W46,
W51 and W46 when the WTP was operational. Dissolved and total lead HQs indicate an occasional potential
unacceptable hazard at station W51 when the WTP was not operational. One fish HQ based on dissolved lead
in surface water and two fish HQs based on total lead in surface water exceeded the HQ of one at station
W51. All other fish HQs for lead were less than one.
Dissolved and total zinc HQs indicate a potential unacceptable hazard at station W51 when the WTP was
operational. All other fish HQs for zinc in water were less than one. Dissolved and total HQs for zinc indicate
a potential unacceptable risk at station W51 when the WTP was not operational. Almost all fish HQs based
on dissolved and total zinc at W51 were greater than one. All other fish HQs for zinc in water were less than
one.
Trends in COPC concentrations at stations W51 and W32 appear to follow seasonal discharges and climatic
conditions. The highest flows at these stations occurred between mid-May to December, and the period
between January and early April had HQs that were on average three to four times higher than during the
high flow period. These winter months represent low flows and the least amount of dilution. The highest HQs
at these stations occurred during late April and early May for all COPCs. This is the result of snow melt and
precipitation and the subsequent annual flush of historic waste dumps into the river.
The hazard quotients were highest at station W51, followed by station W32. The HQ results at stations W10
and W46 (upstream of the mine) rarely exceeded one for any sampling event. Station W51 appears to be the
station most affected by discharge from the mine site. At this location, the HQs for all COPCs were greater
than one under both WTP operational and nonoperational conditions, with the exception of lead, which had
an HQ consistently less than one under the WTP operational period. At each station, HQs were considerably

lower during WTP operational conditions compared to when the WTP was not operating, indicating that the
WTP has a positive influence on water quality.
Until such time that the historical waste rock is capped to reduce infiltration, it does not appear possible to
prevent occurrences of HQs exceeding one during the spring freshet. This is true whether or not the WTP is
operating or not. The HQ was less than one for the majority of the year, with or without WTP operation,
including the critical time periods when Chinook, sockeye and coho salmon are migrating to spawn and eggs
are incubating and hatching.
The most sensitive life stage of Chinook and steelhead trout for acute toxicity for cadmium, copper and zinc
is the juvenile form (Chapman, 1978). Juvenile salmonids in the receiving environment are very unlikely to
rear in the fast-flowing, turbid water of the main channel. As such, the juvenile forms of the receptors of
concern are less likely to be exposed to the episodic loadings of COPCs from mine discharge. Potential impacts
to salmon spawning are one of the key issues to evaluate in this risk assessment. The highest concentrations
of COPCs coincides with the period of the annual snowmelt and rainfall in late April to early May. Both
spawning periods of the coho and sockeye salmon coincide with a time of low HQs for COPCs and therefore
result in relatively lower exposure levels. Coho enter the river between mid-July and November and spawn
in the watershed between August and December. Sockeye spawn between mid-June and August. Moreover,
all of these salmonid species are more likely to be found in the clear-water side channels than the main stem
of the river.
Given the frequency (annual) and size of jökulhlaup events combined with high turbidity, unsuitable
substrate (cobble) and very cold temperatures (< 1°C), it can be reasonably predicted that the receiving
environment only provides marginal or transitional habitat and that fish utilisation of the main stem is
transient and primarily for migratory purposes. As is the case during the rest of the year (late spring through
early fall), most fish (including fall spawners) would likely be found in the clear-water side channels along the
river margins.
This risk assessment focused on direct surface water contact. Knowledge of the tissue residue levels in food
consumed by the salmonid species could provide information about whether dietary uptake is the major
route of exposure. Generally, the higher the metal concentration in the water, the more metal is taken up
and accumulated by fish. The relative importance of waterborne versus dietary uptake differs greatly among
metal species and may differ between fish species (Miller et al., 1993; Sappal et al., 2009; Franklin et al.,
2005; Mount et al., 1994). Tissue residue concentration in fish depends on more than just dietary and
waterborne metal concentrations. Time of exposure, environmental conditions (water temperature, pH,
hardness, salinity) and intrinsic factors (fish age, feeding habits) must also be considered (Jezierska and
Witeska, 2006). Additional detailed studies would need to be undertaken in this receiving environment to
determine whether dietary metals uptake or waterborne uptake is the key uptake mechanism. However, it
is highly likely that resident fish are largely drift feeders and that direct surface water contact is the key
exposure source.
Based on the seasonal trends of metal concentrations and the life cycles and habitat preferences of the
receptors of concern, the risk is considered low for anadromous species. The risk to resident receptors of
concern (if any are present in the main stem during the prefreshet period) is greater (i.e., moderate) due to
increased potential for exposure to COPCs.
The objective of the study was to determine the potential for risk to fish species in the receiving environment
from discharge from a historical mine site. A second objective was to attempt to assess the effectiveness of
the WTP that operated briefly in 2012. A systematic screening of all measured surface water quality
parameters resulted in the identification of four contaminants of potential concern: cadmium, copper, lead
and zinc. Using the hazard quotient methodology, evaluation of mine effluent showed that the highest HQs
in the receiving environment coincided with the period of site snowmelt. This is believed to be the result of

Risk Assessment
the annual flushing of the historic mine waste rock during the spring thaw. During the annual flushing period,
most juvenile salmonids will be overwintering in the preferred habitats of the clear-water side channels.
With respect to the effectiveness of the WTP, surface water quality monitoring during its operation did show
that the HQs were lower at sites downstream from the points of discharge. However, during the annual flush
period, HQs for copper and zinc were still greater than one. Until such time that the historic waste rock is
capped to reduce infiltration, it does not appear that the WTP is capable of reducing mine discharge to levels
where resulting HQs do not exceed the threshold of one. Regardless of whether the WTP was operating or
not, the HQs at all sites except W51 were less than one for the majority of the year.
Overall the potential risk to aquatic receptors as a result of mine discharge is considered low (HQ < 1 most of
the year). As HQs at some sites were greater than one, the risk to main stem aquatic receptors would be
considered moderate (HQ < 100) during those times. However, as most migratory species are known to utilise
clear-water side channels removed from direct influences of the mine discharges, and there were no effects
on tissue metal concentrations in the resident species (Dolly Varden/bull trout), the moderate risk
designation for the selected aquatic receptors is considered conservative.
It is important to note that the goal of the AERA was to determine the relationship between the mine
discharge and the risk to aquatic life. The goal was not to provide a specific description of the potential
impacts of aquatic receptors. One important application of the AERA is help identify data gaps and focus data
collection as part of future monitoring plans. This targeted data collection can then be used to refine risk
assessments and help appropriately monitor treatment and management strategies to reduce off-site
impacts under changing conditions. For example, monitoring is now focused on the snowmelt period in April,
recognising that there is no or only low potential for adverse effects during the open-water summer season.
This case study outlines an approach at a mine where available background information was put to good use
in helping to define potential risk to the aquatic environment and focus resources for future monitoring and
management strategies. Such an approach is recommended at other sites, particularly remote sites or closed
sites where resources may be limited. Risk should be determined and only appropriate environmental
protection measures put into place.
We thank Marc Cameron, Core6 Environmental Ltd., for the original risk assessment and Mike Whelan, Triton
Environmental Consultants Ltd., for compiling the fish presence information.
Chapman, G.A. (1978) Toxicities of cadmium, copper and zinc to four juvenile stages of Chinook salmon and steelhead, Transactions
of the American Fisheries Society, Vol. 107(6), pp. 841–847.
Franklin, N.M., Glover, C.N., Nicol, J.A. and Wood, C.M. (2005) Calcium/cadmium interactions at uptake surfaces in rainbow trout:
Waterborne vs dietary routes of exposure, Environmental Toxicology Chemistry, Vol. 24, pp. 2954–2964.
Government of Canada (2001) Pacific region integrated fisheries management plan, salmon Taku River, B.C., June 1 2001 to May 31
2002, Department of Fisheries and Oceans, viewed September 2013, http://www.dfo-mpo.gc.ca/Library/327688.pdf.
Government of British Columbia (1999) The recommended guidance and checklist for tier 1 ecological risk assessment of
contaminated sites in British Columbia, Ministry of Environment, Lands and Parks, viewed September 2013,
www.env.gov.bc.ca/epd/remediation/policy_procedure_protocol/protocols/tier1/.
Hitselberger, J.P. (2012) Tulsequah Chief mine acid rock drainage: whole body metals concentrations in Dolly Varden char, Alaska
Department of Fish and Game, Technical Report No. 11-09, Douglas, AK.
Jezierska, B. and Witeska, M. (2006) The metal uptake and accumulation in fish living in polluted waters, in I. Twardowska, H.E. Allen
and M.M. Haggblom (eds), Viable methods of soil and water pollution monitoring, protection and remediation, Springer,
Dordrecht, The Netherlands, pp. 3–23.
Miller, P.A., Lanno, R.P., McMaster, M.E. and Dixon, D.G. (1993) Relative contribution of dietary and waterborne copper to tissue
copper burdens and waterborne copper uptake in rainbow trout (Oncorhynchus mykiss), Canadian Journal of Fisheries and
Aquatic Science, Vol. 50, pp. 1683–1689.
Mount, D.R., Barth, A.K., Garrison, T.D., Barten, K.A. and Hockett, J.R. (1994) Dietary and waterborne exposure of rainbow trout
(Oncorhynchus mykiss) to copper, cadmium, lead and zinc using a live diet, Environmental Toxicology Chemistry, Vol. 13, pp.
2031–2041.

Sappal, R., Burka, J., Dawson, S. and Kamunde, C. (2009) Bioaccumulation and subcellular partitioning of zinc in rainbow trout
(Oncorhynchus mykiss): Cross-talk between waterborne and dietary uptake, Aquatic Toxicology, Vol. 91(4), pp. 281–290.
Science Advisory Board for Contaminated Sites in British Columbia (2008) Detailed ecological risk assessment (DERA) in British
Columbia, technical guidance, September 2013, Golder Associates Ltd., Burnaby, BC, viewed http://www.sabcs.
chem.uvic.ca/ DERA2008.pdf.

N.J. Healey Azimuth Consulting Group Partnership, Canada

B.A. Power Azimuth Consulting Group Partnership, Canada
D.M. Kristoff Saskatchewan Ministry of the Environment, Canada
Public sector accounting standards require governments to estimate and report financial liabilities associated
with contaminated sites for which they are responsible. Abandoned mines represent a large potential financial
liability for governments, but it is difficult to quantify these liabilities accurately without investing significant
monetary resources in environmental investigations at the sites. Additionally, without the benefit of a site-
specific environmental risk assessment, liabilities may be booked on the basis of generic environmental quality
criteria such as national or provincial soil, water and sediment quality guidelines or standards. However, these
generic criteria were not intended to signal a need for clean-up, and their use in this context results in an
overestimate of environmental risk and the associated financial liabilities. An environmental risk assessment
framework was developed to assist in deriving estimates of financial liability for five abandoned hard-rock
metal mines in northern Saskatchewan, Canada, for which there were limited existing environmental data.
The framework accounts for potential risks to human health, plants and invertebrates, and fish and wildlife.
Risks are estimated based on an evaluation of the potential magnitude, spatial extent and duration of
exposure; the relative sensitivity of the exposed organism; and the relative potency and mechanism of toxicity
of the contaminant. The framework is systematically applied to categorise potential human health and
ecological risks associated with environmental contamination according to one of four possible outcomes: (1)
low risk with low uncertainty in the conclusion; (2) low risk with high uncertainty in the conclusion; (3) high
risk with high uncertainty in the conclusion; and (4) high risk with low uncertainty in the conclusion.
Contamination associated with high risk and low uncertainty conclusions was considered an environmental
liability, and cost estimates to manage the contamination were developed. Contamination associated with
low risk and low uncertainty conclusions was not considered an environmental liability, and no cost estimates
were developed. Risks with high uncertainty conclusions were identified as data gaps for further investigation.
This presentation will use the five mine sites in Saskatchewan to illustrate the application of the environmental
risk assessment framework and demonstrate its potential application to other contaminated mine sites for
estimating financial liabilities and identifying relative priorities for further investigation or risk management.
The costs involved in assessing and managing the health and environmental risks associated with abandoned
mines are a significant burden in many jurisdictions. For example, the estimated liability to the Canadian
federal government for abandoned mines in northern Canada is in excess of C$ 2 billion (Office of the
Parliamentary Budget Officer, 2014). Even the process of assembling an inventory of abandoned mine sites
and developing initial estimates of the associated potential financial liabilities can be costly and time
consuming; reducing the time and costs of developing financial liability estimates for abandoned mines can
have significant financial benefits.
A body of critical literature is developing on how to meet the challenges of assessing and managing
abandoned mine sites effectively and efficiently. Commonly recommended best practices include two key

Environmental risk assessment framework for financial liability estimates for abandoned metal mines N.J. Healey, B.A. Power and D.M. Kristoff
elements: (1) using a systematic process; and (2) using a risk-based approach. One of our co-authors (Power)
was previously involved in the development of a systematic, risk-based method of ranking abandoned mine
sites according to risks to human health and the environment (Power et al., 2010). This paper extends that
effort by describing a risk-based framework created to support the development of financial liability
estimates for five abandoned metal mines in northern Saskatchewan, Canada. The framework can potentially
reduce the time and cost of assessing abandoned mine sites, produce more accurate estimates of the
financial liabilities and provide a defensible method for prioritising sites according to risks to human health
and the environment.
Our objective is to describe the key elements of the risk-based framework in sufficient detail that the
concepts can be applied or adapted for the purposes of supporting liability estimates and identifying relative
priorities among other abandoned mines or other contaminated sites. This paper provides background
information on the five mine sites and the overall methods used to develop financial liability estimates for
the sites, describes the key elements and the process of the risk-based framework and concludes with a
summary of the lessons learned from this experience. We do not present the results of the risk assessments
for the five sites to which we applied the framework.
Before introducing the overall project and the subject sites, it is important to define the following terms:
 Abandoned mines refers to mine sites that have not been remediated and for which the
government is solely responsible.
 Generic criteria refers to regional or national environmental quality guidelines or standards (e.g.
the Canadian Council of Ministers of the Environment soil, water and sediment quality guidelines).
Generic criteria are differentiated from site-specific criteria, which are based on a site-specific
environmental risk assessment.
 Risk management refers to any active management of environmental contamination – this could
range from monitoring to engineered or administrative controls that limit exposure (e.g. fencing
or covers) to clean-up.
The overall project objective was to develop estimates of financial liability for the risk management of
environmental contamination at five abandoned metals mines in northern Saskatchewan, Canada. The
locations of the sites are illustrated in Figure 1.
Liability estimates were developed according to the Public Sector Accounting Board’s accounting standard
on liability for contaminated sites (Section PS3260). An important feature of the PS3260 standard is that it
allows financial liability estimates to be based on professional judgment and knowledge from experience
with similar sites. Other liability accounting standards or guidance may not allow for these methods, and
such limitations would preclude the application of the approach described in this paper.
Key attributes of the five mine sites are summarised in Table 1. The five sites are relatively small, but they
are in relatively sensitive settings: four of five sites are adjacent to lakes, one is within a provincial park and
another is located near a fly-in fishing lodge, and three are within close proximity to human settlements and
are frequently used by the public for recreation, fishing and hunting. Two of the five sites are relatively
remote, which increases the costs of assessment and risk management.
Five primary project tasks were completed as part of the broader project to develop financial liability
estimates for the sites.
The first project task was to review the files and existing data for each site. There was missing or incomplete
information on the location of mine workings, including openings, and existing environmental data for the
sites were few. Typically, data on metals concentrations were only available for less than five soil samples
and two or three surface water samples for each site. No sediment chemistry data were available. One of the
five sites lacked any representative soil chemistry data.

Risk Assessment
The second project task was to conduct a reconnaissance-level site inspection. The site reconnaissance was
conducted by a mining geologist, a geotechnical engineer specialising in mine reclamation and remediation,
a mining geochemist and the site owner. While in the field, the geochemist conducted a limited number of
rinse tests to assess the potential for waste materials to produce acid rock drainage metals leachate
(ARD/ML), and the engineer and geologist conducted preliminary evaluation of the of possible remedial
options. The site reconnaissance team spent up to a day at each site.
Site 1 Site 2 Site 3 Site 4 Site 5

Primary metals Copper and Nickel and
Gold Gold Gold
recovered gold copper
Open pit and
Type of mine Underground Open pit Underground Underground
underground
1931–1940,
1965-68,
Operating period(s) 1939-42 1966-72 1943-47 1981–1982,
seasonally
1987–1988
Free gold and Sulphide-
Gold- Gold-
Ore mineralisation in coatings on Chalcopyrite rich
arsenopyrite arsenopyrite
pyrite pyroxenite
Crushing,
milling,
Crushing,
Crushing, Crushing, Crushing, floatation,
On-site processing milling,
milling milling milling separation,
roasting
cyanide
leach
Ore processed
1.4 Mt 2 Mt 0.04 Mt 0.3 Mt 0.03 Mt
(approximate)
Site footprint (m2,
approximate) 75,000 250,000 50,000 14,000 45,000
Unmaintained
Access Road Air Road Road
road
Distance to nearest
permanent human
settlement (km) 35 30 140 1.4 2.4
The third project task was the development of site overviews. Site overviews included a description of the
site history and geology, a preliminary assessment of the geochemical stability of mine wastes and their
potential as a source of ARD/ML, and the identification of areas of potential environmental concern (APECs),
including areas that could be sources of environmental contamination (e.g. waste rock, ore, milling wastes,
concentrate and tailings) as well as receiving environments, such as nearby lakes or other surface water
features, that may have been impacted by contaminants migrating from the source areas. This method was
used because risks associated with contamination of the receiving environment may be distinct from risks
associated with contamination at source areas. The available environmental quality data were also reviewed
for quality assurance, tabulated and screened against provincial and national generic soil and water quality
criteria.

The fourth project task was to conduct a human health and ecological risk assessment for each of the sites.
The framework developed to conduct the risk assessments is described in detail below. The output of the
risk assessments was the identification of APECs likely to require risk management and the potential
exposure pathways that needed to be addressed by risk management. If the need for risk management could
not be determined with sufficient certainty, risk assessment data gaps and the costs associated with
addressing those gaps were identified. While the overall project included the development of cost estimates
to address physical public safety hazards such as mine openings, the scope of the risk assessment framework
was limited to assessing health and environmental risks associated with environmental contamination at the
mines.
The final project task was the development of remedial options and associated cost estimates for all of the
APECs identified in the risk assessments as likely to require risk management.

Risk Assessment
This section of the paper describes the risk assessment framework used to support the development of
financial liability estimates for the five sites. The methods used in the human health risk assessment (HHRA)
and ecological risk assessment (ERA) of the abandoned mine sites are consistent with currently available
guidance on contaminated sites risk assessment from Environment Canada (2012), Health Canada (2010a,
2010b, 2012), the Canadian Council of Ministers of the Environment (CCME, 1996, 1997), and the U.S.
Environmental Protection Agency (EPA) (1998, 2007).
A preliminary, or screening level, qualitative risk assessment was undertaken to systematically evaluate and
characterise the potential human health and ecological risks associated with known or suspected
environmental contamination at the sites based on the available data. The preliminary qualitative risk
assessments follow the same process as more detailed, quantitative risk assessments. However, they (1) rely
on existing data only, despite scarce data for some areas of some sites, (2) do not reflect site-specific effects
information (e.g. field-collected data on effects to ecological receptors), and (3) rely on professional
judgment to an appreciable extent. These risk assessments are useful for the purpose of liability assessment,
but they would not likely be viewed as sufficient on their own to obtain regulatory release from liability for a
contaminated site.
Contaminated site risk assessment is an iterative process. As a first step, risks are typically evaluated
qualitatively or semi-quantitatively in a preliminary or screening level risk assessment using conservative (i.e.
health-protective) assumptions where there is uncertainty. A preliminary risk assessment is unlikely to
underestimate risks, and “no risk” or “low-risk” conclusions from a preliminary risk assessment are generally
considered sufficiently certain to support risk management decision-making. Where a preliminary risk
assessment concludes that risks of significance may exist, risk managers must weigh the relative costs of
further work to reduce the uncertainty in the risk assessment conclusions (e.g. additional site investigation
and detailed site-specific risk assessment) against the costs of simply taking risk management measures to
address the sources of potential risk (e.g., remediation).
The objective of the preliminary qualitative risk assessments undertaken for the five abandoned mine sites
was to characterise potential risks associated with environmental contamination according to the two-
dimensional framework illustrated in Figure 2. Low-risk conclusions that also have low uncertainty are
unlikely to represent a financial liability for the site owner. All other risk conclusions represent a potential
financial liability — for either (1) further work to address identified data gaps and reduce the uncertainty
associated with the risk conclusions or (2) risk management measures to address the potential risks. Risks
were expressed as the likelihood that a site-specific detailed quantitative risk assessment would trigger the
need to consider risk management. Risks were evaluated separately for each APEC, but the risk assessment
considered the potential cumulative effects of exposure to multiple APECs, if applicable.
The first step of the risk assessment involved developing a conceptual exposure model for each APEC. A
conceptual exposure model for a risk assessment identifies the contaminants of potential concern, the
receptors of potential concern, and the exposure pathways by which the receptors of concern could
potentially exposed to the contaminants of concern. A chemical exposure risk requires the presence of all
three of these elements; this interrelationship is illustrated in Figure 3. Scenarios with no plausible existing
or future exposure pathway were excluded from further evaluation in the risk assessment. The first step of
the risk assessment was equivalent to a qualitative or screening-level risk assessment, which is accepted in
some Canadian regulatory jurisdictions as sufficient for making risk management decisions for contaminated
sites (e.g. BC MOE, 2008).

Where an exposure pathway was open for a receptor (i.e. there is a connection, or exposure, between the
contaminant and the receptor), the exposure scenario was carried forward to the second step of the risk
assessment process — risk characterisation. Potential risks were characterised for all receptors with operable
or potentially operable exposure pathways (i.e. a means by which a receptor could be exposed to
contamination). Risk characterisation was achieved by systematically evaluating a series of variables related
to contaminant exposure, the sensitivity of the exposed organism and the properties of the contaminant.
2. Low Risk, 3. High Risk,

High Uncertainty High Uncertainty
Uncertainty
= Lower Priority = Higher Priority

Data Gaps Data Gaps
1. Low Risk, 4. High Risk,

Low Uncertainty Low Uncertainty
= No Liability = Environmental
Liabilities
Risk

Risk Assessment
Available site chemistry data for each APEC were compiled, tabulated and compared to appropriate generic
environmental quality criteria. This served to identify contaminants of potential concern (COPCs) for each
APEC, as well as quantify the magnitude by which each COPC exceeds its respective environmental quality-
screening criterion. These “screening quotients” (i.e., the ratio of a chemical concentration to its respective
environmental quality-screening criterion) were used in the risk assessments to support a semi-quantitative
evaluation of the potential magnitude (the margin by which measured concentrations exceed a criterion
concentration) and extent (proportion of samples failing a criterion) of exposure.
Separate human health- and ecological-based screening criteria were identified and applied for screening soil
and surface water chemistry. Therefore, the HHRA could have COPCs for a given APEC that differ from those
of the ERA. For example, copper often screens as a COPC for ERAs because of the susceptibility of aquatic life
and some ruminants to copper toxicity, whereas copper is relatively non-toxic to humans, so unless
concentrations are very high, copper is not often screened into HHRAs.
Note that screening quotients are not a direct expression of the magnitude of toxic effects, because the
screening criteria they are based on include neither information on the nature of the toxic effect (an
environmental quality guideline, for example, could be based on lethal effects or sub-lethal effects unrelated
to fitness or survival of the organism) nor the slope of the underlying dose-response relationship. For these
reasons, a contaminant with a screening quotient of 10 should not be interpreted as 10 times more “toxic”
than a contaminant with a screening quotient of one.
None of the five sites have measures in place to control public access. Therefore, it was assumed that all
human receptors, including vulnerable subpopulations such as toddlers and subsistence harvesters, could
potentially be exposed to contaminants associated with the sites. A toddler was assumed to represent the
critical (i.e. most sensitive) of all potential age classes for assessing potential risks associated with exposure
to non-carcinogenic contaminants, whereas adult receptor characteristics were used to estimate a lifetime
average daily dose for assessing potential risks associated with exposure to carcinogenic contaminants. This
approach is consistent with Health Canada (2010a) guidance on HHRA. The likelihood and frequency of
human exposure were considered in the risk characterisation stage of the assessment, and human health
risks were considered less likely for sites that are remote or otherwise difficult to access.
The following general ecological receptor groups were evaluated in the ERA: invertebrates, flora (plants),
wildlife and fish. It was not possible to identify ecological receptors of concern in more detail (e.g. to the level
of individual species) with the available site information. Both listed (i.e. rare or endangered) wildlife species
and common wildlife species were identified as general receptors for the ERA. These receptor groups were
evaluated separately because listed species are typically afforded a higher level of protection against risks
from contaminant exposure than are common species.
The conceptual model stage of the risk assessments included a systematic review of the potential for
identified receptors of concern to be exposed to chemicals of concern. Both direct and indirect exposures
were evaluated. Direct exposures include inhalation, ingestion and dermal contact with contaminated soil,
water, sediments or vapours. Indirect exposure includes consumption of contaminated food or prey that
have bioaccumulated contaminants from contaminated soil, water or sediments. If an exposure pathway was
open or potentially open for a receptor (i.e. an operable pathway), that exposure scenario was carried
forward for risk characterisation.

Risks were characterised for receptors with operable exposure pathways or potentially operable exposure
pathways. Risks were characterised by systematically evaluating variables that are important in determining
human or ecological risks associated with exposure to environmental contamination. These variables are
described below and can be grouped into those that relate to exposure, those that relate to the sensitivity
or the vulnerability of the exposed receptor and those that relate to the characteristics of the contaminant.
In most cases the variables were evaluated qualitatively; however, in some cases they were evaluated semi-
quantitatively.
The following variables were considered in evaluating the exposure of a receptor of concern to
environmental contaminants associated with an APEC:
 screening quotients (described above) to assess the potential magnitude (the margin by which
measured concentrations exceed a criterion concentration) and extent (proportion of samples
failing a criterion) of exposure for human and ecological receptors;
 the frequency and intensity of public use of the sites based on the proximity of the site to human
settlements, site access, information in the site files and evidence from the site reconnaissance
(e.g. the presence of berries, tire tracks, campfire rings and discarded shell casings);
 the site history and how representative the available environmental data are of potential or
expected extent and magnitude of contamination;
 the estimated size of the APEC relative to the foraging range of ecological receptors (e.g. the size
of contaminated habitat at the mine sites would only represent a very small percentage of the
foraging range of receptors like moose and bears, and risks to receptors with large foraging
ranges, therefore, were considered low),
 the extent of vegetative cover on areas of contaminated soil;
 the anticipated temporal stability of mine wastes and the potential for ARD to increase metals
contamination in the future.
The following variables were considered in evaluating the sensitivity of a receptor to the potential adverse
effects associated with environmental contaminants at an APEC:
 the conservation status of potential ecological receptors (risks to common species were evaluated
as potential risks to population viability, whereas risks to listed species were evaluated as
potential risks to individual organisms);
 the proportion of contiguous habitat potentially adversely affected by the contamination;
 the uniqueness of the ecological function of the exposed ecological receptor (e.g. niche or
keystone species were considered more sensitive than generalists).
The following variables were considered in evaluating the potential for a contaminant to cause adverse
effects to human or ecological receptors:
 the toxicokinetics of the contaminant and the expected potential for bioaccumulation in
organisms or biomagnification in the food web;
 the geochemistry of the mine wastes and the speciation and bioavailability of metal (e.g. the
known or potential presence of arsenic trioxide was considered a higher risk than the known or

Risk Assessment
potential presence of arsenic as an arsenopyrite, because arsenic trioxide is known to have much
higher oral bioavailability than arsenopyrites);
 the toxicological mode of action of the contaminant.
Ruminants and aquatic plants and invertebrates are more sensitive to the toxic effects of copper than are
other ecological receptors or humans. Human exposure to carcinogens, such as arsenic, can generally be
defensibly averaged over time and therefore, for carcinogens, human exposure risks correspond to the
frequency and duration of exposure. On the other hand, relatively short periods of human exposure to some
metals can cause developmental effects (e.g. lead and mercury) or other lasting health effects (e.g. elicit an
allergic sensitivity to nickel), and it may not be defensible to characterise risks for these metals based on
longer term average exposures. Therefore, the frequency of exposure was not given much weight in deriving
conclusions about human health risks from exposures to contaminants with these modes of action.
Risks for receptors with operable exposure pathways were systematically evaluated by considering the above
variables and were characterised as: (1) low, (2) low but uncertain, (3) high but uncertain, or (4) high. As
discussed above, these conclusions express the likelihood, based on professional judgement and our
experience with detailed quantitative risk assessments at other abandoned mine sites, that a site-specific
detailed quantitative risk assessment would trigger the need for risk management. Risks characterised as low
are relatively low priorities for further assessment or risk management and are unlikely to present a financial
liability for the site owner. Risks characterised as high are those for which a detailed quantitative risk
assessment is likely to show that the potential risks to the identified receptor require risk management; these
risks are likely to present a financial liability for the site owner. Risks that are uncertain represent data gaps
that require additional investigation before a decision about the need for risk management can be made with
sufficient certainty. The costs of addressing these data gaps and the probability that they will identify issues
that require risk management can be weighed against the costs of simply risk managing the sources of
exposure contributing to these uncertain risks. To facilitate these trade-offs, cost estimates were developed
to address the risk assessment data gaps and remediation options, and associated cost estimates were
developed for all risks except those characterised as low.
Table 2 presents a summary of the number of APECs evaluated at each of the five sites, the number of
operable exposure pathways identified by the conceptual exposure models (step one of the risk assessments)
and the number of exposure pathway-receptor combinations in each of the respective risk characterisation
categories. As illustrated in Table 2, the risk assessment framework significantly narrowed the scope of issues
that may require risk management. Despite the sparse and uneven environmental data available to inform
the risk assessments, risk assessment conclusions were supported in all cases except for the ecological
receptors at site number four. There were no soil chemistry data available for this site, and the source of
elevated metals in the adjacent lake water was uncertain. Therefore, the risk assessment team felt that
conclusions regarding potential ecological risks attributable to the site were not supported and risks were
characterised as “undetermined.”

Site Site 1 Site 2 Site 3 Site 4 Site 5

Number of APECs 5 6 5 4 4
Human health risk assessments
Number of exposure pathways evaluated 30 36 30 24 24
Number of operable exposure pathways 17 26 15 11 2
Number of exposure pathway-receptor 10 6 0 9 2
combinations with low risks
combinations with low but uncertain risks
combinations with high but uncertain risks
combinations with high risks
Ecological risk assessments
Number of exposure pathways evaluated 30 36 30 24 18
Number of operable exposure pathways 22 27 21 18 10
Number of exposure pathway-receptor 11 16 2 Undeter- 8
combinations with low risks mined

combinations with low but uncertain risks mined

combinations with high but uncertain risks mined

combinations with high risks mined
A risk assessment framework was developed to support the development of financial liability estimates for
five abandoned mine sites in northern Saskatchewan. The risk assessment framework was designed to guide
the systematic application of risk assessment principles combined with professional judgment of experienced
practitioners to derive defensible risk conclusions based on very limited site-specific data. Risks to human
and ecological receptors from metals contamination at the mines were characterised according to the
likelihood that a detailed site-specific risk assessment would indicate the need for risk management.
Remedial options and associated cost estimates were developed to address high risks, and the associated
costs were identified as potential financial liabilities for the site owner. Risks characterised as low were not
considered financial liabilities. Uncertain risk conclusions were identified as data gaps requiring additional
investigation. Remedial options and associated cost estimates were also identified for uncertain risk
conclusions so that the costs of addressing the associated data gaps can be weighed against the costs of
remediation.
A number of challenges were associated with this project. The approach to developing liability estimates
described here required an experienced, multi-disciplinary project team of geologists, geotechnical

Risk Assessment
engineers, geochemists, toxicologists and risk assessors. This multi-disciplinary project team was supported
by a knowledgeable site owner with the flexibility to develop a novel approach to deriving liability estimates.
A limitation of the approach is that the risk assessments derived to support the liability estimates may not
be sufficient to support risk management decision-making in some contexts, such as those where a risk
assessment is required to demonstrate regulatory compliance. For example, the BC Ministry of Environment
requires that site contamination be fully delineated before it will accept a remedial action plan based on a
risk assessment.
Despite these challenges and limitations, the approach to developing liability estimates described here offers
some significant potential advantages. First among these advantages is the potential for significant savings
in the time and money required to develop financial liability estimates. It is not uncommon for contaminated
sites to progress linearly through site investigation, risk assessment and remedial option development stages
– each requiring several iterations of investigations and reporting. This is because the focus of the early stages
of traditional contaminated site investigation includes delineation of contamination. The cumulative time
required to progress through these stages of assessment often amounts to several years and sometimes
requires a decade or more. The duration required to progress through these stages is prolonged in
environments, such as northern Saskatchewan, where there are climate-related limitations to the timing of
field investigations. Each of these stages of investigation, depending on the size and complexity of the site,
can cost in the tens to hundreds of thousands of dollars. This level of effort may be justified in the context of
needing to meet very prescriptive regulatory requirements, but in most cases is not required to support a
defensible risk assessment or to develop liability estimates for sites expected to remain under the
stewardship of the current owner. A streamlined method of site investigation to support risk-based
prioritisation of abandoned mine sites has been developed and is described by Power et al. (2010). This or
similar methods may also be suitable for supporting risk-based liability estimates.
Generic environmental quality criteria are often used in isolation as decision criteria for the need to risk-
manage environmental contamination. An issue with this approach is that comparisons of environmental
concentrations of contaminants to generic environmental quality criteria, such as national soil quality
guidelines, inherently overestimate risks for most sites because the generic criteria are designed to account
for a wide range of possible receptors, exposure conditions and contaminant properties. In addition to this
inherent conservatism, generic environmental quality criteria are usually not intended to be used as clean-
up standards – even though this is how they are typically interpreted and applied. The outcome of this
approach is that potential risks tend to be overestimated, which can, in turn, artificially inflate the associated
financial liabilities.
A second advantage of the approach described here is that it provides a method for systematically prioritising
risks, which has been identified by independent reviews as a critical requirement of robust and successful
management of contaminated sites (Unger et al., 2015; Auditor General of British Columbia, 2003). While
the approach described here is largely qualitative, it considers the key variables that determine the risk of
environmental contamination to human and ecological receptors. This leads to defensible conclusions about
the relative risks associated with environmental contamination. While more quantitative methods may
appear more rigorous and subjective, they can also be less transparent about trade-offs and assumptions.
For example, as discussed above, the magnitude by which the measured concentration of a contaminant
exceeds its respective generic environmental quality criteria should not, in isolation, be used to make
inferences about the risk of toxic effects to human or ecological receptors exposed to that contamination.
More quantitative methods also demand more data which, as described above, can be time consuming and
expensive to collect. The risk assessment framework presents information about why certain APECs or
contaminants present greater potential risks than others. This information may be used to support risk
communication requirements such as responding to concerns about risks of consuming fish or other country
foods harvested in the vicinity of the mine sites, and allocating limited resources for further investigation or
risk management.
Finally, the integration of risk assessment with remedial options development and financial liability estimates
allowed for side-by-side comparison of risks, data gaps and their associated costs, and risk management

options and their respective costs. This illustrated the relative costs and potential benefits of different
management options, such as, whether it was less costly to simply manage a potential risk rather than invest
further in investigation with a likely outcome that management would be required anyway.
The project described in this paper was funded by the Saskatchewan Ministry of the Environment. SRK
Consulting Canada Ltd. was the prime consultant for the project and led all project tasks except for the risk
assessments. Azimuth Consulting Group Partnership developed the risk-based framework, conducted the risk
assessments, and assisted with the design of the remaining tasks. The authors gratefully acknowledge the
support and assistance of their colleagues at the Saskatchewan Ministry of the Environment (Marnie Kinzel),
SRK Consulting (Canada) Ltd. (Mark Liskowich, Kelly Sexsmith, Saskia Nowicki, Maritz Rykaart, Warren
Medernach and Cam Scott) and Azimuth Consulting Group Partnership (Eric Franz) in the successful
completion of the project. Warren Medernach (SRK) produced Figure 1 for this paper.
Auditor General of British Columbia. 2003. Managing contaminated sites on provincial lands. Report 2002/2003 (5). Victoria, BC,
Canada.
BC Ministry of Environment (MOE) (2008) Screening level risk assessment, protocol 13 for contaminated sites, viewed 16 January
2015, http://www.env.gov.bc.ca/epd/remediation/policy_procedure_protocol/protocols/pdf/protocol_13.pdf.
Canadian Council of Ministers of the Environment (CCME) (1996) A framework for ecological risk assessment: General guidance,
National Contaminated Site Remediation Program, Winnipeg, Canada.
Canadian Council of Ministers of the Environment (CCME) (1997) A framework for ecological risk assessment: Technical appendices,
National Contaminated Site Remediation Program, Winnipeg, Canada.
Environment Canada (2012) Federal contaminated sites action plan (FCSAP) — Ecological risk assessment guidance, report prepared
for Environment Canada by Azimuth Consulting Group Inc., Vancouver, Canada.
Health Canada (2010a) Federal contaminated site risk assessment in Canada, part V: Guidance on human health detailed quantitative
risk assessment for chemicals (DQRAChem), Safe Environments Directorate, Contaminated Sites Division, Ottawa, Canada.
Health Canada (2010b) Federal contaminated site risk assessment in Canada: Supplemental guidance on human health risk
assessment for country foods (HHRAFoods), Safe Environments Directorate, Contaminated Sites Division, Ottawa, Canada.
Health Canada (2012) Federal contaminated site risk assessment in Canada, part I: Guidance on human health preliminary
quantitative risk assessment (PQRA), Version 2.0, Safe Environments Directorate, Contaminated Sites Division, Ottawa,
Canada.
Office of the Parliamentary Budget Officer (2014) Federal contaminated sites cost, Ottawa, Canada, viewed 16 January 2015,
htt://www.pbo-dpb.gc.ca, last updated 10 April 2014.
Power, B.A., Tinholt, M.J., Hill, R.H., Fikart, A., Wilson, R.M., Stewart, G.G., Sinnett, G.D. and Runnells, J.L. (2010) A risk-ranking
methodology for prioritizing historic potentially contaminated mine sites in British Columbia, Integrated Environmental
Assessment and Management, Vol. 6(1), pp. 145–154.
Unger, C.J., Lechner, A.M., Kenway, J., Glenn, V. and Walton, A. (2015) A jurisdictional model for risk management, accountability,
and continual improvement of abandoned mines remediation programs, Resources Policy, Vol. 43, pp. 1–10.
US Environmental Protection Agency (EPA) (1998) Guidelines for ecological risk assessment, EPA/630/R-95/002F.
US Environmental Protection Agency (EPA) (2007) Framework for metals risk assessment, Office of the Science Advisor, Risk
Assessment Forum, EPA 120/R-07/001.

R. Montpellier The Mining Industry Human Resources Council, Canada
Due to the cyclical nature of the mining industry and the finite lifespans of mines and mills, it is not uncommon
for a worker to be employed at up to six different mines throughout the course of his or her career. In light of
this, and to facilitate the redeployment of a workforce following a mine closure, some employers are turning
to national worker certification as a way to support employees in their transition to new and continued
employment in the mining industry. By training and certifying workers to a national occupational standard
(NOS) for their respective occupations, employers are providing employees with nationally recognised,
portable credentials – documentation of their skills and experience they can take with them to prospective
employers.
The Mining Industry Human Resources Council (MiHR) has launched Canada’s first and only national mining
worker certification program, the Canadian Mining Certification Program (CMCP), which has certified over
1,000 workers. This paper examines the strategic approaches of Agrium Inc.’s Kapuskasing operation, Xstrata
Zinc’s Brunswick mine and Northgate Minerals Corporation’s Kemess mine to illustrate how employers are
leveraging worker certification as one component of a responsible mine closure strategy that will help to
support the process of transition and make redeployment easier. Certification provides departing workers
with a professional credential that is proof of their skills and competencies, so potential employers have a
clear indication of what the employee is capable of before he or she is hired. By ensuring workers are prepared
for transition in the period leading up to or following mine closure, companies strengthen their legacy within
communities. A case study analysis of how worker certification was implemented as a part of responsible
mine closure at these three operations offers insights into this approach and how it supports the overall
retention of workers in Canada’s mining industry.
Since its inception in 1996, MiHR has convened industry to develop stakeholder-led human resources (HR)
programs that focus on the sustainability of the mining industry. MiHR’s most recent analysis shows that the
industry’s hiring needs over the next decade, despite the current market conditions, are forecast at 120,000
workers. Trades and production occupations continue to make up a large proportion of the mining industry’s
hiring requirements, and there is significant pressure to replace retiring workers. This category includes high-
demand occupations where mining must compete against other sectors for talent (e.g., haul truck drivers
and heavy equipment operators). It also includes a number of occupations where there are simply not
enough new entrants to meet needs projected for the next 10 years (e.g., underground production and
development miners and supervisors in mining and minerals processing).
MiHR’s labour market information is recognized as the authoritative resource for the Canadian mining
industry because it is sector specific and has been validated since 2007. It provides accurate and timely labour
market information to mining industry stakeholders, forecasting future hiring requirements in the sector by
occupation and region based on fluctuations in commodity prices, which are the largest driver of employment
in the sector. This information enables stakeholders to take practical measures to mitigate the risks
associated with labour shortages (or surpluses).
Figure 1, showing the age distribution in the Canadian mining industry, highlights the significant number of
mining employees that are eligible to retire in the next five to 10 years. In a mine closure situation, if these
workers are downsized, a huge amount of valuable knowledge is lost and not transferred. This practical

Worker: certification a tool for responsible mine closure and worker redeployment R. Montpellier
knowledge cannot be replaced by education, so a different kind of skills gap in the industry is created, where
supervisory and management staff have to be promoted without the knowledge and experience they need
to succeed. Tools like the MiHR-developed National Occupational Standards (NOS) can help transfer
knowledge associated with the roles of underground miner, surface miner, minerals processing operator and
diamond driller. In 2016, MiHR will also add supervisors, trainers and hoist operators to this list. The NOS for
supervisors will be particularly valuable to facilitate knowledge transfer in a mine closure or downsizing.
Source: Mining Industry Human Resources Council, 2013
The top 10 occupations with the greatest forecasted hiring requirements across the next decade are listed
below and detailed in Table 1:
 Underground production and development miners
 Heavy equipment operators (except crane operators)
 Construction millwrights and industrial mechanics (except textile mechanics)
 Truck drivers
 Welders and related machine operators
 Primary production managers (except agriculture managers)
 Machine operators, mineral and metal processing
 Heavy-duty equipment mechanics
 Industrial electricians
 Supervisors, mining and quarrying
These hiring requirements, which are produced by the MiHR labour market information model and validated
by an industry committee annually, indicate a baseline scenario where there is no growth or contraction in
the industry. The full annual report (MiHR, 2014) also shows the hiring requirements in contractionary and
expansionary scenarios. MiHR has also completed provincial custom analyses for Ontario, Saskatchewan,
British Columbia, the Yukon and the Northwest Territories; these are all available on request.

Risk Assessment
MiHR’s research has shown that employers who are downsizing find that workforce transition support
instituted prior to layoffs results in better attendance for affected employees and less disruption to quality.
Remaining employees are positively affected, since they have concrete evidence of the employer’s concern
for workers. Thinking of the long term, companies may be currently downsizing, relocating or closing, but in
future may want to retain certain employees or rehire them.
In the context of long-term strategic workforce planning, providing employees with a credential that smooths
and enables their transition to another mine also retains them in the industry, reducing the chance of further
shrinking the future labour pool. This is essential, considering the hiring requirements over the next decade.
NOS provide a detailed inventory of the skills, knowledge and competencies required to perform specific jobs
safely and productively in the mining industry. They also form the foundation of the Canadian Mining
Certification Program (CMCP), which has now certified over 1,000 workers from across the country as surface
miners, underground miners, diamond drillers and minerals processing operators. The first four NOS for
mining – underground miner, surface miner, minerals processing operator and diamond driller – were
prioritised by industry and developed from 2006 to 2008. In January 2014, MiHR received funding from
Employment and Social Development Canada (ESDC) through its Sectoral Initiatives Program (SIP) to develop
three new standards for the roles of frontline supervisor, industry trainer and hoist operator, as prioritised
by industry through extensive research.
NOS are developed by a committee of industry subject matter experts (job incumbents, managers and
trainers) and validated by a broader group of industry representatives. Committee members and validation
participants are selected to represent a broad range of commodities, company sizes and regions. The NOS
development committees (NOSDCs) meet in person and by teleconferences and, through facilitated sessions,
conduct the occupational analysis and create the occupational standards. Meeting facilitators are trained to
guide the discussions, ensuring that the NOS are consistent and broadly representative of industry’s
requirements for the occupation being defined.

Once the standards have been developed and the validation process is completed, the committees are
reconvened to finalise the NOS and authorise the release. At this time, the NOSDC can also develop
certification schemes, outline training programs or set training standards for the occupation.
In developing NOS for the mining industry, MiHR has so far focussed on occupations that have been described
as “semi-skilled” because they have lower requirements for formal post-secondary education or because
there is a lack of formal recognition of these occupations at the provincial/territorial or national level. NOS
development at MiHR has also focussed on those occupations for which industry demand or hiring
requirements are anticipated to be the greatest in the future.
To date, MiHR’s four NOS have been used in certifying employees and, to a lesser extent, in accrediting post-
secondary education programs. Nevertheless, as outlined above, the application of NOS can be much
broader, supporting workforce planning and training program and curriculum development and facilitating
recruitment, hiring and retention.
NOS establish clear, objective benchmarks and are used in a variety of industries in Canada and around the
world to do the following:
 Support personnel certification.
 Facilitate recruitment and expedite hiring processes by informing job descriptions and providing
the basis for job applicants.
 Inform training program development and assess third-party training providers.
 Identify career paths to support employee retention.

Risk Assessment
 Evaluate and determine the qualifications of potential employees, including internationally

trained workers.
 Establish performance criteria and develop training plans.
 Support succession planning by identifying the critical competencies of experienced workers who
are transitioning out of the industry and identifying the training and experience required for more
junior workers to assume more senior positions.
 Assist with mine closure or downsizing.
Canada’s NOS are garnering international attention amongst global mining employers and Canadian
companies operating abroad and are often viewed as the gold standard to which other countries aspire – a
great compliment to the Canadian mining industry.
As custodians of the NOS, MiHR receives inquiries from across the globe from countries such as the United
States, Ghana, Turkey, Indonesia, Peru, Colombia, South Korea, Chile and Kazakhstan. Some nations are
referencing the standards for research and comparison or as a model to develop their own NOS, while others
are interested in certifying their workers against the Canadian NOS. This global interest points to a growing
commitment to standardised skills and training, safety and productivity at mine sites, and the development
of a highly-skilled mining workforce. As the program expands, this will open up global opportunities for
transitioning workers affected by mine downsizing or those who are able to secure work with a contractor.
NOS and certification also play an important role when market conditions result in downsizing or mine
closure. Xstrata’s Brunswick mine, Kemess mine and Agrium’s Kapuskasing phosphate operations have all
used the CMCP as one tool to assist in responsible mine closure. Workers who have participated and been
certified under the CMCP have a professional credential that is recognised throughout the mining industry in
Canada and can be used to demonstrate their transferable skills to other industries. This is also beneficial to
mining employers and workers as the market recovers, as it enables workers to transition back in to roles
that reflect their skills and experience. Prior to the creation of the NOS for underground miner, surface miner
and minerals processing operator, these occupations were referred to as “undesignated.” The lack of a
formal, national recognition system led to high turnover rates, inconsistent training and an inefficient labour
market. As it can take anywhere from two to five years to train a skilled worker for the mining industry,
understanding the future need and having the tools in place to plan for this is essential.
For mines that are downsizing and have employees choosing retirement, NOS provide a foundation for
effective succession planning in high-demand jobs. Clarity regarding the important skill sets that will be lost
through retirement attrition is essential so that employers do not create unmanageable skills gaps on site.
The following process for responsible mine closure is based on MiHR’s Mining Workforce Transition Kit,
developed by MiHR and industry experts to assist employers in times of transition, and on lessons learned
from the responsible mine closures of the case study companies detailed in Section 4: Northgate, Xstrata and
Agrium:
1. A transition committee is established: this should include representatives from HR, training, the
union and employees.
2. The transition committee meets with MiHR to discuss timelines and the use of NOS and
certification.

3. The transition committee appoints a CMCP implementation team, which should include
representatives from the training department, HR and recruitment.
4. An action plan and certification timeline is created and agreed by the CMCP implementation team
and MiHR’s Office of the Registrar.
5. The transition committee establishes and equips an information centre for transitioning
employees with materials on certification (MiHR supplies materials). This information centre is
usually integrated with other resources that assist transitioning employees, such as interview
guides, resume writing tips, job hunting advice, etc.
6. The training lead in the committee aligns the company training standards with the relevant NOS.
This will expedite the certification process.
7. A workplace assessor is nominated by the employer and trained by MiHR on how to assess
candidates on the job. Timeline is based on the mine closure plan and employees’ skills and
experience.
8. Employees are enrolled into the certification program.
9. All employees who have applied for certification conduct self-assessments, which are based on
the NOS, and submit these to the workplace assessor.
10. Workplace assessments are conducted with candidates on site, using the CMCP record of
assessment (ROA), based on the NOS.
11. Employees are certified once they perform tasks at the level of competency required in the
certification scheme and have achieved the required number of hours in the workplace, as
documented by a workplace assessor in the ROA and validated by the MiHR Office of the
Registrar.
12. Upon certification, employees receive recognition items: a skills passport, lapel pin, hardhat
sticker and certificate.
13. MiHR provides employees with tools to show recruiters and potential employers how to interpret
the skills passport and ROA. These tools benefit employees and employers by enabling faster
integration into the new workplace, eliminating the need to duplicate training and validating
workers’ skills by a nationally recognised organisation.
14. The ROA is a valuable, detailed inventory of the employee’s skills and experience; combined with
the skills passport and other certification items, it assists employees in their job search and in
marketing their skills.
The ROA is built from the NOS and is a tool for trainers to assess the skills and competencies of a worker
against the standard and against the required levels for certification in that occupation. Figure 3 shows
examples of the ROA.

Risk Assessment
Once employees are certified, they receive recognition items (pins and certificates) and the tools to
demonstrate their skills and competencies to their potential employers (skills passport, ROA). All workers’
records, once they are certified, are held by the MiHR Office of the Registrar, a resource that helps employees
to maintain their career growth and helps employers to validate candidates who are applying for work.

Figure 4 shows the skills passports for each occupation, certificate and pins.
We are currently gathering data on using the skills passport and ROA instead of a resume in the hiring process
to identify any efficiencies gained through the new system. A second metric that is being explored is whether
the skills passport and ROA reduce the time required to integrate new employees, given that their skills are
assessed and validated so training dollars and time are not duplicated.
Agrium Inc. had a closure plan in place for the Kapuskasing phosphate operation very early on. When the
mine first opened in 2000, Agrium had estimated that the mine would close around 2014. That target moved
slightly, and in 2010 it was announced the mine would close in 2013. Agrium implemented the CMPC to
complement a long-term mine closure and transition plan that emphasised the marketability of departing
employees’ skills and competencies in support of their effective transition to new employment in the mining
industry.
Pierre Plamondon, human resource business analyst at Agrium, was part of the transition team at
Kapuskasing. He said the company had been very open and engaged with employees about the closure:
We were working through many planning activities that involved preparing employees for
employment transitions. We really worked with all employees to plan those transitions as far back
[as] a couple of years before closure. Some employees completed training and courses in
preparation. A few accepted Agrium’s offer to relocate to our other operations in Saskatchewan
and Alberta. However, most were rooted in northern Ontario and planning to stay in the region.
Agrium’s goal once the closure was announced shifted to packaging departing employees’ work portfolios in
ways that best allowed them to easily share them and showcase their years of training and experience and
the competencies they had each acquired in their respective fields of work.

Risk Assessment
Agrium first learned about the CMCP only a few months prior to the closure, but because it so closely aligned
with the company’s existing transition plan and provided employees with a portable portfolio of their skills
and experience through national certification, they invested in the training and resources to implement the
program in a compressed timeline.
To achieve certification through the CMCP, workers went through prior learning assessments and/or on-the-
job demonstrations of skills with an MiHR-qualified workplace assessor, who captured and documented their
level of competency in the required areas, as outlined in the minerals processing operator NOS. Employees
worked closely with the HR team to access the information (HR records, training certificates, and so on) that
would serve as supporting evidence in their application for certification. Agrium then combined these
certification efforts with its other transitional support programs, such as the creation of training portfolios,
employee profiles and resume updates. This information was leveraged when Agrium reached out to local
employers and other interested mining sector employers ahead of closure to showcase and introduce the
employees. Plamondon notes that a program like this requires team commitment:
All of our leaders, from site manager to front line supervisors had employee transitions as a main
priority. We supported employees through their transition; they supported Agrium through a very
successful closure. Our focus to help front line workers in this area allowed them to focus on safely
operating the plant to the very end. This was beneficial for everyone including our community
partners and the mining sector in the area and it allowed us to retain all of the critical skills
necessary to operate the plant, to process all of the mined ore and to successfully close the mine
site.
A total of 19 minerals processing operators from the Kapuskasing operation received their Level 1
certification, and Plamondon says the initial impact was twofold. First, employees working through the
certification process were able to strategically organise their acquired skills, knowledge and experience,
which has helped prepare them to market themselves. Second, they were given a national certification, which
is something that had eluded minerals processing operators in the past. “It put them on equal footing with
peers in other professions,” Plamondon explains. “They were able to share this certification with employers
interested in their skills (ahead of closure, in many cases) and the prospective employer was able easily assess
their qualifications and overall value as an applicant.”
Plamondon says certifying the operators as one component of the company’s overall closure strategy helped
to support the smooth transition of the Kapuskasing workforce; over 85% of the employees had secured work
for after closure even before their jobs were completed at Agrium.
For Xstrata Zinc, the decision to offer the CMCP was of strategic importance from a corporate social
responsibility standpoint. As part of responsible mine closure, it was important to the company to ensure
workers were transitioned and prepared to begin seeking new employment. Here as well, the CMPC provided
workers at a closing mine with nationally recognised, portable credentials to support their transition to new
employment and their continued career in the mining industry. By implementing the CMCP at their site,
Xstrata Zinc’s Brunswick mine had workers trained and certified to the NOS, which provided them with
documentation of their skills and experience to take with them to prospective employers.
Prior to the development of the CMCP, many occupations in mining were considered “undesignated,”
meaning they lacked a nationally recognised professional credential, so a worker’s skills and competencies
were tied to the mine at which they were employed. MiHR’s research has indicated that the turnover rate
for these undesignated occupations, such as underground miner, minerals processing operator and diamond
driller, is twice as high as the average in all occupations in mining.
After operating in the city of Bathurst, New Brunswick, for close to 50 years, Xstrata Zinc wanted to preserve
its legacy in the community by providing its workforce with the means to demonstrate the skills and

experience that many residents had earned throughout the duration of their careers at the mine, and MiHR’s
certification program provided the most value for the workforce because of its national recognition.
Brunswick mine certified 190 workers as underground miners and minerals processing operators over a span
of two and a half years, as the mine’s operations wound down. These workers became some of Canada’s first
nationally certified mining workers, and they now have recognition for their skills and experience in the
industry, giving them a portable credential that supports applications for future employment and validates
what is on their resumes. The certification includes a skills passport, a certificate, a lapel pin and a wallet
card. “I think it’s a win, win,” says Georges Ouellette, who was the mine’s production support superintendent
before the closure in early 2013. “There’s no doubt about that in my mind. It’s something that’s been lacking
in the industry for many years. The employees obviously love it because they are going to be recognised.”
Euclide Haché, a certified Level 2 underground miner, and formerly a mobile equipment miner at Brunswick,
agrees:
I have always been proud to be a miner. I’ll say to the person who claims that I’m not competent
that it’s not true because in my occupation it takes skills and aptitudes to be able to come to work
in the morning and come out at night exactly the same way you were in the morning. I also have in
my wallet a card and a [skills] passport that can prove it.
Even before the Brunswick mine closed, Cementation Canada had conducted job fairs in Bathurst to recruit
some workers from the mine, and many offered their CMCP passports as proof of their skills, which
Cementation, another participating CMCP employer, readily accepted. This highlights another key advantage
of the CMCP: savings in time and training. A certified employee is trained to a national standard, and the
skills passport documents the skills and competencies the worker has demonstrated on the job to achieve
his or her certification, so potential employers have a clear indication of what the employee is capable of
before he or she is hired.
The employer can also easily identify any gaps in training and/or experience and target their training dollars
toward getting that experienced worker mobilised in their active workforce faster and more efficiently.
The Kemess Mine (now owned by AuRico Gold) was in operation for 14 years prior to its closure in 2011. A
transition plan was implemented a year before the projected time of closure, once management knew the
ore body was depleted, but planning was underway long before that, says Linda Hodgson, who was the HR
superintendent with Northgate Minerals at the Kemess Mine and is currently an HR consultant with AuRico
Gold. Northgate made use of the certification program to supplement the transitional support offered to
workers at a closing mine in order to reward long service and help them secure employment at their current.
Employees were kept informed about the projected time of closure and how they would be supported in
their transition to new employment. Hodgson says she thought offering national certification was an
excellent way to assist their departing surface miners in securing employment at their current level, rather
than having to start at the bottom with a new employer:
From my experience, I have seen so many people leave a mine and go to another mine (maybe
they are a shovel operator and have to go down to a truck driver) because there are no credentials
they can take with them that [are] recognised across the country, or even across the province.

Risk Assessment
Several employees had given many years of service to Kemess, having been employed for the full 14-year
lifespan of the mine as shovel operators, and Hodgson says being able to offer national certification to some
of the departing workers was a way to reward this long service and to help increase their likelihood of finding
employment and staying in the mining sector.
The certification program was implemented alongside other transitional programs Kemess was offering,
which included training updates in mine rescue, first aid and fall arrest. The company hired a workplace
transition coordinator to lead a transition committee and help workers update their training and resumes.
Equipment trainers worked with the certification candidates to assess their skills and competencies against
the surface miner NOS and help them compile HR and training records that would form the evidence required
to support their application for certification. The workplace transition coordinator then worked with the
certification candidates to compile this information into updated resumes, making the processes
complementary and efficient.
Hodgson says certification is an excellent tool to work into an effective transition plan to support a mine
closure – one that will help to retain employees throughout the transition and keep them engaged:
One of the things that is so important is that people continue to work at the same level right to
mine closure, and that you don’t lose the engagement of the employees … Because when you get
to the end of a mine closure, people are often asked and sometimes expected to work outside of
their designated area, and so if you can’t keep these individuals engaged and they are just putting
in time, your chances of having a successful closure are comprised.
If you treat people with respect and provide them with the tools and support to carry on, you are
more likely to retain your workforce throughout a closure – and offering individuals the
opportunity to obtain national certification in their occupation is certainly one component of a
transition plan that will make them feel valued and supported, and help keep them engaged.
The feedback we have received from employers and a transitioning worker survey indicates that certification
is a valuable tool for mine workers affected by mine closure or downsizing. The program was launched fully
in 2011, and we are starting to see examples of how the program is facilitating recruitment and retention
and assisting workers affected by mine closures or downsizing. In 2015, we are researching key metrics with
our implementing sites to gather data and metrics on any impacts on retention and on the role of certification
documents in facilitating recruitment and making training and integration more efficient.
This data will enable employers to understand the business case for certification, whether they are in
production or moving toward closure. For more information please contact rmontpellier@mihr.ca.
Mining Industry Human Resources Council (MiHR) (2014) Canadian mining industry employment, hiring requirements and available
talent 10-year outlook, Mining Industry Human Resources Council, Kanata, Ontario, http://www.mihr.ca/en/resources/
Hiring_Requirements_Available_Talent_10_year.pdf.
Mining Industry Human Resources Council (MiHR) (2013) Canadian mining industry employment, hiring requirements and available
talent 10-year outlook, Mining Industry Human Resources Council, Kanata, Ontario, http://www.mihr.ca/ en/resources/MiHR
_10_Year_Outlook_2013.pdf.
Mining Industry Human Resources Council (MiHR) (2010) Mining workforce transition kit, Mining Industry Human Resources Council
and Government of Canada’s Sector Council Program, http://www.mihr.ca/en/publications/ MiHRPublications.asp.
Mining Industry Human Resources Council (MiHR) (2014) National occupational standards and certification, http://www.mihr.ca/
en/priorities/NationalStandards.asp.


S.R. Pearce O’Kane Consultants, Australia

S. Lehane O’Kane Consultants, Australia
The characterisation and assessment of waste rock has been the subject of considerable research, and many
guidance documents have been published internationally and within Australia related to operational and
closure risk management. While these documents provide detailed information on how to characterise waste
rock, there is not a great amount of detail on how to determine optimal waste-management options from the
perspectives of operations and closure risks and outcomes. Indeed, much closure risk mitigation has focused
on risk-mitigation measures that can be applied close to the end of landform construction, such as cover
systems, rather than on how risk mitigation can be more progressive in nature (i.e., incorporated into
operational variables such as waste placement technique).
A multitude of interrelated factors and variables contribute to assessing the closure risks of waste
management. This paper uses a semiqualitative and quantitative methodology to approach these
assessments. Quantitative methodologies improve confidence in these assessments by providing logical,
measureable assessments.
As part of a strategic mine-closure risk assessment for waste management at an active mine site, the authors
recently completed a detailed risk-based investigation of the influence of waste placement factors on mine-
closure risk. These factors include placement method, placement height, sulphur grade determination, and
material conditioning (e.g., compaction) on the associated risks of waste material storage. The assessed risks
include temperature assessments (the potential for thermal take-off as a result of spontaneous combustion),
toxic gas production and flow, and acid production potential. Applying a numerical modelling approach to
assessing the different waste material placement strategies allows us to graphically present and
communicate the variation in risk through risk matrices and histograms.
Producing and integrating multiple risk matrices as part of a site-specific, semiqualitative method of risk
assessment provides significant benefits for optimising waste placement options and managing closure risks
in a more progressive manner.
The characterisation and assessment of waste rock forms the initial stage of waste-management strategy
planning that extends through the life of the mine to assessment of closure risk. Prior to operational decision
making, waste-management strategies should evaluate both short- and long-term (closure) risks associated
with the exposure, stockpiling and placement of waste materials. This includes risks such as spontaneous
combustion, toxic gas production, and acid and metalliferous drainage (AMD). These risks are complex and
interrelated and are associated with air and water entry into the waste material, where subsequent oxidation
reactions can occur (Lottermoser, 2010).
Because the time horizon of risk assessments for mine closure is extended, and the horizon related to
operational risk management is short term, risk-mitigation strategies can be implemented at both
operational and closure phases. However, much mine closure risk mitigation has typically focused on
measures that can be applied close to the end of landform construction, such as cover systems, rather than
on measures that can be incorporated into operational variables, such as waste placement techniques that
allow for a more progressive risk management approach.

Quantitative risk assessment tools to assist with waste management and placement guidelines S.R. Pearce and S. Lehane
It is common practice in waste characterisation to classify material primarily on the basis of geochemical risk
factors: for example, as potentially acid forming (PAF) or non-acid forming (NAF). Material that is determined
to pose significant risks of AMD, such as PAF, is then prescribed a specific management method such as
encapsulation as part of a placement strategy to reduce potential AMD risks. However this method of
assessment is very prescriptive and polarised, and materials are categorised into very few categories, such
as PAF or NAF. These can become catch-all categories, which can result in polarised decisions, such as that
all PAF must be managed in a set manner. In addition, construction and placement techniques are not often
specified or do not vary depending on the PAF or NAF classification. This method of assessment does not, in
reality, have a close association to field conditions, as there will be a broad range of risk, from low through
high. Several variables control the level of AMD risk:
1. Sulphide content of material
2. Physical properties of material (grain size and distribution, weathering rate)
3. Structure of waste rock due to placement (pathways for air and water movement)
4. Climate (temperature and rainfall)
5. Closure mitigation measures (such as cover systems)
On a typical site, operators will have little control over Variables 1 and 4, which are fixed to an extent, and
they will have limited control over Variable 2; the main variables that can be controlled are Variables 3 and
4. Variable 5 is often considered the only variable over which control and thus risk mitigation can be
employed, as this does not impact on operational decisions. However, Variable 3 (the structure of waste rock
due to placement) is a very significant risk factor. Although it is common practice for this variable to be
considered as fixed based on the placement of waste rock being a cost-driven operational decision, this paper
demonstrates how critical this variable is with respect to risk management (in both short term and closure).
Evaluating risk based on multifaceted variables requires application of semiquantitative analysis, such as the
operation of an analytical model. This paper presents an assessment process based on a risk matrix that
captures these multifaceted inputs and employs an analytical model to provide semiquantitative analysis and
outputs. This method of assessment allows operational and closure risks to be assessed on the basis of
operational placement technique and not just on material properties in isolation.
WRSF landforms are typically constructed by end-tipping, paddock dumping, high wall tipping or a
combination of these methods. The specific method used on a given site to construct a landform is generally
based on the availability of equipment, cost and scale of construction, and construction methods are far from
uniform across all sites. Construction methods will also vary depending on engineering design parameters;
for example, the requirement for a low-permeability layer to help manage oxygen ingress may require
compaction of a layer of paddock-dumped clay-rich material. Similarly, the placement of PAF materials in
cells may require the use of short tipheads (<10 m).
Based on site-specific variables, there are therefore multiple variations of construction methods used on a
site. Examples of typical construction methods employed where PAF materials are being codisposed into a
WRSF are shown in Figure 1.

Risk Assessment
The term “encapsulation” refers to the placement of specific materials within confined cells. There is often
no engineering specification for this encapsulation process, and therefore the use of the term in an
engineering context should be viewed with caution.
What is common to all sites is that the construction method is a significant factor in determining closure risks,
and it can be considered as an independent variable, along with materials characterisation. This is because
the structure of the WRSF will dictate variables that directly control closure risks with respect to AMD risks.
Examples of these variables include
 Oxygen ingress
 Net percolation rates
 Erosion and stability
 Gas flux
 Seepage flux
When assessing closure risk, the construction of a WRSF is best viewed from the perspective of progressive
closure risks, which in turn requires consideration of both construction method and materials
characterisation. Detailed consideration of how these risks are influenced by operational construction
methods may therefore indicate how cost-effective risk-mitigation measures can be implemented during
construction.
The potential for excess heat production to lead to spontaneous combustion and for acid generation to
require mitigation and remediation were considered the focal risk elements for the development of a risk
matrix approach based on semiquantitative analysis. For the purpose of this study, contributions to these
risks were limited to the following factors:
 Waste material placement strategy (paddock dumping or end-tipping, with variable tiphead
heights)
 Geometry of the waste rock storage facility (WRSF) (height, width, length)
 Sulphur grade of the material
 Geological classification and potential for acid neutralising capacity (ANC)

 The mine schedule (waste material types and volumes over time)
 Application of risk-mitigating engineering techniques (treatment and conditioning of material
such as compaction, placement of compacted layers of material)
While these factors are often considered in a qualitative manner, this assessment applied iterative numerical
modelling based on conceptual models developed with the client from a large database of geological and
geotechnical data and in situ monitoring data.
The specific objectives of the analytical model are as follows:
 Provide a basis for comparing potential risks associated with placement of materials of different
geochemical properties using different WRSF construction techniques.
 Inform upon potential waste material management strategies.
 Enhance understanding of potential risks (both short-term operational and long-term closure).
 Inform closure risk-management options.
 Assist with decision making on waste material disposal.
The following methods were adopted to achieve these objectives:
 Complete quantitative oxygen ingress modelling into the waste material under different disposal
scenarios.
 Complete long-term sulphide oxidation-rate modelling of waste mass under different disposal
scenarios.
 Calculate the potential spontaneous combustion risk related to disposal of black shales.
 Calculate acidity generation based on sulphide oxidation with various sulphur grades.
The potential for spontaneous combustion of waste rock materials is well documented in the literature.
Lottermoser (2010) describes the inter-relationship between oxygen availability (both atmospheric oxygen
and dissolved oxygen allow for the exothermic reaction to take place) and fine particle size, which insulates
and retains generated heat, initiating a positive feedback reaction whereby increased heat increases reaction
rates. Numerical models designed to identify potential heating zones and increased risk areas have been
developed specifically for underground coal mines; these can be expanded upon for WRSFs (Humphreys,
1998).
AMD risk is arguably the most well-defined and researched risk associated with WRSFs, and it has more
extensive and prolonged impacts. The industry and research communities alike are working to minimise and
mitigate long-term costs and potential legacy issues of this issue (Pearce et al., 2012). The ubiquity of pyrite
throughout geological environments associated with coal and metal deposits has led to extensive research
into its oxidation and effects, namely heat generation and decreasing pH, both of which increase the reaction
rate of oxidation and further acid production and heat generation (Lottermoser, 2010).
While the issues above are identified as risks, it is still difficult to determine how this relates to operational
controls and how mine planning and operations can utilise this data to inform and assist with prioritising the
characterisation, scheduling and placement of materials to minimise these risks. Waste material placement
and sulphur grade of waste material are two factors that are most likely to be monitored and controlled in
relation to heat and acid generation.

Risk Assessment
Heat generation within a WRSF is largely attributed to the heat generated by exothermic oxidation reactions
known to occur within the waste material. For waste material that contains both carbon and pyrite, the
consumption of oxygen by the oxidation of organic matter and pyrite must be considered. Bennett et al.
(1995) state that, in coal wastes, the consumption of oxygen by the oxidation of carbonaceous material must
be accounted for. Field pyrite and carbonaceous shale oxidation rates (POR and COR) are based on column
leach test experiments to derive the intrinsic oxidation rate (IOR) for pyrite based on sulphate release from
the test. A scaling factor (SF) is then applied to estimate actual field oxidation rates from these laboratory
tests.
A temperature factor (TF) is intended to account for the increased reaction rate of carbonaceous shale, which
occurs as temperature increases. In the waste material containing both pyrite and carbon, initial temperature
in the AMD is usually driven by pyrite oxidation. Anecdotally, it has also been observed that temperature
increases rapidly once the temperature reaches 70°C; this is termed “thermal runaway.” Relevant laboratory
tests are required to understand how oxidation rates increase with temperature.
PORs are based on the IOR determined from column leach tests of individual samples. A general rate has
been calculated for the IOR based on sulphur content, which provides a good correlation for samples having
<4 wt% S (Figure 2). This provides a reasonable estimation of the oxygen consumption rate with changing
sulphur grade. The COR is based on a literature review for coal, which has 31–41 wt% C. Further work is
required to confirm CORs for the carbon contained within the site-specific material, particularly across a
range of temperatures.
The use of laboratory kinetic rates for pyrite and carbon oxidation indicates that pyrite oxidation kinetics
dominate when oxygen and water are freely available; this in turn implies that it is the heat generated by
pyrite oxidation that causes the initial heating of the pyritic shale, leading to spontaneous combustion.

Research indicates that transferring kinetic rate data from laboratory column leach tests to a commercial
scale WRSF, where oxygen is freely available, reduces reaction rates by an order of magnitude. For example,
the calculated IOR for waste rock was an order of magnitude lower in trial dump experiments compared to
the 500 tonne trial and lab columns in the work of Andrina et al. (2012). Most importantly, Andrina et al.
(2012) note that the trial dump was often well oxygenated, with oxygen content at 20%, which is comparable
to atmospheric concentrations. Based on these studies, when oxygen is generally freely available, it may be
proposed that the laboratory-based IOR can be reduced by an order of magnitude due to scaling effects,
which we refer to here as a laboratory-to-field conversion factor (LFCF).
Reactivity of material also refers to the propensity for acid generation. Total sulphur content is often used as
an analogue for calculating pyrite content and maximum potential acidity (MPA). Therefore, understanding
the anticipated sulphur schedule allows us to plan appropriate risk management on site to minimise the
potential for environmental damage and/or harm to personnel.
Geochemical characterisation of material can then be included in a more general material characterisation
for production scheduling and life-of-mine (LOM) planning, as discussed in the following section.
Any assessment of waste placement scenarios must consider the production schedule and the availability of
materials, potentially for staged or phased placement. Identifying the waste rock lithologies and varying
material properties is fundamental to developing a waste materials placement plan. Particularly with PAF
waste material, it is important to identify the physical properties of materials and to understand how each
waste material behaves with regard to potential material segregation. Scheduling of each waste material in
relation to the other waste materials should consider both materials’ geotechnical and geochemical
characteristics, as the WRSF and this material placement arrangement can potentially promote or prevent
the risks associated with the PAF waste material.
Calculating the proportions of each material during the production schedule allows for a more complete and
robust waste placement design, with materials classified and assigned for specific engineering purposes, as
well as geochemical reactivity. Materials in this study were classified as shown in Table 1.
Figure 3 shows the proportions of materials over time, highlighting the importance of mine life planning;
Class 1 and 2 materials decrease while Class 3 and PAF materials increase in the excavated proportions.

Risk Assessment
Class Description
1 NAF waste material that is suitable for use in the outer
zones of a WRSF due to its inherent hardness and size
distribution material properties, which promote
construction of a stable landform
2 NAF waste materials that have characteristics similar to
Class 1 materials but are associated with lithologies that
may contain elevated sulphur levels. They are not
necessarily PAF, but will have a sulphur level <0.2 wt% S
and will likely have unfavourable impacts on the success of
landform recovery.
3 NAF waste material that does not possess material
characteristics that allow for successful rehabilitation of
WRSFs. Properties include dispersiveness (low electrical
conductivity, clays); low pH potential, although not
classified as PAF; high degree of oxidisation; or fine-grained
poorly-cemented lithologies.
PAF Material with the potential to generate AMD or
spontaneously combust
Reactivity of materials is defined by both the physical and geochemical properties of each material type, as
grain size and moisture retention properties are integral to the holistic understanding of the material type
(Weber et al., 2014).
One current method for waste material placement is by end-tipping directly over the tiphead rather than
push-dumping. Examples of such tipheads are shown in Figure 4; these suggest that the slope can be
comprised of alternating coarser and finer textured materials and a basal rubble layer of coarse boulders.
The variances in grain sizes seen in Figure 4 also seem to be related to the particle size distribution (PSD) of
the waste rock type after blasting and disposal to the tiphead.

Based on site-specific observations of the general PSD for the lithologies present, it is reasonable to assume
that simply dumping these materials together will cause the formation of significant coarse- and fine-
textured layers down the tiphead; often these may be discontinuous, creating lenses.
Increasing tiphead height will also contribute to grain size segregation processes, leading to a general trend
of finer to coarser textured particles from the tiphead crest to the toe. Based on the literature available, it
appears that end-dumping may create greater segregation of particles compared to push-dumping.
In summary, end-tipping or push-dumping of waste rock results in the segregation of coarse- and fine-
textured materials and the formation of interbedded materials with coarse- and fine-textured particle sizes.
The resulting structure is a coarse basal rubble layer overlain by (finer) interlayered coarse- and fine-textured
waste rock at the angle of repose and perpendicular to the direction of the advancing OSA face. Such OSAs
can be divided into three zones of material, which are formed down a slope profile:
• Zone A (fines) occurs in the top 15%–30% of the slope and comprises >40% fines (where fines
are <2.3–4.3 mm), with layers of coarse- and fine-textured materials.
• Zone B (graded material) occurs at mid slope and comprises a well-graded mixture of material in
alternating coarse- and fine-textured layers.
• Zone C (coarse rubble) occurs in the bottom 15%–30% of the slope and comprises mainly cobbles
and boulders (>40%) and very little fines (<10%).
In developing an adjustment factor for modelling purposes to manage air ingress and the potential for AMD,
gas, and spontaneous combustion, it is reasonable to assume the greatest risk is associated with codisposal
of PAF materials of distinctly different grain sizes onto a high tiphead. The risk is lowest when materials of
similar PSD are dumped using short tipheads or paddock dumping.
If WRDs are well constructed to minimise convective ingress of oxygen, such that IOR cannot be achieved for
a high proportion of the WRD, it is reasonable to assume that the IOR can be reduced further when the main
transport mechanism for oxygen is due to diffusion. The work of Brown et al. (2014) indicated that convection
of oxygen accounts for 90% of the oxygen transport into a WRD and diffusion accounts for 10%. This is close
to an order of magnitude difference between the two oxygen ingress processes. Hence, if waste rock is
dumped and the management of air ingress is not considered, then the IOR would not be reduced further
than the IOR adjusted by the LFCF, as oxygen would be freely available. This would be comparable to the
condition observed by Andrina et al. (2012). If the WRD is built to eliminate (control) convective transport of
oxygen, then the IOR for that WRD could be reduced by another order of magnitude.
A sliding scale (Table 2) has been developed as part of this conceptual work to compare a well-constructed
OSA, where oxygen is limited and the IOR is decreased by another order of magnitude, and a poorly
constructed OSA, where the IOR remains unchanged except for the initial tenfold decrease due to the LFCF.
Thus we propose a waste rock management factor (WRMF) ranging from 0.1 to 1.0. The WRMF is 0.1 for a
well-constructed OSA that minimises tip height; uses paddock and push-dumping methods and has minimal
fines segregation, preventing the formation of a basal rubble layer; and compacts lifts in 1 m to 2 m intervals.
The WRMF is 1.0 for a poorly constructed waste rock dump that has a well-developed basal rubble zone and
a strong segregation of coarse- and fine-textured materials.
A WRMF of 0.1 could also represent an OSA where diffusion is the dominant transport mechanism for oxygen
into the OSA; a WRMF of 1.0 could represent an OSA where convection is the dominant transport mechanism
for oxygen into the OSA.

Risk Assessment
Waste
rock Principal Combined
Lift Plume PSD effects Segregation General
tipped form of adjustment
height width adjustment adjustment compaction
layer oxygen factor
(m) (m) factor factor of WRD
thickness ingress (codisposal)
(m)
2 10 6.25 0.1 0.1 High Diffusion 0.10
4 10 3.13 0.2 0.1 Moderate Diffusion 0.13
6 9 2.31 0.3 0.1 Moderate Diffusion 0.16
8 9 1.74 0.4 0.4 Low Convection 0.40
10 8 1.56 0.6 0.6 Low Convection 0.60
20 7 0.89 0.8 0.9 Poor Convection 0.85
30 7 0.60 0.9 1.0 Poor Convection 0.93
40 6 0.52 1.0 1.0 Poor Convection 0.98
50 6 0.42 1.0 1.0 Poor Convection 1.00
60 5 0.42 1.0 1.0 Poor Convection 1.00
70 5 0.36 1.0 1.0 Poor Convection 1.00
80 5 0.31 1.0 1.0 Poor Convection 1.00
90 5 0.28 1.0 1.0 Poor Convection 1.00
100 5 0.25 1.0 1.0 Poor Convection 1.00
110 5 0.23 1.0 1.0 Poor Convection 1.00
120 5 0.21 1.0 1.0 Poor Convection 1.00
Application of this conceptual model indicates that end-tipping of waste material at lift heights greater than
30 m provides no reduction in oxidation rates other than the initial laboratory-to-field reduction.
Furthermore, application of the concept assumes that paddock dumping and subsequent compaction will
produce an order of magnitude reduction in oxidation rates such that the oxidation rates obtained in the
laboratory can be reduced by two orders of magnitude.
Several processes occur simultaneously within WRSFs. These processes include but are not limited to the
following:
 Chemical reactions: pyrite oxidation, oxidation of organic carbon, dissolution of salts, dissolution
of carbonates and reactive silicate minerals, subsequent neutralisation of acidity, etc.
 Physical change: compaction, segregation, particle size shrinkage, etc.
 Water flow: rainfall infiltration, seepage flow, matrix and/or preferential flow, vaporisation
 Gas transport: diffusion, convective and advective gas transport
 Heat transport: conduction and convective heat transport

These processes work in combination, and together they form a complex system in the WRSF. To model these
complex processes using one-dimensional modelling analyses, we made a number of assumptions, which
were based on site-specific data where possible. Table 3 shows the key modelling parameters.
Modelling parameters Consideration in modelling Key points

PAF/NAF ratio Ratio varies from 10:90 to 90:10. Ratio of PAF to NAF within any
one specific area of the backfill
varies with scenario.
PAF/NAF material Material properties are assumed to be Class 1 NAF may be disposed
properties consistent in most of the models as the with PAF in early stages of the
majority of materials are shale-rich Classes 2– schedule. Consideration of this
4. Class 1 competent materials are likely to potential is undertaken as part
have the most significant geophysical of comingled short tiphead
properties. scenario.
Segregation of materials Segregation of materials into course-fine- Segregation factor is used as
(coarse-fines) intermediate grade profiles is considered as input to modelling. Factor
part of the segregation factor function. ranges from 0.1 (paddock
dumping, 2 m lift) to 1.0 (30 m,
end-tipping).
Compaction (traffic Traffic compaction is considered as part of Traffic compaction considered
compacted layers) numerical modelling. as layers within the layered
PAF/NAF construction
scenario.
Sulphide content (% Analytical model considers sulphur grades of POR varies depending on
weight and total mass) 1 wt%, 3 wt%, 5 wt%, 10 wt% and 18 wt%. sulphur grade.
Carbon content (% weight Analytical model considers fixed organic COR fixed using 5% carbon as
and total mass) carbon grade of 5 wt%. reference.
Internal/external Analytical model calculates heat generation Gradient based on climatic

temperature gradient from oxidation reactions of pyrite and organic data for site.
carbon, which sets the temperature gradient.
AMD risks are estimated based on acidity production. Figure 5 shows the output from the model. The high
levels of potential stored acidity are apparent as only one case (paddock dump with 1 wt% S content) may
have a NAG less than 100 t (H2SO4)/day. Paddock dump construction poses the lowest AMD risk and may
limit NAG to less than 1,000 t (H2SO4)/day for all evaluated sulphur contents.

Risk Assessment
Although the model predicts the results of oxidation of sulphides as acidity generation, this may not be
released in a soluble form with seepage. Acidity may be stored within the waste mass as secondary sulphate
minerals and may form dissolved phase acidic discharges. On this basis, the calculation of tonnes of acidity
being released by the model may not accurately determine the potential for acidic seepage discharges (that
is, acid loading rates will likely vary significantly with respect to infiltration rates and therefore climatic
variability). Seepage models would need to be linked to this load model to determine the impacts on seepage
quality. However, the model produces a useful output as the magnitude of acidity generation for each
dumping method is displayed on a comparative basis, so the relative risks of the different methods can be
assessed. With respect to closure risks, the assessment is important because acidity generation may progress
at a sufficient rate such that by the time closure is achieved, a large amount of stored acidity may have built
up within the WRSF. In this scenario, the adoption of risk-mitigation measures such as cover systems may be
of limited benefit to reduce long-term AMD seepage risks.
Temperature results for different WRSF deposition completion scenarios are shown in Figure 6. The
maximum temperature in the WRSF can reach approximately 450˚C when end-tipped scenarios are
considered at higher sulphur grades; this is consistent with spontaneous combustion. Relatively lower
temperatures are obtained with paddock dumping and codisposal 5 m tipheads, indicating that these
methods of construction have a significantly lower risk profile.

A coloured decision matrix, which relates directly to analytical model outputs, can aid in visually presenting
relative risks with respect to screening of WRMS. The matrices are based on the criteria shown in Table 4.
Risk element Visual classification Limits

White <100 t (H2SO4)/day
Net acidity
Grey Between 100 t (H2SO4)/day and 1,000 t (H2SO4)/day
generation
Black >1,000 t (H2SO4)/day
White <80˚C
Temperature in the
Grey Between 80˚C and 120˚C
WRSF
Black >120˚C
Matrix outputs are shown in Tables 5 and 6: each colour zone within the matrices corresponds to a different
level of risk. The white zone has a lower relative risk, the grey zone has certain risk but may be manageable,
and the black zone has high risk and will likely require more effort to manage. This analysis suggests that
waste material management strategies should fit in the white zone or grey zone, and occurrences within the
black zone should be limited.

Risk Assessment
Layered Codisposal
PAF sulphur Paddock Codisposal Codisposal Codisposal
PAF blocky
content dumping 5 m tip 10 m tip 30 m tip
codisposal NAF
<1%
1%–3%
3%–5%
5%–10%
10%–18%
>18%
PAF Layered
Paddock Codisposal Codisposal Codisposal Codisposal
sulphur PAF
dumping 5 m tip 10 m tip 30 m tip blocky NAF
content codisposal
<1%
1%–3%
3%–5%
5%–
10%
10%–
18%
>18%
The method of constructing WRSFs has a significant and direct impact on operational and closure risks such
as AMD; however in many cases, specific construction methods are derived based on mining operational
parameters (cost, haul distance, adoption of standard procedures, etc.) and are not assessed in a quantitative
or semiquantitative manner with respect to their impact on closure risks. In this manner, operational and
closure risks can be assessed based on characterisations of materials in isolation. Risk-mitigation strategies
under these circumstances are at a high risk of being inherently biased toward the adoption of closure
mitigation strategies such as cover systems rather than more progressive management strategies such as
materials placement.
The benefits of progressive placement management strategies such as paddock dumping can be quantified
based on the matrices generated as part of this assessment. These make it clear that codisposal by end-
tipping methods shows a more rapid progression into higher-risk categories, and this is more pronounced
with increases in tiphead height. The effectiveness of measures implemented at closure, such as covers, can
be assessed using this information; it is possible that larger end-tipped WRSFs may have generated significant
stored acidity before the cover can be placed. The effects of temperature on risk are also shown to be
significant, and this is not often considered; the use of tipheads greater than 5 m for higher-grade material,
for example, may mean that elevated temperatures (over 80°C) will develop in the WRSF over time, which
will significantly increase long-term acidity risks, as reaction rates increase with temperature, and may also
lead to spontaneous combustion risks.

The data obtained from the matrices allows us to carry out a risk-weighted cost-benefit analysis of
construction methods and to assess closure scenarios and the requirement for closure mitigation measures
such as cover systems. For example, paddock dumping may be too costly for all materials, but a sulphur grade
cut-off can be established such that the material with high acidity and/or temperature risks could be
selectively managed in this manner.
Overall, material with higher sulphur content (greater than 3 wt% S) has fewer acceptable disposal options.
This is because oxidation rates for higher-grade material are significantly higher than for lower-grade
material, and so the level of risk is much greater for a given mass of sulphur. Therefore, identifying and
characterising this material should be considered a high priority when planning and establishing appropriate
management strategies.
The risk assessment involves several assumptions, so the analytical model has limitations; however, there
are valuable conclusions regarding oxygen ingress, estimated internal gas compositions, potential
spontaneous combustion and AMD. The assessment also identifies significant observations that relate
directly to operational controls and waste placement planning. While a one-dimensional model cannot fully
address the complexity of these multifaceted, time-dependant problems, this work shows developments
toward a quantitative assessment of risk. Analytical modelling identifies a direct link between WRSF
construction methods and the potential development of AMD risks and impacts, highlighting the importance
of comprehensive material characterisation and WRMS for material placement. Investing in research and
design prior to construction should make the target of designing WRSFs with reduced risks of AMD through
to closure achievable.
Andrina, J., Wilson, G.W. and Miller, S.D. (2012) Waste rock kinetic testing program: assessment of the scale up factor for sulfate and
metal release rates, in Proceedings of the Ninth International Conference on Acid Rock Drainage (ICARD), 20–26 May 2012,
Ottawa, Canada, W.A. Price, C. Hogan and G. Tremblay (eds), Curran Associates, Red Hood, NY, pp. 882–893.
Bennett, J.W., Comarmond, M.J., Clark, N.R., Carras, J.N. and Day, S. (1995) Intrinsic oxidation rates of coal reject measured in
laboratory, in Proceedings of Sudbury 1995: Conference on Mining and the Environment, pp. 9–17.
Brown, P.L., Logsdon, M.J., Vinton, B., Schofield, I., Payne, K., 2014. Detailed characterisation of the waste rock dumps at the
Kennecott Utah Copper Bingham Canyon Mine – Optionality for Closure. Proceedings of the Eighth Australian Workshop on
Acid and Metalliferous Drainage (Eds. H Miller and L. Preuss), Adelaide May 2014; pp. 1–12.
Humphreys, D.R. (1998) The nature of underground heatings as indicated by numeric modelling, in Proceedings of First Australasian
Coal Operators Conference, COAL ’98, 18-20 February 1998, University of Wollongong, NSW, E.Y. Baafi (ed), The Australian
Institute of Mining and Metallurgy, Illawarra Branch, pp. 604–616.
Lahmira, B., Lefebvre, R., Aubertin, M. and Bussière, B. (2007) Modelling the influence of heterogeneity and anisotropy on physical
processes in ARD-producing waste rock piles, in Proceedings of 60th Canadian Geotechnical Conference and the 8th Joint
CGS/IAH-CNC Groundwater Conference, 21–24 October 2007, Ottawa, Canada, pp. 1486–1493.
Lottermoser, B.G. (2010) Mine wastes characterization, treatment and environmental impacts, third edition, Springer, Berlin
Heidelberg.
Pearce, S.R., Beavis, F.R., Winchester, S.J. and Thompson H. (2012) Managing closure risks by integrating acid and metalliferous
drainage assessments with mine scheduling – real world applications, in Mine Closure 2012 Proceedings of the Seventh
International Conference on Mine Closure, 25– 7 September 2012 A.B. Fourie and M. Tibbett(eds), Australian Centre for
Geomechanics: Brisbane, Australia, pp. 801–814.
Weber, P., Olds, W., and Pizey, M. (2014) Geochemical and geotechnical investigations at the Reddale coal mine, Reefton, New
Zealand, in Proceedings of the Eighth Australian Workshop on Acid and Metalliferous Drainage, 29 April – 2 May 2014,
Adelaide, South Australia, H. Miller and L. Preuss (eds), JKTech Pty Ltd, Brisbane, Australia, pp. 445–466.

M.D. Henschel MDA Geospatial Services Inc., Canada

S. Sato MDA Geospatial Services Inc., Canada
R. Ahola MDA Geospatial Services Inc., Canada
M.A. McParland MDA Geospatial Services Inc., Canada
Satellite based Interferometric Synthetic Aperture Radar (InSAR) ground deformation monitoring has been
effectively used to study ground change at active mine sites throughout the world. Frequent acquisitions allow
ground change to be monitored on a routine and timely basis, providing mine operators with critical insight
into current ground motion patterns which may negatively impact operations.
Monitoring of mines undergoing decommissioning stages is no less important. Risks, such as slope failures on
pit walls, waste piles and tailings containment structures, continue to be present as long as these structures
exist. In the absence of mining activity, human observation of ground motion no longer occurs regularly,
increasing the importance of remote monitoring operations.
While the importance of remote monitoring remains, the frequency and detail of ground motion
measurements for decommissioned mines need not be as great as for active mines. Recording periods may
be extended from weeks to months or even years, as the lack of active mining activity will lessen the likelihood
of rapid changes in ground motion. Measurement resolution may also be reduced in order to allow for wider
scale monitoring at a reduced cost. This paper discusses strategies to provide effective ground monitoring of
mines undergoing closure phases. The effects of reducing the frequency of data acquisitions and using low
resolution monitoring methods are discussed in an effort to understand the engineering trade-offs between
coverage expense and accuracy.
Interferometric Synthetic Aperture RADAR (InSAR) is a well proven technique for monitoring ground surface
movement in many diverse environments. The techniques have been applied widely to mine environments
for many years with increasing acceptance due to the reliability of contemporary InSAR processing
techniques.
InSAR studies have advanced in capability greatly since their introduction (Gabriel et al., 1989). Early InSAR
work considered principally the complex differences between two images. That is, the studies considered
phase differences between the RADAR returns for every resolution cell of the imagery. In regions where
ground conditions are suitable, such as arid regions and where smooth ground deformation occurs, the
returned waves are coherent (Zebker and Villasenor, 1992), allowing very precise measurements of ground
movement to be made within the resolution cell of the RADAR.
As the desire to make precise measurements of ground movement grew to regions where distributed RADAR
returns provided changing scatterers, InSAR methodologies changed to identify point targets. That is, the
measurement technique was adapted to identify stable targets that were coherent over time (Van der Kooij,
2003). These methodologies exploited the availability of time series of images to capture those points in
space and time that could be used as robust measurement points. The benefits of coherent or point scatterer
methodologies made InSAR possible in regions where trees, ground water, or even crop growth made the
original methodologies difficult to interpret.

Effective, long-term monitoring of ground deformation throughout mine closure phases M.D. Henschel, S. Sato, R. Ahola and M.A. McParland
Techniques that consider the generation of a network of interferograms from a time series of images (Lanari
et al., 2004) appeared and were quickly adopted. These methods attempt to increase the suppression of
image to image noise by considering all possible interferometric pairs with suitable imaging geometry.
With the introduction of high resolution spaceborne RADAR sensors, it was possible to demonstrate
submillimetre accuracy in InSAR measurements, as was done for instance with RADARSAT-1 (Ferretti et al.,
2007).
The combination of high-resolution SAR sensors (with ground resolutions on the order of 3 m) and the
requirement to continue to expand areas where robust interferometric measurements can be made, led to
the development of methods that consider both stable points and distributed neighbourhoods of scatterers
with homogeneous movement (Eppler and Rabus, 2011).
The growth in application areas where robust InSAR measurements can be made has grown with each
generational improvement in InSAR techniques and provides for operational monitoring of active mine areas,
for instance Shanker et al. (2011) and Henschel et al., 2014b; 2014c.
As generational improvements in InSAR methodologies occurred, conditions under which accurate
measurements of ground movement can be expected have expanded. Recent work, applying the current
generation of InSAR methods, has demonstrated that the accuracy of InSAR surveys can be shown to be
better than ±2 mm (Henschel et al., 2014b). That work was conducted in the Swedish Arctic over a mine with
significant changes in ground conditions over time. Early assessments of the same mine with InSAR (Henry et
al., 2004) showed episodic measurements of ground movement. The current generation of operational
monitoring records continuous time series of information over the entire region impacted by the mining
operation.
Constructing a time series of observations with ±2 mm of height accuracy and ground resolutions of
approximately 3 m may be more than is necessary for a mine in closure phase. Ground movement in a mine
that is dormant may be expected to occur from continued settling, ground water changes, or other slow
(when compared to block caving, say) processes. In the following we describe a methodology for monitoring
change associated with slower ground deformation processes. The use of reduced observation frequency
and resulting consequences for the accuracy of monitoring are discussed.
From sublevel caving, to long wall, to open pit mining, InSAR measurements have been shown to provide
high density, wide area point coverage. The technique is routinely used to monitor active mines throughout
the world. The operational tempo for monitoring these mines can vary greatly depending on the mechanism,
the satellite used and, to some extent, reporting requirements. For example, subsidence over room and pillar
mines may occur only at a slow rate as pillars degrade (Harrison, 2011; Samsonov et al., 2014). In contrast,
sublevel caving used at LKAB’s mining operation in Kiruna, Sweden (Stöckel et al., 2013) causes rapid
deformation (metres per year) in the caving zone (above the ore body) and more moderate deformation of
a few millimetres per year in the deformation zone that extends out through the hanging wall. Long wall
mining practices can produce metres of ground deformation in a month, followed by many months of slow
ground settling after the fact (Jeran and Trevits, 1995; Bauer 2008). In open pit mines, ground movement
during operations may be too large for InSAR monitoring of pit walls, but the methodology provides very
accurate measurements of movement around mining infrastructure and along slopes near the pit. When the
pit slope is visible to the satellite RADAR, open pits (such as Bingham Canyon, near Salt Lake City, USA) can
be monitored very precisely with InSAR (Styles et al., 2011).
Considerations in the design and implementation of a robust program for active mine monitoring include:
terrain characteristics, types and rates of ground movement expected, and requirements for data reporting.

Landforms and Evolution in Reclamation
InSAR measurements are made possible by considering the path distance two waves travel. In classical
interferometry, the wave-wave differences produce constructive and destructive interference along the
peaks and troughs of the two waves – much like the effect of dropping two pebbles into a pond. By
considering the resulting pattern of illumination, very precise measurements of the distance the waves have
travelled can be made. In repeat pass interferometry, the microwaves are measured from the same point in
space at different times. The waves travel through space to a point on the ground and back to the satellite
where both the phase and amplitude of the wave is recorded. The amplitude of the returned radiation is
dictated by the reflectivity characteristics of the imaged terrain. The phase of the return is related directly to
the path distance travelled.
Since the wave amplitude is affected by the terrain, the ground conditions can be a major consideration in
the design of an effective monitoring program. Microwave observations can be affected by seasonal
variations in snow and ice cover (Wickramanayake et al., 2014), ground water, and local vegetation. While
the third generation of InSAR techniques have demonstrated enhanced control over noise levels associated
with ground conditions, there remain regions where ground conditions require the use of installed targets to
ensure robust monitoring. We have been able to repeatedly demonstrate the accuracy and efficiency of
installed measurement monuments (Henschel and Lehrbass, 2011) and (Henschel et al., 2014b). However,
our primary design consideration is to use ground infrastructure and ground points wherever possible. While
active mine monitoring may require installed targets, monitoring mines that are in closure phases should not
require additional infrastructure (except perhaps in far northern regions).
The type and rate of ground movement are critical factors in the design of a successful monitoring program
for an active mine. The rate of expected ground movement dictates the required time between images and
the wavelength necessary for imaging. Contemporary RADAR satellites have repeat orbits that vary from 8,
11, to 24 days with observation wavelengths from 3.1 cm, 5.6 cm, to 15 cm. The observation wavelength
dictates the length of deformation that can be imaged from one observation to the next. For instance, an X-
band satellite (at 3.1 cm) can unambiguously measure ground motion of 1.6 cm (in the line of sight) from one
pass to the next. When continuous, adjacent points are available the amount of ground deformation can be
followed to show many centimetres or even metres of motion. If the region or the ground motion has
discontinuities, the ambiguity distance becomes important. Figure 1 shows the ambiguity distance for both
C-Band and X-Band satellites.
Essentially the ambiguity distance means that for discontinuous observations, the rate of imaging with X-
Band RADAR needs to be twice as often as with C-Band RADAR. The frequency of observation for active

mining applications can also be enhanced by the use of repeated independent measurements of the ground
deformation. This can be accomplished by using complimentary orbits or by using multiple satellites
(Henschel et al., 2014a).
Figure 2 shows ground movement projected to the vertical from July 2011 to June 2013 at an open pit mine
site in Southern USA. The region was imaged with the RADARSAT-2 satellite with an exact repeat orbit of 24
days and a RADAR wavelength of 5.6 cm. The ground resolution of the RADAR beam used is approximately
5 m. The image provides a fairly typical description of the ground deformation available from third generation
InSAR techniques using our proprietary methods. The image shows a very high density of points and a large
range of observations throughout the mine area. Measurements are present on tiers, open slopes, and
throughout the neighbouring region. White areas in Figure 2 are not measured due to RADAR layover or
shadow (i.e. they are not visible to the satellite). In operational monitoring, beams that look at the mine from
the east and from the west are often used to provide full areal coverage (Henschel et al., 2014c).

Figure 2 is included as a demonstration of the type of information available with a high rate of observation
and a high ground resolution. Our goal in this article is to consider the effects of monitoring such an area with
lower resolution in both time and space. The operational measurement requirements for active mines are
likely to be more stringent than for mines in closure phase. The next section discusses the effects of the
reduction of resolution on the resulting information set.
The period of remediation and final closure of a mine takes place after the ore body ceases to be economically
viable for mining. The obvious corollary is that the physical forces on the rock change. Large motions
associated with ore and waste rock removal, dewatering and other activities are replaced by slower settling,
or stopped altogether. The monitoring of continuing (or stopped) ground movement should therefore not
require the level of accuracy or detail provided for active mine monitoring.
In active mine monitoring high resolution and high accuracy is necessary to differentiate types of motion, to
assist the mine’s geotechnical team in understanding risks posed by sometimes subtle movement. In a
remediation period, the use of lower resolution methodologies may provide the necessary information, but
with a lower cost to monitor.
To demonstrate the use of lower cost monitoring methodologies, we considered the case of the active mine
shown in Figure 2. That monitoring product was created with image observations occurring on a 24-day basis
at high ground resolution. Alternative satellite beams and alternative satellite systems are available that
provide lower ground resolution (at reduced cost). Furthermore, the operational tempo of monitoring can
conceivably be reduced when the physical forces moving the rock are reduced.
Figure 3 shows the results of monitoring an active region with lower resolution data in time and space. The
spatial resolution of the data has been reduced to approximately 25 m and the frequency of observation has
been reduced by one third (image observation every 72 days). The coverage of the region is similar to the
high resolution monitoring product. In fact the wider area coverage is more comprehensive due to the larger
pixel size. The ground motion estimates are collocated with the high resolution product but may be more
difficult to interpret.
Some regions in Figure 3 show as uplift, or a combination of uplift and subsidence, while the high resolution
product in Figure 2 shows clear subsidence. The rate of image collection and resolution is clearly too low for
an accurate product in an active mine. However, the low resolution product helps us to understand what to
expect in a reduced resolution monitoring program over an abandoned mine.
The reduced resolution product still clearly highlights regions where slope movement is occurring. The sense
of the direction of motion, however, may not be correct. This may be due to single observations in time
approaching the ambiguity distance and therefore not being resolved in the correct sense – in these cases,
we can anticipate more easily interpretable results for a mine that is not operating.
In general we note that the location of the motion identified is clearly similar in both the high and low
resolution products. The use of low resolution techniques that exploit the benefits of third generation InSAR
methods is a continuing subject of our research and operational development.

Since the introduction of ground measurement techniques using satellite RADAR, techniques for information
extraction have increased generationally. The combination of contemporary techniques and the availability
of high resolution satellites provide for robust and accurate measurements over active mine sites.
The operational tempo of a mine in closure phase does not require the robustness of an operating mine.
When a mine is monitored after closure, the risk to life or equipment is not the same and the accuracy and
timing of observations can be safely reduced. We have shown one methodology to reduce costs, but maintain
the availability of information in monitoring.
We would like to acknowledge the support of MDA Geospatial Services Inc. for this work and permission to
publish the materials.
Bauer, R.A. (2008) Planned coal mine subsidence in Illinois: a public information booklet, Illinois Department of Natural Resources
and Illinois State Geological Survey, Champaign, Illinois.
Eppler, J. and Rabus, B. (2011) Monitoring urban infrastructure with an adaptive multilooking InSAR technique, in Proceedings of the
8th International Workshop on Advances in the Science and Applications of SAR Interferometry: FRINGE 2011, 19–23
September 2011, Frascati, Italy.
Ferretti, A., Savio, G., Barzaghi, R., Borghi, A., Musazzi, S.M., Novali, F.… Rocca, F. (2007) Submillimeter accuracy of InSAR time series:
experimental validation, IEEE Transactions on Geoscience and Remote Sensing, Vol. 45(5), pp. 1142–1153.

Gabriel, A., Goldstein, R. and Zebker, H. (1989) Mapping small elevation changes over large areas: differential radar interferometry,
Journal of Geophysical Research, Vol. 94, pp. 9183–9191.
Harrison, J.P. (2011) Mine subsidence, in P. Darling (Eds.), SME Mining Engineering Handbook Third Edition, p. 632.
Henry, E., Mayer, C. and Rott, H. (2004) Mapping mining-induced subsidence from space in a hard rock mine: example of SAR
interferometry application at Kiruna mine, CIM Bulletin, Vol. 97.
Henschel, M.D. and Lehrbass, B. (2011) Operational validation of the accuracy of InSAR measurements over an enhanced oil recovery
field, in Proceedings of the 8th International Workshop on Advances in the Science and Applications of SAR Interferometry:
FRINGE 2011, 19–23 September 2011, Frascati, Italy.
Henschel, M.D., Deschamps, B., Robert, G. and Gailing, R. (2014a) Multi-satellite InSAR deformation analysis for pipeline monitoring:
operational application, results and validation at the Belridge Oil Field, 2001–2012, in Proceedings of the International
Pipeline Conference, 29 September–3 October 2014, Calgary, Canada.
Henschel, M.D., Dudley, J., Lehrbass, B., Sato, S. and Stöckel, B.-M. (2014b) Monitoring ground deformation in arctic and sub-arctic
mines, To be published in Proceedings of Slope Stability 2015, 12–14 October, Cape Town, South Africa.
Henschel, M.D., Sato, S., Ahola, R. and McParland, M.A. (2014c) Designing InSAR survey for mine monitoring with accuracy
assessment, in Proceedings of AusRock 2014: Third Australian Ground Control in Mining Conference, 5–6 November, Sydney,
Australia.
Jeran, P.W. and Trevits, M.A. (1995) Timing and duration of subsidence due to longwall mining, United States Department of the
Interior and United States Bureau of Mines.
Lanari, R., Mora, O., Manunta, M., Mallorqui, J., Berardino, P. and Sansosti, E. (2004) A small-baseline approach for investigating
deformations on full-resolution differential SAR interferograms, IEEE Transactions on Geoscience and Remote Sensing, Vol.
42(7), pp. 1377–1386.
Samsonov, S.V., González, P.J., Tiampo, K.F. and d’Oreye, N. (2014) Modeling of fast ground subsidence observed in southern
Saskatchewan (Canada) during 2008-2011, Natural Hazards and Earth System Sciences, Vol. 14, pp. 247–257.
Shanker, P., Casu, F., Zebker, H. and Lanari, R. (2011) Comparison of persistent scatterers and small baseline time-series InSAR results:
a case study of the San Francisco bay area, IEEE Geoscience and Remote Sensing Letters, Vol. 8(4), pp. 592–596.
Stöckel, B.M., Mäkitaavola, K. and Sjöberg, J. (2013) Hanging and footwall slope stability issues in sublevel caving, in Proceedings of
the 2013 International Symposium on Slope Stability in Open Pit Mining and Civil Engineering, 25v27 September, Brisbane,
Australia, pp. 1045-1060.
Styles, T., Stead, D., Eberhardt, E., Gaida, B.R. and Bloom, J. (2011) Integrated numerical modelling and InSAR monitoring of a slow
moving slope instability at Bingham Canyon mine, in Proceedings of the 2011 International Symposium on Rock Slope Stability
in Open Pit Mining and Civil Engineering, September 18–21, Vancouver, Canada.
Van der Kooij, M. (2003) Coherent target analysis, in Proceedings of the 3rd International Workshop on ERS SAR Interferometry:
FRINGE 2003, 1-5 December, Frascati, Italy).
Wickramanayake, A., Henschel, M.D., Hobbs, S., Buehler, S., Ekman, J. and Lehrbass, B. (2014) Seasonal variation of coherence in SAR
interferograms in Kiruna, Northern Sweden, International Journal of Remote Sensing, Vol. 1(17).
Zebker, H. and Villasenor, J. (1992) Decorrelation in interferometric radar echoes, IEEE Transactions on Geoscience and Remote
Sensing, 30(5), pp. 950–959.


C.D. Tiemann Newcrest Mining Limited, Australia

M. Wealleans Environmental Consultant (formerly Newcrest Mining Limited), Australia
The Telfer gold mine (Telfer) is located approximately 1,800 km north-northeast of Perth, Western
Australia (WA), in an arid environment that experiences a hot wet season followed by a warm dry
season. In 2004, Telfer embarked on a 10-year research program involving the design and
management of large waste rock dumps. Under existing legislative conditions, Telfer is required to
design and construct waste rock dumps that are compatible with the regional physiography, stable in
the long term and do not present future acid mine drainage risks. In order to meet these conditions,
the company initiated a mesa landform project, which involved an extensive ecohydrology program
and progressive rehabilitation trials focused on replicating the natural mesas (concave slopes and flat-
topped landforms) of the Pilbara region.
The project represents a significant divergence from the traditional approach of surface ripped, berm-
and bench-style slopes typical of older mines in WA. This approach has been generally affected by
tunnel erosion and rill and gully erosion, both evident in places at Telfer. The project aimed to
determine whether these erosion features could be overcome on artificially constructed mesa-styled
landforms and, if so, whether they could be integrated into Telfer’s rehabilitation strategy for eventual
mine closure.
Project researchers began by collaborating with the University of Western Australia to gain an
understanding of ecohydrological processes in the Telfer region. The ecohydrology study aimed to
assess the interaction of soils, vegetation and hydrology in the arid zone by analysing water fluxes in a
range of natural ecosystems. This information was then used to develop designs and assessment
techniques for rehabilitation trials using the mesa concept.
The rehabilitation trials were completed in three stages. Stage 1 was completed in 2007 on a 30-metre-
high waste rock dump. The dump was battered down to create a concave slope at angles of 16, 25 and
37 degrees. Four slope treatments were then applied using a combination of soil and competent rock.
A series of hummock dumped treatments, for controlling surface water flow, were constructed on the
dump’s top surface. Stage 2 trials completed in 2009; this stage involved the rehabilitation of a satellite
waste rock dump (20 hectares in area and 25 metres high). It used concave slopes at angles of 7, 16
and 37 degrees with application of hummocks as the surface treatments. The Stage 3 trial was
completed in July 2013 on a 60-metre-high waste rock dump with three slope treatments applied. The
slope was battered to concave angles of 12, 20, 25 and 37 degrees.
The ecohydrology study and progressive rehabilitation trials showed that a mesa-styled waste rock
dump landform incorporating a concave slope and hummock dumped capping system was stable,
supported vegetation and met government approval criteria.
Overall, the mesa concept has many additional advantages over the standard berm-batter-berm
approach, including a smaller footprint, reduced surface area to rehabilitate, potentially easier
landforming and enhanced environmental and aesthetic qualities.

From mines to mesas: a rehabilitation journey at a gold mine in the Pilbara, Western Australia C.D. Tiemann and M. Wealleans
After successfully constructing the three mesa rehabilitation trials and obtaining regulatory approval
in 2015, a progressive rehabilitation plan was developed. This included a three-year program
encompassing up to 50 hectares, which have now been integrated into the surface mining operations.
The Telfer gold mine (Telfer) is located approximately 400 kilometres east-southeast of Port Hedland
in the Great Sandy Desert, part of the East Pilbara region in Western Australia (WA).
Telfer commenced mining and processing gold there in 1975, and the mine was placed on care and
maintenance in October 2000 due to oxide recoveries being exhausted. Following the completion of
a comprehensive feasibility study in 2002, construction of a new operation to exploit the primary
deposits commenced in early 2003. Production from the open pit recommenced in 2004 and from the
underground mine in 2006. In the 12 months to 30 June 2014, the milling rate was 21.3 million tonnes
per annum (Mt/a), and Telfer produced 536,342 ounces (oz.) gold and 25,506 t copper (NML, 2014a).
Mining is both open pit and underground. Current reserves are 4.8 million ounces of gold, and
resources are 12 million ounces of gold (NML, 2015).
The climate of the region is described as arid and is characterised by hot wet summers and warm dry
winters. Meteorological data is available for Telfer since 1974 and is considered reliable. The average
daily temperature rises to 40.4°C during January and drops to 10.6°C during July. Mean annual rainfall
for the area is 372.1 mm and is strongly seasonal, with almost all the rain occurring in the summer
months, between December and March; little to no rainfall occurs during the intervening months.
These climatic conditions present a considerable challenge to successful rehabilitation and were key
considerations in developing rehabilitation strategies for Telfer (Australian Government, 2015).
In 2004, Telfer embarked on a plan to design and build large waste rock dumps (WRD) that would
resemble the dominant land form of the region: mesas. Mesas are typically concaved slopes and flat-
topped landforms. Although this was conceptually possible, there remained a number of challenges
to overcome. To begin with, regulations required the company to “design and construct waste rock
dumps that are compatible with the regional physiography, stable in the long term and [do not]
present future acid mine drainage risks” (NML, 2002). The project was also constrained by the fiscal
imperative of all mine sites, that is, to work within agreed budgetary forecasts. The mesa landform
project was initiated to address the regulatory criteria; including an extensive ecohydrology program
and progressive rehabilitation trials focused on replicating the natural mesas of the Pilbara region.
This meant the design, construction and operation of waste rock dumps as mesa landforms could be
no more costly that the traditional berm and bench technique.
In broad terms, the project represents a significant divergence from the traditional approach of deep
surface ripping and berm- and bench-style slopes that are typical of older mine sites in WA. The
traditional approach has been generally affected by tunnel erosion associated with surface water
ponding and often by deep (> 1 metre) rill and gully erosion down slopes. Early rehabilitation trials
(pre-1999) at Telfer have shown that these traditional techniques are not suitable for available waste
material types or climatic regimes present at the site. The project was developed to determine
whether these erosion features could be overcome on artificially constructed mesa-style landforms,
thereby meeting regulatory approval conditions and achieving stable, aesthetically pleasing and
nonpolluting structures.

Up to this time, the berm and bench rehabilitation technique dominated the mining reclamation
landscape in Australia and, as a result, moving to a mesa landform approach represented a shift and
a degree of risk for state government regulators in that it had not yet been proven. To overcome this
risk, government regulators allowed Telfer to proceed with mesa-style landforming, but subject to
successful research and development. If this was successful, approval from the government would be
sought to integrate the mesa style into Telfer’s rehabilitation strategy for eventual mine closure. To
this end, Telfer approached potential partners including industry, academia and government
departments and regulators.
The ecohydrology research project was designed to integrate research from soil physics and plant
biology in understanding the mesa environment within the Telfer region. Together with the Australian
Research Council (ARC) and the Minerals and Energy Research Institute of Western Australia
(MERIWA), Newcrest funded this work. The study aimed to assess how soils/regolith, vegetation and
hydrology interact in the arid zone by analysing stores and fluxes of water in the plants and soil of a
range of natural ecosystems (e.g., mesa systems, interdunal swales, sand ridges and rock outcrops).
The research program focused on the following objectives:
 understanding the vegetation/soil associations as a function of the landscape position for
the natural analogue systems;
 assessing the water balance dynamics of the analogue system in space and time, linking
them to distribution of vegetation and soil;
 quantifying the water use of vegetation as related to different communities and plant
functional types;
 determining the large-scale pulse response of arid ecosystems to rainfall by analysing
rainfall, cyclone tracks, and remotely sensed data on vegetation and soil moisture storage;
 assessing self-organised vegetation patterns by modelling soil/vegetation feedbacks in arid
ecosystems; and
 comparing the natural mesa system to the existing WRDs in terms of surface characteristics
such as rock armour contributing to landform stability.
These objectives were achieved through establishing a monitoring program at nearby natural mesa
sites within a hundred kilometre radius of Telfer. Various hydrological and ecophysiological
measurements were collected, allowing plant/soil water balances across mesa and sandy plain
topographies to be calculated. The monitoring program was further complemented by assessing the
plant-soil association with different soils and subsoils and the physiological characterisation of soils
and plants in different landscape positions.
This information was then used to develop waste dump designs through the application of native
vegetation ecophysiology, landform evolution and ecosystem hydrology factors (NML, 2010). This
ultimately allowed modelling of rehabilitation designs and assessment criteria to be incorporated into
the project (Lambers et al., 2012).
The rehabilitation trials were completed in three stages beginning in 2006, with the learnings from the
ecohydrology study and each subsequent trial being incorporating into the following trial. Each
iteration of the rehabilitation trial expanded on its predecessor, and the parameters of design and
construction were tested to build confidence in the mesa concept for eventual regulatory approval.

Approximately one year into the ecohydrology study, and with some of the early study results
available, planning to design and construct the first mesa surface and slope trials began. The trials
concentrated on testing the physical mechanics of construction and the viability of topsoil in
combination with rock armour. Because Telfer had only recently commenced mining after a four-year
care and maintenance period, there were no new outer waste dump slopes available nor any final
surfaces to perform the trial on, so the decision was made to utilise a small but appropriately sized
waste rock dump in the west dome area, which was planned for mining in 2012. The dump had a
combination of waste rock types (competent rock and silt stone material) and an available surface
area for hummocks of approximately 0.25 ha. Figure 1 shows the size and angle configuration of the
WRD 10 mesa design (NML, 2014b).
Throughout construction, the mechanics of building the structure were refined. A D9 dozer was used
to complete the three slope angles. The topsoil and rock armour was delivered to the tipping edge
using side-tippers, and the topsoil and rock armour was applied using a push-through approach with
the D9, thus allowing homogenous mixing when applied down slope. The hummocks were constructed
using a 966 loader in single drops, and a safety berm was constructed at the mesa edge that more or
less blends in with the hummock heights. No excavators were used. The 966 loader was used for
hummocks as a first pass trial, in preparation for using single truck dumps from Cat 777, 785 or 963s
in similar 1:1 ratios. Three slope angles (16, 25 and 37 degrees) plus four slope treatments and four
surface treatments were tested, all in combination with the construction mechanics. One of the key
outcomes of the first trial was an understanding of how to physically build these structures and which
of the available materials would be most appropriate for final construction. Figure 2 shows the slope
of WRD 10 immediately after application of final rock armour, and Figure 3 shows the level of
vegetation growth nine years later.
10 m
20 m
Slopes were treated as follows:

 Treatment 3 – Old topsoil (> 10 years old) was applied as a surface 10 cm layer over
competent rock.
 Treatment 4 – Old topsoil was mixed in situ with competent rock and applied as a
combined ~ 30 cm layer.
 Treatment 1 – Direct return topsoil (< 6 weeks old with no rainfall event from harvest to
application) was applied as a 10 cm layer over competent rock.
 Treatment 2 – Direct return topsoil was mixed with competent rock in situ and applied as a
combined 30 cm layer.
Surfaces were treated as follows:

 Hummocks were dumped from the bucket of a 966 loader (~50 cm in height) in a mosaic
pattern.
 Treatment 5 – Old topsoil was applied in a 1:1 ratio of one bucket of topsoil and one bucket
of competent rock, all interlocking to form a sequence of humps and catchments.
 Treatment 6 – Old topsoil was premixed with competent rock and dumped in a 1:1 ratio with
competent rock.
 Treatment 7 – Direct return topsoil was applied in a 1:1 ratio of one bucket of topsoil and
one bucket of competent rock.
 Treatment 8 – Direct return topsoil was premixed with competent rock and dumped in a 1:1
ratio with competent rock.
Given the relative success of the 2007 initial mesa trial, the decision to progress the mesa concept to
the next stage was made. The new trial site needed to incorporate Stage 1 results in addition to again
testing the mechanics of construction. The second test site was a mined-out satellite pit located about
5 km south of Telfer. It was considered appropriate in terms of elevation (~25 m) and area (20 ha). A
contract earthmoving company was selected to complete the earthworks, and similar sized equipment
(D9 dozer, side-tippers and loader) was used to complete the trial (see Figure 4). The construction of

hummock surface cover, while effective, was the most time-intensive part of the process (NML,
2014b).
The Stage 2 trial site was different from the Stage 1 site in several ways:
 The large surface area had some relief, but no engineered surface water drains were
established, as the interlocking hummock mosaic was to provide hundreds of
microcatchments, therefore preventing surface water from ponding, laterally flowing or
ultimately tunnelling under saturated flow (Figure 5 shows the hummocks immediately
after construction and with four years of vegetative growth).
 The elevation across the waste dump averaged 20–25 m, which meant it was unlikely that
the same three slope angles used at Site 1 would be mechanically feasible; therefore,
alternate concave slopes at 7-, 16- and 37-degree angles were used, as this combination
best matched the regional physiography. Figures 6 and 7 depict WRD 13 immediately after
construction and four years later.
 The surface and slopes had numerous weeds, which were being eradicated as part of a
parallel study. This meant the area required premechanical weed treatment, which could
potentially mobilise the dormant seed.

Stage 3 was constructed on a 60-metre-high double-benched waste dump (WRD 6) comprised of

competent NAF waste material. The design incorporated the combined learnings of the research
program and two previous slope trials. In addition to this, an aerial survey (including LIDAR) of natural
mesas of this magnitude was also undertaken. The LIDAR survey data of natural mesa slope angles
was used to determine the final slope angles of the trial. These involved angles of 12, 20 and
25 degrees, with a final 10 m cover added at 37 degrees, battered down across a 200-metre-wide
section of the waste dump (shown in Figure 8) (NML, 2014b).
20 m
Once the bulk earthmoving was completed, three surface treatments on the slope face were applied.
In this trial, a D9 dozer was also used for bulk pushing; however, topsoil and NAF rock material was
sourced using a Newcrest fleet consisting of a 996 loader and 793 dump trucks. Each treatment used
varying combinations of materials, topsoil, NAF waste rock and outer siltstone member (OSM), a
dispersive clay-like waste material, to a depth of 50 cm. Figure 9 shows the initial slope surface of

WRD 6 before and after the application of surface treatments. The treatments were applied in the
following ratios:
 Treatment 1 – 2:1 (NAF rock: topsoil), blended;
 Treatment 2 – 4:1:1 (NAF rock: topsoil: OSM), blended; and
 Treatment 3 – 4:1:1 (NAF rock: topsoil: OSM), with OSM applied in the first layer and
blended NAF and topsoil applied over the top.
The aim of the Stage 3 trial was to construct a 60 m mesa waste dump slope and top cover with the
following objectives:
 Test the slope and top cover interaction.
 Prove, over approximately a two-year period, depending on rainfall, that the design is
robust and suitable for building waste dumps at Telfer up to 60 m using available material
and within the local climate conditions.
 Provide sufficient confidence to integrate the design into LOM planning.
The Stage 3 trial assesses the long-term stability of the mesa landform with controlled erosion,
vegetation establishment and growth, and similar ecological functioning (diversity and density) as the
natural mesas around Telfer. Figure 10 shows the final mesa slope with three months of plant growth.

Further rehabilitation of the 60 m mesa trial surface, involving a hummock dump treatment, will be
considered. This would likely be NAF rock waste and topsoil mixed at ratios of 3:1, respectively,
dumped at heights of approximately two metres to form a series of interlocking hummocks.
Rehabilitation at all trial sites has been monitored since their establishment. This has involved a
number of different techniques including photo point, transect monitoring and soil chemistry analysis.
From 2011, monitoring at WRD 10 and WRD 13 has utilised ecosystem function analysis (EFA) to
expand the collection of rehabilitation performance data and enable the development of closure
criteria for eventual regulatory approval. The rehabilitation monitoring program comprises a
combination of visual assessments, measurements of geochemical and geotechnical characteristics
and ecological indicators at permanently established transects. In addition to these rehabilitation
trials, a number of analogue sites were also chosen for ongoing EFA monitoring programs; these
provide baseline measurements to compare and contrast monitoring performance results (NML,
2014b).
–
Shortly after completion of the trial in November 2007, the site experienced three cyclonic rain events
that deposited approximately 650 mm over a three-month period. This turned out to be the perfect
“road test” as the intensity and duration of the rainfall events were typical of a worst-case scenario
for the region. After each event, the slopes and surface treatments were visually inspected, and the
observations backed up initial expectations: there was no tunnel erosion on the surface and no gully
or rill erosion on the slopes. Only Treatment 1 on the slopes showed signs of slight rill erosion; this
was consistent with the prediction that topsoil applied as a top layer would not benefit from the stable
function of embedded rock armour. All other treatments have remained stable. EFA, employed on the
site from 2011, has been conducted on both slopes and hummocks.
When compared against analogue equivalents, the landscape function indices for the slope have
shown an ever-increasing trend to target levels. Increased vegetation cover has also improved stability
indices over time, showing a trend towards equivalent analogues. Rain splash protection is a key
measure of stability, and both analogue and rehabilitation sites are comparable; however, while
vegetation cover drives the outcome on WRD 10, the presence of rocks is the main determinant at
analogue sites. This important distinction has been incorporated into Stage 3 trials, where the ratio of
rock to topsoil in cover material has been increased. Both infiltration and nutrient cycling indices have
been variable over the monitoring period, but despite this, they have remained on track with analogue
sites. Litter cover has been noticeably lower on the rehabilitation site than on the analogue site, and
this may have influenced the formation of a stable physical crust of cryptogram (COOE, 2012).
Only a single monitoring exercise has been conducted on WRD 10 hummocks, where four sites have
been selected to represent the four different cover trials. Stability, infiltration and nutrient cycling
indices are tracking towards analogue equivalents. Although vegetation cover is well established on
both the slope and cover, the hummocks have established litter cover and physical crust stability to a
greater extent. This is attributed to the very low levels of sheet erosion on the surface cover, which is
a function of the hummock composition and design (COOE, 2012).

Canopy volume has continued to increase marginally at all four slope monitoring sites on WRD 10, and
values are trending towards analogue values. Across the four soil treatments applied to the slope of
WRD 10, vegetation densities increased from Treatment 1 (eastern end) to Treatment 4 (western end).
Vegetation on Treatments 3 and 4, consisting of old topsoil (> 10 years) obtained from topsoil
stockpiles, included a greater cover of canopy species, mainly Acacia species. The canopy volume
across the analogue sites was predominantly low ground cover species. Much of the rehabilitation
was dominated by grass and herbaceous species such as Ptilotus exaltatus (tall mulla mulla),
Sclerolaena densiflora (hairy bindii) and several other grass species. A total of 20 species were
identified across the trial site (COOE, 2012).
Canopy volume at the WRD 10 hummock sites is dominated by ground cover species (0.5 m height
class) and is comparable to the volume at the analogue sites. After six years, much of the rehabilitation
is dominated by low cover species. Vegetation densities and volumes differed with each hummock
treatment, with the highest growth observed at the two treatments with fresh (< six-week-old) topsoil.
Vegetation cover on the hummocks is mature and comprises grasses, herbs and shrubs, with many of
the same species observed on the slope treatments (COOE, 2012).
Measurements of rill erosion along the WRD 10 slope have increased slightly since 2011; however, the
total number of rills formed has decreased. This indicates that the slope is becoming more stable,
although as only two cycles of monitoring have been conducted, there is insufficient evidence to
confirm this assertion. In relation to the slope profile, the lower portion has remained most stable,
and the upper sections show a small increase in erosion from year to year (COOE, 2012).
–
Immediately after completion of the Stage 2 trial, a cyclonic rain event followed by a series of
noncyclonic rain events again tested the slope and cover design. In each instance, the surface and
slopes performed according to design, with time-lapse monitoring results indicating healthy and
functional rehabilitation in line with proposed rehabilitation criteria. The trial also allowed a greater
understanding of the mechanics of completing mesa-styled rehabilitation to enable more cost-
efficient techniques to be incorporated in the future.
In 2009, a comprehensive rehabilitation monitoring program was initiated to track landform stability,
document the establishment and growth of vegetation and demonstrate that potential acid
metalliferous drainage (AMD) presents no risk to the natural ecosystem. The monitoring program
included photo point monitoring, chemical analysis of cover materials, monitoring transects (for the
EFA program) and soil moisture sensor stations. The results of the five-year monitoring program are
summarised below.
Landscape function indices have generally improved or remained comparable throughout the
monitoring program. Stability indices have improved since 2012 and show trending towards the
analogue site. Incorporation of rocks in the soil surface has contributed to stability levels, and with the
increase in litter and the formation of crusted soils, stability is predicted to improve in subsequent
years. Indices for infiltration and nutrient cycling are still comparable with the analogue site over the
last five years. The increase in vegetation at WRD 13 has contributed to increased litter, subsequently
adding elements of infiltration and nutrients to the soils (COOE, 2013).

Vegetation at WRD 13 is dominated by grass seedlings including Enneapogon caerulescens (limestone

grass), E. purpurascens (purple nineawn) and Paraneurachne muelleri (northern Mulga grass). Canopy
species observed throughout the site include Acacia ancistrocarpa (Fitzroy wattle) and Acacia
translucens (poverty bush).
Vegetation cover is notably sparse in the middle sections of the waste dump slopes, whereas upper
slopes have consistently shown a higher percentage of cover. In comparison, vegetation on the
hummocks is higher in both density and diversity and has germinated following suitable rains, not only
in topsoil but also in a mix of topsoil and waste rock material. In early years, the weed species Aerva
javanica (kapok bush) was found in prolific numbers; however, it has subsequently been out-
competed by native colonisers such as Ptilotus sp. (mulla mulla). Kapok continues to prevail on the
lower sections of the WRD 13 slope and, to a lesser extent, among hummocks on the surface; however,
it has recently been observed to contribute significantly to organic matter across the rehabilitation
site (COOE, 2013).
Rill erosion measurements on the slope face of WRD 13 have increased slightly over time, with the
greatest increase seen between 2012 and 2013. Two sinkholes were also observed on the top of the
waste dump in 2012. These occurred at a low point along the perimeter access track and on a topsoil
trial that did not have hummocks. Despite the observed erosion on WRD 13, a comparison of stability
indices between analogue sites indicates that the landform is stable and behaving similarly to natural
mesas in the area (COOE, 2013).
Observations from the soil moisture sensors show that water percolates through the upper 800 mm
of soil of WRD 13 following a significant rain event of more than 20 mm in a day. This has occurred
between two and four times a year over a consecutive three-year period.
The surface of WRD 13 is bowl shaped, and it appears that most rain is trapped by the hummocks and
infiltrates, but excess water during significant events runs towards the low points on the dump. An
examination of the data suggests that 15 mm of rain over three days will generate runoff towards
these low points (COOE, 2013).
Soil pH rose marginally between 2009 and 2010, but since then it has maintained a stable range in line
with natural levels. Soil salinity has remained stable in all materials, apart from waste rock on the
dump surface. The waste rock hummocks appear to have become slightly more saline over time, which
is attributed to leaching from the waste rock hummock. All other indicators of soil fertility recorded
no significant change or improvement over time. In contrast, cation-exchange capacity decreased as
did exchangeable sodium percentage in soils containing waste rock. Overall topsoil on the hummocks
was statistically indistinguishable from the baseline survey.
Nitrogen increased in all soils tested, which strongly indicates that nutrient cycling has established on
all cover materials on WRD 13. Phosphorus remained statistically similar to baseline and increased in
waste material.
Arsenic, copper and iron decreased over time in the mixed soil but remain at similar levels in topsoil
and waste rock. Cobalt and manganese remained unchanged in all soil types (COOE, 2013).

–
Following completion of the WRD 6 mesa slope trial in June 2013, baseline surveys were carried out
in October, and EFA transects were established for each treatment on the upper and lower slope,
providing a total of six transects.
The initial survey showed that landscape function indices were already comparable to that measured
at the analogue site, due to the high level of rock cover across all three treatments. Seedlings were
already appearing across the site, and this is predicted to assist with further stabilisation of the slope
surface. Further monitoring will track the landscape function indices in comparison to the undisturbed
analogue site to determine which of the slope treatments performs best, and this will be incorporated
into future rehabilitation programs (COOE, 2013).
As the mesa slope trial construction was completed only four months prior to monitoring, vegetation
cover was sparse and dominated by recently germinated seedlings. Flora identified were
predominantly Acacia and Senna species. With further monitoring and plant growth, more accurate
identifications will be possible. No vegetation was observed along the EFA transects within Treatment
2. Treatment 1 recorded a higher density of plants in comparison to Treatment 3, which is likely to be
the result of the higher percentage of topsoil within this soil treatment (COOE, 2013).
To effectively determine the performance of all three trials, a number of analogue sites were also
included in the monitoring program. This enabled development of baseline landform indicators that
were used to compare results from our rehabilitation sites. These included stability, infiltration,
nutrient cycling, plant cover and plant density.
At all rehabilitation sites, stability and infiltration indices achieved natural ranges within three years,
and in the case of WRD 6, attainment was observed within the first year. Comparable levels of nutrient
cycling have also been achieved at WRD 10 and WRD 13, when compared to baseline targets. Despite
good vegetation growth at all sites, plant cover and plant density have shown to be the most difficult
indices to reach natural levels.
The major difference in vegetation composition is the presence of taller plant species on rehabilitation
trials. Natural mesa slopes are predominantly Spinifiex species monocultures with higher plant density
and cover rates than those of constructed landforms. Investigating plant species observed at each trial
shows that Spinifex is generally found in low densities, which correlates to its slow growing
characteristic. As expected, pioneering species are most prevalent at trial monitoring sites; however,
observations have shown a shift towards vegetation communities more common on analogue sites.
Table 1 shows the landform indices monitoring results of all three trial sites compared against
analogue sites (COOE, 2013).

Landform WRD 10 WRD 10 WRD 13 WRD 13 WRD 6

hummock slope hummock slope slope
Stability 2012 2011 2011 2011 2013
Infiltration 2012 2011 2011 2011 2013
Not yet
Nutrient cycling 2012 2011 2011 2011
achieved
Not yet Not yet Not yet
Plant cover 2012 2012
achieved achieved achieved
Not yet Not yet Not yet Not yet
Plant density 2012
achieved achieved achieved achieved
Statutory requirements from the WA Environmental Protection Authority (EPA) and the WA
Department of Mines and Petroleum (DMP) required Newcrest to revise and submit the mine closure
plan (MCP) for Telfer in June 2014. Newcrest undertook a thorough review of rehabilitation
monitoring data and supporting studies in compiling this document. This analysis allowed the
company to conclude that the proposed project had successfully proven that this method of
rehabilitation would achieve desired closure outcomes if implemented at Telfer.
In submitting the MCP to the regulators, Newcrest acknowledged that full-scale implementation of
the mesa landform rehabilitation strategy represented a significant move away from traditional WRD
rehabilitation methods employed in the WA mining industry. The original approved closure strategy
employed the berm and bench methodology, which had already proven unsuccessful at Telfer.
In summary, our research programs and trials to date have shown that the mesa landform has a
number of advantages:
 The WRD footprint is significantly reduced, with less land disturbance as a consequence
(when compared to an equivalent berm and bench design).
 Material haulage, rehabilitation and follow-up repair costs are also reduced (when
compared to an equivalent berm and bench design).
 This method achieves a more sustainable landform; key criteria of stability and vegetation
support can be achieved over a short period of time (potentially less than five years).
The importance of cost-effective closure strategies at Telfer cannot be overstated, considering that
the current mine disturbance footprint exceeds 4,000 ha. While the financial implications of
constructing final landforms is crucial, thought must also be given to the effectiveness of those
rehabilitated structures, as rework from slope failure and erosion is costly (Smith, 2008). The major
benefit of the mesa landform trialled at Telfer has been its resilience to the climatic conditions of the
region (erosion resistance) and its ability to support a vegetation community comparable to analogue
sites.
Another element of the Telfer MCP was the use of Newcrest rather than contractor fleet equipment.
When the final cost analysis was completed, the Newcrest fleet was considerably more cost effective,
in the order of 25%. However, as this was still theoretical, the company needed to be able to
demonstrate that this could be shown in practice. Subsequent rehabilitation trials at Telfer need to
prove that our current fleet of surface mining equipment is able to complete landforms as per the
mine closure plan, achieving both technical and financial objectives. Figure 11 shows the final mesa
design published in the Telfer MCP.

Upon receiving final approval for the mesa rehabilitation strategy, Telfer developed a progressive
rehabilitation plan for the following three years. One key to building the plan was identifying WRDs
that could be rehabilitated without compromising future mine expansion plans. To this end, a detailed
life-of-mine plan was essential. Newcrest identified two WRDs available for rehabilitation along with
the lower batters of its operating tailings storage facility. A total of up to 50 hectares was included in
the rehabilitation plan, and a continuation of WRD 6 was to begin in 2014–2015. This would be
followed by another waste dump adjacent to WRD 6 and the construction of a cover trial incorporating
the hummock methodology. A key aspect of this plan was that Newcrest surface mining equipment
had to be used to implement it.
Before rehabilitation activities on WRD 6 began, a D11 dozer with a universal blade was prepared.
Dedicated operators were selected, and the rehabilitation designs were completed by Newcrest
surface mining engineers. Oversight and direction were provided by the Telfer Environment
Department; however, day-to-day operations were managed by the surface mining team. Figures 12
and 13 show the slope before and after it was reprofiled.

While the progressive rehabilitation earthworks continue, initial analysis has shown that available
Newcrest equipment is suitable for the task. A cost comparison of Newcrest and contract equipment
has also demonstrated significant financial savings when site equipment and operators are used. This
is an important outcome of the progressive rehabilitation program and will provide greater confidence
in closure cost estimates in the future.
The ecohydrology and progressive rehabilitation studies and trials have shown that the needs of a
mesa-styled waste rock dump landform incorporating a concave slope and hummock dumped surface
treatment in the Telfer regional context “are compatible with the regional physiography, stable in the
long term and [do not] present future acid mine drainage risks” (NML, 2002). This landform will also
support sustained vegetation diversity and abundance. In short, the mesa concept can meet all
government approval requirements and Newcrest operating criteria for mine closure.
The mesa concept has many additional advantages over the standard berm-batter-berm approach,
including a smaller footprint, reduced surface area to rehabilitate and reduced construction and
follow-up repair and maintenance costs. The rehabilitation trails undertaken at Telfer have been
easier from a landforming perspective and have shown enhanced environmental and aesthetic
qualities. The trials have also proven that mesa landforms can be successfully constructed using in-
house mining equipment.
After successfully constructing the three progressive mesa rehabilitation trials and obtaining
regulatory approval, Telfer developed a further rehabilitation plan (Stage 4). This included a three-
year program (post 2015), encompassing up to 50 hectares of slope and surface areas and capturing
the key learnings from the Stages 1–3 trials and supplementary studies. Integration into surface
mining operations has now been achieved, and this will continue to provide verification of mesa slope
and surface hummock construction using existing Newcrest equipment to maximise future savings.
Finally, mesa-style rehabilitation may not be applicable for all mine sites, but it does provide a viable
and cost-effective option for waste rock dump rehabilitation in the Pilbara.
Australian Research Council (ARC) Linkage

Emma Ryan Reid, Environment Superintendent (Telfer 2009–2011)
Joe Mifsud, Principal (COOE)
Matthew Devlin, Environment Superintendent (Telfer 2011–2013)
Minerals and Energy Research Institute of Western Australia (MERIWA)
University of Western Australia

Australian Government (2015) Climate statistics for Telfer Aero, Bureau of Meteorology, viewed 20 February 2015,
http://www.bom.gov.au/climate/averages/tables/cw_013030.shtml.
Lambers, H., Hinz, C., Veneklass, E., Blaby, T. and McGrath, G. (2012) Ecohydrological characterisation of the natural and
rehabilitation ecosystems at Newcrest’s Telfer gold mine, Report 297, MERIWA Project M381, Minerals and Energy
Research Institute of Western Australia (MERIWA), Perth, Australia.
Newcrest Mining Limited (NML) (2002) Telfer project: mine and borefield extensions: notice of intent/additional referral
information, Newcrest Mining Limited, East Perth, Western Australia.
Newcrest Mining Limited (NML) (2014a) Newcrest Mining annual report 2014, viewed 19 February 2015, http://www.
newcrest.com.au/media/annual_reports/FINAL_Annual_Report_2014_72dpi_web.pdf, released on 30 September
2014.
Newcrest Mining Limited (NML) (2015) Annual mineral resources and ore reserves statement 31 December 2014, viewed 19
February 2015, http://www.newcrest.com.au/media/resource_reserves/2015/Newcrest_-_Market_Release_-
_December_2014_Resources_and_Reserves_Statement.pdf, released on 13 February 2015.
Smith, B. (2008) Mining for closure: sustainable mine practices, rehabilitation and integrated mine closure planning (thesis),
University of New South Wales, Sydney, NSW, 84 p.
COOE (2012) Rehabilitation monitoring report 2012, December 2012.

COOE (2013) Rehabilitation monitoring report 2012, December 2013.
Newcrest Mining Limited (NML) (2010) Telfer gold mine – mine closure plan, June 2010.
Newcrest Mining Limited (NML) (2014b) Telfer gold mine – mine closure plan, June 2014

T. Chirinda University of the Witwatersrand, South Africa

Industrial sands mining is a crucial economic activity undertaken by medium-scale miners in Zimbabwe.
Unfortunately, this mining practice leaves open pits, which have to be rehabilitated in order to return the
degraded land to an economically usable condition after mining is completed. This paper addresses two
competing priorities: minimising reclamation costs and maximising community benefits by returning
productive land to communities after mining. Industrial sands mining produces silica sand for use in glass
manufacturing, abrasives, foundries, and municipal water purification and construction purposes; it also
produces pebbles for construction. This business has a potential to hand over competitive goods and services
to mining communities if reclamation costs are optimised. This paper seeks to investigate a cost-effective
method of land reclamation at industrial sands mines in Zimbabwe that is compliant with the existing
environmental legislation. Land reclamation is the process of ecological restoration whereby the land surface
is rehabilitated for a beneficial end use like agriculture. Adequate financial reporting regarding the
reclamation process can create investment opportunities for industrial sands mines by reducing the
environmental burden on investors.
The reclamation technique proposed in this paper involves excavating clay and silt tailings from sand traps at
the mines and at old dump sites scattered around the vicinity of the mines, using these materials as backfill
and applying subsidised and locally available biosolids to the backfill material. This would be a cost-effective
means of implementing land reclamation. We used a TreePlan® decision tree model to draw a cost
comparison to investigate whether outsourced or in-house labour was the most viable option for the
reclamation process in terms of expected monetary values (EMVs). A cost benefit analysis was also done to
investigate whether reclamation was economically viable and worthwhile. The decision tree analysis
indicated that outsourcing was the most cost-effective option for the entire reclamation process. The cost
benefit analysis showed that costs were low and embarking on the reclamation project would be favourable
for the mines. It is encouraging to realise that communities can be economically empowered by this approach
as they get to offer the outsourced services. One of the most important benefits shown in this land reclamation
study is the economic benefit to communities, which are at times relegated to the peripheries of economic
development. Reclamation also brings immense environmental and social benefits, as explained in the paper.
The paper also proposes adequate environmental accounting and reporting measures for the mines’ financial
statements, which can ensure carefully considered reclamation strategies that minimise costs and create
opportunities for the communities in the mining areas.
Industrial sands mining in Zimbabwe is a booming economic sector that serves several market needs,
including providing building materials for communities. The mining results in opened out pits, which need to
be reclaimed in a cost-effective manner that minimises costs to the mines and maximises benefits to
communities; these benefits include providing reduced cost products for sale on the market. The reclamation
technique proposed and optimised in this study provides socio-economic benefits for communities,
government and the mining companies themselves. Reliable reclamation approaches help companies avoid
environmental-related penalties and afford them an enhanced financial outlook, equipping them with the
capacity to extend considerable financial rewards to their employees. More employment is created as the

Reclaiming mined-out pits at industrial sands mines in Zimbabwe: Cost minimisation and benefit maximisation T. Chirinda and T. Zvarivadza
reclamation technique involves outsourcing partnerships. This forms strong linkages between the mines and
the outsourcing partners and develops a network of revenue streams for the government through taxes. The
communities’ domestic animals are protected as scattered open pits are closed. As the mine can afford to
operate at low costs, the community gains access to cheaper high-quality building materials necessary for
building decent accommodation. Steady and reliable water supplies are increased as silted dams are
reclaimed. The disturbing legacies of acid mine drainage (AMD) are also addressed, protecting the
environment and biodiversity, the absence of which is at the centre of innumerable risks to life on earth.
Small- to medium-scale mining companies in Zimbabwe and worldwide are faced with the burden of
implementing effective reclamation techniques. This is especially applicable in surface mines, where a lot of
stripping or excavating of land leaves the surface tarnished and results in further environmental problems
such as AMD and siltation of rivers and dams. The absence of an ongoing and effective reclamation plan
results in long-term and immediate environmental and financial problems. Sand mining processes are cyclical
and involve relatively minimal losses of input raw materials, but the mining operations are heavily dependent
on the efficiency of environmental cycles. For instance, water used for washing silica sand and pebbles is
recovered and cycled back to the main dams for reuse. As a result, siltation of the dams is detrimental not
only to the environment but also to the business, as no sand mining operation can sustain itself without an
adequate supply of fresh water. In the case where there are erratic rains in the rainy season, water supply
becomes a very big problem in addition to the problem of siltation of the dams.
Siltation of dams at industrial sands mines in Zimbabwe reduces the dams’ capacity to hold vast amounts of
water for production purposes. A typical example happened in 2010, when there was a critical shortage of
water supply due to erratic rains and siltation. As a result, water had to be pumped from all the available
reserve dams to the main dams near the operations. This resulted in extra pumping costs, which were not
anticipated. These financial and operational constraints could have been avoided had proper measures been
taken to avert the problem of siltation. It is necessary to remove piles of clay and silt tailings lying in old
dumps and use them for reclamation, since they are being eroded, exacerbating the problem of dam siltation.
One dam at one of the mines has completely silted and cannot be used for water storage for production
purposes. The silt material derived from dam excavation can be incorporated in backfilling material for
reclamation. The historical costs for the dam excavation task were useful in cost analysis for this study.
The most significant challenge to ongoing reclamation efforts is the exorbitant cost involved. In the case of
most medium-scale mines, these costs would render their whole mining venture unprofitable. It can be
argued that the current economic crisis in Zimbabwe has led the mines to adopt the view that the most logical
step is to prioritise production and stay operational, although they are fully aware of their obligation to
implement some sort of land restoration measures.
Apart from the failure to implement cost-effective and business-oriented environmental management plans,
reporting of these was also noted to be a problem in annual reports. Ndaba (2010) asserts that disclosure of
financial information relating to the environment in corporate annual reports at the Zimbabwe Stock
Exchange (ZSE) was poor and was the result of ignorance on such matters. This resulted in the ZSE attracting
fewer investors (Ndaba, 2010) not only because of increased awareness of environmental issues in the mining
industry but also because of investors’ fear of inheriting enormous environmental liabilities where there is
poor disclosure of such information. This proves that poor environmental management is detrimental not
only to the company in question but also at the national level.
It is usually difficult for mining companies with relatively small operations to establish and implement cost-
effective and practical plans for land restoration of the degraded environment. It is also challenging to
implement an environmental management plan that restores the environment to its original state, let alone
one that makes it available for a different beneficial socio-economic purpose like agriculture.
Appreciation of the importance of environmental accounting and sustainability reporting as a normal
accounting procedure is on the rise. This indicates a shift towards better ways of managing and reporting of

environmental plans in order to contain environmental costs in mines (Ndaba, 2010). Awareness of
environmental issues in the mining industry is increasing, and investors are afraid of inheriting enormous
environmental liabilities where there is poor disclosure of such information. It is thus imperative that
medium-scale mines follow suit by improving land reclamation practices and financial reporting on
environmental issues, as there is a lack of compliance on their part. The economy of the nation would also
benefit significantly from a chain of efficient and profitable medium-scale mines that have the capacity to
expand and attract foreign direct investments.
In order for environmental management strategies to be cost effective and practical, they must differ from
mine to mine. A fully integrated approach that encompasses economics of the particular company in question
should be implemented. This will enable the company to craft a cost-effective environmental management
plan that is practical and that, in turn, provides socio-economic benefits for the mine and the local
inhabitants. Local inhabitants would benefit from the transformation of sandy areas into arable land for
subsistence living. The results of implementing an ongoing environmental management plan in the course of
mining would be essential to avoid cumulative environmental burdens at or towards the end of the life of
the mine. In terms of operations and finances, the mines benefit from reduced levels of siltation in the dams
and larger water carrying capacity. Water supply will become more efficient, resulting in more reliable and
viable operations even in times of successive poor rainfall seasons.
The cost-effective reclamation strategy proposed by this research complies with the environmental
legislation of Zimbabwe as enforced by the Environmental Management Agency (EMA). Complying with this
legislation would save companies from the burden of stiff penalties for noncompliance and would also uphold
a favourable reputation for the mines. Implementation of a cost-effective method that incorporates the
sound financial valuation techniques presented in this paper, such as cash flow interpretations and net
present value (NPV) analysis, would introduce better operational and financial viability in the process of
implementing the proposed techniques. The strategy would result in a benefit of minimal environmental
burdens for the mines at mine closure and would also help companies attract financial investors and foreign
customers through better environmental accounting and reporting. The reclamation method offsets land
degradation by making land suitable for future agricultural use, thereby promoting corporate social
responsibility and sustainable development for the company.
Industrial sands mining is a medium-scale mining operation undertaken by light-duty surface mining
companies that mine industrial sands in Zimbabwe. The core business of these companies is to mine, market
and sell silica sand for use in glass manufacturing, abrasives, foundries, and municipal water purification and
construction purposes; the mines also produce pebbles for construction purposes. The mine products are
also used to produce a clear glass called flint glass and a coloured glass called amber, which Zimbabwe Glass
Industries (Zimglass) uses for bottle making. Mining involves excavating 2.5 m deep pits of various sizes,
depending on the mineralisation of the area, mostly with a single excavator; mine products are then carried
out with two loaders at each of the mines.
Several mines are conveniently located approximately 5 km from the highway, which provides transportation
to various customers by road; the railway, which runs parallel to the highway, transports pebbles to the
National Railways of Zimbabwe.
Climate aspects such as rainfall and temperature have a significant impact in surface mining operations and
are an important consideration for reclamation processes. They affect the timing of backfill material removal,
pit water volumes in the rainy seasons, and visibility for operators in inclement weather.
Land reclamation differs from rehabilitation; it can be defined as the process of ecological restoration
whereby companies prepare the land surface for a beneficial end use like agriculture rather than merely

filling up mined-out pits (Cooke and Johnson, 2002). Reclamation obviously requires relatively more careful
planning and implementation to ensure that the desired results are obtained.
This research focuses on reclamation of mined-out pits at industrial sands mines in Zimbabwe –a better
technique of offsetting degradation caused by mining activities than rehabilitation. Transforming the land for
an economically and socially beneficial postmining use like agriculture would yield better results. However,
it should be noted that Zimbabwe’s environmental legislation, which was used as the legislative reference
for this study, is vague on issues pertaining to reclamation. The legislation puts more emphasis on
rehabilitation in general. Thus the definition of reclamation was derived from the Canadian mining land
restoration system, as expressed by Lehigh Hanson Canada (2014).
Significant environmental damage is done to the environment through mining processes. There is land
erosion and siltation of rivers, and the most detrimental effect is land degradation through mine dumps that
are scattered over the vicinity of the mining claims and mined-out pits that have been left unreclaimed. The
mining companies are well aware of their obligation to implement sound reclamation strategies that ensure
sustainable development. However, the significant challenge companies have faced in implementing
reclamation since the early 2000s is that of finances. The current harsh economic climate in Zimbabwe is
limiting the capacity to undertake such efforts, calling for more efficient and cost-effective reclamation
strategies.
Surface mining operations inherently cause considerable visible damage to the environment. As such,
responsible companies are obligated by the environmental management bodies in the host countries to
implement sound environmental impact assessments (EIAs) and land restoration initiatives. In this case, the
enforcing body is the Environmental Management Agency of Zimbabwe (EMA). By complying with these
requirements, companies avoid stiff annual environmental penalties, thereby realising more revenue and
making companies attractive to investors, who are always reluctant to inherit environmental financial
burdens.
Financial constraints are the main challenge facing the implementation of sound, practical and cost-effective
reclamation techniques at the mines. Reclamation procedures must be cost effective so that they can be
practically implemented by medium-scale mines, especially in the current harsh economic climate of
Zimbabwe. The companies hope to market their business ventures throughout Africa in order to expand their
customer base. So they need to boost their image by engaging sustainability measures that reflect their
commitment to corporate social responsibility. The mining companies also need to improve proper reporting
standards with regard to environmental issues in order to make them more attractive to investors on the
Zimbabwe Stock Exchange. Investors are risk averse regarding exorbitant environmental burdens; therefore,
effective reclamation strategies need to be embraced to minimise this risk.
According to the Zimbabwe Environmental Law Association (ZELA), legislation in place to control
environmental damage by mines and quarries is embodied in the Environmental Management Act No. 13 of
2002 (ZELA, 2003, as cited from SADC Environmental Legislation Handbook, 2012). Before 2002, the
constitution of Zimbabwe had no specific clause that provided for the protection of the environment, nor did
the Natural Resources Act of 1941 provide a comprehensive effort that catered to environmental
management (Walmsley and Tshipala, 2014). As a result, EIAs were not a legal requirement in Zimbabwe,
and no sound framework was in place to manage environmental issues. The Environment Management Act,
which the Government of Zimbabwe promulgated in 2002, deals with issues pertaining to environmental
protection and management and sustainable development.
Fines for causing environmental damage were very low, so companies and individuals resorted to committing
environmental crimes and easily paying the fines instead of taking due responsibility to avoid the damage
(ZELA, 2003). This situation formed part of the basis for the establishment of the Environmental Management

Act of 2002. Nowadays, stiff penalties are imposed for failure to comply with environmental legislation;
hence companies need to implement cost-effective methods for reclaiming mined-out pits at industrial sands
mines.
The EMA is a key institution that deals with issues of environmental management and encompasses such
issues regarding mining activities. Thus it has been used as the legislative authority that governs
environmental reclamation issues for this project report. Section 4:2 of the Act stipulates that “Any person
who causes pollution or environmental degradation shall meet the cost of correcting such environmental
pollution or degradation” (ZELA, 2003). In addition, section 114 of the same Act states that ‘the minister may
order an owner, occupier, or user of any land to undertake measures or steps to protect the environment,
which include rehabilitation of a mining site’ (ZELA, 2003,). The Act has been used as the legislative basis for
discussion of reclamation issues in this project report despite the slight shortcoming that it only emphasises
rehabilitation in general and is vague regarding reclamation issues.
There is a constant call to revise penalties for environmental degradation and noncompliance with EMA
regulations. For instance, it has been claimed that the fine for operating without the proper environmental
documentation, currently US$5,000, was not prohibitive to most companies that fail to comply with
environmental legislation. On the other hand, this fee is significant for relatively small mining companies.
Although penalties for noncompliance with environmental legislation are high for smaller and light-duty
mining companies, they are not significant compared to the cost of degrading the environment (The Source,
2014). This is demonstrated by the fact that mines decommissioned decades ago are causing adverse
environmental, sociological and economic challenges today. EMA Director General Mutsa Chasi advocates
for an increase in penalties for pollution and calls for legislators to synchronise mining and environment laws
to emphasise the importance of EIAs before mining operations can be allowed to start (The Source, 2014). In
the wake of the EMA advocating for stiffer penalties, the strategy developed by this research goes a long way
in addressing EMA concerns.
Traditionally, mining has been known for the inherent environmental burdens it leaves behind in the host
communities. According to Musa (2009), sand and gravel mining has been one of the most serious
environmental problems globally, leading to land degradation, loss of agricultural land and increased poverty.
Questions have been raised regarding the extent to which mining activities are beneficial to their host
communities and the environment, given this poor record of environmental sustainability. Musa (2009)
further notes that, although there is widespread occurrence and potential impact on agricultural lands,
attempts to address these challenges have paid little attention to sand mining activities. These challenges
contribute to a bad reputation for the mining industry, which otherwise provides huge domestic earnings for
many countries worldwide.
Mines that were decommissioned as early as the 1930s in Zimbabwe are reported to be causing adverse
sociological, ecological and economic problems. According to the EMA director general, the organisation is
currently faced with decommissioned mines that are causing problems such as death of domestic animals
due to acid mine drainage, collapse of slime dams and uncovered open-cast mines, which present hazards to
communities. He notes that just four of the 15 decommissioned mines requiring urgent attention would
require US$32 million just to make them safe for people who are still living there in a bid to correct the
problems they are causing after mining (The Source, 2014).
The scope of this research mainly emphasises the cost effectiveness and practicality of the proposed method:
reclamation through recycling tailings from sand traps and silt from the silted dams and sand trap dams to
refill mined-out pits. The paper also addresses issues pertaining to environmental management, sustainable

mining development and corporate social responsibility. It presents financial and technical calculations to
support the proposed reclamation technique; these include net present value (NPV) and expected monetary
value (EMV) calculations based on TreePlan® decision tree software.
The basis of this study is reclamation through recycling tailings material that has accumulated in sand trap
facilities and has been dumped over many years in old dumps. The backfill will be mixed with subsidised
locally available biosolids to enhance soil fertility and increase backfill volumes. Since the mines are mostly
located conveniently close to the cities, subsidised biosolids can be sourced from the municipal biosolids
generators.
For detailed analysis, one of the mines’ operations was studied in depth. The basis for the costs is the 2012
cost schedule for excavation of a dam that had silted. Currently, there is another silted dam at the mine, and
it is proposed that the silt excavated from the dams will form part of the backfill material over the years as
dam excavation occurs.
The study is a cost-effectiveness analysis (CEA) of the proposed method of reclaiming mined-out pits, which
involves backfilling the pits using tailings material from old dumps, silted dams and sand trap facilities at the
mine. The study also draws a cost comparison between outsourced services and in-house services for the
task of excavating silt and clay tailings from silted dams, sand traps and old dumps for backfilling.
The study used a TreePlan® decision tree to evaluate the optimum option between outsourced and in-house
services. Overall, the most cost-effective option was evaluated based on the EMVs computed at different
nodes in the software. The total estimated reclamation costs were evaluated for dam excavation that was
done in 2012. We determined that the mine would require a loan to finance the reclamation project. The
costs involved in the loan application were estimated, whereas some costs of excavating the silted dam were
based on the company’s 2012 cost schedule.
Costs were optimised by eliminating unnecessary costs of the land reclamation process. It was assumed that
the outcome from the decision tree model for the dam excavation is applicable to the entire reclamation
process. A simple qualitative cost benefit analysis was done to affirm the benefit of embarking on the
reclamation project. This cost benefit analysis was based on the various cost-cutting measures that were
employed for this study as well as the potential economic, environmental and socio-economic benefits that
can be gained.
The following sources were useful in obtaining the necessary values and statistics for analysis in this study:
 Chamber of Mines Journal of Zimbabwe
 Historical and current publications on land reclamation processes
 Company cost schedule (2012) for silted dam excavation from the mine
 Site investigations and measurements
 The African Development Bank, for historical inflation rates to be used in adjusting costs for
inflation
We based the overall cost effectiveness of the reclamation on the excavation costs incurred in the excavation
of the dam that had silted in 2012. We used a decision tree to analyse the optimum decision that resulted in
the highest EMV and resultant cash flows.

A decision tree was used to model all the possible outcomes using assumed probabilities and estimated costs
at various nodes in the decision tree. Decisions were modelled at decision nodes, and the probabilities were
taken into account at event nodes. We used a trial version of the TreePlan software to evaluate the resultant
cash flows and EMVs at the nodes and ascertain the best decision to take regarding the use of in-house or
outsourced labour for excavating the silted dam.
TreePlan software is based on a Microsoft Excel platform, with decision tree, sensitivity analysis and
simulation add-ins. It uses embedded logical and mathematical algorithms to evaluate the desired outcomes
in decision making, namely expected monetary values and resultant cash flows. TreePlan software was
adequate to make the comparative study and evaluation of the most cost-effective option between
outsourcing and using in-house labour.
Modelling in TreePlan commenced with a decision node whereby an initial decision had to be computed. This
initial decision was whether or not to apply for a loan to cover the excavation costs that are part of the
reclamation process. After the loan application decision, the modelling progressed by computing an event
node. At this event node, the outcome of the loan application is not independent of the company’s actions
but is probabilistic, depending on the lender’s perception of the company’s profile. Hence, for control and
comparative purposes, three probabilities with an equal distribution were assumed. Probabilities of 0%, 50%
and 100% were modelled. The 100% probability is a relatively optimistic expectation of getting the loan,
whereas the 50% probability is a moderately conservative expectation. The 0% probability is too pessimistic;
it was used as a control.
After computing the event node, we computed a decision node for the primary decision of interest: choosing
in-house or outsourced labour. The total costs incurred for outsourced labour were obtained from the 2012
cost schedule for dam excavation. The final event node was computed for the outsourcing decision; for the
in-house decision, the options were either to hire machinery for the reclamation tasks or to use existing
machinery used in the day-to-day core business. The final decision was based on the outcome with the
maximum resultant cash flow and EMVs. This represents the most cost-effective option.
The following assumptions and judgements were used in the study:
 Inclement weather reduces the efficiency of reclamation operations like dozing because rain
water fills up pits and the soil becomes muddy and slippery. It is therefore assumed that
reclamation activities are carried out in the months of April to October, when there are no heavy
rainfalls expected. This increases visibility and efficiency for operators.
 Only up to 20% of the original sand bulk is retained in the sand traps and trap dams as fine silts
and clays to be used for reclamation. This figure is based on previous experiences at the mine.
Thus, in the latter years towards the end of the life of the mine, additional backfill material would
have to be sourced to effectively complete the land restoration procedure. However, biosolids
will play a role in increasing backfill volumes.
 No fertiliser or mulching was required in the reclamation process; site investigations indicated
that clays in the sand traps and old dumps supported vegetation. Municipal biosolids were taken
as the cheaper alternative, which would result in more cost-effective reclamation than a process
using fertilisers and lime.
 The decision tree analysis used U.S. dollars at 2012 values, since the cost schedule was created for
2012. This was sufficient for comparing between in-house and outsourced labour. Adjustments to
2014 monetary values were carried out on the resultant NPV (resultant cash flow) derived from
the decision trees as a practical cost-reporting practice. The currency used is U.S. dollars since this
is currently the official currency of Zimbabwe. It is expected that the U.S. dollar will continue to be
the official currency in Zimbabwe for numerous years up to or beyond the possible recovery of
the country’s economy (Sapa-AFP, 2013).

 A loan of US$35,000 would be sufficient to cover the costs of excavating the silted dam based on
the total cost incurred when a company was contracted in 2012 to carry out a dam excavation;
the cost for this job was US$30,174. The US$30,174 that was quoted did not include the site
mobilisation/demobilisation fee of approximately US$2,000. Hence, an additional cost of
US$2,000 was added to give a total cost of US$32,174 for the excavation.
 The in-house cost for the excavation job was scaled down to US$20,000 based on the assumption
that the company would be able to implement some cost-cutting measures on sundry expenses
and labour costs.
 A cost of US$17,269 for hiring machinery was based on the 2014 cost of hiring excavators, which
was US$85 per hour. This figure was multiplied by eight working hours per day and 28 days per
month, and the result was adjusted by 5% for inflation to provide an estimated equivalent in 2012
dollars so that all costs would be consistent.
 For the decision to use the mining company’s existing machinery, the cost incurred was scaled up
by approximately 1.5 times to an estimated cost of US$25,000. This was because using existing
machinery would significantly reduce the availability of the equipment for the day-to-day core
business of silica sand mining. Moreover, there would be additional maintenance costs, including
tire replacement, repairs and scheduled maintenance. This is a burden since the company is
already experiencing machinery breakdowns and shortages of equipment and has only one
excavator currently operational for the day-to-day core business of silica sand mining.
An estimated inflation factor of 5% was used to adjust the 2012 monetary terms to 2014 terms. This inflation
factor is based on the 2013 inflation rates and 2014 projections by the African Development Bank. It is
reported that inflation in Zimbabwe averaged about 4.1% in 2013, and this rate was projected to slightly slow
down to 4.0% in 2014 (African Development Group, 2014). We opted for a slightly higher rate of 5% for
inflation in this study as a simple contingency plan allowance. An inflation factor of 5% was also used to adjust
the current excavator hiring cost to the 2012 value for adequate comparisons of amounts in the same base
year.
Probabilities of 0%, 5% and 100% were used in the decision trees to investigate all possible outcomes as the
authors had no information regarding the probability of obtaining the loan. We used a loan amount of
US$35,000 (in 2012 monetary terms) for the excavation of the silted dam in the decision tree modelling. The
excavator hiring cost used was US$17,269 per month (in 2012 monetary terms). A cost of US$25,000 was
used in the decision to use the mining company’s existing equipment.
This section focuses on the analysis of decision trees, which were modelled to compare the resultant cash
flows and EMVs for two main scenarios using different probabilities. The two scenarios were used to
determine whether an in-house or an outsourced service would be more cost effective. Using a rough
estimate, it was assumed that a cost of US$1,000 would be incurred in applying for the loan; this is the first
amount shown in blue in the decision trees. The analysis decision trees for a 0%, 50% and 100% probability
of getting the loan are shown in Figures 1–3.

 The decision to apply for the loan results in an EMV of US$−1,000. This negative EMV shows that a
loss of US$1,000 (shown in black at the loan application decision node) will result if the mining
company does not get the loan. The amount of US$35,000 is the loan amount that the company
would acquire if it succeeded in getting the loan.
 If the loan were obtained, the costs incurred for outsourcing were taken to be US$32,174 based
on the US$30,174 cost as determined in the company cost schedule and the mobilisation fee of
US$2,000. In the event that the loan was obtained, the optimum decision taken in the TreePlan
modelling is decision 1 (shown in the small square at the third node). An ultimate cash flow of
US$1,826 is obtained at the terminal node.
 If the company decides to utilise in-house services and hire machinery for the job, a resultant cash
flow of US$−3,269 (US$3,269 loss) results at the terminal node. If the company decides to use its
own equipment, the resultant cash flow further diminishes to US$−11,000 (US$11,000 loss).
 If the company does not apply for the loan, the resultant cash flow is zero, based on the
assumption that no costs would be incurred as reclamation is assumed to be implemented if and
only if the company acquires the loan.
 At the first node, the decision not to apply for the loan (decision 2) was chosen by TreePlan. This
is because, based on the 0% probability, it is impossible for the company to acquire the loan.
Therefore, this would be the best decision at 0% probability of getting the loan.
The resultant cash flows are independent of probabilities of getting the loan. It is only the EMVs that are
dependent on probabilities, as these reflect the inherent risk related to the various probabilities on the cash
flow. Thus only the EMVs will be analysed henceforth.

 In the case where a 50% chance of getting the loan was modelled, the EMV at the second node
increased to US$413. This increased EMV is because there is now a 50/50 chance of getting the
loan, so the risk of eroding wealth by applying for the loan is reduced.
 At the first node, decision 1 was chosen in TreePlan as there is now a positive EMV of US$413. In
addition, if the loan were obtained and the company decided to outsource, decision 1 results in
the highest cash flow of US$1,826.
 In the event that the company chose in-house services, decision 1 (hire machinery) would be a
better decision as this results in a loss of US$3,269 rather than a loss of US$11,000. This lower
cash flow is attributed to the higher maintenance costs that would result if the company’s
equipment, already in poor condition, were further subjected to reclamation work.
 In the case where there is a 100% chance of getting the loan, the EMV at the second node further
increased to US$1,826. This increased EMV is because there is now a 100% chance of getting the
loan; thus there is no risk of eroding wealth by applying for the loan.
 At the first node, decision 1 was chosen in TreePlan due to the further increase in the EMV to
US$1,826. If the loan were obtained, the decision to outsource would be chosen.
These cash flows represent the actual money gained or lost in each scenario. On the other hand, the EMVs,
since they are based on probabilities, indicate the risk associated with the various option models. EMVs are
thus useful for comparing the relative risk involved in the scenarios. A higher EMV represents a lower risk,
and vice versa.
The results above can be illustrated in a table for simplicity and clarity. Table 1 summaries the cash flows and
EMVs for all scenarios modelled. The best decision to take is to outsource because this decision results in an
EMV of US$913. It is important to note that the EMVs were calculated based on a 50% probability of getting
the loan and based on the assumption that the company will apply for the loan. The “do not apply” decision
was eliminated in the EMV calculation, resulting in a different EMV from that shown on the decision trees.
Nevertheless, the principle is that the highest EMV resulted from outsourcing, and this is thus the best
decision, as illustrated in Table 1.

Loan Application Who does Equipment Cash

Event Cost Probability EMV
amount cost it? cost flow
Contractor −32,174 0 1,826 0.5 913
Own with −20,000 −17,269 −3,269 0.5 −7,134.5
Getting hired
the 35,000 −100 equipment
loan
Own with −20,000 −25,000 −11,000 0.5
own
equipment
As a standard reporting procedure, it is crucial to calculate NPVs to translate the 2012 monetary terms to
2014 real terms to get an appreciation of the real monetary values allowed. A 5% rate was used based on the
African Development Bank forecasts for 2014 (African Development Group, 2014). In addition, the cost of
hiring an excavator was calculated to be US$19,040 per month based on a figure of US$85 per hour from Dig
It Contracting (2014). Table 2 shows NPVs for excavating the silted dam.
2012 amount 2014 amount

Item
(US$) (US$)
Loan amount 35,000 38,588
Loan application cost 1,000 1,103
Outsourced service cost 32,174 35,471
In-house service cost (hired equipment) 20,000 22,050
In-house service cost (own equipment) 20,000 22,050
Excavator hiring cost per month 17,269 19,040
Cash flow (outsourced) 1,826 2,013
Cash flow (in-house; own equipment) 3,269 3,604
Cash flow (in-house; hired equipment) 11,000 12,128
The calculations above illustrate a slight increase in the monetary figures from 2012 to 2014 estimates. This
increase is due to adjustments for inflation made to establish a sense of the costs and revenues in current
real terms. Thus the time element of any reclamation procedure is important as time plays a crucial role in
determining the purchasing power of the bond amount and inflates the costs of reclamation.
The industrial sand mine analysed is a medium-scale mine operation that is currently facing financial
challenges to remain operational. Hence, any reclamation method should be optimised such that the costs
are as low as possible and there is a favourable benefit to reclamation. The method of using silt and clay
tailings from sand traps, silted dams and old dumps to backfill mined-out pits at the mine was optimised for
cost effectiveness. The optimisation of costs was done by outsourcing reclamation services and eliminating

all unnecessary costs in reclaiming the land so that it is suitable for future agricultural purposes. TreePlan
software was used for choosing the most cost-effective option between in-house services and outsourcing.
Outsourcing was found to be the most cost-effective option as it is the only option that did not result in
diminishing wealth due to the reclamation process. Not hiring equipment for reclamation would involve
higher costs for servicing the company’s existing equipment. In addition, it proved to be too costly to hire
equipment and embark on reclamation using in-house resources. Moreover, if the in-house decision was
taken, the availability of the mine’s single excavator and dozers for silica sand production would be
significantly reduced, thereby reducing the company’s revenues.
This study illustrates how financial calculations can be done to find the optimum budget requirements for
land reclamation initiatives. In the cost analysis, it was shown that the time element of any reclamation
project plays a crucial role. With the progression of time, the purchasing power of the bond amount
diminishes and costs become inflated. Reclamation should thus be done with adequate resources that ensure
no unnecessary delays are incurred in the reclamation project. Outsourcing was chosen as the most viable
option for dam excavation due to higher resultant cash flows and EMVs. This result indicated that the
remaining activities like grading and backfilling should also be outsourced. This would result in lower costs
and better mining productivity. Hence it can be inferred that, for the entire reclamation process, the company
should hire external expertise to result in better optimisation of reclamation costs. Considerable reclamation
cost benefit emanates from the land restoration technique employed by this research. The further cost-
cutting measures that were employed for this study, such as eliminating fertiliser and mulch and using
cheaper and readily available municipal biosolids, ensure a more favourable cost benefit when the technique
is implemented.
One particular practical challenge in reclamation techniques for surface mines regards the soil chemical
composition from tailings. The soil might contain chemicals and might also lack enough organic matter to
support plant growth. According to Cooke and Johnson (2002), the most successful land restoration cases
have been witnessed in the industrial minerals sector. This is because there are fewer chemical toxins in
these soil tailings. At the mine studied for this paper, the old dumps and the tailings in sand traps were found
to support plant growth.
The cost benefit of this research reclamation technique is also coupled with immense socio-economic
benefits that would be reaped for the mining company, the community and the government. The company
gets an enhanced financial outlook and can manage to extend competent rewards to its employees. The
outsourcing partnerships create employment and more revenue streams for the government in terms of tax.
The communities have their domestic animals protected as pit traps are reclaimed and the disastrous
consequences of acid mine drainage (AMD) are controlled. The environment, which is the fuel of healthy
living, is also protected for the benefit of all. The reduced operational costs translate into cheaper building
products from the mine, enabling the community to build decent and durable accommodation. Restoration
of silted dams increases water supply in the communities.
The use of biosolids in reclamation renders the reclamation technique proposed in this study a success in
terms of practicality and cost effectiveness. Industrial sands mines supply municipalities with sand for water
purification processes. Adequate bilateral relations between mines and municipal governments mean that
mines should be able to obtain the required biosolids at a highly subsidised rate. The Zimbabwean legislation
does not encompass the use of biosolids in land reclamation since there has not been extensive research or
industrial practice on the subject in Zimbabwe. Extensive research on the use of biosolids for reclamation in
Zimbabwe needs to be carried out as it proves to be beneficial.
The benefits of acquiring investments and partnerships with local and international investors can only be
achieved if adequate environmental accounting and reporting standards are implemented. An environmental
and sustainability conference for corporate decision makers exposed a serious skills and efficiency crisis
pertaining to environmental accounting and sustainability reporting in Zimbabwe (Ndaba, 2010). This has led
to overall poor disclosure of financial information pertaining to the environment at the ZSE. One of the major
implications of this poor reporting is to make the ZSE less attractive to investors due to undisclosed

environmental risks of many companies (Ndaba, 2010). Investors are likely to be highly sceptical when it
comes to investments where they could incur enormous environmental burdens. Thus, we propose that
adequate reporting procedures be employed for reclamation efforts in order to reap financial gains of
goodwill, investments and business partnerships for companies. Ndaba (2010) lists the following
recommendations pertaining to environmental amounting and sustainability reporting:
 Modify and develop current accounting systems to include environmental issues reporting.
 Include environmental issues in financial planning, including eco-budgeting, cost consideration
and capital appraisals.
 Introduce environmental issues in performance measurement when reporting to interested and
affected shareholders and stakeholders.
 Use decision-making models that encompass environmental issues.
The use of decision-making models, as highlighted in one of the recommendations above, is crucial in
optimising costs and analysing various decisions. The use of TreePlan software to model the cost-effective
decisions for the reclamation process in this study enabled an analysis of different combinations of inputs,
such as the probabilities of getting a loan and whether or not to outsource services, in our analysis of possible
outcomes. Another benefit of using TreePlan software in the cost comparison is that it provides a visual
representation of the output values, namely expected monetary values, while preserving the original input
Several socio-economic benefits are to be realised from the reclamation technique proposed by this research,
as shown by the cost-effectiveness study. The method of using silt and clay tailings from sand traps, silted
dams and old dumps to backfill mined-out pits was optimised for cost effectiveness. This makes the process
feasible and practical for industrial sands mines currently facing financial difficulties. Hence, in the future,
they may be able to implement the reclamation technique and prevent stiff land degradation penalties from
the Environmental Management Agency of Zimbabwe. This study noted that the environmental legislation
in Zimbabwe is vague on reclamation as a means of land restoration and only explicitly defines rehabilitation.
This ambiguity could result in slight deviations regarding the outcomes of this study from the expectations of
the EMA.
Adequate environmental cost accounting and reporting in companies’ financial reporting standards is
necessary to help companies avoid loss of investments and partnerships due to poor environmental
accounting. Thus we propose that adequate reporting procedures be employed for the reclamation
techniques outlined in this study. This would result in companies reaping financial benefits such as goodwill,
investments and business partnerships. The use of TreePlan software in decision trees was useful for finding
the most cost-effective option for the whole reclamation process, as derived from the modelling for the
smaller task of excavating the silted dam.
The study further enhanced cost effectiveness through cost-cutting measures such as the use of cheaper and
locally available municipal biosolids to enhance soil fertility and increase the volume of backfill material. A
mining technique that involves excavating topsoil separately and piling it close to the pits should be used as
this would leave behind a significant amount of topsoil that could later be incorporated in backfilling. This
technique would result in significant cost cutting in transportation and working times when the resulting pits
are to be backfilled. The expensive activity of transporting the topsoil to the plant for washing and processing
only to recover the clay as tailings in sand traps is also averted.
The sociological and ecological impacts of sand and gravel mining can be mitigated if proper land restoration
measures are employed. In fact, positive benefits of mining can be reaped by all interested and affected
parties, namely the local inhabitants, the government and the mining company itself. The local inhabitants
benefit from the ecological restoration of the land for their benefit. The mining company benefits from the
reduced costs of reclamation and the improved image of the company, which will be more likely to attract

financial partners and investors. The company also avoids loss of funds due to environmental penalties. The
improved financial outlook of the company enables it to extend enormous financial rewards and benefits to
its employees and the community at large. The reduced operation costs permit the company to sell its
products, such as high-quality building material, at affordable prices to the communities. This empowers the
communities to build decent accommodation. Employment creation is enhanced through partnerships with
companies offering outsourcing services, creating a strong network of viable interdependent networks
through which government tax revenue streams are opened up. Domestic animals are protected, as open
pits are closed and the anticipated damaging consequences of AMD are addressed beforehand. The
environment, the firm foundation on which life is sustained, is also protected, using a cost-effective and
reliable reclamation strategy. Water supplies for communities, animals and sand silt mines are improved, as
the method involves the restoration of silted water dams.
African Development Group (2014) Zimbabwe economic outlook, viewed 22 May 2014, http://www.afdb.org/countries/southern-
africa/zimbabwe/zimbabwe-economic-outlook/.
City Press (2013) Zimbabwe central bank says no return to local currency, viewed 20 May 2014, http://www.
citypress.co.za/politics/zimbabwe-central-bank-says-no-return-to-local-currency/.
Cooke, J. and Johnson, M. (2002) Ecological restoration of land with particular reference to the mining of metals and industrial
minerals: a review of theory and practice, Environmental Reviews, Vol. 10, pp. 41–71, viewed 16 April 2014,
http://globalrestorationnetwork.org/uploads/files/LiteratureAttachments/115_ecological-restoration-of-land-with-particul
ar-reference-to-the-mining-of-metals-and-industrial-minerals.pdf.
Dig It Contracting (2014) Rates and pricing for equipment hire, viewed 29 July 2014, http://www.digitcontracting.com.au/equipment-
hire/price-list.htm.
Lehigh Hanson Canada (2014) Stewardship and biodiversity: paying it forward - The Sechelt biosolids program, viewed 18 May 2014,
http://lehighhansoncanada.com/stewardship-and-biodiversity-paying-it-forward.
Musa, J. (2009) Assessment of sociological and ecological impacts of sand and gravel mining – a case study of East Gonja District
(Ghana) and Gunnarsholt (Iceland), Land Restoration Training Programme final project, Agricultural University of Iceland,
Reykjavik, Iceland, viewed 15 April 2014, http://www.unulrt.is/static/fellows/document/musah-ja.pdf.
Ndaba, R. (2010) Watching the environment for sustainable business, Chamber of Mines Journal of Zimbabwe, Aug–Oct 2010, pp.
50–52, viewed 20 February 2014, http://www.chamberofminesofzimbabwe.com/publications/category/8-chamber-of-
mines-journal.html?download=27%3Achamber.
SADC Environmental Legislation Handbook (2012) Chapter 16: Zimbabwe, viewed 24 March 2014,
http://www.saiea.com/dbsa_handbook_update2012/pdf/chapter16.pdf.
The Source (2014) Disused mines require millions for rehabilitation – environment agency, viewed 16 April 2014,
http://source.co.zw/2014/03/disused-mines-require-millions-for-rehabilitation-environment-agency/.
Walmsley, B. and Tshipala, K.E. (2014) Zimbabwe, in Handbook on Environmental Assessment Legislation in the SADC Region,
Development paper 179, Development Bank of Southern Africa (DBSA) and Southern African Institute for Environmental
Assessment (SAIEA), pp. 405–417, viewed 24 March 2014, http://www.saiea.com/dbsa_book/zimbabwe.pdf.
Zimbabwe Environmental Law Association (ZELA) (2003) Summary of the Environmental Management Act (Chapter 20:27) for use by
community groups, as cited from SADC Environmental Legislation Handbook (2012).

C. Melchers University of Applied Sciences TFH Georg Agricola Bochum, Germany

P. Goerke-Mallet University of Applied Sciences TFH Georg Agricola Bochum, Germany
L. Henkel University of Applied Sciences TFH Georg Agricola Bochum, Germany
M. Hegemann University of Applied Sciences TFH Georg Agricola Bochum, Germany
In recent decades, the mining industry in Europe has been characterised by a very high rate of mine closure.
One of the affected branches is the underground hard coal mining industry. There is substantial demand for
environmentally compatible mine closure programs, especially in light of the implementation of the new EU
Water Framework Directive. Hence, acid mine drainage is a specific challenge.
By 2018, the last remaining active coal mines in Germany’s largest hard coal mining area, the Ruhr industrial
region, will be closed. With a maximum population density of 2,800 people per km², the Ruhr district is the
largest urban agglomeration in Germany and one of the most densely populated coal mining areas
worldwide. Early coal mining activities date back to the twelfth century. Since then, a total of about 10 billion
t tonnesof coal has been produced. The maximum depth of the mining process is 1,500 m below the surface,
and the maximum subsidence ranges up to 25 m.
It is necessary to sustain the drainage system to prevent built-up areas from being flooded or impacted by
mine water. Hence, water management measures will continue in perpetuity after coal mining has ceased:
about 70 million m3/a of mine water and an additional 900 million m3/a of water in the polder areas have to
be pumped permanently.
Numerous shafts, deep cavities and shallow openings call for remedial activities in order to prevent damage,
stabilise new structures or restore former geomorphology. For these reasons, risk management was
established a couple of years ago. A special challenge will be the protection of subsurface groundwater
horizons, especially drinking water horizons, against infiltration of mine water.
The paper also summarises the results of a study that compiles numerous European hard coal mining areas’
experiences with mine closures. In particular, the investigation compares the effects of rising mine water level,
impacts on the ground surface resulting from mine subsidence/uplift and effects of the mine water
hydrochemistry at the different sites.
On this basis, advice is given to reduce closure risks and optimise closure operations. The purpose is to achieve
lasting safety for the ground surface, the population and the water bodies at a reasonable cost.
The mining industry has played an important part in the economic and social development of Europe. In the
nineteenth century, the extraction of coal and lignite provided the basis for the industrialisation of many
European regions. However, the production of mineral raw materials in Europe has permanently decreased
in recent decades. Between 1992 and 2012, a total decrease of 300 million tonnes took place (Reichl et al.,
2014). The metal ore and hard coal mining industries have been particularly affected by this trend. The main
reasons for this include reservoir conditions, production costs, popular resistance and environmental
regulations. The European hard coal mining industry has therefore had to cope with a massive reduction in
production from more than 350 million tonnes in 1990 to 110 million tonnes in 2012 (European Association

Experiences with mine closure in the European coal mining industry C. Melchers, P. Goerke-Mallet, L. Henkel and M. Hegemann
for Coal and Lignite , 2013). The current annual production of hard coal in the European Union comprises 100
million tonnes. As a result, the past 50 years have seen a significant reduction in the number of active mines,
so that in Germany alone more than 250 mines have been shut down. Approximately 20% of this total relates
to metal ore mines, and more than 50% to underground hard coal mines. For comparison, in 1960, around
140 million tonnes of hard coal were extracted in Germany. After the mine closures in 2012, the last
remaining three operating collieries produce a combined total of barely 7.5 million tonnes a year.
Corresponding to the decline in production, the number of personnel decreased from 130.3 thousand in 1990
to 14.5 thousand in 2013 (Figure 1). The drop in production during recent decades was primarily caused by
sales-related problems and reduction of government subsidies. In 2007, in agreement with the European
commission, the federal government of Germany and RAG AG, the last remaining German hard coal mining
company, decided that subsidised hard coal production would be phased out by 31 December 2018
(Steinkohlefinanzierungsgesetz, 2007; GVST, 2014). Similar agreements have been made for the other coal-
producing states of the EU.
As it is not just Germany or Europe that is affected by mine closures, the subject is of global significance,
posing a challenge for cities, regions and governments all over the world. Structural change in mining districts
is a complex process comprising social, ecological, economic and cultural aspects. Therefore, the demand for
environmentally compatible concepts for the back filling and plugging of mines, shafts and caverns is
continuously increasing. These concepts apply to both active production plants and mines that are already
closed. The handling of mine water in active mining as well as in post-mining areas yields specific challenges.
Directive 2000/60/EC from the European Parliament and the Council from 23 October 2000 established a
framework for community action in the field of water policy. This EU water framework directive commits EU
member states to achieve good qualitative and quantitative status for all water bodies by 2027 (EC Directive,
2000). Unfortunately, in numerous cases it is still not clear how to achieve this objective. The following

chapters reveal the current and prospective situation of the German coalfield regions and summarise the
main research activities of the Post-Mining Research Institute at the University of Applied Sciences TFH Georg
Agricola Bochum (TFH Bochum) concerning mine closure and post-mining in Europe.
In Germany, three coalfields are currently in the transition phase between production and closure. The
largest of these is the Ruhr-area, followed by the Saarland and the coalfield around Ibbenbüren (Figure 2).
Production in the Saar coalfield ceased in 2012. The last remaining collieries are the Auguste Victoria mine in
Marl (Ruhr-area), which will be closed by the end of 2015, and the Anthrazit Ibbenbüren colliery in
Ibbenbüren and the Prosper-Haniel colliery in Bottrop (Ruhr-area), which will both shut down by the end of
2018.
The Ruhr-area, named after the Ruhr River, covers parts of the Federal State of North-Rhine Westphalia
(NRW). According to the Ruhr Regional Association (Regionalverband Ruhr), an administrative association of
cities and districts, the Ruhr metropolitan region covers an area of about 4,400 km2 and is populated by more
than five million people living in 11 cities and four districts. It is the largest urban agglomeration in Germany
and the third largest in the European Union after Paris and London. The average population density is 1,161
people per km2. In the five largest cities – Duisburg, Essen, Gelsenkirchen, Bochum and Dortmund – the
highest population density is 2,800 people per km2 (Figure 3) (Regionalverband Ruhr, 2012).

The beginning of the retreat of the mining industry 50 years ago triggered a crucial structural change in the
region. Today, the overall economy is dominated by service industries, high-tech operations, logistics sites
and research and educational institutions. In the course of this industrialisation, the population of the Ruhr-
area increased very quickly. The infrastructure of the whole region is therefore very well developed.
The extraction of hard coal in the Ruhr coalfield region dates back to the twelfth century. It started in the
south, where the coal seams crop out directly at the surface or underneath thin overlying quaternary
sediments (Pfläging, 1999). The coal-bearing Carboniferous strata was extensively folded during the Variscan
orogeny and reshaped by syn- to post-Variscan break tectonics. North to the river Ruhr, the thickness of the
caprock increases continuously, and the carboniferous rocks are unconformably overlain by massive post-
Variscan deposits of Permian to Quaternary age. According to its extensive distribution, the Upper
Cretaceous strata comprises the most important caprock unit. In the stratigraphical sequence, massive,
fissured and local karstified limestones of Cenomanian and Turonian age are overlain by the Emscher Mergel,
Coniacian to lower Santonian clay marlstones.
The active mining sites operate further in the north, where the target depth reaches up to 1,500 m. The
mines in the southern part of the area were closed gradually in recent decades (Figure 4). In total, about 10
billion tonnes of coal have been exploited from the Ruhr deposit (personal communication with T. Kalisch
(GVST), 2015). The exploitable areas influence a zone of more than 3,000 km2 at the surface, and the mining-
induced subsidence reaches an absolute number of 25 m locally (Harnischmacher, 2010). The last remaining
coal mine will shut down by the end of 2018.

As the Saar coalfield was formed under limnic conditions, the waters draining from this deposit are less
enriched in chlorides than those of the Ruhr coal basin. The hard coal mining in the Saar region dates back to
the fifteenth century. In the first half of the twentieth century, 60,000 people worked in 18 hard coal mines.
The annual production in the year 1960 accounted for 17 million tonnes of hard coal (Slotta, 2011). As a
reaction to a number of mining-induced seismic events (the strongest of which, in 2008, had a maximal
vibration velocity of 93.5 mm/s), the production of hard coal ceased in 2012 (Preusse et al., 2010). Since last
year, the mine water level has been rising.
The hard coal deposit Ibbenbüren, located in the northwest of Germany, is characterised by a horst-structure.
As a result of the upturning underground, the caprock is missing, and the coal layers strike out at the surface.
The colliery is situated on a hill that overtops the surrounding area by more than 100 m. The extraction of
anthracite coal began in the fifteenth century and currently reaches a depth of 1,600 m. The production of
hard coal will be phased out by the end of 2018. The production in the western part of the 90 km² coalfield
ceased in 1979, and the underground mine workings were subsequently flooded (Goerke-Mallet, 2000;
Goerke-Mallet and Drobniewski, 2013; Rudakov et al., 2014).
The perpetual mine management obligations (also referred to as “inherited liabilities with unlimited
duration”) are water management measures that will continue in perpetuity even after coal mining has
ceased. In order to reduce the costs for the public sector and thus for the taxpayers, RAG-Stiftung has
committed itself to funding perpetual management measures for the coalfields of the Ruhr-area, Saarland
and Ibbenbüren. The foundation will obtain the funds for these tasks by selling its shares in Evonik Industries
AG and through income from its holdings and diversified financial assets. Past and future incomes from the

sale of shares, as well as the annual Evonik dividends, supply a large part of the funds the foundation needs
to perform its tasks. At the end of 2013, the foundations’ shares in Evonik and in the real estate firm Vivawest
had a total value of around € 10 billion. It is estimated that starting in 2019, around € 220 million will have to
be invested annually to finance the perpetual mine management (RAG-Stiftung, 2014). It should be noted
that the correction of mining-related damage is not a perpetual management measure. The financing of
mining-induced damage to buildings, properties and roads is and will be covered by the financial reserves of
the RAG Corporation. This will also be the case after 2018 (Fischer, 2014).
Over the centuries, mining has also caused changes to landscapes. Entire regions have subsided – in extreme
cases, by up to 25 metres (Harnischmacher, 2010). As a result, 30% of the morphological plane region
between the rivers Emscher and Lippe is made up of polder areas – potential flooding areas – with no outlet
(Figure 5). Diking and deepening of water bodies resulted in a hydrological system with a dominant
anthropogenic character. The surface water will always have to be actively regulated in these areas in order
to prevent it from collecting in the depressions. More than 200 special pumping facilities are currently
operating in order to ensure that the water is properly drained (EGLV, 2013).
The installation of an underground channel system for wastewaters was not possible because of sustained
mining-induced ground motion. This is why the mining industry and the associated rising population density
turned the untamed river Emscher into a manmade system of open waste waterways. In order to handle the
wastewater and to ecologically transform the watercourse, the Emschergenossenschaft and Lippeverband
(EGLV), supported by the NRW state and the EU, initiated a huge project called the Emscher conversion. In
future, the wastewater will be channelled through closed conduits, and the river and its tributaries will be
converted into natural waterways. Converting a river system this size is a project extending across several
generations and involves far more than turning former open areas used to absorb water in the event of
flooding into attractive recreation areas. The objective is to upgrade the Emscher region decisively through
projects extending well beyond the river. The processing has been developed in continuous dialogue with

neighbouring cities and districts, industry and business, the relevant government authorities and many other
organisations and institutions. The Emscher conversion serves as a model project for numerous regions in
the world (EGLV, 2013).
To avoid flooding or pollution around and in urbanised areas, it will be necessary to maintain water
management for an unforeseeable period beyond the production phase. Annually, around 70 million m 3 of
mine water and about 900 million m3 (RAG Aktiengesellschaft, 2014) of ground- and rainwater in polder areas
have to be pumped and drained (EGLV, n.d.; LINEG, 2014). Fifteen pit water management facilities currently
operate in NRW and five in Saarland. These facilities are located at the deepest points of the shafts, often
more than 1,000 metres underground. A complex system of pump stations collects the pit water from the
underground excavations and adits and transports it to the main water management facilities, where it is
pumped to the surface through metre-wide pipes. The pump system has a redundant arrangement so that if
a pump fails it can be quickly replaced by another (RAG-Stiftung, 2014). Beyond that, contaminated water
(e.g. around former coking plants) has to be collected and cleaned to prevent it from mixing with drinking
water reservoirs.
As water management represents a high expense factor, thought should be given to how to reduce costs of
water drainage measures by partially or completely flooding open underground mine workings. Thus, after
active mining has ceased by the end of 2018, the task is to extensively flood the mine workings in the whole
Ruhr-area. The rising mine water level may lead to various impacts on the caprock groundwater reservoir
and on the ground surface. The protection of important drinking water reservoirs against infiltration of mine
water will be a special challenge in this context (Terwelp, 2014). Therefore, potential risks and the level of
mine water recovery have to be assessed at an early stage in the process in order to minimise these risks.
In the long term, the flooding concept provides an average mine water level between 250 m below zero in
the south and 600 m in the west. This will keep the mine water level far below the lowermost boundary of
sub-surface groundwater horizons in order to avoid any risk of mixing. After the mine water table has reached
the predicted level, which is expected to be in 2035, it is planned to operate six main water management
facilities in the Ruhr region (RAG Aktiengesellschaft, 2014). At these localities, well water management
facilities (brunnenwasserhaltungen) will be installed within the existing shafts (Figure 6). With the help of
powerful submersible pumps, the water will be discharged to the surface and then pumped into the receiving
watercourse.
To date, Germany has been characterised by extensive abandoned surface and underground mine workings.
Around 60,000 abandoned shafts and adits are expected in NRW alone, with 28,000 of these being registered
at the mining authority NRW. According to Welz (2014), 52% of the total land surface of NRW is affected by
abandoned mining. In addition, a huge number of the presumed 60,000 openings to the surface have to be
classified as not or inadequately secured. On the basis of the present evaluation of the available
documentation (e.g. mine maps and authority documents) on the mining activities of the past centuries, the
mining authority NRW identified approx. 11,500 abandoned openings to the surface in the Ruhr region (Welz,
2014). If the mine workings are flooded after active mining has ceased, in the worst case scenario, an
estimated 5,300 abandoned openings to the surface (including 4,000 shafts and 600 adits) will be affected
by the rising mine water (Welz and Weissbeck, 2006). Therefore, numerous shafts and near-surface openings
require long-term securing measures in order to avoid any damage to the surface in order to enable
subsequent use of the area or to reconstruct the former morphology.
A few spectacular sinkholes brought the maintenance required to secure the surface was brought to the
public’s attention (Figure 7).

In order to establish the long-term stability of insufficient backfilled shafts and near-surface extraction fields,
high-performance technologies have been developed to be flexibly tailored to the specific requirements in
each case. In this context, it should be noted that according to the court decisions of recent years and the
German mining law (Bundesrepublik Deutschland, 1980) the last mining operator is obliged to deal with the
legacies and consequences resulting from all mining activities within the respected excavation site in the case
of potential endangerments. For the reasons mentioned above, the mining authority NRW and numerous
mining companies established a risk-management program (Sikorski and Reinersmann, 2010).

Recognising the significance of securing the surface and optimising the long-term water management in the
Ruhr-area, the RAG-Foundation funded a chair for the new academic course “Geotechnical Engineering and
Post Mining Management” at TFH Bochum. In this masters program, the students obtain the competence to
support industrial companies, consulting engineers and administrative bodies in the field of post-mining.
In addition to organising the new masters program, the holder of the endowed chair and his team are
entrusted with establishing the Post-Mining Research Institute. One of the basic research projects is the
development of an organised process for the perpetual obligations in the Ruhr-area. In this context, the
institute is dealing with the retention of specific know-how in the mining industry, with the processing of the
key research areas and the establishment of collaborations. Some of the current research projects are:
 acquisition of density stratification in flooded mine workings;
 monitoring of underground flooding processes;
 analysis of water-bearing adits or drainage adits;
 analysis of terminated flooding processes of mines in Germany, Europe and worldwide.
In flooded shafts of the Ruhr-area, distinct boundaries between differently mineralised water bodies have
been observed at diverse localities. The influx of higher mineralised water from the rock mass and from the
mine workings on the one hand, and the penetrating less mineralised surface water on the other hand causes

significant density differences in the water column. The formation of a stable stratification can be proven by
measurements of temperature and conductivity. These measurements confirm that columns of water in
flooded shafts display stratification with distinct boundaries between individual, homogeneous layers. The
stable conditions are driven by convection currents generated from the influx of higher tempered and
mineralised mine water. Therefore, the observed density stratification can be considered the result of a
complex flow pattern governed by the mineralisation and temperature of the water in the shafts. There is
substantial evidence that density stratification exists in other mining regions as well. There are also
indications of missing convection cells in other localities. Figure 8 shows the temperature and conductivity
profile logged in a shaft located in the southern Ruhr-area. The jump of the measured values for conductivity
and temperature at a depth of between 848 and 850 m displays the boundary layer.
Density stratification is an important criterion for the flooding process of mine workings after closure.
Considering the significance of this phenomenon for the flooding of different mines, proceeding site
investigations are required; they are the subject of a current research project (Melchers et al., 2014).
The flooding of a hard coal mine in the Ruhr region will be attended by intensive monitoring underground
and above ground within the context of a research project. With the help of distinct sensors, the observation
of the rising mine water is carried out via telemetering. The measurement parameters are mainly water
pressure, salinity, temperature, flow velocity, flow rate and methane content. Among other measurements,
the reaction of the surface to the rising mine water will be controlled by satellite geodesy.
For the first time in Germany, the flooding of the Saar coalfield will be accompanied by the implementation
of an environmental impact study. In this context, the uplift of the surface connected to the rising mine water
table and the flooding-induced seismic events will attract public attention.

In the Ruhr-area, as well as in numerous other mining sites, mine water is discharged to the surface via adits.
In many cases, the emerging mine water registers as reddish colour. This phenomenon can be attributed to
increased iron contents due to the oxidation of pyrite. The precipitates may form inside the mine or several
miles downstream. There are 50 drainage adits in the southern Ruhr district, some of which reach a total
length of more than 10 km and may drain extensive mine workings and large areas at the surface. In
cooperation with the responsible mining authority, these adits are currently analysed according to their
water supply, hydrochemistry and possible effects on the surface.
There is evidence that coal mines and ore mines are affected by the oxidation of pyrite, resulting in acid mine
drainage. As part of a research project, the deferrisation systems are to be optimised. In the caprock-free
coalfield of Ibbenbüren the deferrisation system has been operating for many years, with mine water rates
averaging around 6 m3 and iron contents of currently approximately 200 mg/l (Goerke-Mallet, 2000). Similar
conditions are prevailing in the abandoned ore mine of Meggen in the city of Lennestadt, district of Olpe,
NRW. In both cases, the sulphate contents are approximately 2,000 mg/l. As the deferrisation is operating
successfully, continuous process optimisation is required. For a similarly successful reduction of water-
compatible sulphate contents, adjacent techniques have to be developed. The EU water framework calls for
a profound analysis considering the overall topic of mine water purification.
In the context of a research project, 21 European hard coal mining regions have been analysed according to
the available information about mine closures. The evaluation concentrated mainly on the flooding process
within the mine workings and the resulting effects. The different upward and downward movements of the
surface and the specific prevailing hydrochemical conditions are compared. The study provides some
interesting results regarding the implementation of the flooding process and the hydrochemistry of the mine
water. For instance, in nearly all cases, there was no efficient monitoring, making it impossible to perform
any further optimisation of the flooding processes. In some of the flooded underground mines, there are
indications of the formation of density stratification. In none of the analysed cases is there evidence of any
pollution of drinking water horizons by mineralised mine water. However, in many regions, the drain water
adversely affects the water quality of the next receiving watercourse.
The rising mine water level has led to uplift of the surface in many mine regions worldwide. The maximum
of the measured upward movement reaches 30 cm. It should be noted that this number is based on the
available data. Higher upward movement might have taken place in regions where the monitoring did not
have sufficient resolution. Massive mining-related damage to buildings in Germany to date is only known to
have occurred in one German hard coal mining district, near Aachen. Here, the flooding mobilised a tectonic
fault, unilaterally inducing the generation of a discontinuity (Melchers and Dogan, 2014).
Based on the previously performed and planned studies, suggestions can be given regarding the reduction of
closure risks and the optimisation of processes related to mine management. The goal of the current research
is to achieve sustainable safety for the ground surface, the population and the water bodies (ground- and
surface waters) in combination with reasonable costs.
For not only the Ruhr deposit but also numerous other mining areas, the development and deepening of
knowledge on the optimisation of closure processes is important for environmental and financial reasons.
The projects of the research institute are designed to provide information about chances and risks in
managing planned floodings of underground mine workings for different mines in the foreseeable future.
This includes dealing with former extraction fields and proposing solutions for an environmentally compatible
purification of mine water.

The current domestically sourced production of raw materials covers approximately 70% of the total demand
for resources in Germany. Germany remains a mining country with an extensive sector of old mining. In a
region with such a high population density, the old mining sector creates special risks for the safety of the
ground surface, the population and the environment. Unstable abandoned mine openings and sub-surface
mine workings, uncontrolled methane emissions and acid mine drainage must be considered. One subject of
the ongoing research at the Post-Mining Research Institute at TFH Bochum is risk management for post-
mining and the optimisation of closure processes in the mining industry. The conclusions to be drawn from
the discussed and further studies will be used to design the processes for closure and safekeeping as
optimally as possible.
Thus, the environmentally compatible flooding of the underground mines of the largest German hard coal
region, the densely populated Ruhr-area, is to be planned and accompanied by an efficient monitoring
program. Beyond pit water management, the so-called perpetual obligations include extensive polder
measures and groundwater purification. The specific need for action in this context is also spurred by
European regimentations.
There is a global need for environmentally compatible closure and safekeeping methods. The objective
should be to enhance the transparency of communication on this subject in order to transfer specific pre-
existing know-how across distinct mining regions all over the world. An important element is the
international exchange of experiences.
Bundesrepublik Deutschland (1980) Bundesberggesetz BBergG, last amended by Art. 4 Abs. 71 G on 7 August 2013, 79 p.
EC Directive (2000) EC Directive 2000/60/EC of the European Parliament and of the Council of 23 October 23rd 2000 establishing a
framework for community action in the field of water policy, Official Journal of the European Communities (L 327), Vol. 43,
72 pp.
Emschergenossenschaft and Lippeverband (EGLV) (ed) (2013) Booklet fließgewässer im emscherraum – biologie, beschaffenheit,
bachsysteme, Essen, Germany, 71 pp.
Emschergenossenschaft and Lippeverband (EGLV) (ed) (n.d.) Pumpwerke – zahlen, daten & fakten, Essen, Germany, viewed 12
January 2015, http://www.eglv.de/wasserportal/impressum.html.
European Association for Coal and Lignite (EURACOAL) (2013) Coal Industry across Europe 2013, B. Ricketts (ed), 5th edition, 80 p.,
viewed 06 May 2015 http://www.euracoal.be/pages/medien.php?idpage=1410
Fischer, P. (2014) Ende des Steinkohlenbergbaus im Ruhrrevier – Zeitplan und Herausforderungen, in Beiträge zum RuhrGeo Tag 2014
Ende des Steinkohlenbergbaus im Ruhrrevier: Realität und Perspektiven für die Geotechnik, T. Schanz (ed), Ruhr-Universität
Bochum, Schriftenreihe des Lehrstuhls für Grundbau, Boden- und Felsmechanik, Vol. 50, pp. 1–10.
Goerke-Mallet, P. (2000) Untersuchungen zu raumbedeutsamen Entwicklungen im Steinkohlenrevier Ibbenbüren unter besonderer
Berücksichtigung der Wechselwirkungen von Bergbau und Hydrologie,PhD dissertation, Rheinisch Westfälische Technische
Hochschule Aachen, Aachen, Germany 226 pp.
Goerke-Mallet, P. and Drobniewski, M. (2013) Planning long-term mine-water management for the Ibbenbüren coal basin, in
Deutscher Markscheider-Verein e.V. (DMV) (ed), Proceedings XV International ISM Congress, (International Society for Mine
Surveying), Aachen, 16 – 20 September 2013 pp. 319–324.
GVST (2014) Gesamtverband Steinkohle e.V. Jahresbericht 2014 – Herausforderungen und Perspektiven, Herne, Germany 2014, 86
pp.
Harnischmacher, S. (2010) Quantification of mining subsidence in the Ruhr District (Germany), Géomorphologie: Relief, Processus,
Environnement, Vol. 3, pp. 261–274.
Kaiser, U. (2002) Das Bergbauinformationssystem (BIS) des Landes NRW, Altbergbaukolloquium, VGE Verlag GmbH, Essen, Germany
pp. 68–96.
LINEG (ed) (2014) Abwasserpumpanlagen, Kamp-Lintford, viewed 12 January 2015 http://www.lineg.de/Abwasser/Abwasserpump
anlagen/index.php.
Melchers, C. and Dogan, T. (2014) Studie zu erfolgten Grubenflutungen in Steinkohlenrevieren Deutschlands und Europas.
Altbergbaukolloquium, VGE Verlag GmbH, Essen, Germany, pp. 300–305.
Melchers, C., Wesche, D. and Coldewey, W.G. (2014) Ausbildung von Dichteschichtungen in Schächten des Steinkohlenbergbaus im
Ruhrrevier, Altbergbaukolloquium, VGE Verlag GmbH, Essen, Germany, pp. 156–162.
Pfläging, K. (1999) Steins Reise durch den Kohlenbergbau an der Ruhr – Der junge Freiherr vom Stein als Bergdirektor in der Grafschaft
Mark, Vol. 6, Geiger Press, Horb am Neckar, Germany, 225 pp.

Preusse, A., Kateloe, H.-J. and Sroka, A. (2010) Assessment of seismic events in German hard coal mining – occurrence and prediction,
in Proceedings 10th Underground Coal Operators’ Conference, N. Aziz (ed), University of Wollongong and the Australasian
Institute of Mining and Metallurgy, pp. 135–138, viewed 06 May 2015 http://ro.uow.edu.au/coal/322/.
RAG Aktiengesellschaft (2014) Neue Wasser – Wege: Die Konzepte zur Grubenwasserhaltung der RAG, Steinkohle – Das Mitarbeiter-
magazin der RAG Aktiengesellschaft, Vol. 12, pp. 4–5.
RAG-Stiftung (ed) (2014) Ewigkeitsaufgaben, Essen, Germany, viewed 12 January 2015, http://www.rag-stiftung.de/ausgelagert/
impressum/.
Regionalverband Ruhr (2012) Kleiner Zahlenspiegel der Metropole Ruhr – Zahlen, Daten, Fakten, Stand Dezember 2011, Essen,
Germany, 9 p.
Reichl, C., Schatz, M. and Zsak, G. (2014) Minerals production/Rohstoffproduktion, in World-Mining-Data, International Organizing
Committee for the World Mining Congresses, Vol. 29, Wien, Austria, p. 261.
Rudakov, D.V., Coldewey, W.G. and Goerke-Mallet, P. (2014) Modeling the inflow and discharge from underground structures within
the abandoned hardcoal mining area of West Field (Ibbenbüren), in An Interdisciplinary Response to Mine Water Challenges,
Sui, W., Sun, Y., and Wang, C.(eds), China University of Mining and Technology Press, Xuzhou, China, pp. 699–705.
Sikorski, A. and Reinersmann, N. (2010) Altbergbau in Nordrhein-Westfalen, Bergbau – Zeitschrift für Rohstoffgewinnung, Energie,
Umwelt, Vol. 61(1), pp. 9–14.
Slotta, D. (2011) Das Steinkohlenrevier an der Saar, RAG Aktiengesellschaft (ed), Herne, Germany, 35 pp.
Steinkohlefinanzierungsgesetz (2007) Steinkohlefinanzierungsgesetz vom 20 December 2007, zuletzt geändert am 11 July 2011 (BGBl.
I), 3086 pp.
Terwelp, T. (2014) Bergwerksstilllegungen unter Berücksichtigung der Grubenwassersituation im Zuge der Stilllegung des
Steinkohlenbergbaus im Ruhrrevier aus Sicht der Bergbehörde, Bergbau – Zeitschrift für Rohstoffgewinnung, Energie,
Umwelt, Vol. 65(10), pp. 454–459.
Welz, A. (2014) Gefahren des Altbergbaus aus der Sicht der Bergbehörde, Praktikerseminar Rechtsfragen des Altbergbaus, Ruhr-
Universität Bochum, 2 September 2014.
Welz, A. and Weissbeck, G (2006) Protokoll der Regionalratssitzung vom 28.09.2006 zum Thema: Hinterlassenschaften des Bergbaus
– Probleme, Aufgaben und Chancen aus der Sicht der Bergbehörde, 48 pp.


V.O. Kofoed Willowstick Technologies, United States

M.L. Jessop Willowstick Technologies, United States
D.F. Bierman Willowstick Technologies, United States
The responsibility to protect, monitor and maintain groundwater quality will always be a chief obligation of
the mining industry. This paper considers a high-speed, minimally invasive groundwater mapping technology
called the Willowstick Method. The method has been specifically designed to map preferential flow paths or
areas of highest interconnected porosity (transport porosity) within the subsurface. The method has proven
effective in delineating and characterising subsurface aqueous systems in many complex hydrogeologic
settings for numerous mining clients in a variety of applications. This paper includes three case studies: a
tailings impoundment seepage investigation, an investigation of increased selenium levels near a mine site
and an acid rock drainage investigation.
One of the greatest challenges facing the mining industry, with regard to active mining practices and eventual
mine closure, is the long-term liability involved in requirements to monitor and protect groundwater. When
a mine is closed, the potential threat to air quality ceases. Following reclamation of the land (i.e., covering,
contouring and reseeding), the mine eventually fades in with the surrounding scenery. Groundwater quality,
however, can be a perpetual concern.
Understanding the source, location and distribution of groundwater into and through mine workings is
becoming increasingly more important for the successful operation and eventual closure of a mine. This
paper considers a high-speed, minimally invasive mapping technology called the Willowstick Method. This
method has been specifically designed for mapping preferential flow paths or areas of highest interconnected
porosity (transport porosity) within the subsurface. The method has proven effective in delineating and
characterising subsurface aqueous systems in many complex hydrogeologic settings for numerous mining
clients in a variety of applications.
The Willowstick Method is a unique application of magnetometric resistivity (MMR) (Edwards and Nabighain,
1991). Most earthen materials are fundamentally electrical insulators with electrical conductivities ranging
between 10-12 and 10-17 S/m. Yet in situ measurements of electrical conductivities range from 10-1 to 10-8 S/m,
which is many orders of magnitude higher. This discrepancy is due to the conduction of electric current by
way of ions dissolved in the groundwater. After determining the groundwater of interest, the method
involves placing electrodes in direct contact with the groundwater to create an alternating current (AC)
electric circuit that follows the groundwater’s natural course. A specific signature frequency is used, 380 Hz,
which avoids the harmonics of the power frequencies that are in common use around the world. With
adherence to the Biot-Savart Law, which precisely describes how magnetic fields are generated by electric
currents, the distribution of subsurface electric current flow can be mapped by carefully measuring the
magnetic field with a portable instrument that has been designed for this purpose. Figure 1 shows an
illustration of the instrument, which is hand-carried to each measurement station. The principle components
of the instrument are a magnetic field receiver, a global positioning system (GPS) and a handheld computer.

Identifying, mapping and modelling infiltration into and through mine workings V.O. Kofoed, M.L. Jessop and D.F. Bierman
The magnetometer consists of three coils arranged orthogonally. The magnetic coils are high-inductance and
yet compact due to a proprietary design. The inductance of each coil is approximately 60 Henries. The size is
about 5.7 cm in length and 3.8 cm in diameter. As a critical element to data processing, measurement station
coordinates are obtained from the GPS and recorded with the magnetic field data. These magnetic field
components are then measured at numerous stations to define the electric current’s subsurface distribution
and flow patterns. Magnetic field data is processed and compared to the predicted magnetic field from a
theoretical homogenous earth model to highlight the deviations from the uniform model. Finally, magnetic
field contour maps and inversion models are created and interpreted in conjunction with other hydrogeologic
data to provide enhanced definition of preferential groundwater flow paths.
Similar to an angiogram, which uses radioactive dye as a contrast agent to image and visualise the flow of
blood through the human body, the Willowstick Method uses a signature electric current as a contrast agent
to image and visualise the location of water through the subsurface (Figure 2). The application of the
methodology is based on the principle that, as a general rule, water infiltrating and flowing through mine
workings substantially increases the conductivity of the earthen materials through which it flows.

A mine located in Canada ceased operations in 1996. As part of reclamation operations, the mine owner
actively monitored groundwater conditions in and around the mine site. A spring south of the mine was
adversely impacted by groundwater flow through the tailings impoundment, while a creek located east of
the mine was not contaminated by groundwater. In 2010, the mine owner contracted Willowstick
Technologies to help characterise preferential groundwater flow in and out of the tailings impoundment with
the expectation that the results would improve future monitoring and possibly help the owner identify
remediation measures.
Historically, mines were abandoned and left without proper cleanup, leaving groundwater contamination
problems. Due to problems in the past with mines, most regulators require mining companies to have mine
closure plans and place multimillion dollar bonds to cover the costs of mine remediation and reclamation.
Water monitoring and treatment is frequently part of these mine closure plans (Lottermoser, 2012).
To map infiltration at this site, an electric current of specific signature frequency was applied by energising
from each of several upstream locations to the tailings impoundment to detect preferential paths coming
into the impoundment (see Figure 3). Then electrodes were placed to energise from the impoundment to
each of several downstream locations to detect preferential paths escaping the tailings impoundment. For
each configuration, as electric current flowed between electrodes, it generated a signature magnetic field
that was measured within the appropriate study area. The methodology also incorporates a prediction of the
magnetic field response for any given electrode configuration, assuming a homogenous subsurface. The
measured magnetic field is compared to this uniform case distribution to bring out any heterogeneity or
preferential flow of electric current. The processed data was used to create a map for a two-dimensional
(surface level) interpretation of preferential flow paths. The data was then processed through an inversion
algorithm to generate a three-dimensional (3-D) model of electric current distribution. This model was
presented within the framework of a 3-D site model, creating a powerful tool to explore and explain how
electric current (interpreted as preferential groundwater paths in this case) potentially infiltrates and escapes
the impoundment.

The investigation provided insight into infiltration and effluent flow patterns through, beneath and around
the impounding structure. As shown in Figure 4, preferential flow paths of groundwater were identified both
into and out of the tailings impoundment. The arrows represent the flow paths interpreted from several
survey configurations. The tailings impoundment captures flow from the north and west sides, but all flow
exiting the impoundment heads in a southern direction; this explains why the spring to the south was
contaminated and the creek to the east was uncontaminated. Although the Willowstick Method can locate
pathways of preferential groundwater flow, it does not directly identify the amount of water along a
particular pathway.
The investigation results were used by a project hydrogeologist to enhance the hydraulic groundwater model
for this site. The results also guided and enabled cost-effective decisions regarding how to prevent
groundwater from infiltrating the tailings impoundment and spreading contaminates to the well located
south of the impoundment. This information will be a guide for proper water treatment for mine closure.


In 1997, the U.S. Forest Service notified phosphate mining companies in Idaho that elevated levels of
selenium had been detected on and near certain phosphate mine sites. As a result, the mine owners became
proactive in minimising the discharge of selenium from their mine. In 2006, an infiltration basin was
constructed up-gradient of an overburden disposal area (ODA) to allow water to penetrate downward,
diverting it away from the ODA and into the regional groundwater aquifer and preventing surface and
groundwater from infiltrating the ODA and leaching selenium. The groundwater contamination problems
must be resolved in order for the mine to continue proper operation and eventual closure.
After the infiltration basin was added to the mine site, regulators questioned the effectiveness of the
infiltration basin and the possibility of flow into the ODA. The mine owners conducted a Willowstick
investigation to determine whether the infiltration basin was working as designed and to identify any other
potential areas of infiltration. The investigation was performed by energising the ODA and surrounding
ground with a signature electric current to map water-saturated zones in, through and around the subsurface
study area.
A Willowstick survey was set up to energise from the infiltration basin to a deep well located down-gradient
of the ODA (Figure 5). The circuit wire connecting the electrodes was positioned in a large loop around the
study area to keep its magnetic field influence at a distance from the survey area. The thin dashed lines in
Figure 5 indicate the general distribution of electric current between electrodes. Measurements were made
at 1,050 stations located on a 33 m grid. The measurement density or grid spacing was adequate to obtain
sufficient detail and resolution for identifying preferential electric current flow paths; at the same time, this
optimised funds available for the investigation in order to adequately explore areas of potential interest.
Figure 6 presents a depth slice of the 3-D electric current distribution (ECD) model obtained by inverting the
magnetic field data. The slice is taken at an approximate 30.5 m depth beneath the topography. The colour
scale of the slice shows the distribution of the electric current. Six areas of interest (labelled 1-6) are noted
in the figure. Areas 1 and 2 indicate where water is pooling and percolating down to the regional aquifer
below. Area 2 is at the north end of a backfilled pit, where water has collected and migrated northward.
Areas 3 and 4 indicate locations where meteoric water accumulates on top of the mineral pile before
infiltrating the ODA. Area 5 indicates where meteoric water accumulates alongside and on top of the ODA
before infiltrating. Area 6 shows where water accumulates and flows along the edge of the ODA at the toe.
To further describe infiltration into the ODA, Figure 7 shows two cross-sectional views cut through the ECD
model (cross-sections A–A’ and B–B’). These cross-sectional views show the majority of the electric current
(dark shading) flows around or beneath the ODA in the deep regional aquifer lying below. The figure shows
the same six locations shown in Figure 6, except area 1, where water contacts the ODA materials. Here again,
areas 3, 4, 5 and 6 are interpreted as meteoric infiltration zones. There is no evidence of significant infiltration
into the ODA from any other location.
The Willowstick technology provided reliable information about infiltration into the ODA study area. The
inversion model identified the locations of preferential electric current flow and the corresponding depths
into and through or around the ODA. The results indicate that the infiltration basin is working as designed to

prevent water from entering the ODA from the west. Areas where water infiltrates the ODA were targeted
for remediation.

The Crystal mine, located approximately 11 km north of Basin, Montana, was in operation from the early
1900s to 1960. While the mine was in operation, it produced 0.1 t of gold, 9.7 t of silver and over 900 t of
lead. The mine includes two long horizontal adits and a sublevel adit (see Figure 8), which are no longer
accessible. A surface trench had been developed along the surface expression of the mineralised zone, but
this has since been filled in and covered over. The subsurface workings are approximately 1.5 km in length.
The depth of the lower adit beneath the study area is roughly 91 m. Conditions were such that groundwater
was infiltrating the lower adit and generating acid rock drainage (ARD). ARD flows out of the entrance of the
lower adit at a rate of less than 1.26 L/sec for most of the year. However, flow increases to as much as
6.31 L/sec during spring snowmelt (May through June). ARD from the lower adit discharges into Uncle Sam
Creek (see Figure 8). The Crystal mine is a good example of a mine that was not properly closed, and
groundwater contamination is a perpetual concern.
The purpose in performing the Willowstick geophysical investigation at this site was to help identify, map
and model preferential groundwater flow paths infiltrating the lower adit in order to monitor areas of
excessive ARD.
Because saturated strata act as good subsurface electrical conductors, the method employed for the
investigation energises the ground with a signature electric current through strategically positioned
electrodes to follow the water-saturated zones through the subsurface study area into the adit. By identifying
preferential electric current flow paths, the technology can successfully answer questions about where
groundwater potentially infiltrates the lower adit.
Three independent electrode configurations were used to bias electric current flow in and out of the lower
adit. The three associated surveys were designed to target shallow, medium and deep reaches of the

subsurface study area above the lower adit. For each survey, predicted magnetic field models were created
for a homogenous subsurface conductivity environment. The measured magnetic field data were compared
to the homogenous models and then modelled using inversion methods to identify anomalous electric
current flow patterns.
Figure 9 presents a summary of the method’s findings at Crystal mine. The map includes the distribution of
electric current density. Four zones of potential infiltration into the lower adit were identified as a result of
the investigation; these are labelled and ranked 1 through 4. Each survey showed supporting evidence, and
the ranking is based in part on the anomalous electric current intensity observed in the three data sets, which
suggests that more groundwater infiltrates through zone 1 than through zone 2, more through zone 2 than
through zone 3 and so on. These four points of interest were determined by close analysis of vertical and
horizontal cross-sections of the electric current density models for all of the surveys.
The solid black lines labelled “left leg,” “cross bar” and “right leg” (roughly forming the letter “H”) are
interpreted as geological features that appear to influence groundwater flow through the subsurface.
Changes or transitions in electric current density across these features tend to obstruct or influence electric
current flow and therefore hydraulic flow to some degree.
The dashed arrows north of the study area represent the inferred direction of shallow groundwater flow.
North of the left leg feature, groundwater following the natural drainage bifurcates. Some flows west and
south away from the adit (dashed line), and some flows inside the “H” and southward toward zone 4. In the
centre of the “H,” a small amount of electric current preferentially flows down from the north toward what
is labelled infiltration zone 3. Groundwater originating north of the right leg feature flows down toward
infiltration zones 1 and 2. The right leg feature is more pronounced at shallower depths; nevertheless, in all
the surveys, electric current tends to flow more strongly and more freely east of the right leg than on the
west side, suggesting an increase in subsurface conductivity east of this feature. Shallow groundwater
observed northeast of the “H” migrates toward Uncle Sam Creek, bypassing the adit.
The results of the Willowstick investigation identified four target areas of possible infiltration into the lower
adit. Three geologic features that impact the electric current flow and hydraulic flow were also noted.
Originally, the client had planned to drain the northwest wetlands to see if that would reduce adit flows.
However, based on the Willowstick findings, the client believes that draining the northwest wetlands would
have little to no effect on the flow rate of ARD. The client is still working on further remediation efforts based
on the information provided by the Willowstick survey in order to clean up contaminates and properly close
the mine.

This paper describes an optimised implementation of the MMR method to track preferential groundwater
flow paths. AC electric circuit was placed in the ground with a signature frequency. The application of the

methodology is based on the principle that, as a general rule, water infiltrating and flowing through mine
workings substantially increases the conductivity of the earthen materials through which it flows. The
distribution of electric current was identified by its signature magnetic field. Three case studies of mines with
groundwater contamination problems were reported in this paper; these include a tailings impoundment
seepage investigation, an investigation into increased selenium levels near a mine site and an acid rock
drainage investigation. In all three studies, the Willowstick Method targeted specific areas to help remediate
problems at reduced expense. The results of the investigations were used by the mine owners and other
consultants to make informed decisions regarding how to best remediate groundwater contamination
problems. One of the mines in this paper is still operating, and the other two are already closed. Resolving
groundwater contamination is a requirement for safe operation and proper closure.
Edwards, R.N. and Nabighian, M.N. (1991) The magnetometric resistivity method, Electromagnetic methods in applied geophysics:
Investigations in geophysics, Volume 2, M.N. Nabighian (ed), Society of Exploration Geophysicists, pp. 47–104.
Lottermoser, B. (2010) Mine wastes: Characterization, treatment and environmental impacts, Springer, New York, 400 pp.


P. Kirsch Sustainable Minerals Institute – Minerals Industry Safety and Health Centre (SMI-MISHC),
University of Queensland, Australia
A. Hine Sustainable Minerals Institute, University of Queensland, Australia
I. Amizlev Montreal Museum of Fine Art, Canada
In mining regions across the world, new topography is being built daily, layer by layer. Symbolic of the primacy
of industry and economics in the past 150 years, this material is a message for future generations, recounting
the history of its makers and reminding us of hundreds of thousands of workers who spent their time digging
massive holes in pursuit of essential resources. The new topography is often absurdist. There are no artists,
designers or community advocates involved in creating alternate shared visions. It is important to reflect on
what this means for the future of these innumerable sites and, more broadly, the state of the land in a
postmining future. Currently, rehabilitation of postmined land focusses on restoring the site to its original
ecological (but not aesthetic) state prior to the mining intervention. With a growing disconnect between
satisfying ecological needs and envisioning an appropriate response to the landscape, there is a critical need
for creative intervention.
This paper investigates perceptions of mining and postmining landscapes, questioning the aesthetically driven
emphasis on the “natural.” It proposes instead an opportunity to reframe the way we look at overburden and
landscape impacts from mining, exploring the possibility of imagined postmine landscapes unburdened by
traditional aesthetic expectations. Concepts from Land Art, and Earthworks in particular, are presented as
models that merit serious consideration for reimagined space and novel place making. These mine works
landscapes, conceived by regional stakeholder round tables, have the capacity to be built by regional mining
consortia as a foundation for sustainable postmining economies. Postmine regeneration offers both the
motivation and the basic materials to create an altogether new environment from a blank slate. Few
circumstances allow the opportunity to truly explore a reenvisioning of the land on the monumental scale
available in these postmined landscapes.
The life of mined land after the life of mine is generally placed in the hands of environmental specialists
grounded in philosophies of ecological restoration or conservation biology. This philosophy bounds the
approach taken at a specific site within a view to repair or restore industrial harm – an approach that
precludes “blue sky,” outside-of-the-box consideration for completely novel uses that would be possible via
construction of new reimagined landforms and place making. Instead of focussing on restoration and
rehabilitation, it is possible to expand postmined land scholarship to embrace consideration of broad
strategies for land planning and management, strategies that include the disciplines of design, art and
architecture.
This narrow focus of contemporary mining practice is documented in the Australian Government handbook
Mine Rehabilitation, which outlines principles and leading practices of mine rehabilitation (Australian
Government, 2006a). The overriding focus is on landform design and revegetation, with particular emphasis
given to restoration of natural ecosystems and reestablishment of native flora. The handbook reviews a series
of very technical topics: rehabilitation objectives, soil handling, earthworks, revegetation, soil nutrients,
fauna return, maintenance, success criteria and monitoring. The publication states that “rehabilitation is the

Earthworks/Mineworks: creating shared visions for postmined landscapes/landshapes P. Kirsch, A. Hine and I. Amizlev
process used to repair the impacts of mining on the environment. The long-term objectives of rehabilitation
can vary from simply converting an area to a safe and stable condition, to restoring the pre-mining conditions
as closely as possible to support the future sustainability of the site” (Australian Government, 2006a, 2).
Within the narrow focus of this handbook, design is restricted to developing and creating appropriate
landforms for the mine site that will behave and evolve in a predictable manner, according to the established
design principles and with resulting appropriate sustainable ecosystems. Consideration for legal
requirements, climate, topography, soils and community views are recommended in premining studies, with
“community views ... clearly most important in deciding the final land use as they are the most likely site
users. The postmining land use for an area should be defined in consultation with relevant interest groups
including government departments, local government councils, nongovernment organisations, Traditional
Owners and private landholders” (Australian Government, 2006a, 3). Yet within this handbook, the
discussion of final land uses has a particular emphasis on natural ecosystems.
This paper will argue that there is a clear opportunity to include the disciplines of design, art and architecture
throughout this consultation process to enable a full exploration of novel and imagined site uses. Stokols
states that:
a fundamental goal of scholarship in the field of landscape architecture is to enhance the practice
of designing, planning, and managing the land. Owing to the inherently multifunctional nature of
landscapes (encompassing both natural or ecocentric, and cultural or anthropocentric
dimensions), many scholars have stressed the importance of developing a strong cross-disciplinary
approach to landscape research and practice–one that moves the field from multidisciplinary
studies of the biological, physical, social, and cultural elements of landscape (which often proceed
in parallel yet isolated fashion), toward interdisciplinary and transdisciplinary research programs
that explicitly integrate the diverse perspectives of multiple fields and facilitate the translation of
research findings into practical guidelines for effective landscape design and management.
(Stokols, 2011, 1)
We argue that, in current mine rehabilitation practice, the multidisciplinary studies of fauna, flora, soil,
precipitation, geological landforms and so on need to be placed within a much broader sphere of
interdisciplinary and transdisciplinary approaches that integrate the design, art and architecture disciplines
together with social and cultural elements of the landscape to facilitate sustainable place making on a
landscape scale.
Mined land is a significant contributor to the topological changes that characterise the Anthropocene as well
as being a strong social symbol (whether interpreted as positive or negative) of the changing relationship
between humans and the earth; and the effects of the mining industry as a physical force will continue long
into the future. As new, deliberate topography is carved layer by layer into innumerable sites across the
globe, it is important to recognise that the values of the time, as well as the history of industry and economy,
will be legible far into the future from the forms that are left behind. This fact must act as a guiding force for
the decisions that surround mining closure and the establishment of postmining landscapes. It is critical that
the creativity of design, art and architecture is harnessed to optimise final outcomes.
Creation of today’s postmining landscapes is torn between microscale biodiversity concerns, medium-scale
slope and irrigation engineering and the pressure to be seen to respond to the needs of surrounding
communities and address public environmental concerns, with mining companies frequently failing to
consider the postmining landscape as a whole. Current approaches to postmining landscape planning, where
there are genuine intentions to rehabilitate or reclaim (there are 50,000 abandoned mine sites in Australia
alone; see Unger, 2014), section the land from the viewpoints of many disciplines, neglecting master planning
and focussing instead on a particular aspect of the rehabilitation. Within the context of land formations, this
is commonly exemplified by absurdist topography, where stability engineering has taken priority, or an
attempt at creating a mimetic landscape, where naturalistic ideals have advocated blending the landscape
with its surroundings (Hine and Kirsch, 2014).

Within the anthropocenic environment, however, such naturalistic ideals are no longer entirely genuine. It is
important at this point to note that there is a difference between the socially constructed image of nature
and the scientific processes of functioning ecosystems. Though nature, like landscape, has gone through
many iterations, it is situated as a cultural idea that stands as the (fictional) antithesis of human-altered
space, and it has no real relationship with the actual environment. It is currently imagined to exist as an
absence of infrastructure (Waldheim, 2006) and as a pristine site that has remained untouched since before
human dominance (Cronon, 1996). Neither of these scenarios can be actualised, as the presence of industry
has permeated every aspect of the natural environment and, in many cases, the mine site is a particularly
concentrated example of irreversible modification.
Despite ongoing disregard for this impossibility and the subsequent emphasis on ecological rehabilitation
evidenced through much of the Australian Government’s mine closure documentation, its own legislation
broadly dictates that land be developed according to an agreed-upon postmining use. General requirements
for the closure plans of Australian mine sites dictate the following:
 future public health and safety are not compromised
 environmental resources are not subject to physical and chemical deterioration
 the post-mining use of the site is beneficial and sustainable in the long term
 any adverse socio-economic impacts are minimised
 the opportunity is taken to maximise socio-economic benefits (Australian Government, 2006b, 4)
Similarly, the International Council on Mining and Metals (ICMM) (2003) recommends broadly that
operations will “plan, design, operate and close operations in a manner that enhances sustainable
development” (Principle 2), “rehabilitate land disturbed or occupied by operations in accordance with
appropriate post-mining land uses” (Principle 6) and “contribute to community development from project
development through closure in collaboration with host communities and their representatives” (Principle
9).
It is evident from these sets of guidelines that while stability and long-term safety are advocated, there are
no direct requirements for rehabilitation that replicates the original site prior to mining intervention. Indeed,
one could go so far as to say that attempting to renaturalise a site will frequently fail to address the socio-
economic impacts of mine closure on the community just as thoroughly as the grass-covered waste piles of
half-hearted rehabilitative attempts. Rather than embracing and attempting to recreate a nostalgic fiction in
order to appease outdated cultural perceptions, then, the reshaping of the land presents an opportunity to
create a cohesive vision that will engage and provide a functional site far into the future, while also
attempting to curb adverse economic and social effects of the mine closure upon the affected communities.
Achieving these visionary landforms requires a comprehensive understanding of the many facets of
landscape development, beyond isolated scientific and engineering solutions. This means including
disciplines that have the capacity for large-scale planning and design, such as landscape architecture, urban
planning and visual arts.
This paper will trace the development of a marginalised but thoroughly interdisciplinary form of postmining
landscape design – one that seeks not to replicate lost topography or neglect master planning in favour of
stability, but instead to create a new site that responds comprehensively to its surroundings, presents a long-
term vision and has the potential to act as a socio-economic stimulant if approached effectively. American
Earthworks, also known as Land or Environmental art, is a visual arts movement originating in the latter half
of the 1960s (both in the U.S. and in England) and exemplified by large-scale, three-dimensional artworks
located outside in remote settings, which were often immobile, ephemeral and site specific. This movement
produced an ecological subset referred to as Reclamation Art, which focussed on repairing industrial sites
through their reinterpretation as earth-based, exterior sculpture. Specifically targeting mines and quarries,
Reclamation Art was a focus early on and enjoyed a brief resurgence in the 1990s, but has since been only

rarely explored as an art form. We assert, however, that there is real underexplored potential in the
interdisciplinary solutions that characterise the movement.
As an extension of minimalism and land art, and as an amalgamation of public art and landscape architecture,
land reclamation takes the form of large-scale reshaping of overburden and onsite materials. Though artists
have varying philosophies, in general, they approach the site not as something to be obscured and replaced,
but instead as something to be enhanced through change that reflects the relationship of humanity with its
environment, thereby highlighting the constant friction between anthropocentrism and environmental
ethics. It also stands in accordance with the postmodern approach to sculpture (Krauss, 1979), broaching
nontraditional mediums and extending beyond the gallery space.
Indeed, the purpose of many reclamation works is not, as one might expect of artists practicing with earth-
based mediums, to condemn the environmental impact of intervention, but instead to take advantage of
industry funding and experiment with landscapes that exist outside the public eye and observe the
interactions between natural and artificial forces over time. The non-landscape and non-architecture of
sculpture in the expanded field must seek a likeness in the land that it alters. Robert Smithson, one of the
earliest exponents of reclamation art, speaking in relation to finding a location in the Netherlands for his
works Broken Circle and Spiral Hill (1971), stated that “I was looking for an area that was somehow raw
because Holland is so pastoral, so completely cultivated and so much an Earthwork in itself that I wanted to
find an area that I could mould, such as a quarry or a disused mining area” (Smithson in Beattie, 2006, 19).
One of Smithson’s reclamation proposals, the Bingham Canyon Reclamation Project, conceived in 1973, was
designed for Kennecott Utah Copper’s copper mining pit in Utah, the world’s largest single open-cut pit.
Apparently ignored by the mining company, this proposal exemplifies the scale and conceptual innovation
that reclamation art exemplifies. It was audacious in its recognition and retention of the extraordinary nature
of the site itself, regardless of the feasibility of its implementation. As described in the Metropolitan Museum
of Art’s online catalogue, “[t]he unexecuted project involved the construction of a massive revolving disk–
reminiscent of the viewing platform of a nineteenth-century panorama–at the pit's base from which to survey
nature's gradual and inevitable reclamation of man's invasive enterprise” (The Metropolitan Museum of Art,
2014).
The work is intended to exist for a significant period of time and thus reveal natural changes in the site.
Consideration of the site’s properties beyond stabilisation and the rehabilitation of non-aesthetic biodiversity
is rarely discussed, and Smithson demonstrates how the introduction of an artist or a designer can assist in
creating a complex solution that synchronises industrial and environmental processes without casting
judgement. As he states, “[a]rt should not be considered as merely a luxury, but should work within the
processes of actual production and reclamation. We should begin to develop an art education based on
relationships to specific sites” (Smithson and Flam, 1996, 380).
Beyond utility, however, Bingham Canyon’s design responds to a recurring goal among artists producing
earthworks, in general, which is to seek to involve visitors both cerebrally and physically, thus providing an
overall experience. Although dependent on the mine’s location in relation to local populations, responding
to visitors and facilitating active interaction is a vital function of the site that frequently goes unaddressed.
One of the keys to creating an honest relationship with the environment is to allow people to dwell within a
landscape that combines the natural with the infrastructural. As it stands, nature has become conceptually
separated from the human environment, with increasing isolation of supposedly pristine areas perpetuated
by geographical and legal demarcation of, for example, national parks and wildlife areas.
While the illusion of nature being separated from human infrastructure may appear to be an effective way
to maintain a healthy planet, it actually marks the transformation from the coexistence of people and place
within the same locale to a mental separation and observance of the landscape from the outside. It has, most
essentially, led to the invention of the landscape itself and the resulting delusion of a balance of nature within
the world represented by this created landscape (Ellison, 2013), which can only be truly penetrated by those
few who exist within it.

By envisaging Bingham Canyon as sculpture, Smithson is able to physically reintegrate human infrastructure
and the natural environment while simultaneously ensuring that this is done in service of the visitor,
reintroducing the idea of the single environ. This is a common theme among reclamation works, whose very
creation is an overlapping of nature and infrastructure, as they are unmistakeably human-moulded earth-
based materials located outside gallery or museum walls.
Reclamation art also stands as an effective solution to the need to incorporate design and landscape planning
into the postmining site. The works produced utilise the existing materials, capitalising on the vast quantities
of overburden and the heavy machinery on-site, and can be constructed alongside excavation processes.
As it stands, however, there are only a limited number of significant examples of either reclamation art or
any form of innovative reclamation on the sites of closed mines. The following catalogue of reclamation
works and projects inspired by the Earthworks movement acts as a historical review, with the works
dispersed over the decades following the Bingham Canyon proposal.
Created for the 1979 symposium Earthworks: Land Reclamation as Sculpture, a landmark event organised by
the King County Arts Commission, Washington, Morris’s work Untitled Johnson Pit #30 took the site of a
discontinued and overgrown gravel pit. Morris adapted it by adding six terraces and removing all vegetation
except for a ground covering of rye grass and a number of truncated tree stumps near the edge of the
property (King County Archives, 2014a).
The work can be viewed as a reminder of the changes that the site has undergone, from preindustrialisation
through to its abandoned state. The reshaping of earth can be perceived as a continuation of the most basic

of human manipulations of the environment – the digging and piling of earth. It may also be a comment on
the changing nature of art, particularly sculpture, in an aesthetic, economic and physical sense.
Morris, in his keynote address during the symposium, communicated the struggle to go beyond the
commercialisation of art as well as the need to move outside the gallery as a result of the growing size of
works and a desire for site-specificity. Increasing scale, he claimed, resulted from the artists’ intention,
derived from minimalism, that artworks share space with viewers; he monumentality enhanced this effect
by including viewers within the work. Thus viewers experience earthworks both temporally and spatially as
they explore their environment (King County Archives, 2014b). Escape from the marketplace through
exterior, often site-specific and ephemeral artworks, and the inclusion of visitors through scale and temporal
observation were all characteristics clearly exemplified within Untitled Johnson Pit #30, but while they fulfil
the artist’s intentions, they also present a compelling case for the beneficial effects of artist involvement in
industrial reclamation.
Commissioned by the Ottawa Silica Company in 1983, Effigy Tumuli are mounds of modelled earth in the
forms of five animals native to the region: water strider, frog, catfish, turtle and snake. Located on land that
had previously been a coal strip mine, the project’s intention was to repurpose the site and create a public
space that was gifted to the state of Illinois as an extension of the existing Buffalo Rock State Park. Heizer’s
animal figures make reference to the surrounding archaeological sites by alluding to the Native American
mounds found throughout the region (The Center for Land Use Interpretation, 2015). Like Morris, Heizer
used scale as a key element within the work, with the figures so large that they could only be viewed in their
entirety and truly understood from aircraft. This necessitates an experiential viewing from the ground, with
physical and “chronological development of perception” (Friederici, 1992), again placing the visitor within
the work and sharing space with it.
Following in the wake of the first pieces of reclamation art typified by Morris, Heizer and Smithson, there
have been a small number of reclamation artworks that follow the same earth-shaping techniques and
experiential ideologies. Notable among these are Maya Lin’s Storm King Wavefield (2009) and Strijdom van
der Merwe’s a.m./p.m. Shadow Lines: Earthwork (2010).

Lin’s work is situated in what was previously a gravel pit within the grounds of the Storm King Art Center, in
New York State. The work is comprised of seven parallel lines of undulating mounds covered in grass, with
the form derived from water waves in both shape and scale. Lin’s work demonstrates a commitment to
achieving an aesthetic, immersive experience through the reshaping of earth, while remaining entirely
compliant with the environmental reclamation policies of the New York State Department of Environmental
Conservation (Maya Lin Studio, 2015).
In contrast, van der Merwe’s work is an earthwork that dismisses ecological rehabilitative efforts in order to
focus on the creative reuse of site-based materials in service of an artistic statement. The work titled
a.m./p.m. Shadow Lines: Earthwork is situated within a mine dump used by De Beers in Namaqualand Mines
on the Northern Cape in South Africa (MacKenny, 2013). Van der Merwe used the heavy machinery already
on site to rearrange the existing overburden into an abstract pattern of parallel lines whose distinct shadows
alter the appearance of the artwork throughout the course of the day, thereby depicting the passage of time.
The retention of the original materials, without introducing vegetation, is reminiscent of Smithson’s Bingham
Canyon project; they share similar immersive intentions for the visitor, facilitating a consideration of the
landscape as simultaneously natural and industrial.
Alongside earthworks that adhere to the traditional medium and intention of reclamation art are a number
of fully functional works that double as environmentally sound rehabilitative processes, including Angelo
Ciotti’s Wingfield Pines (2006– ), Harriet Feigenbaum’s Erosion and Sedimentation Control Plan for Red Ash
and Coal Silt Area – Willow Rings (1985) and Herbert Bayer’s Mill Creek Canyon Earthworks (1982). Each of
these three works focusses on the specific requirements of its site, developing an environmental solution
that, at the time of implementation, was in keeping with reclamation standards.
The most recent project, Wingfield Pines, was a collaborative effort, with Ciotti contributing as a professional
reclamation artist to the aesthetic vision, working with environmental engineers and the Allegheny land trust.
Though a complex site, the central feature is the passive treatment system for the abandoned mine drainage,
which consists of settlement ponds and wetlands (Chilcott et al., 2007). The result is an aesthetic solution to
the required treatment system, which enables visitors to traverse the whole area and observe physical
changes in action through the colour alterations (see Figure 3).
Similarly, Bayer’s Mill Creek Canyon Earthworks (also resulting from the King County Earthworks symposium
of 1979) was designed as a multifunctional public park, storm water retention facility and landscape artwork
(The Cultural Landscape Foundation, 2014).
Bayer, like Ciotti, exemplifies the movement from public art as discrete object to designed space. Mill Creek
Canyon Earthworks provided a practical solution to the buildup of silt and debris that was entering the site’s
waterways from surrounding developments (King County Archives, 2014a). The work sought to retain the
site’s original beauty while simultaneously controlling the flow of water. Bayer chose to use geometry to
inform the work, establishing an unusual site that was designated a new landmark for the city of Kent.


Feigenbaum’s practice in land art was based around tree planting patterns. Willow Rings is a reclamation
project below a mine access road that echoes the experience provided by Bayer and Ciotti’s works, although
it is less overtly geometric. It features a silt pond surrounded by two rows of planted trees. “As the trees have
grown, the scarred terrain has taken on a “settled” appearance” (Marter, 1994, 33). Feigenbaum focussed
more on the reclamative aspects of the work, apparently conceiving the work itself as an act of art in the
same vein as the European conception of land art as a process of creation.
Completed in 2013, the Australian Garden in Cranbourne, has embraced the ecological and structural
limitations of its location within a former sand quarry and established itself as a highly successful sustainable
redevelopment. The site’s industrial operations changed the soil, necessitating a selection of flora adapted
to the new conditions and reflecting the distinctly arid environment that much Australian flora thrives in.
Taylor Cullity Lethlean (TCL) designed an entirely new spatial experience through the abstraction of this
environment. Although not an Earthwork in the strictest sense, the landscape architects, TCL, took direct
inspiration from the land artists’ practice of moulding earth in a number of areas. The space also echoes the
immersive experience sought by many land art practitioners and could be considered to some extent a hybrid
of art and landscape architecture. Indeed, it follows in the footsteps of the original earthworks, as “[t]he
garden highlights the tension between the natural landscape and our human impulse to steadily change it”
(Lethlean, 2014).
In Canada, landscape architect, Martha Schwartz, worked with Barrick Gold to rehabilitate and reshape
14 million tons of tailings into a 170 hectare earthwork, the McLeod Tailings Project, at the closed Geraldton
Mine in Ontario, Canada. The transformation was the result of a cooperative effort of a community, a mining
company and the government ministries that shared a vision. The town’s local population decided to
transform the existing tailings piles found at its main entrance into a more appealing arrival sequence for
visitors. The new landform was intended to become a visual feature as well as a cultural artefact, highlighting
the role of mining in the life of the town and serving to spur economic redevelopment by reshaping access
to the area and facilitating a variety of popular activities within the space itself (Martha Schwartz Partners,
2014; Martschuk, 2001).
One of the more commonly discussed innovative mining reclamation projects was that run by the
Internationale Bauausstellung (IBA) Fürst-Pückler-Land. Based in Lusatia in eastern Germany, the project
operated from 2000 to 2010, identifying 30 individual sites to redevelop across nine zones throughout the
region. The projects, acknowledged due to the work they have produced and the sheer scale of the area that
has been masterfully planned and reclaimed, fall into several categories:
 industrial heritage
 waterscapes
 energy landscapes
 new territory
 border landscapes
 urban landscapes
 transitional landscapes (IBA Fürst-Pückler-Land, 2014)

The aim of these projects was to revitalise a region that had suffered from a negative image, as a result of
industrialisation during the German Democratic Republic, and then through a loss of livelihood, as many
lignite strip mines closed. IBA’s purpose was to oversee the development of a regional plan that encouraged
a rediscovery of the region. One of the largest projects undertaken was the creation of lakes in 30 closed
mine sites, ten of which link together via artificial canals.
Many elements of the Lusatian regional plan resonate with the intentions of pioneer land reclamation artists,
including the manipulation of the existing mine sites without a complete obliteration of their industrial
history; in some cases, this involves an active appreciation for and desire to retain the changed landscape,
with the radical difference between the “moonscape” of sites and the surrounding lush countryside seen as
advantageous (e.g., Landscape Project Welzow, which hosts industrial tourism site visits). IBA also worked
with unusual visions of the landscape, seeking to create new experiences for visitors that allowed them to
overcome their preconceptions of the Lusatia region. Examples include the floating discovery centre known
as “The Sun” and the sculptural “Lusatian Lake Landmark,” which is both sculpture and viewing tower and
allows visitors to survey the lake region.
The projects discussed above share a long-term goal for industrial sites: the intention of reestablishing
biodiversity while presenting broad solutions that can be applied to the site throughout the process of mine
closure and even throughout the life of a mine. Even without reclaiming land through artistic expression,
these considerations must be addressed.
The duality of a site’s identity, both infrastructural and naturalised, during its regenerative phase is an
important and rare feature within our increasing cartographical and psychological separation of
environmental and human landscapes. The retention of this duality in several examples given above responds
to an understanding of and curiosity about this separation. Because of the unusual outcomes that this
response facilitates, however, it can be difficult to bridge the gap between traditional and alternative
conceptions of postmining landscapes and create productive discussions that embrace the full potential of
the land. Despite this, and as a number of organisations have documented (notably IBA and the Eden Project,
located in Cornwall, U.K.), it is an achievable goal that simply requires a detailed planning process and a
willingness to experiment and engage with the public, stakeholders, internal employees and contractors.
In order to demystify interdisciplinary engagement and the creation of an atypical site, the following list
documents the stages of development.
 Identify community of stakeholders (could be a community of practice): mining company,
government, interested parties, impacted communities (leaders, members, youth, schools).
Alongside stakeholders typically identified as relevant to the decision-making process for mined land
end use, youths and design- and arts-based community members can also become involved in
alternative solutions. These groups present an opportunity to create and personalise a local identity
that will grow in conjunction with the contributors and be fully realised far into the future.
 Develop the site vision.
Develop several landscape approaches that not only address the full set of needs for each site
(technical, social, environmental) but also creatively present the competing opportunities available
for repurposing the postmined land, potentially including sustainable, climate-resilient production of
food and fibre; watershed management; biodiversity conservation; alternative energy production;
tourism and outdoor recreation; aesthetic creations and rural/urban livelihoods. Part of this process
could be to implement a series of action-research workshops (or town-hall meetings) to develop,
communicate, debate, revise and fine tune the site vision by bringing together many of the diverse
organisations and communities of practice established and/or already engaged in integrated
landscape initiatives in the region. Stremke et al. (2012) review several landscape visioning
approaches that could be adapted for this activity.

 Enable a participatory visioning process for multiple stakeholders.

The more general involvement of stakeholders is essential, however, as planning for the site requires
visionary input from all directions in order to achieve an outcome that is both effective and satisfying
for all. If ideas about the end-of-mine use are explored early in the mining lifecycle rather than during
mine closure, possibilities for design become attainable. Discussions can take the form of workshops
facilitated by professionals presenting design and/or artistic proposals. Juarez and Brown (2008)
provide a sound review of the benefits and approaches to citizen participation in planning and
design.
 Design, accept and build the designs as the mine is built.
The design process requires a commitment to multidisciplinary teamwork, with designers or artists
joining the traditional teams of engineers and scientists. Objectives must be understood by all
involved, promoting open dialogue and cross-disciplinary communication. If such discussions are
initiated when the mining begins, the workload and cost will be significantly diminished as most
designs require the movement of material via heavy machinery, and such activities can be
accomplished as part of the original operations.
 Include community of stakeholders throughout the life of the mine.
Keeping community members involved in the design and construction process builds a sense of
ownership and acceptance towards the site and demonstrates an intention for an outcome that will
leave a lasting legacy, thereby justifying the significant interruption to the community’s ordinary
social and economic operations.
 Encourage placemaking and final ownership.
The ultimate objective of the atypical postmining landscape is not only to challenge simplistic and
narrow solutions to reclaiming damaged land but to create an environment that facilitates
relationships: between visitor and space, visitor and place and between visitor and visitor. It must
aim towards placemaking, using the creation of effective space as a catalyst for improving
community relations and offering people a harmonious interaction with the site. More important
even than an aesthetic outcome and the ensuing human involvement, the new place must resonate
with its stakeholders and, ideally, create an opportunity for continued economic stimulation. Like the
MacLeod Tailings project, the successful site must be functional and user friendly, encouraging
visitation and use. Furthermore, by being involved and watching the site evolve throughout the
process, stakeholders feel a profound sense of ownership and appreciation for the final result.
While the theoretical underpinnings of the alternative postmining landscape may be diverse and somewhat
complex, the practicalities of implementing such a process and outcome are hardly beyond the scope of an
average site. The scale of the mine site and the unusual alterations it experiences are a clear opportunity to
explore new ways of interacting with and experiencing the environment, and these experimentations
potentially lay the groundwork for a new landscape that responds to the modern environment and offers
long-term vision for affected communities.
By engaging with creative and scientific professionals outside the fields usually associated with mining,
innovative approaches and ideas can be investigated. If viable, these can be implemented to repair disused
land by restoring it ecologically and beautifying it, thereby transforming it into an accessible place for
enjoyment, so that the outcome is effective on aesthetic and utilitarian levels while fulfilling a communal
need. In her review of the Nine Mile Run Greenway project that rehabilitated a very debilitated postmining
watershed in Pennsylvania, Lora Carney makes an emphatic point: “If there is a single lesson to be learned
from the narrative of the Nine Mile Run Greenway Project in Pittsburgh, it is that artists have the potential
to make extraordinary contributions through transdisciplinary ecological initiatives” (2010, p. 63).
This expands the narrow focus of ecosystem restoration into a broader scheme of placemaking. As
Schneekloth and Shibley state, “placemaking ... seeks to create relationships between people and places, and

relationships among people in places. Making relationships and communities is the goal, and making places
is the vehicle for this practice. Beautiful places will emerge as a result of the relational aspects of making
places” (2008, p. 207).
Australian Government (2006a) Mine rehabilitation, Department of Industry, Tourism and Resources, Commonwealth of Australia,
viewed 10 May 2014, http://www.industry.gov.au/resource/Documents/LPSDP/LPSDP-MineRehabilitationHandbook.pdf.
Australian Government (2006b) Mine closure and completion, Department of Industry, Tourism and Resources, Commonwealth of
Australia, viewed 10 May 2014, http://www.industry.gov.au/resource/Documents/LPSDP/LPSDP-MineClosureCompletion
Handbook.pdf.
Beattie, A.H. (2006) The crossroads of industry and ecology: the new landscapes of Robert Smithson, Edward Burtynsky and Susan
Leibovitz Steinman, MA thesis, Department of Art History, Concordia University, Montreal, Canada, 140 pp.
Chilcott, B., Perkovich, J. and Walton, M. (2007) Reinventing Wingfield Pines: creating a regional showcase for floodplain
rehabilitation in a post-mined landscape, Master of Landscape Architecture practicum project, School of Natural Resources
and Environment, University of Michigan, http://deepblue.lib.umich.edu/bitstream/handle/2027.42/50479/FINAL_Wing
field_Pines_Chilcott_Perkovich_Walton_04_17_07_.pdf?sequence=1.
Carney, L.S. (2010) Ecology and the ethics and aesthetics of collaboration, Canadian Art Review, Vol. 35(1), pp. 63–72.
Cronon, W. (1996) The trouble with wilderness: or, getting back to the wrong nature, Environmental History, Vol. 1(1), pp. 7–28.
Crutzen, P.J. (2006) The anthropocene, in E. Ehlers and T. Krafft (eds), Earth system science in the anthropocene, Springer, Berlin
Heidelberg, pp. 13–18.
Ellison, A.M. (2013) The suffocating embrace of landscape and the picturesque conditioning of ecology, Landscape Journal: Design,
Planning, and Management of the Land, Vol. 32(1), pp. 79–94.
Friederici, P. (1992) Art under fire: tourist attraction closed due to flying bullets, viewed 10 January 2015,
http://www.chicagoreader.com/chicago/art-under-fire-tourist-attraction-closed-due-to-flying-bullets/Content?oid=879752.
Hine, A. and Kirsch, P. (2014) Dreams of futures past: curating post-mined landscapes, in Proceedings of AUSIMM Life of Mine
Conference, July 2014, Brisbane, Australia, Australasian Institute of Mining and Metallurgy, Melbourne, Australia, pp. 121–
137.
Hine, A., Kirsch, P., and Amizlev, I. (2014) Red mud – art and the post-mining landscape, Artlink, Vol. 34(4), pp. 32–37.
IBA Fürst-Pückler-Land (2014) IBA projects in Europe's biggest building site overview, viewed 10 January 2015, http://www.iba-
see2010.de/en/projekte.html.
International Council on Mining and Metals (ICMM) (2003) 10 principles, International Council on Mining and Metals,
http://www.icmm.com/our-work/sustainable-development-framework/10-principles.
Juarez, J.A. and Brown, K.D. (2008) Extracting or empowering? A critique of participatory methods for marginalized populations,
Landscape Journal, Vol. 27(2), pp. 190–204.
Lethlean, T.C. (2014) Australian garden, Taylor Cullity Lethlean Landscape Architecture Urban Design, viewed 10 January 2015,
http://www.tcl.net.au/projects/cultural-interpretative/australian-garden.
King County Archives (2014a) Earthworks: land reclamation as sculpture, Exhibits, viewed 10 January 2015, http://www.
kingcounty.gov/operations/archives/exhibits/Earthworks.aspx.
King County Archives (2014b) Robert Morris earthworks seminar 79 1, http://vimeo.com/78577040.
Krauss, R. (1979) Sculpture in the expanded field, October, Vol. 8, pp. 31–44.
MacKenny, V. (2013) Land matters - a visual exploration of land, mining and resources in South Africa, Art South Africa, Vol. 11(4),
pp. 42–45.
Marter, J. (1994) Nature redux: recent reclamation art, Sculpture, Vol. 13(6), pp. 30–33.
Martha Schwartz Partners (2014) McLeod Tailings, Geraldton, Canada, viewed 10 January 2015, http://www.marthaschwartz.
com/projects/reclamation_tailings.php.
Martschuk, J. (2001) Facelift for the future, Canadian Mining Journal, viewed 10 January 2015, http://www.canadianminingjournal.
com/news/facelift-for-the-future/1000184473/?&er=NA.
Maya Lin Studio (2015) Storm king wavefield, 2009, Art, viewed 10 January 2015, http://www.mayalin.com/.
Ruddiman, W.F. (2013) The anthropocene, Annual Review of Earth and Planetary Sciences, Vol. 41(1), pp. 45–68.
Schneekloth, L.H. and Shibley, R.G. (2008) The public realm – weaving a regional civic life, Landscape Journal, Vol. 27(2), pp. 205–218.
Smithson, R. and Flam, J.D. (1996). Robert Smithson, the collected writings. Berkeley: University of California Press.
Stremke, S., van Kann, F. and Koh, J. (2012) Integrated visions (part I): methodological framework for long-term regional design,
European Planning Studies, Vol. 20(2), pp. 305–319.
Stokols, D. (2011) Transdisciplinary action research in landscape architecture and planning prospects and challenges, Landscape
The Center for Land Use Interpretation (2015) Effigy tumuli, (land use database), viewed 10 January 2015, http://clui.org/
ludb/site/effigy-tumuli.
The Cultural Landscape Foundation (2014) Mill Creek Canyon earthworks, Landscapes, viewed 10 January 2015, http://tclf.org/
landscapes/ mill-creek-canyon-earthworks.

The Metropolitan Museum of Art (2014) Bingham copper mining pi t– Utah/reclamation project, The Collection Online, Metropolitan
Museum of Art, viewed 10 January 2015, http://www.metmuseum.org/collection/the-collection-online/search/284572?i
mg=0&imgNo=0&tabName=label.
Unger, C. (2014) What should we do with Australia's 50,000 abandoned mines? The Conversation, http://theconversation. com/what-
should-we-do-with-australias-50-000-abandoned-mines-18197.
Waldheim, C. (2006) The landscape urbanism reader, Princeton Architectural Press, New York.


R.E. Breadmore Aboriginal Affairs and Northern Development Canada, Canada

G.J. Lafferty Aboriginal Affairs and Northern Development Canada, Canada
Aboriginal Affairs and Northern Development Canada (AANDC) has custodial responsibilities for many
contaminated sites situated on federal lands in Canada’s orth. Through the Northern Contaminated Sites
program (NCSP), the Department manages legacy impacts from mining, oil and gas, and military activities.
The objective of the NCSP is to reduce and, where possible, eliminate, risk to human health and the
environment along with associated financial liabilities to Canada. Working within the Treasury Board’s
Federal Approach to Contaminated Sites with funding through the Federal Contaminated Sites Action Plan
(FCSAP), the NCSP is delivered in the Northwest Territories by the Contaminants and Remediation Division
(CARD), which must function within complex regulatory requirements for health and safety and
environmental protection. Among the numerous approvals, permits and licences required for project
execution, one that must be obtained early in the project life cycle is the “social licence.”
Social license can be defined as a community’s acceptance of an undertaking they believe has the potential
to have an effect on their wellbeing. For the NCSP and CARD, social licence with aboriginal stakeholders is
obtained through active community engagement throughout all project phases, from remedial planning to
the execution of remedial measures through to closure and long-term monitoring. Community engagement
within CARD has been modelled largely after the successful Discovery and Colomac mine remediation projects.
Project planning for those projects involved up to two years of front-end community meetings, leadership
updates, site tours and remedial options selection and remedial action plan finalisation. Community capacity
was developed through traditional knowledge studies and high-school-level science camps as well as job
shadows and apprenticeship programs during the remediation phase. These models helped shape CARD’s
engagement strategies including those adopted for the Great Slave Lake (GSL) Remediation Project.
The results obtained to date from the GSL engagement strategy indicate a sound understanding of the project
among the stakeholder groups and strong community support. This suggests that CARD has obtained its social
licence for the GSL Project; the GSL Project Team must now maintain this licence.
Aboriginal Affairs and Northern Development Canada (AANDC) established the Northern Contaminated Sites
Program (NCSP) in 1991 in response to environmental degradation in Northern Canada resulting from historic
resource extraction projects (mineral, oil and gas) and military operations. The NCSP functions within the
Federal Contaminated Sites Management Framework and is funded by the Federal Contaminated Sites Action
Plan (FCSAP). Within the Northwest Territories, the NCSP is delivered by the Contaminants and Remediation
Division (CARD). The primary objective of the NCSP is to mitigate negative impacts from contaminated sites
on northern communities and ecosystems by reducing and, where possible, eliminating risks to human health
and the environment. Community participation at all stages of an NCSP project life cycle is critical to the
successful execution of the project and will ensure that program and project objectives are met.
A “social licence” or “social licence to operate” (SLO) is defined by Gunningham et al. (2004) as “the demands
on and expectations for a business enterprise that emerge from neighbourhoods, environmental groups,
community members and other elements of the surrounding civil society” – in other words, a community’s
acceptance of an undertaking they believe has the potential to have an effect on their wellbeing. This is

Social licence strategies for effective community engagement R.E. Breadmore and G.J. Lafferty
particularly true within the resource extraction industry, where there have been increasing demands for
mining and petrochemical companies to develop more effective engagement strategies with local
stakeholders, not just at project initiation as was the practice in the past, but at all stages of a project life
cycle through project closure. Unlike typical regulatory authorisations, social licenses are intangible
agreements between a proponent and community that are achieved incrementally over time and require the
proponent to go beyond traditional project management approaches and due diligence exercises. Social
licenses are inherently dynamic and, as such, they can be vulnerable to relapse if they are not maintained in
a consistent and continuous fashion. The key outcome of obtaining a social license is a reduction of project
risk. The strength of a social licence is inversely related to both the legal and financial risk factors associated
with environmental degradation. The stronger the social license, the lower the risk that a community will
resist a project (Yates and Horvath, 2013).
Trust and open communication between the proponent and community allows for the sharing of traditional
knowledge, which Parks Canada defined as:
[k]nowledge derived from, or rooted in the traditional way of life of Aboriginal people. Traditional
knowledge is accumulated knowledge and understanding of the human place in relation to the
universe. This encompasses spiritual relationships, relationships with the natural environment and
the use of natural resources, relationships between people, and is reflected in language, social
organisation, values, institutions, and laws. (Parks Canada, 2008)
This exchange can provide the proponent with invaluable information related to seasonal weather patterns,
wildlife migration, harvesting patterns and culturally significant areas. In return, community capacity is
increased with the exchange of science and technology related to the undertaking. Employment, training and
school outreach programs gradually build local skills and awareness and can reduce future project costs by
creating a skilled local labour force. The proponent must make concerted efforts to build trust within the
community and “gain and maintain” a social licence to work in a region (Calette, 2014).
Social licence within the NCSP is highlighted by the successes of the Discovery and Colomac mine remediation
projects, located in the Northwest Territories, Canada. Strong partnerships were developed early with local
Aboriginal communities through active community participation in remediation planning processes, site
tours and the incorporation of traditional knowledge into hard science and engineering approaches. The
partnerships were instrumental in streamlining the finalisation of remediation plans and regulatory
processes, thereby greatly reducing cost and timelines. The successful engagement approaches employed on
the Discovery and Colomac projects form the foundation of the Great Slave Remediation Project Community
Engagement Strategy, which to date has included: (1) stakeholder identification, (2) community engagement
plan development, (3) community engagement plan implementation, (4) technical review and traditional
knowledge study (mapping exercise, aerial site reconnaissance and community-delivered questionnaire), (5)
a remedial options analysis workout with the community, (6) Elder and land and water board site tours, (7)
the Copper Pass trail clearing program with a community-based forestry crew and (8) project update
meetings (Breadmore et al., 2013).
1 Background
The Discovery Mine is an abandoned underground gold mine situated within the Akaitcho region of the
Northwest Territories, located approximately 85 km north of the City of Yellowknife, on the edge of Giauque
Lake. The mine operated through the 1950s and 1960s and was, at the time, one of the most profitable gold
mines in the country. The mine was closed permanently in 1969, and upon abandonment, the mine property
reverted to the federal government (Silke, 2009). The primary stakeholders affected by the Discovery
Remediation Project are the Yellowknives Dene First Nation, whose concerns were related primarily to the
surficial and sub-aqueous mercury and heavy metal–bearing tailings left uncontained at the site and the risks

First Nations and Communities
posed to human health and the local ecosystems. Between 1998 and 2000, AANDC remediated the Discovery
Mine site, managing the tailings onsite using an engineered cover design.
The Colomac Mine is an abandoned open pit gold mine situated within the Tlicho region of the Northwest
Territories, located 220 km northwest of the City of Yellowknife, just north of Indin Lake. The mine operated
between 1990 and 1991 and then again between 1994 and 1997. A drop in gold prices permanently halted
production, and in 1999, the Colomac property reverted to the federal government (Silke, 2009). The primary
stakeholders affected by the Colomac Remediation Project are the Tlicho First Nation, who were deeply
concerned with the state of the mine site at the time of abandonment, primarily because of cyanide-bearing
tailings and tailings water, petroleum hydrocarbon-contaminated soil, sediments and groundwater, waste
rock dumps, dilapidated mine infrastructure, waste chemical inventories and the risk to downstream
environments. The Colomac Mine site was remediated by AANDC between 1999 and 2011.
The Great Slave Lake Remediation Project consists of three abandoned mine sites within the East Arm basin
of Great Slave Lake, Northwest Territories. Gold, copper and tungsten were mined at the Outpost Island Mine
site (1941–1942 and 1951–1952); cobalt and nickel at the Blanchet Island Mine site (1950s through 1970s);
and nickel at the Copper Pass site in the late 1960s (Silke, 2009). The Outpost and Copper Pass sites were
partially remediated in the 1990s under the Arctic Environmental Strategy program, with a focus on surface
clean-up and site stabilisation. The primary stakeholders for the GSL Remediation Project are the Akaitcho
Dene and Metis, whose site concerns include the metal-impacted sediments, tailings and waste rock/ore,
mine openings, impacted soils and hazardous/non-hazardous wastes. Environmental site assessments were
completed between 2005 and 2012, with the Human Health and Ecological Risk Assessment and Remedial
Action Plan (RAP) finalised in 2013.
YELLOWKN#688143 - v1
Engagement strategies must be delivered in a manner appropriate to the community. For example, does the
community prefer to be engaged through the local executive or band council office, lands and environment
committee, technical advisor or Elders’ committee? Do they prefer site tours, site meetings, public meetings,
workouts or perhaps all of the above? Engagement methods that do not consider a community’s wants or
needs will at best be poorly received by the community; at worst, they could derail the social license process.
While it is desirable to develop standardised approaches and procedures for engagement strategies, subtle
differences between projects and stakeholder groups require scalable and adaptable approaches.
Engagement strategies should be finalised based on feedback gathered during the stakeholder
identification/assessment process.
The Discovery Mine Remediation Project marked the first NCSP project to use a formalised, “front-loaded”
community engagement strategy. The project team recognised the need to “engage early, engage often,”
and public meetings and site tours were organised early on in the project life cycle. It was understood that
community members wanted to share their concerns and knowledge about the mine site and that the project
team needed to actively listen to community concerns and include community members in the decision-
making process. The project team was faced with the challenge of evaluating qualitative remedial options
and initially attempted a quantitative “9-3-1” approach where nine points were assigned to the best fit
option, three points to the next best and one point to least. After limited success, the project team adopted
a “high-medium-low” method that better fit the community’s preference for ranking the options; that is, the
qualitative ranking was more easily understood by the Elders and increased the likelihood of achieving the
desired outcome of the option being selected.
Shortly after AANDC completed emergency care and maintenance measures at the Colomac site (1999
through 2002), a partnership agreement was struck with the Tlicho leadership to develop the Colomac
remediation plan. Four Tlicho Elders were selected to represent Tlicho leadership on the Colomac project
management team (PMT); they became actively involved with the remedial planning process, including
participation in numerous site visits, mapping exercises, traditional knowledge studies and a remedial options
selection process based on the discovery model. The selection process was revised slightly from the discovery
model to an “acceptable-unacceptable-preferred" approach as requested by the Tlicho members. This
approach provided greater clarity on the ranking and the desired outcome of the option being selected.
The remediation plan was completed in 2004, setting the groundwork for the major remediation effort that
was carried out between 2005 and 2011.
Throughout the engagement process, the Colomac Project Team listened to the Tlicho’s concerns regarding
the need to raise awareness among Tlicho youth about environmental science and engineering, particularly
how they relate to contaminated site remediation within the Tlicho region. Discussions with local district
education authorities suggested that a science camp approach might be appropriate, and plans were set in
motion for the inaugural Tlicho youth science camp in 2007. The camps were staged annually until 2012, at
which time a new community-based approach was developed and delivered jointly with the Government of
the Northwest Territories. The Colomac Mine Remediation Project entered the post-closure phase in 2012,
and the Tlicho continue to be very much involved with the long-term monitoring activities.
The 2004 Colomac Site Remediation Plan was inherently conceptual, with specific components established
as “design-build” approaches that would encourage innovation from the remediation contractor. One such
component was the waste rock perimeter berm around the three open pits. The design intent of the berm
was to prevent caribou and hunters/trappers from accessing the open pit area post closure.

During a site tour shortly after the perimeter berm was proposed, the Tlicho Elders observed that caribou
would likely not expend the energy to scramble up the steep waste rock slopes where there was no food
source awaiting them. The Elders suggested that the PMT consider using the waste rock slopes themselves
as barriers. The only additional remedial effort would be the selective placement of 2.5-metre-high waste
rock berms at access points to the south and north waste rock and open pit areas.
The placement of the seven shorter waste rock berms (indicated as black circles in Figure 3) across access
points to the waste rock dumps and open pits was effective in achieving the remedial objective of preventing
wildlife and hunter/trapper access to the waste rock dump and open pit areas as well as creating cost savings
to the Crown.
Mine Design Cost Adaptive design Actual cost Cost savings

component
Waste rock Continuous berm Unit rate of Use slope of waste C$ 150,000 Cost savings to
around open pits C$ 29/m3 × rock dump as Crown =
(2.5 m × 6,000 m) to 53,500 m3 physical barrier C$ 1.65 M
prevent caribou = C$ 1.8M plus placement of
from entering pits selective waste
rock berms at
access points
The GSL Remediation Project benefitted greatly from the strategies and lessons learned from the Discovery
and Colomac models and adopted these models as best practices. Still, the GSL remediation project was
functioning within a newer regulatory regime than the earlier models. The Mackenzie Valley Land and Water
Board created guidelines for proponents to follow and required that regulatory applications be accompanied
by a community engagement plan and a consultation log.
With the GSL project moving into the remediation planning phase in 2012–13, there was a need to develop
a rigorous community engagement strategy. Stakeholders were identified through consideration of site
proximity and existing comprehensive land claim agreements and interim measures agreements
(assumptions validated through AANDC’s Aboriginal and territorial affairs directorate) and included the
Yellowknives Dene First Nation (N'Dilo and Dettah), Lutsel K’e Dene First Nation, Deninu Kue First Nation and
Fort Resolution Métis Council. With Copper Pass falling within the Mǫwhì Gogha De Niitlee boundary (Tlicho
Final Agreement), the Tlicho were also consulted. Other stakeholders included our federal partners: Parks
Canada, Health Canada, Fisheries and Oceans Canada and Environment Canada; the Mackenzie Valley Land
and Water Board; the Government of the Northwest Territories; Environment and Natural Resources and the
Great Slave Yacht Club.
The GSL Project Community Engagement Plan was developed in late fiscal year 2012–13. Community
leadership and members were invited to project update meetings to provide an introduction to the new
project team, an update on work completed to date and a description of next steps for the project. The
meetings also provided a forum for the communities to share their knowledge of the sites, express their
concerns and inform the PMT how they would like to be engaged on the project. Through the meetings, it
was determined that the communities possessed solid knowledge of the sites and had built some capacity
through earlier clean-up efforts, mine remediation/monitoring training sessions and more recent
community-based monitoring programs.
Follow-up meetings were held with the Lutsel K’e Dene First Nation, Deninu Kue First Nation and Fort
Resolution Métis Council executive councils in April 2013 to respond to questions received by email and
raised during the project update sessions, discuss socio-economic opportunities and map out next steps for
the project. Coming out of these meetings, it was determined that the communities preferred the GSL

Options Analysis workout to be held in Fort Resolution to bring the communities together for a collaborative
approach.
The GSL traditional knowledge study was conducted in May 2013 and included a mapping exercise with
community-selected Elders in Yellowknife, aerial site reconnaissance with selected Elders via helicopter and
a community-delivered traditional knowledge questionnaire. The RAP was provided to the communities for
review by their technical advisor around the same time. The remedial options selection workout was
conducted over two days in June 2013. The sessions opened each day with an opening prayer recited by a
selected Elder, followed by technical presentations, in-depth discussion among working group members
related to site-specific risks and the merits of each remedial option being tabled. Engineering and
environmental science information was exchanged for traditional knowledge of the sites and surrounding
areas and, in the end, preferred options selection was achieved through round table consensus.
The remedial options selection process involved a number of steps:
 discussion of all technically sound options (including a “base case” or “do nothing” option);
 populating the options selection matrix posters based on group consensus (acceptable,
unacceptable, preferred) as a visual indication of the option decision;
 documenting the option decision on flip chart/posters as well as digital recording to provide
clarity to the group at the time of discussion and become part of the permanent record of
proceedings;
 actively listening to Elders’/community members’ concerns regarding traditional land use in the
area and commitment to follow up.
Mine Remedial Remedial Remedial Remedial Preferred remedial

components option 1 option 2 option 3 option 4 option
Mine openings Leave as is Retrofit caps Upgrade existing “Hybrid” “Hybrid”

and raise fill-in concrete caps Concrete Shaft #1  Option 2
and seal main and foam Shaft #2  Option 3
raise
Tailings/waste Leave as is Consolidate and No preferred option
rock re-slope identified
Following the remedial options selection workout, an Elders and water board site tour was completed in July
2013. The site tour proved effective in “ground-truthing” the workout discussions; for example, for the
tailings/waste rock component at Outpost Mine, a preferred option could not be achieved during the options
workout (italicised in Table 2), but during the site tour, the area around the north bay was thoroughly
examined and discussed with the Elders and technical advisors. Following the site tour, there was a dinner
meeting in the community of Lutsel’ Ke, at which time the preferred option for the Outpost tailings waste
rock component was formally decided.
With input from the community’s technical advisors and completion of the options selection workout and
site tour, the GSL RAP was finalised. The final RAP document included proceedings from the RAP options
selection workout as a record of all of the decisions made.
One size does not necessarily fit all, and that certainly applies to remedial options for NCSP projects. Using
the example of waste rock management, the remedial options selected for Discovery, Colomac and GSL
varied considerably. Waste rock at Discovery Mine was managed as part of the underground mining
operations (stabilisation) or crushed for airstrip and road construction and was not included as a remedial
component. The waste rock inventory at Colomac amounted to 35 million tonnes deposited in the north and
south waste dumps, and the cost to remove the material would have been prohibitive. Fortunately, the
geochemistry of the waste rock permitted in situ management. At GSL, waste rock volumes total
approximately 23,000 tonnes across the three sites with potentially acid-generating geochemistry at
Blanchet and Copper Pass. The waste rock at these two sites will therefore be managed onsite beneath an
engineered cover.
Waste rock piles – comparisons

Mine projects Option 1 Option 2 Option 3 Option 4
Discovery Waste rock inventory was managed as part of the underground

mining operations (stabilisation) or crushed for airstrip and road
construction and was not included as a remedial component
Do nothing Contour sop Contour top and Contour with

Colomac sides (partial full cover
sloping, field fit
final contour
locations)
Leave as is Deposit in
Outpost
surface
excavations
Great Slave
Lake Leave as is Re-slope Re-slope and
Blanchet
cover
Leave as is Grading Consolidation in
Copper Pass trenches and
capping
It is apparent that obtaining social licence requires significant effort on the part of the proponent, but the
benefits provide both immediate and long-lasting rewards. Project-specific engagement will benefit the
project at all stages, while longer term benefits will carry over to other projects and eventually to the program
level. It is much more cost effective to fund robust engagement programs than to fund reactive mitigations
or address conflict during project execution. From a project delivery perspective, social licence will ensure
higher quality projects through the incorporation of technical and traditional approaches, increased technical
and cultural awareness, establishment of trust networks, facilitation of Elder–youth information exchange
and mitigation of unexpected events and impacts on a site, allowing the project to move ahead more fluidly
with adaptive strategies.

The remediation planning phase of an NCSP project helps build community capacity through information
exchange and decision making. The remediation phase provides the highest degree of socio-economic
benefits through the hiring of local heavy equipment operators, camp staff, site foremen, labourers and
wildlife monitors. The post-closure and monitoring phase provides a less intense but prolonged degree of
community involvement, with community members remaining actively involved in site monitoring activities.
Aboriginal youth may benefit from science camps and decide to pursue a post-secondary education in the
environmental sciences and engineering fields. Science camp graduates may have opportunities to assist with
environmental monitoring on NCSP projects as field assistants and job shadows. Typical fieldwork programs
include archaeological studies, revegetation surveys, water quality, hydrology, geo-technical stability, and
aquatic and terrestrial sampling. Where possible, site visits are coordinated with Elders and local regulatory
boards such that youth benefit from traditional knowledge and regulatory perspectives.
To date, none of the NCSP projects (with the exception of the Giant Mine remediation project) have triggered
an environmental assessment under the Canadian Environmental Assessment Act owing in large part to the
engagement effort during the early stages of the projects and development of robust RAPs. Subsequent
amendment requests and renewals are easier to obtain because of the relationships with the local
communities and regulatory boards, as has been the case with the GSL remediation project.
Through the NCSP, positive relationships and partnerships have been forged with many Aboriginal
governments and communities. CARD’s relationship with the Yellowknives Dene First Nation on the Discovery
Mine project continues today through long-term monitoring involvement and third-party activities at the
site. The partnership formed between AANDC and the Tlicho in the early stages of the Colomac project
remains strong to this day, as evidenced by a letter of support received from the Tlicho for the aquatic-
terrestrial sampling permit AANDC requested in 2013 and through data sharing under the Marian Lake
watershed stewardship program. The GSL remediation project has benefitted from these past relationships
and partnerships and has strong project support within the Akaitcho Dene and Metis communities. It is
anticipated that these relationships will strengthen as the project progresses.
The results obtained to date from the GSL engagement activities indicate a sound understanding of the
project and strong community support. These results suggest that CARD has obtained social licence for the
GSL project; it is now up to the project team to maintain the licence. Retaining social license on the GSL
project will require active participation of the Akaitcho Dene and Metis during the execution of the RAP and
throughout the post-closure and monitoring phase. The project team will adhere to the community
engagement plan throughout the project life cycle and will adopt new strategies if required based on
feedback from the communities. Outreach programs will consist of school presentations and community-
based science camps where possible, and community meetings and site tours will be conducted throughout
the life of the project.
Obtaining an SLO (social licence) is critical to the successful execution of any project. Effective community
engagement builds trust among stakeholders and forms the foundation from which partnerships and working
relationships are built. Engagement strategies must be delivered in a manner appropriate to the communities
and must be adaptable as the project evolves. Social license will produce high-quality projects through the
incorporation of traditional and technical approaches and will help streamline overall project processes. It is
not enough to simply gain social license; it must be maintained throughout all project stages to realise
maximum benefits. Through AANDC’s NCSP, effective community engagement has provided the social licence
required to continue to deliver the program in Canada’s North.
The authors acknowledge the guidance and contributions of current and past AANDC and Public Works and
Government Services Canada staff: Carole Mills, Mike Nahir, Emma Pike, Andrew Richardson, Melissa
Tokarek, Jane Amphlett, Jessica Wilson, Myranda Bolstad, Lisa Dyer, Octavio Melo, James Edwards, Rebecca
Vanderspeigel, Bruce Friesen, Bill Coedy, Dawn Curtis, Jennifer Whittaker, Christina Gray, Giselle Cotta, Janice
Lee and Jessie Hoyt.
Thanks to all of our Aboriginal partners – the Tlicho Government, Yellowknives Dene First Nation
(communities of N'Dilo and Dettah), Lutsel K’e Dene First Nation, Deninu Kue First Nation and Fort Resolution
Métis Council – as well as our Aboriginal contractors and suppliers: Tlicho Logistics, Aboriginal Engineering
Ltd., Tlicho Engineering and Environmental Services and Air Tindi Ltd. Thanks to our numerous technical
resources over the years, including AECOM, BGC, Dillon, Flat River Consulting, Franz, Holubec Consulting,
SENES, SRK and Stantec, to name but a few.
And, finally, a special thanks to all of the Aboriginal Elders for their contribution to the planning, remediation
and monitoring of the aforementioned projects:
Tlicho Region – Eddie Camille, Robert Mackenzie (deceased), Francis Williah, Harry Apples, Harry Mantla
(deceased), Mary Adele Rabesca, Joe Champlain, Isadore Charlo, Mary Charlo, Pierre Beaverho, Francis
Simpson, Jimmy Nitsiza, Harry Simpson (deceased), Joe Black, Fred Mantla, Alphonse Apples, Joseph Pea’a
(deceased), Jimmy Kodzin, Joseph Judas, Joe Rabesca, Eddie Paul Rabesca (deceased), Eddie Erasmus, Alexis
Arrowmaker (deceased), Joe Migwi (deceased), John Nitsiza (deceased) and Francis Blackduck (deceased).
Akaitcho/Drygeese Region – Alfred Baillargeon, Angus Martin, Peter Sangris, Jonas Sangris, Isadore Tsetta,
Ernest Boucher, August Enzoe, Arthur Beck, Pete King, Paul Smith, Stanley Beck and Robert Beaulieu.
AANDC (2007) Revised design drawing for waste rock berm placement at Colomac, Contaminants and Remediation Directorate,
viewed http://mvlwb.com/.
Bourke, P. (2013) GCCR 802: Community development for the mining industry, Community relations for the extractive industries
graduate certificate, Queen’s University, Kingston, Canada, viewed February 2015, http://www.queensu.ca/sgs/community-
relations-extractive-industries.
Bourke, P. (2014) GCCR 803: Community engagement and mining, Community relations for the extractive industries graduate
certificate, Queen’s University, Kingston, Canada, viewed February 2015, http://www.queensu.ca/sgs/community-relations-
extractive-industries.
Breadmore, R., Gowman, J. and Lafferty, G. (2013). Great Slave Lake project: Community engagement and capacity building strategy,
in Proceedings 2013 Geoscience Forum, 20 November 2013, Yellowknife, Canada.
Calette, M., (2014) Community engagement: The process can be very rewarding – if you want to do what it takes, in Proceedings Real
Property Institute of Canada Federal Contaminated Sites Conference, April 14-16, 2014, Ottawa, Canada.
Gunningham, N., Kagan R. and Thornton, D. (2004) Social License and environmental protection: Why businesses go beyond
compliance, Law & Social Inquiry, Vol. 29(2), pp. 307–341. February 2015.
Parks Canada (2008) An approach to Aboriginal cultural landscapes – Aboriginal world views – traditional knowledge, viewed February
2015, http://www. pc.gc.ca/docs/r/pca-acl/sec1/sec1b_e.asp.
Silke, R. (2009) The operational history of mines in the Northwest Territories, Canada, Ryan Silke, Yellowknife, Canada, “Colomac”,
pp. 103–190.
SRK (2004) SRK Consulting, Water licence MV2000L2-0018 Colomac Site remediation plan final report, Contaminated Sites Office,
March 2004, drawing by S. Schultz. http://mvlwb.com/
Yates, H. and Horvath, C.L. (2013) Social license to operate: How to get it, and how to keep it, 2013 Pacific Energy Summit working
papers, viewed 9 February 2015, http://www.nbr.org/downloads/pdfs/eta/PES_2013_summitpaper_Yates_Horvath.pdf.

C.M. Jeffrey Amec Foster Wheeler, Canada

M.L. Unger Teck Resources Limited, Canada
J.C. Pugh Amec Foster Wheeler, Canada
The Sä Dena Hes mine, which produced lead and zinc concentrates, is located approximately 70 km north of
the Town of Watson Lake, Yukon Territory, Canada and lies within the traditional territory of the Liard First
Nation. The decision to enter the site into permanent closure was made in 2013. Under the mine’s licenses,
reclamation must be substantially completed by 31 December 2015. Teck Resources Limited retained Amec
Foster Wheeler Environment & Infrastructure in 2013 to carry out the implementation of Teck’s closure plan.
The involvement of First Nations personnel was one of the main project priorities.
During the short 2013 construction season, First Nations personnel worked over 60% of all on-site man-hours.
The bulk of construction activities needed to be completed in 2014 to prepare for final reclamation activities
to be carried out in 2015 during the final construction period. The strategy for completing the aggressive 2014
project phase within a limited timeframe included engaging larger general contractors that could handle the
large scopes of work with stipulations during the tendering process of required First Nations engagement as
part of the contractor selection process. Some of the site works, including items such as road maintenance
and pipeline removal, were specifically left out of tender packages and awarded directly to local First Nations
contractors to further promote their engagement in the project. Other First Nations engagement included
hiring environmental monitors and first-aid attendants. The 2014 season saw more than 50% of all Teck-
contracted man-hours on-site worked by First Nations personnel. Overall, First Nations personnel have
worked 53.2% of all man-hours (26,754 out of 50,242 total man-hours) on Teck-contracted project aspects.
The Sä Dena Hes (SDH) Mine, located approximately 70 km north of the Town of Watson Lake in Canada’s
Yukon Territory, is a lead/zinc mine that operated between August 1991 and December 1992. The detailed
decommissioning and reclamation plan (DDRP, hereafter referred to as the closure plan) for the SDH mine
site was developed and submitted to the Yukon Ministry of Mineral Resources and Water Board for approval.
The closure plan addresses the socio-economic expectations of maximising the use of local resources to
complete the project and promote the region’s economy (Teck, 2013). The decision was made to enter the
SDH mine into permanent closure in January 2012, and the final closure plan was submitted and approved in
July 2013. The project needs to be completed by 31 December 2015, as the existing Type “A” water license
QZ99-045 and quartz mining production license QML-0004 expire at that time.
The SDH mine decommissioning and reclamation project is the first major reclamation project of its kind to
be carried out in the Yukon Territory by a private mining company (rather than completed by the territorial
government). The owner, Teck Resources Limited (Teck), retained Amec Foster Wheeler Environment &
Infrastructure (Amec Foster Wheeler) to implement the closure plan for the site within the license expiry
timeframes. The site lies within the traditional territory of the Liard First Nation, making engagement of local
First Nations personnel a project priority and a key component of its successful execution.
With the project approval not granted until summer 2013, only minor works could be carried out during that
construction season. Since the 2015 season is reserved primarily for final reclamation activities such as
revegetation and site access road decommissioning efforts, the bulk of decommissioning and reclamation

First Nations engagement in mine closure: Sä Dena Hes mine decommissioning and reclamation C.M. Jeffrey, M.L. Unger and J.C. Pugh
activities needed to be carried out in 2014 to ensure the project was completed on schedule. In order to
accomplish the work within the established timeframe while maximising the use of local resources, a strategic
approach was developed: general contractors would bid on the project with the understanding that the
inclusion of First Nations personnel would be a key component of the selection process. Certain site works
would purposely be left out of major works contracts to allow First Nations the chance to complete these
works.
To date, the implementation approach has proven effective in completing the scope of work outlined in the
closure plan. During this time, engaging local First Nations resources has remained a priority. The final stages
of the SDH mine decommissioning and reclamation project will be executed in 2015. A similar
implementation approach will continue to be used during the final reclamation activities to ensure the
project objective of engaging First Nations personnel is maintained.
The SDH mine is currently owned by a joint venture between Teck (50% ownership) and Pan Pacific Metal
Mining Corp. (a wholly owned subsidiary of Korea Zinc, 50% ownership). Teck is the operator under the joint
venture agreement. Formerly a zinc/lead mine, most of the mineral occurrences range in elevation from 1200
m to 1500 m above sea level. Prior to decommissioning and reclamation activities, on site features included
mill infrastructure, tailings management facilities, mine workings including open pits and underground
portals, waste rock dumps, electrical infrastructure, and tailings and water supply pipelines.
Exploration activities at the SDH mine began as early as 1962 and continued until construction of the mine
site began in 1991. Between 1979 and 1982, diamond-drilling programs were carried out and estimates
outlined 250,000 tonnes of zinc and lead mineralisation. During the mid-to-late 1980s, further geological and
geochemical prospecting was conducted, and estimates suggested a zinc-lead-silver mineral inventory of

more than five million tonnes scattered among many areas. Further drilling was carried out in 1989, and this
program refined the assessment of mineable reserves to 3.9 million tonnes at 11.5% zinc, 3.8% lead and 53
grams/tonne silver. Curragh Resources constructed the mine in 1991.
During the 16 months of production (August 1991 to December 1992), approximately 700,000 tonnes of ore
were mined and processed. Production included approximately 120,000 tonnes of zinc concentrates with a
grade of 59% zinc and 54,000 tonnes of lead concentrates at a grade of 77% lead. A sharp downturn in metal
prices forced the mine to shut down in December 1992, when the site was put on a care and maintenance
program. The current owners purchased the property in 1994.
In 2000, the first closure plan for the site was developed and subsequently updated as required under the
mine’s licenses. The closure plan outlines the closure measures and objectives to be carried out for the site.
The site remained in care and maintenance until 2013 subsequent to the approval of the closure plan. A full-
time caretaker resided at the mine site to oversee the care and maintenance until 2013, when
implementation of the decommissioning and reclamation activities began. The following is an excerpt from
the closure plan detailing pertinent activities from the care and maintenance period:
Teck held a Water License for the site, allowing the regulated volumes of water to be discharged
from within the site’s Tailings Management Area to the environment. Teck also holds a Quartz
Mine Production License for the site. The Company currently holds a Type A Water License QZ99-
045, and Quartz Mine Production License (QML-0004) for quartz mining, both of which were
issued in 2003 and expire on December 31, 2015. In 2005 and again in 2010, extensions to the
Temporary Closure clauses were obtained for both Licences with the 2010 extension specifying
that Temporary Closure ends on January 27, 2013. As of January 28, 2013, the site is deemed to be
in “Permanent Closure” which requires implementation of the site closure plan. (Teck, 2013)
Throughout the care and maintenance period, Teck was committed to involving First Nations and local
community members in the site’s activities. In the closure plan for the site, Teck also committed to the
continued involvement and engagement of First Nations and local community members throughout the
reclamation activities at the site. Regarding First Nations and local community engagement on the site
throughout the care and maintenance period and into closure:
When the Sӓ Dena Hes Operating Corporation purchased the Sӓ Dena Hes Mine in 1994, the
venture worked with Liard First Nation and formulated a Socio Economic Agreement that was to
guide their joint relationship during the operation of the mine. Unfortunately, the mine did not
reopen and remained dormant. Upon the decision to permanently close the Sӓ Dena Hes Mine,
Teck sought to update its relationship with the Liard First Nation. The intent of the relationship
between the two parties is for both parties to obtain value from their relationship as well as the
optimization of the economic opportunities within the Nation and locally in the Watson Lake area.
(Teck, 2013)
The decision to enter the site into permanent closure was made in 2012. Under the mine’s licenses, closure
must be completed by 31 December 2015: “Upon review and acceptance of this update to the DDRP, the
plans contained will be implemented beginning in the summer of 2013 and will be substantially complete by
December 31, 2015” (Teck, 2013).
The approved closure plan schedule included three construction seasons; however, internal and external
project approvals were only received in summer 2013, which required the schedule to be compressed.
Following the decision to enter the site into permanent closure, Teck’s legacy properties division hired Amec
Foster Wheeler to perform project management, procurement and construction management services, and
to provide health, safety and environmental oversight associated with the implementation of Teck’s
approved closure plan for the site. Site services Teck requested also included emergency medical
transportation (EMT) services complying with regulatory requirements associated with working in isolated
workplaces, as well as surveying services that would be carried out on-site. To fulfil the requirements, Amec

Foster Wheeler employed two primary sub-contractors. Polar Medical Services was contracted to carry out
the remote first-aid services at the mine site, including establishing and maintaining a first-aid room,
providing an EMT vehicle and providing first-aid attendants. The first-aid attendants also assisted with site
security and served to document and control personnel accessing the site on a daily basis over the course of
the project. Amec Foster Wheeler also partnered with Yukon Engineering Services (YES) in a sub-contractor
role. YES, based out of Whitehorse, provided on-site surveying and inspection services over the course of the
project. The addition of YES to the project team was also beneficial because of the firm’s firsthand local
knowledge of the area, including the climate and socio-economic environment, and familiarity with the local
general contractors that would be participating in the delivery of the services during the execution of work
at the site.
In implementing the project, Teck made it clear to Amec Foster Wheeler that engagement of First Nations
personnel was a top project priority. This message aligns with commitments made in the approved closure
plan:
Teck’s approach to working with the Liard First Nation in 2013 and into the future will be based
upon the solid working relationship established and maintained between Teck and the LFN to
date.
As part of the ongoing consultation with LFN, including the ecological and human health risk
assessment processes, traditional land uses and values will be incorporated into the end land use
activities considered. (Teck, 2013)
The general strategic model for the project was developed in 2013, and local First Nations contractors were
hired to carry out minor reclamation works prior to the majority of construction activities taking place in
2014. In 2014, Teck and Amec Foster Wheeler worked together to develop an implementation methodology
and strategy that addressed the need to hire local community members to contribute to the successful
execution of the project. The 2014 project phase was carried out on multiple concurrent fronts. A large on-
site management team was provided to oversee the implementation of the closure plan. Owing to scheduling
considerations and the large scope of many of the objectives, general contractors with the capacity to carry
out the work needed to be hired. Inclusion of First Nations personnel in the project was a major consideration
during contractor selection and the tendering process. Certain site works were purposely left out of major
works contracts to ensure small local contractors were engaged in the project and had the opportunity to
complete these works.
For the purposes of this project, First Nations engagement is defined as follows:
𝐹𝑖𝑟𝑠𝑡 𝑁𝑎𝑡𝑖𝑜𝑛 𝑤𝑜𝑟𝑘𝑓𝑜𝑟𝑐𝑒 ℎ𝑜𝑢𝑟𝑠

= % 𝐹𝑖𝑟𝑠𝑡 𝑁𝑎𝑡𝑖𝑜𝑛𝑠 𝑒𝑛𝑔𝑎𝑔𝑒𝑚𝑒𝑛𝑡
𝑇𝑜𝑡𝑎𝑙 𝑤𝑜𝑟𝑘𝑓𝑜𝑟𝑐𝑒 ℎ𝑜𝑢𝑟𝑠
The level of engagement is presented as a percentage using this calculation.
The SDH mine closure plan, dated March 2013, details the closure plan for the mine and associated
infrastructure. The key reclamation activities associated with the project are as follows:
 demolish and dispose of site infrastructure;
 seal underground mine workings;
 re-slope waste rock dumps;
 dewater and remove two dams;
 decommission, cap and reclaim the tailings management facility;

 cap and re-vegetate mine facilities;

 final closure works related to the reclamation and closure of the mine site.
As presented in Section 2, the original closure plan implementation schedule was for three seasons, but since
the final submission/approval was delayed, the schedule was compressed. A short construction window was
available in 2013, but the majority of the work needed to be carried out in 2014 and 2015.
In 2013, following receipt of the necessary approvals, a limited demolition and water pumping program was
carried out near the end of the construction season (12 September to 10 October). Key work activities were
carried out that would prepare the site for the majority of the work to take place in 2014. Work included
demolition of the mill camp facilities and office complex as well as associated on-site landfill operations.
Pumping of water within the tailings management area was also performed during this time. The pumping
operations carried out in 2013 were critical to the success of the 2014 construction season, as they reduced
the number of days of pumping required in 2014. Discharge of water from the tailings facility is only
permitted for 90 cumulative days on an annual basis and is limited by a maximum hourly discharge volume;
hence the importance of doing as much pumping as possible in 2013.
The 2013 work phase was executed by awarding work activities directly to local First Nations contractors.
This was done following a pre-project meeting in the town of Watson Lake that involved the Teck project
manager, the Amec Foster Wheeler project manager, the Amec Foster Wheeler construction manager and
the owners of two local First Nations contractors. Teck/Amec Foster Wheeler (hereafter referred to as the
construction management team) had prepared a list of work activities that needed to be accomplished in
2013, and these two contractors were invited to the meeting with the construction management team to
discuss the work and their availability to carry out the work. The work primarily involved heavy equipment
with some labour to support the effort for pumping operations and pipeline removal. The construction
management team presented the work activities to the two contractors and requested a collaborative effort
between them to accomplish the work. Following the meeting, both contractors provided a rate schedule for
their equipment. The work was carried out on a time and materials basis using these rates, and the
construction management team used every effort to use equipment from both contractors to execute the
work. Additionally, Amec Foster Wheeler, through sub-contractor Polar Medical, was able to engage
members of the Liard First Nation to carry out first-aid attendant/site security services for the duration of
the 2013 construction program.
Additionally, Teck finalised a formal communication and engagement plan in 2013 that included updating
the community with meetings, funding traditional knowledge studies and forming a project working group.
The traditional knowledge studies involved providing funding to the community to gather knowledge from
the elders and to identify trails, traditional uses of the area, including those involving plants and animals, and
potential end land uses for the mine site. The information gathered was also important for completing the
human health and ecological risk assessment studies undertaken to guide the overall remediation of the mine
site.
Owing to the requirement to implement the closure plan activities by the end of 2015, the work schedule
was compressed and included critical path tasks that needed to be completed in 2014. The activities
scheduled for the 2015 construction season are primarily revegetation and road removal efforts across all

site locations. This meant that construction works identified for 2014 had to be completed in the limited six-
month construction season. This would ensure that reclamation was sufficiently advanced to permit final
tasks to be executed in 2015 prior to the expiration of the site licenses.
The 2014 component of the decommissioning and reclamation project involved a large scope of work to be
accomplished. Reclamation activities in 2014 included:
 dismantling and salvage operations, including demolition of the mill and associated infrastructure;
 decommissioning of tailings management area including removal of dam structures, development
of a rock quarry for production of rip rap, construction of drainage channels and capping tailings
areas;
 permanently sealing mine workings, infilling open pits and reshaping waste rock dumps;
 decommissioning and removal of all electrical infrastructure;
 decommissioning of all on-site tanks not associated with the mill dismantling and salvage
operations;
 other site activities including, but not limited to, capping the mill site once the infrastructure had
been removed, road maintenance, removal of pipeline, demolition of shacks, landfill operations,
general site clean-up and construction of helipads.
The 2014 construction program included several critical path challenges. Water accumulated within the
tailings management area needed to be discharged within a 90 cumulative day timeframe, and pumping
could only be carried out to a specified maximum hourly discharge. If pumping was not completed within this
timeframe, the remainder of the works could not be completed. To ensure the pumping was completed,
penalties for loss in pumping production were built into the contract documents such that the contractor
was required to maintain a minimum hourly discharge and would be monetarily penalised for lost time
associated with pumping. Additionally, the contractor was to provide a full set of backup equipment. The
isolation of the site meant breakdowns would result in significant overall delays to the construction schedule,
which could not be tolerated owing to the overall critical path nature of the 2014 construction season. Portal
closures and construction of reinforced concrete caps over vertical vent shafts needed to be completed in an
extremely tight construction window. This work needed to be completed in 2014 to ensure revegetation and
road decommissioning could be carried out in 2015. Substantial completion dates with associated penalties
for delays were incorporated into the contracts to ensure the contractors adhered to the project schedule.
Other schedule challenges included completing the removal of the mill and associated infrastructure by mid-
season to allow concrete breaking and reshaping efforts to be carried out. Additionally, all exposed tailings
needed to be covered with clean dam fill material before winter to ensure vegetation efforts could be
undertaken early in 2015.
One of the primary activities in 2014 was the dismantling and salvage operation of the mill and associated
infrastructure. Teck was able to avoid demolition and landfilling of the facilities by selling the assets. The
nature of the sales contract meant the mill purchaser was the construction manager in charge of collecting
assets and removing them from the work site. The mill purchaser was only on-site to collect assets and did
so under its own management and direction.
The construction management team developed a work breakdown strategy for completing the Teck-
controlled aspects of the 2014 phase of the project within the limited timeframe while attempting to
maximise the local First Nations workforce. The strategy involved engaging general contractors who were
capable of handling large scopes of work and using a contractor selection process that took into account the
extent of involvement and inclusion of First Nations labour. To ensure maximum First Nations engagement
was achieved, tender documents indicated that favourable consideration would be given to tenders
containing “meaningful” First Nations content. It was indicated that bidders were encouraged to provide

employment opportunities for local First Nations equipment and resources. In their bid submissions,
contractors were required to supply plans demonstrating their efforts to fill positions with First Nations
personnel, indicating the types and quantities of such resources they would use and explaining how they
intended to use these resources in the execution of the work, and to provide documentation supporting
these commitments.
In the tender documents, it was stated that letters of recommendation from local First Nations leadership
would be given additional consideration. It was indicated that, upon award, the contractor would be required
to provide confirmation of the use of the resources outlined in the bid by providing the names of personnel
and the types of equipment required. It was stated that if the contractor were unsuccessful in acquiring First
Nations engagement prior to the bid submission date, the contractor would, at a minimum, explain its efforts
to do so and provide supporting phone logs or communication records. It was indicated that the contractor
would only be relieved of the commitments to First Nations engagement if the resources declined to
participate post-award or were unable to execute the work required as determined by the construction
management team.
During the analysis of all bid submissions for these large contracts, careful consideration was given to the
commitment to First Nations engagement. Each contractor received a score for its proposed level of
commitment, which contributed to the overall scoring of each contractor’s bid.
Another key component to successfully engaging First Nations personnel was the decision to leave some of
the site works out of the scope of the larger contracts. These works were awarded directly to local First
Nations contractors. The on-site work activities set aside for direct award included:
 road maintenance;
 concrete breaking at mill site;
 mill site capping and shaping;
 removal of pipeline;
 landfill maintenance;
 decommissioning of monitoring wells;
 installation of erosion protection materials;
 construction of helipads for future monitoring;
 demolition of site exploration camp infrastructure and other small shacks;
 reclamation of a dyke;
 general site clean-up.
These work activities were specifically chosen as items that could easily be left out of the larger contracts.
These items were awarded directly to the same First Nations contractors who conducted work in 2013, as
discussed in Section 3.1. These work items were chosen for several strategic reasons. The road maintenance
activity was an ideal item for the local First Nations personnel to take care of because site access roads were
shared by several of the general contractors. To avoid disputes between contractors executing the larger
scopes of work regarding who should be maintaining the road (grading, watering, etc.), this task was given
to the First Nations contractors. It was indicated in the various tender documents that this task would be
taken care of by “others” (in reference to First Nations personnel). In the event that substantial road
maintenance was required in certain areas or that all First Nations contractors were busy on other tasks, a
piece of equipment could be used by one of the larger general contractors, and time spent on this activity
was simply billed by the hour. However, First Nations personnel generally performed the road maintenance.
Another strategic example of work provided to First Nations contractors was landfill maintenance. All
permitted waste was hauled to the on-site landfill for deposition. Most of the waste came from demolition

of small shacks and structures, removal of damaged pipeline, debris from historical mining operations and so
on. This material was taken to the landfill, where the First Nations contractors carried out landfill operations.
In the event that general contractors encountered debris while executing contracted work, they could notify
management personnel, and the First Nations contractors on-site were equipped to handle the situation.
Another project aspect awarded to First Nations contractors was the breaking of concrete at the mill site.
Once the mill infrastructure demolition materials were removed from the site to be salvaged or hauled to
the landfill for deposition, First Nations contractors were also responsible for capping the remaining
foundations and shaping the cap. The First Nations contractors rented a hydraulic hammer to carry out
breaking of concrete foundations and walls at the mill site. First Nations labour forces were also provided
appropriate respirator fit testing and training to allow them to enter the work area during demolition of
concrete and to suppress dust using a water truck while fully clothed in appropriate personal protective
equipment (PPE). Once the concrete breaking was complete, the remaining foundations required a minimum
1 m of capping material to be placed and shaped overtop of them. This work kept the First Nations heavy
equipment busy for a period of more than a month.
First Nations labour crews were kept busy with miscellaneous site tasks including assisting in general site
clean-up and pipeline removal. Because of the heavy traffic and demolition activities on-site, a small labour
crew was tasked with driving the access roads daily and collecting debris that had fallen during hauling of
material. This helped to keep the site clean and clear of debris as reclamation activities progressed. This also
prevented stray material from causing any danger on roadways.
Teck also hired Liard First Nation personnel to act as environmental monitors. The monitors received training
from consultants, which enabled them to monitor site activities from an environmental perspective as well
as perform water quality sampling for comparison with regulatory criteria. The monitors also assisted
consultants in carrying out soil investigations and other aspects of overall site monitoring where required.
The monitors reported directly to Teck and kept the community apprised of site activities through their
engagement in the project. Additionally, over the course of the 2014 construction phase, members of the
Liard First Nation provided most of the first-aid attendant/site security services.
First Nations personnel worked more than 25,000 man-hours during the 2014 season. Throughout the
season, the First Nations contractors gained valuable health and safety training and awareness through
various programs conducted on site by Amec Foster Wheeler health and safety monitors and the rest of the
construction management team. The team worked directly with the First Nations contractors to help
promote a safe and healthy work environment and to assist them in becoming more knowledgeable about
general safety, hazards and mitigation measures. Highlights from the season included conducting a respirator
fit testing and training session with the First Nations contractors and attendance by at least one member of
the construction management team during all First Nations contractors’ daily toolbox meetings to provide
further insight and discussion topics related to safety.
The 2015 construction season will involve, at a minimum, the following activities:
 road decommissioning;
 re-vegetation;
 regrading and capping of disturbed site areas;
 other final reclamation efforts at various site locations.

The 2015 construction program will be tailored so that First Nations contractors are again awarded much of
the work directly. The scope of work planned for 2015 contains fewer critical path milestones and schedule
constraints. Therefore, as much work as possible will be awarded to local First Nations workforces. It is
projected that upwards of 75% of man-hours in 2015 will be worked by First Nations personnel.
Over the course of the reclamation project, First Nations engagement has been tracked by means of man-
hours worked on-site. As discussed in Section 3.2, Teck was not able to control the level of First Nations
engagement during the mill dismantling and salvage portion of the work executed in 2014.
However, for all other work phases, significant effort was made to achieve the objective of employing as
many First Nations personnel as possible. Table 1 breaks down the level of First Nations engagement by major
work phase.
Project aspect First Nations engagement (%)

All 2013 work 62.9
Mill infrastructure demolition and salvage 12.8
(2014) — not Teck controlled
Tailings management area decommissioning 56.9
(2014)
Mountain works and tank decommissioning 30.6
(2014)
Smaller site work activities and other (2014) 83.3
Overall project to date 37.6
Overall project to date (excluding mill 53.3
infrastructure demolition and salvage)
As shown in Table 1, the overall project-to-date percentage of First Nations engagement is 37.6%. However,
the lowest contributor to First Nations engagement was the mill infrastructure demolition and salvage work,
which saw only 12.8% engagement. This aspect of the project also contributed the largest number of man-
hours out of all tasks listed in Table 1. To put this into perspective, thus far, 81,874 man-hours have been
worked on the project, with 31,632 (38.6%) coming from the mill dismantling and salvage operations portion
of the work. Because Teck was able to sell the mill infrastructure to allow it to be salvaged rather than
demolished, limited requirements for First Nations engagement could be stipulated in this contract. The mill
purchaser managed the dismantling and salvage of its purchased assets.
However, for the Teck-controlled project aspects, which account for the other 50,242 man-hours on the
project, First Nations engagement of 53.3% (26,754 man-hours) was achieved. Breaking down the data even
further, during the limited 2013 construction season, only 2,548 man-hours were worked on-site. Of these,
First Nations personnel worked 1,602 man-hours, for a percentage of 62.9%. In 2014, Teck-contracted work
involved 47,694 man-hours, with 25,151 worked by First Nations personnel, for a percentage of 52.7%. In
seeking to achieve one of its main objectives, Teck was able to achieve more than 50% First Nations
engagement in these aspects of the project.
It is anticipated that the 2015 season will also see a very strong percentage of First Nations engagement, as
the smaller scope of work planned will allow most of the work to be directly awarded to local contractors. It

is forecasted that by the end of the project (fall 2015), Teck will have achieved First Nations engagement on
the overall project of nearly 60%.
In achieving more than 50% engagement in the project aspects Teck could control, this closure plan objective
was achieved. The critical 2014 phase of the reclamation project was also completed on schedule, and the
site is now ready for final reclamation works to be carried out in 2015 prior to the expiration of the site’s
licenses.
Reflecting on the implementation of the project to date, setting aside specific tasks for First Nations
contractors to carry out was a positive project aspect. This provided a level of certainty for these contractors
throughout the construction seasons that they would have work and be needed on a daily basis. The
accelerated construction schedule resulted in the need to hire large general contractors to ensure work was
completed within the specified timeframes. Comparing the Teck-controlled work with the non-Teck-
controlled work (i.e. mill dismantling and salvage) shows the significance of establishing First Nations
engagement requirements within tender documents for large contracts. When these requirements were
implemented, Teck was successful in providing further employment opportunities to First Nations personnel
beyond the work items specifically set aside for them. Teck held contractor interviews during the selection
process evaluated contractors’ commitment to First Nations involvement. This helped to ensure positions for
First Nations personnel throughout the course of the project.
An employment level of more than 50% First Nations personnel on Teck-controlled aspects of the project
was achieved and was a contributing factor to the successful implementation of the aggressive 2014
construction schedule.
The SDH Mine was dormant for approximately 20 years before the decision to enter the site into permanent
closure was made. The site lies within the traditional territory of the Liard First Nation. The SDH Operating
Corporation has, throughout the existence of the mine site, worked collaboratively with the Liard First
Nation. When the decision to enter the site into permanent closure was made, one of Teck’s primary
objectives for implementing the approved closure plan was to engage First Nations personnel and local
community members throughout the course of the project while being mindful of the critical path nature of
the schedule.
Although the project will not be completed until fall 2015, the majority of work activities were completed by
the end of 2014. To date, Teck has achieved more than 50% engagement of First Nations personnel on the
project. To ensure completion within the schedule, a work breakdown structure was developed to tender
the large work items to general contractors. However, the requirement for general contractors to employ
First Nations personnel was set out in the tender packages, and heavy consideration was given to bidders
that committed to a high level of such employment.
Another key approach to ensuring First Nations personnel were engaged in the project was to set aside many
of the project work items (road maintenance, site clean-up, pipeline removal, etc.) that would be ideal for
smaller contractors. First Nations contractors were awarded this work directly. This decision ensured that
First Nations workforces would be involved throughout the duration of the project. This concept will continue
throughout the remainder of the project in 2015.
In conclusion, efforts to complete the project within the scheduled timeframe, budget and limited Yukon
construction windows, while maximising the engagement of First Nations personnel, have been successful to
date. In 2015, with primarily smaller works planned, Teck will be seeking mostly First Nations efforts to see
the project through to completion. By the end of the project, it is anticipated that First Nations personnel will
have worked approximately 60% of all man-hours on-site.

Amec Foster Wheeler and Teck would like to acknowledge Iyon Kechika Contracting and James Magun
Contracting, the local First Nations contractors that worked on-site to conduct all work in 2013 and that were
on-site in 2014 to carry out the work items set aside from the tendered work. Amec Foster Wheeler and Teck
would also like to acknowledge both of Amec Foster Wheeler’s sub-contractors used during the execution of
the project: YES and Polar Medical Services. YES, based out of Whitehorse, Yukon, provided on-site inspection
services and surveying services for the duration of the project. Polar Medical Services provided first-aid
attendant personnel and an EMT vehicle, and established and maintained a first-aid room on-site for the
duration of the reclamation efforts to-date. Both suppliers will continue in these roles through 2015.
Teck Resources Limited (2013) Sӓ Dena Hes Operating Corporation, Sӓ Dena Hes Mine detailed decommissioning & reclamation plan
March 2013 update, pp. 10–25, viewed 16 April 2015, http://emr.gov.yk.ca/mining/pdf/mml_sa_ddrp_march_2013_
update_july_15_version.pdf.


Diamonds were discovered in the Marange diamond fields (MDF) in June 2006, leading over 30,000 fortune
seekers from different parts of the world to flock there. In November 2006, the Zimbabwe government
launched “Operation No Illegal Mining,” which led to the violent displacement of artisanal and small-scale
miners (ASM) and the takeover of the fields by government forces. Corruption ensued, and the number of
illegal ASMs increased to over 36,000 by 2008, prompting the government to launch another operation, coded
“Operation You Will Not Return.” In July 2009, official mining licenses were issued to three large-scale mining
(LSM) companies to mine in the MDF. Rampant corruption and patronage characterised the license issue such
that only those inclined to the ruling party got licenses. As observed by several researchers, the owners of the
licensed companies are mainly from the Zimbabwe Central Intelligence Organisation, the National Army, the
Air Force, the National Police, and the ruling party’s top brass. The companies do not contribute significantly
to the treasury.
Following an assessment of how events have transpired since the discovery of the diamonds, this paper
identifies several constraints to deriving maximum value from the exploitation of the Marange diamonds for
the benefit of all Zimbabwean citizens, especially the distressed communities within the Marange mining
district. This paper explains how a political environment characterised by corruption and patronage leads to
complications in proper mine closure planning and severe damage to the environment. Compelling
achievements that are necessary for proper closure planning, envisaged for mining operations in an enabling,
free, and fair political environment, are not achievable when mining is done in the opposite political
environment. These include redistribution of proceeds from mineral wealth to ensure the establishment of
health facilities, including efficient hospitals, clinics, and sewer systems; promotion of education and
investment; provision of reliable water supplies, energy, and a sound transportation infrastructure; creation
of sports facilities to facilitate growth of sportsmanship; and extension of employment opportunities at all
levels in LSM mining operations to communities in which they operate. This lack of redistribution of proceeds
from mining presents several issues that negatively impact mining communities after mine closure. This paper
details constraints on maximum value realisation and effective closure plan implementation, including
political instability, corruption, patronage and rent seeking, bad governance and institutions, unstable
policies, lack of strategic planning, and unfavourable relationships between large-scale and small-scale
miners. The author proposes several strategies to overcome the constraints for the benefit of all. The
strategies include using Zimbabwe’s comparative advantages: the presence of the MDF, the biggest find of
alluvial diamonds in the world covering an area of 60,000 hectares and with a value of over US$800 billion;
sound infrastructure; and one of the best educations systems in the world, capable of ground-breaking
research and development. Other strategies explained in detail in the paper include improving political
stability; rooting out corruption, patronage, and rent seeking behaviour; implementing favourable mining
policies, good governance, and institutions; considering society and the environment in mining decisions; and
making use of linkages and clusters that form around the diamond value chain, to mention just a few.
The ultimate goal for any mining company should be to leave a net positive benefit to the surrounding
communities and the country in which it operates when mining is completed. This achievement can be elusive
if unfair politics come into play, as discussed in this paper. Communities bear the brunt of mining; as such,
mining has to be conducted considering the best interests of communities. Strong partnerships have to

Making the most out of Zimbabwe’s Marange diamonds: leaving a lasting positive legacy for distressed communities T. Zvarivadza
develop between communities and large-scale miners (LSM) to foster an environment in which minerals are
exploited for the benefit of all. This cooperation leads to the implementation of community empowerment
projects, which can sustain communities well beyond mine closure. In return, the LSMs gain from a conducive
working environment, where possible conflict with the communities is minimised. LSMs fulfil their corporate
social responsibility, which leads the companies’ share value on the stock exchange to increase. This win-win
situation becomes unattainable when LSMs resort to violence and plunge communities into distress.
This paper looks at the turn of events in Zimbabwe’s Marange diamond fields that led the discovery of the
diamonds to become a source of sorrow for the communities, as tycoons resorted to violence and plundered
the diamonds. Several researchers cited in this paper note that, because it is the custodians of the law who
are involved in unsustainable exploitation of the diamond resources, appropriate closure planning is
complicated, as the environment is severely impacted under the direct watch of the supposed law
enforcement agents. Furthermore, the lawless plunder of resources leaves communities severely impacted
after mine closure, as the mining companies do not make significant investments in communities’ survival
after mine closure. Significant proceeds from diamond mining in the MDF hardly make it to the treasury for
investment in community development due to the corruption and patronage tendencies of the mining
companies. Lack of redistribution of proceeds presents several issues that impact mining communities after
mine closure.
This paper argues that many compelling achievements, necessary for proper closure planning and envisaged
for mining operations in an enabling, free, and fair political environment, are not achievable when mining is
done in the opposite political environment. These unattained achievements include redistribution of
proceeds from mineral wealth to ensure the establishment of health facilities, including efficient hospitals,
clinics, and sewer systems; promotion of education and investment; provision of reliable water supplies,
energy, and a sound transportation infrastructure; creation of sports facilities to facilitate growth of
sportsmanship; and extension of employment opportunities at all levels in LSM mining operations to
communities in which they operate, among many others. Not realising these achievements means that
communities are left poorer at mine closure, since responsible mine closure is dealt a severe blow. This calls
for careful closure planning regulations that demand independence for the mining industry regulators and a
reduction of political influence.
The recognition that mineral resources are nonrenewable, considering that billions of years are needed to
regenerate them, leads to a realisation that effective resource exploitation must derive the maximum
possible value for investment in community development and survival well after mine closure. Research by
Forum for the Future (2015) explains that, to ensure sustainability in minerals development, natural capital
(mineral resources) should be appropriately converted to other capitals (social, human, financial, and
manufactured) through investment in different sectors of the economy, leading to economic diversification.
This ensures that, even when the mines have been closed, the mining proceeds have been used to develop
other sectors of the economy, which can sustain life well after mine closure. Such a scenario is challenging
to achieve when political tycoons encourage the plunder of resources without investing in economic
diversification. A nontransparent political environment interferes with value maximisation and negatively
impacts responsible mine closure.
This paper discusses ways to realise maximum value from the world’s biggest find of alluvial diamonds and
positively impact communities after mining is completed. One way to realise the most value from mineral
exploitation is to apply principles of beneficiation. Beneficiation is a complex concept that mineral resource-
rich countries need to comprehend in order to attain maximum benefits from their mineral resources. It is a
crucial instrument through with countries can attain economic growth and development. This paper applies
the concept of beneficiation to Zimbabwe’s Marange diamonds in order to propose the most effective ways
to leave a lasting positive legacy to the distressed communities in Marange. Factors hampering the best value
realisation from the diamond fields are identified, and strategies to overcome the constraints and make the
most out of these diamonds are proposed.

The Marange diamond fields (MDF) are located near Mutare, Zimbabwe’s third largest city, in Manicaland
Province. Figure 1 shows the location.
The total size of the MDF is 60,000 hectares (Sokwanele, 2011). The MDF contain the largest diamond
deposits in the world (Imara Africa Securities Team, 2011; The Zimbabwean, 2011) and are a billion years old.
Mzaca and Mangena (2010) report that the value of the diamonds, depending on the value per carat, reaches
US$800 billion. Partnership Africa Canada (2010) notes that the diamond output at the MDF can sustain three
million carats a month for 14 years. This report also points out that the mine life can go beyond 30 years if
mining is done at a steady and more responsible rate.
Diamonds were discovered in the Marange fields in June 2006 by African Consolidated Resources (ACR),
leading over 30,000 diamond panners from different walks of life to converge there (Global Witness, 2012).
Lawlessness ensued as the government initially declared the diamonds to be free for all as a campaign
strategy to gain political mileage. Disorderly mining (illustrated in Figures 2 and 3), harmful to society, the
economy, and the environment, led to the deaths of many panners. The severe destruction of the
environment due to this “free for all” political influence significantly affects closure planning, as the level of
damage may not be rehabilitated to a sustainable level.

An operation coded “Operation Chikorokoza Chapera (End of Illegal Panning)” was launched in November
2006 by the Zimbabwe Police and the Zimbabwe Defence Forces to rid the MDF of ever-increasing numbers
of diamond panners. Corruption and greed then took centre stage, as the police and defence forces retained
some of the panners to mine diamonds for them. The diamonds were smuggled out of the country through
its porous borders. The panners who were retained were forced into hard labour for little or no pay. Word
of the existence of diamonds in Marange spread like a veld fire, even beyond the borders of Zimbabwe.
Human Rights Watch noted that “when the scramble peaked in October 2008, more than 35,000 people from
Zimbabwe, South Africa, Botswana, the Democratic Republic of Congo, Mozambique, Equatorial Guinea,
Nigeria, Lebanon, Pakistan, the United Arab Emirates, Belgium, and India were either mining or buying
diamonds in Marange” (Human Rights Watch, 2009). Concerned by the magnitude of the disorder, the
government moved in swiftly in October 2008 to take absolute control of the MDF. An operation coded
“Operation Hakudzokwi (You Will Not Return)” was launched, using the Zimbabwe Central Intelligence
Organisation (CIO), the National Army, Air Force, and National Police. Excessive force was used to drive away
the panners, leading to the violent deaths of 200 people, beatings and torture that caused grievous bodily
harm to 5,000 more, and the displacement of about 36,000 people without government assistance (Human
Rights Watch, 2009). Some of the evicted panners resorted to staying in barns, as shown in Figure 4.

After the expulsion of artisanal miners, three companies – Mbada, Anjin, and Transfrontier – were given
licences to mine different claims of diamonds in the MDF in July 2009. The allocation of licences was marred
by corruption and patronage. All three companies are totally owned by stakeholders aligned with the ruling
party, and the tax regimes applied to them see them contributing almost nothing to the treasury. The main
owners of these companies are the top brass of the police, national army, air force, CIO, and the ruling party
(Global Witness, 2012; Human Rights Watch, 2009, 2010b; Hunt, 2012; Mwonzora, 2011; Crook, 2010).
Masiya and Benkenstein (2012) observe that the MDF are still associated with poverty, killings, torture,
corruption, rape, forced labour, and other human rights abuses reminiscent of conflict situations. They note
that the case of Zimbabwe highlights the reality that human rights abuses in the exploitation of diamonds
can be committed by sovereign governments much as they are by illegitimate rebel groups. The elites pocket
almost all the proceeds from the diamond mining, while little is left for community development, a situation
which leaves mining communities undeveloped and destitute after mining is completed. The disregard for
environmental regulations and proper mine closure plans due to negative political influence may lead to an
environmental disaster in the MDF and the surrounding mining communities. The failure to invest diamond
proceeds in economic diversification leaves the community without survival projects after mine closure,
leading to economic vulnerability. These potentially negative socio-economic mining legacies compound
mining communities’ struggles for survival and need to be blocked through effective mine closure planning
that is independent of any sort of political influence.
Diamonds are currently exported raw, and no efforts are underway to cut and polish them in the country.
Several factors make it difficult to realise maximum benefits for all from Zimbabwe’s diamonds. The main
factors that stand out are bad governance and institutions, political instability, corruption, rent seeking,
sanctions, inflation, patronage, lack of transparency, and lack of strategic planning. These are discussed in
the following sections.

Roe and Siegel (2011) present compelling evidence that political instability impedes financial development
and needs to be taken as one of the major determinants of financial development. Aisen and Veiga (2011)
concur with this observation and mention that political instability is regarded by economists as a serious
malaise harmful to economic performance. They say that it is likely to shorten policymakers’ horizons, leading
to suboptimal short-term macroeconomic policies, and it may also lead to a more frequent switch in policies,
creating volatility and thus negatively affecting macroeconomic performance.
Zimbabwe was plunged into political instability when the main opposition party came into existence in 1999.
Zimbabwe had been almost a one party state since independence in 1980 due to insignificant opposition.
The arrival of a strong opposition led to stiff resistance from the ruling party. This eroded the rule of law such
that the law enforcement agencies became more inclined to the ruling party and did their best to suppress
opposition. The government undertook a fast-track land reform programme in 2000 coded “The Third
Chimurenga (Third Liberation War),” taking land from prominent farmers who supported the main opposition
party and giving it to supporters of the ruling party. This severely undermined the economy, and Zimbabwe
went from being the bread basket of the Southern African Development Community (SADC) to being a basket
case itself, since the new farmers could not produce enough to feed the country or the region. This land
reform program forced the West to impose sanctions on the country, which led the Zimbabwean currency to
rapidly lose value and led to record inflation. The consequences of the land reform program drove off
investors, as the investment climate in Zimbabwe was shrouded in uncertainty. From 1999 to the present,
each time there are elections in Zimbabwe, there are violent and bloody clashes between the ruling party
and the main opposition party supporters.
The major sources of instability in Zimbabwe are political and regulatory in nature (Chanakira, 2011).
International Crisis Group (2014) argues that Zimbabwe’s growing instability is exacerbated by dire economic
decline, endemic governance failures, and tensions over ruling party succession. It infers that the country
could become a failed state if major political and economic reforms are not implemented. This is an important
and accurate observation, as the political influence of the ruling party since 1980 has enabled its mining
affiliates in the MDF to mine with disregard for environmental regulations and to avoid making any significant
contributions to the treasury for economic development. Unregulated mining leaves several negative mining
legacies, such as acid mine drainage, environmental degradation, social ills, economic meltdown, endangered
health and safety, and ultimately, destitute communities. To avert these legacies, impartial governance is
needed to encourage appropriate mine closure planning that extends maximum benefits to the mining
communities.
The value chain of Zimbabwe’s Marange diamonds has been dominated by officials who abuse power and
public trust for personal gain. As mentioned, to enter the industry, one has to be highly connected in the
ruling party; no fair considerations are made in allocating claims. The tax regimes are such that little revenue
is directed to the treasury while most revenue flows to the police, army, CIO, and other ruling party faithful.
It is sad to note that the very company that discovered the MDF, ACR, was denied a license to mine on a
partisan basis. ACR’s hard-earned investment went down the drain, with no hope for recovery. A 2009 Human
Rights Watch investigation into events at the MDF points to rights abuses and evidence of corruption,
extortion, and diamond smuggling. This organisation also exposed the news that the security forces present
in the MDF formed syndicates with key public officials and buyers in order to smuggle diamonds, while some
private sector actors and relatives of high-ranking officials were also implicated in diamond smuggling
(Human Rights Watch, 2010a). More than $2 billion realised from Marange diamond sales between 2012 and
2013 has disappeared amid outrageous salaries and benefits for executive officers running diamond mining
ventures in Marange (CRD, 2013). There has been massive accumulation of wealth by executives of these
mining ventures in recent years, and some of them are suspected to have channelled huge financial resources
realised from Marange diamond sales to multiple private business investments (CRD, 2013).

The arrival of a strong opposition led the ruling party to innovate to retain power. It adopted bad governance
tactics that discriminated against members of the opposition. The government lost transparency,
responsibility, accountability, discipline, fairness, and respect for its own citizens, all to remain in power,
whether good or bad. The institutions that oversee diamond mining in the MDF are also partisans of the
ruling party. Bad governance ignites several social ills. Farr (2009) notes that, though Zimbabwe is mineral
rich, it has been plagued with heavy social tribulations: 85,000 people have been affected by cholera, which
claimed 4,000 lives; 1.3 million people are affected by HIV; the country has one of the shortest life
expectancies in the world; and the economy suffers under hyperinflation.
Unstable policies undermine investor confidence. Once policies are agreed on, there should be little or no
change, so that situations live up to investors’ expectations. In 2011, Zimbabwe enforced an indigenisation
program, which compelled all foreign-owned businesses in Zimbabwe to cede 51% of shareholdings to locals,
despite the fact that the locals did not invest in the establishment of the businesses. CRD (2013) indicates
that the mining industry in Zimbabwe is saddled with archaic laws and policy inconsistencies that are fuelling
corruption in government and aiding the plundering of mineral resources by foreign companies; the failure
by government to overhaul the mining sector and introduce policies that guarantee revenue inflows to the
treasury in order to sustain the country’s developmental needs has impacted negatively on the economy and
increased the vulnerability of communities living in resource-rich areas like the MDF.
Due to the uncertain investment environment, current investors wish to recover their investments as soon
as possible. This leaves no time to establish diamond cutting and polishing facilities to achieve more value
per carat of diamond exported. This is an untenable situation in a country where sanity could be restored
through an effective overhaul of the political system based on a government established on the principles of
transparency, accountability, responsibility, fairness, and mutual trust.
Despite the authorisation large-scale miners (LSMs) have to mine in the fields, there are running battles
between LSMs and illegal artisanal and small-scale miners (ASM). ASMs consider the fields to be their
heritage and birthright, while LSMs consider themselves to be the lawful owners, as they hold licences to the
claims. ASMs intrude into the LSM operations and jeopardise plans by the LSMs. Zvarivadza (2014) argues
that the relationship between the two needs to be appropriately managed if maximum value is to be realised
in the extractive industry. The relational approach should be proactive, as reactive approaches only worsen
the situation.
The huge diamond endowment could turn around the fortunes of Zimbabwe to the benefit of all if it were
well managed. The strategies to do so need to be realistic and implementable. Zimbabwe already has several
comparative advantages, which it needs to utilise for it to be competitive in the world diamond industry.
Strategies to overcome the discussed constraints are presented in the following sections.
Zimbabwe has a well-developed infrastructure, conducive for trade. There are good road, rail, and air
networks linking Zimbabwe and the world. The buildings are in shape and safe. There is a ready market for
diamonds in the world. The country’s baseline and tertiary education systems are among the best in the
world, with Zimbabweans rated among the best in many professions, including engineering. There are
universities undertaking ground-breaking research and development necessary for economic growth and

development. The African Economist (2013) ranks Zimbabwe as number one in Africa in terms of its literacy
rate. Above all, it has diamond resources in abundance, compared to other nations.
Zimbabwe formed a coalition government, the Government of National Unity (GNU), in February 2009 in a
bid to stop politically motivated violence. Unfortunately, the coalition did not work in good faith, and it
collapsed in July 2013 elections, which saw the return to power of the old ruling party. The elections were
widely reported to have been rigged, but the opposition parties did not win the court contest, indicating that
the ruling party had a firm grip on power. There is an urgent need for the country to break this 35-year
autocratic rule and conduct free and fair elections with internationally accredited observers to flush out
dictatorial tendencies. The electorate needs to be educated, through volunteer nongovernmental
organisations, that political violence worsens their plight; as the proverb says, “When elephants fight, it is
the grass that suffers.” People need to work together and tolerate different views and diversity amongst
themselves for the good of the nation.
With an elected government of the people and for the people, a thorough review of mining contracts and an
audit of the allocation procedure must be undertaken to ensure that mining is permitted for companies that
will benefit all citizens. The institutions that oversee the administration of MDF mining must have
representatives from all political parties, together with civilian representation, to ensure that licence
allocation and collection of revenues is done transparently, without fear or favour. An independent public
protector must be appointed to investigate the expenditure of taxpayers’ money, ensure transparency, and
hold public office holders accountable. Mungiu-Pippidi (2013) points out that collective action in fighting
corruption is key to economic development. He laments the fact that democratisation is increasingly
producing a new type of regime – one in which rulers who monopolise power and treat the state as their
own patrimony are succeeded by competing political groups or parties that practice a similarly nonuniversal
allocation of public resources based on patronage, nepotism, and the exchange of favours. All measures need
to be in place to ensure the delivery of a truly democratic government through democratic elections.
A major overhaul of mining policies in Zimbabwe is necessary to attract foreign direct investment (FDI), which
is necessary for establishing diamond beneficiation facilities. FDI can inject much needed funding to meet
the dictates of proper closure plans throughout the life of mines. Mining policies need to be explicitly stable,
clear, equitable, neutral, and efficient. Responsible mine closure must ultimately have strong roots in
effective execution of all the previous mining phases; it is not a standalone entity.
A democratically elected government, together with its institutions, must have the interests of citizens at
heart. It should be transparent, responsible, accountable, disciplined, environmentally friendly, fair,
innovative, and respectful of the will of the majority in all its dealings. Good governance can open up trade
networks for the nation and break trade barriers such as sanctions.
Mining is intended to extend a maximum return on investment to investors and uplift the quality of life in
society while managing the resultant negative impacts on the environment. It is crucial that consideration is
made for these latter intentions and not only for returns on investment. Pegg (2004) reiterates the need for
meaningful public participation and free prior informed consent by society before a mining company is
allocated a licence to mine in an area. Companies carry a duty of care to society, and it is this duty that earns
them the much needed social licence to operate. In the MDF, there was no free consent by the locals; they
were violently evicted from their ancestral land with no compensation after the discovery of diamonds. Stiff

penalties must be levelled against those who contravene agreed legal provisions to protect society and the
environment.
The natural formation of linkages along the value chain leads to diversification of the economy, from primary
industries to manufacturing industries. The clusters formed around the value chain encourage productivity
and innovation and enhance competitiveness as stakeholders share resources and co-invest in capital-
intensive projects. The clusters foster an element of trust, strong relationships, and efficiency, as partners
frequently interact and enjoy just-in-time delivery of goods and services necessary for business success. The
parties to the linkages share lines of communication, experiences, and vision and cultivate a culture of peace,
love, and harmony, which culminates in a social fabric that holds cluster members together. Parties work
together for a common good to fight the “resource curse” (the paradox of poverty in the midst of plenty) and
the so-called “Dutch Disease” (investment in one sector of the economy to the extent that, if that sector fails,
the country’s economy crumbles). Effective mineral resource management and diversification of the
economy through linkages and clusters are crucial to averting these dangers. Figure 5 shows the linkages that
form around a mining operation like that of the MDF; the main stages of a mineral value chain are illustrated
in Figure 6.

Beneficiation of MDF diamonds naturally leads to the establishment of downstream industries, as shown in
Figure 5; the results include employment creation, poverty alleviation, the closure of inequality gaps, and
economic growth and development. Communities are empowered when the value of resources is invested
in the communities to ensure the survival of projects and initiatives even well after mine closure. These
benefits embody the spirit of proper closure planning and ensure that proceeds from mining are properly
utilised in environmental management programmes, thereby averting any environmental disasters.
This paper has indicated that an unstable political environment and influence in the mining industry can lead
to plunder of resources and severe damage to the environment, resulting in complications in mine closure
and closure planning. If the custodians of the law are involved in plundering resources through a negative
political environment, it makes it difficult to close the mines appropriately, since the very enforcers of the
closure planning regulations are the ones breaking them. When mining companies become immune to
closure planning regulations as a result of political influence, it becomes an enormous challenge to close the
mines properly, as the regulations are not followed appropriately. This paper proposes an independent
mining industry regulator to ensure that mining companies abide by mine closure regulations and
appropriately invest in the future of mining communities. This would ensure community survival projects the
communities can rely on well after mining is completed. The principle of sustainability in the mining industry
may not be realised if an enabling, free and fair political environment is not realised. While mineral resources
are finite and nonrenewable, it is crucial for companies to invest in other modes of capital, such as human,
social, financial, and manufactured capital, as means to sustain life after mines close. This paper expresses
dismay at the fact that compelling achievements envisaged for mining operations in free and fair political
environments are not achievable in a compromised political environment. These missed opportunities
include redistribution of proceeds from mineral wealth to ensure the establishment of health facilities,
including efficient hospitals, clinics, and sewer systems; promotion of education and investment; provision
of reliable water supplies, energy, and a sound transportation infrastructure; creation of sports facilities to
facilitate growth of sportsmanship; and extension of employment opportunities at all levels in the LSM
mining operations to communities in which they operate. This lack of proceeds redistribution presents

several issues that impact mining communities after mine closure. The political influence and net
environmental damage in the MDF definitely go against the spirit of proper closure planning, since the
plunder of resources is destined to leave these communities destitute and severely impacted, without the
benefits of responsible mining. A mine should leave the communities in which it operates, and the country
at large, with a net positive benefit, a situation that is far from the reality at the MDF.
The MDF are a major source of wealth for Zimbabwe. The current plunder of the returns from diamond sales
by a few elite Zimbabweans needs to be halted. The 35-year autocratic governance model needs to be
replaced by free and fair elections with internationally accredited observers to install a democratically
elected government. These changes are crucial to address politically motivated violence; root out corruption,
patronage, and rent seeking; establish good governance and institutions; protect public funds; and break
trade barriers like sanctions. They are necessary to derive maximum value from the exploitation of the MDF
and meet proper closure planning expectations. All diamond mining decisions should consider, equally, the
economy, society, environment, and the future of the mining communities after mine closure to ensure
sustainable development in MDF mining. Linkages and clusters that form around the diamond value chain
must be wisely used to ensure maximum returns from the diamond trade. Transparency in the exploitation
of the MDF and allocation of revenue from this resource is the main launching pad for economic development
aimed at community survival beyond mine closure. The revenue can be used to invest in community
diversification projects like farming and horticulture, where products meeting international market standards
are produced. MDF income can be used to diversify the economy and strengthen Zimbabwe’s infrastructure.
Services including education, health, sports, and other recreational facilities can be also established in the
communities. The value addition strategies discussed in this paper go a long way in meeting the requirements
of proper mine closure, which leave communities with a sustainable alternative livelihood after mining has
been completed. The strategies also ensure the protection of the environment, which is the pillar of life on
earth.
Aisen, A. and Veiga, F. (2011) How Does Political Instability Affect Economic Growth? IMF Working Paper, WP/11/12. Washington
DC: International Monetary Fund, viewed 06 January 2015, http://www.imf.org/external/pubs/ft/wp/2011/wp1112.pdf.
Centre for Research and Development (CRD) (2013) Raising the community voice in the extractive sector: challenges and future
prospects of the mining sector in Zimbabwe, viewed 17 April 2014, http://www.crdzim.com/challenges %20and%20future%
20prospects%20of%20the%20mining%20sector%20in%20Zimbabwe.CRD%20Report%20January%202014.pdf.
Chanakira, N. (2011) Political instability affects business, viewed 08 January 2015, http://www.thezimbabwean.co/business/
business-analysis/52261/political-instability-affects-business.html.
Crook, J.R. (2010) United States urges strengthened Kimberly process controls on diamonds from Zimbabwe, American Journal of
International Law, Vol. 104(1), pp. 133–139.
Farr, C.D. (2009) Struggles in Zimbabwe: a brief history, viewed 07 January 2015, http://www.care2.com/causes/struggles-in-
zimbabwe-a-brief-history.html.
Forum for the Future (2015) The Five Capitals, viewed 10 March 2015, http://www.forumforthefuture.org/project/five-capitals/
overview.
Global Witness (2012) Diamonds: a good deal for Zimbabwe? Who controls revenues from Marange diamonds? A case study of
Mbada and Anjin companies, viewed 06 October 2014, https://www.globalwitness.org/sites/default/files/ library/A_GOOD_
DEAL_FOR_ZIMBABWE.pdf.
Hammond, R. (2009) Photographs of Marange diamond mining, in Zimbabwe’s blood diamonds (transcript), viewed 01 October 2014,
http://www.sokwanele.com/thisiszimbabwe/archives/5093.
Human Rights Watch (2009) Diamonds in the rough, viewed 26 January 2015, http://www.hrw.org/reports/2009/06/26/diamonds-
rough.
Human Rights Watch (2010a) Deliberate chaos: ongoing human rights abuses in the Marange diamond fields of Zimbabwe, viewed
26 January 2015, http://www.hrw.org/reports/2010/06/21/deliberate-chaos-0.
Human Rights Watch (2010b) Kimberley process: demand end to abuses in diamond trade, media release, viewed 26 January 2015,
http://www.hrw.org/news/2010/10/29/kimberley-process-demand-end-abuses-diamond-trade.
Hunt, P. (2012) Zimbabwe’s ‘conflict diamonds’ controversy and the flawed Kimberley process in Marange fields, viewed 26 January
2015, http://iloapp.waalmdiplomacy.org/blog/journal?ShowFile&doc=1338966230.pdf.
Imara Africa Securities Team (2011) Has the time arrived to get optimistic about Zimbabwe’s diamond resources? viewed 1 October
2014 http://www.howwemadeitinafrica.com/has-the-time-arrived-to-get-optimistic-about-zims-diamond-resources/13246
International Crisis Group (2014) Zimbabwe: waiting for the future, Africa Briefing N°103, 29 September 2014, viewed 06 January
2015, http://www.crisisgroup.org/en/regions/africa/southern-africa/zimbabwe.aspx.

Lydall, M. (2010) Getting the basics right: towards optimizing mineral-based linkages in Africa – contribution to the International
Study Group (ISG)’s review of African mining regimes,cited from The International Study Group on Africa’s Mineral Regimes
(2011).
Masiya, T and Benkenstein, A. (2012) Zimbabwe’s Marange diamonds and the need for reform of the Kimberley process, SAIIA Policy
Briefing 43, Governance of Africa’s Resources Programme, South African Institute of International Affairs (SAIIA).
Mukwazhi, T. (2006) Zimbabwe diamond mining: journalism, photojournalism, storytelling (blog entry), viewed 26 January 2015,
http://www.behance.net/gallery/ZIMBABWE-DIAMOND-MINING/1527381.
Mungiu-Pippidi, A. (2013) Controlling corruption through collective action, Journal of Democracy, Vol. 24(1), pp. 101–115.
Mwonzora, G. (2011) ‘Diamond rush’ and the relocation of the Chiadzwa community in Zimbabwe: a human rights perspective, M.A.
thesis, International Institute of Social Studies, The Hague, Netherlands.
Mzaca, V. and Mangena, Z. (2010) Fabulous wealth in Marange diamonds, viewed 23 January 2015, http://www.timeslive.co.za/
africa/article591181.ece/Fabulous-wealth-in-Marange-diamonds.
Partnership Africa Canada (2010) Diamonds and clubs: the militarized control of diamonds and power in Zimbabwe, viewed 8 January
2015, http://www.pacweb.org/Documents/diamonds_KP/Zimbabwe-Diamonds_and_clubs-eng-June2010.pdf, p. 14.
Pegg, S. (2004) Mining and poverty reduction: transforming rhetoric into reality, Journal of Cleaner Production, Vol. 14(1), pp. 376–
387.
Roe, M.J and Siegel, J.I. (2011) Political instability: effects on financial development, roots in the severity of economic inequality,
Journal of Comparative Economics, Harvard Business School, viewed 06 January 2015, http://www.hbs.edu/
faculty/Pages/item.aspx?num=39886.
Sokwanele (2011) The Marange diamond fields of Zimbabwe: an overview, viewed 01 October 2014, http://www.sokwanele.com/
system/files/The-Marange-Diamond-Fields-Report-Oct-2011.pdf.
The African Economist (2013) Ranking of African countries by literacy rate: Zimbabwe no. 1, The African Economist, viewed 14 January
2015, http://theafricaneconomist.com/ranking-of-african-countries-by-literacy-rate-zimbabwe-no-1/#.VN9eyv7LeX9. The
International Study Group on Africa’s Mineral Regimes (2011) Report on Minerals and Africa’ Development, viewed 02 March
2015, http://www.eisourcebook.org/cms/files/attachments/other/Minerals%20and%20Africas%20Development.pdf.
The International Study Group on Africa’s Mineral Regimes (2011) Report on Minerals and Africa’ Development, viewed 02 March
2015, http://www.eisourcebook.org/cms/files/attachments/other/Minerals%20and%20Africas%20Development.pdf.
Tozvireva, S. (2009) Families evicted from Zimbabwe’s diamond fields forced to live in barns (photographs), viewed 23 January 2015,
http://investigativeafrica.com/2010/11/16/families-evicted-from-zimbabwes-diamond-fields-forced-to-live-in-barns/.
The Zimbabwean (2011) Zims diamond deposits largest in world, viewed 23 January 2015, http://www.thezimbabwean.co.uk/news/
39695/zims-diamond-deposits-largest-in-world.html.
Zvarivadza, T. (2014) Artisanal and small-scale mining as a challenge and possible contributor to sustainable development, in
Proceedings, 9th International conference on mine closure, I.M. Weiersbye, M. Tibbett, A. Fourie, and K. Mercer (eds), 1–3
October 2014, Sandton Convention Centre, Johannesburg, South Africa, Australian Centre for Geomechanics (ACG), The
University of Western Australia.

T. Charette CPP Environmental, Canada

J. Martin Suncor Energy, Canada
Over the next century, hundreds of wetlands will be reclaimed in The Mineable Oil Sands Region of Alberta,
Canada. They will become common features of the final reclaimed landscape. Environmental Protection and
Enhancement Act (EPEA) operating approvals for oil sands mining stipulate that “the reclaimed soils and
landforms are (to be) capable of supporting a self-sustaining, locally-common boreal forest, regardless of the
end land use.” To achieve desirable end states, there is recognition that an integrated approach, including
watershed hydrology, water quality prediction, and biological, or food-web approaches, are required.
There are presently seven open pit oil sands mines operating in the 475,000 ha surface mineable Athabasca
Oil Sands Area. Located entirely within the Boreal Plain, this area consists of common geology, climate, and
vegetation. Because of this, regional best practices and guidance documents are developed jointly by
operating companies, government and stakeholders. One of these documents, the Guidelines for Wetlands
Establishment on Reclaimed Oil Sands Leases (the “Wetlands Guide”), was published at the end of 2014 as a
significant update from the previous version. It presents an integrated approach to the planning, design,
construction, monitoring, adaptive management, and certification of wetlands. The Wetlands Guide is meant
to be a resource for planners, landform design teams, regulators, stakeholders and Aboriginal peoples with
respect to reclaimed wetlands in the oil sands region.
The purpose of this paper is to demonstrate how ecological research and knowledge on natural landforms
and wetlands in the Boreal forests of northern Alberta support closure planning and design. The focus of the
information in this paper is on depressional wetlands, that is, marshes and open water wetlands, reflecting
the knowledge base for these types of wetlands.
At present, seven oil sands mining projects are operational in the Athabasca Oil Sands Region of north-
eastern Alberta, Canada, where there is a concentration of mineable oil sands. Approximately 763 of 4,800
km2 of the Mineable Oil Sands Area has been disturbed by oil sands mining as of January 2013, of which
approximately 63 km2 is reclaimed or is undergoing active reclamation (Alberta Energy, 2013). Several
additional projects are pending regulatory approval. Oil sands mines operate in close proximity to each other
and have very similar mining processes. Because of this, and the need for consistency, operating approval
conditions related to reclamation are uniform among the operators. This consistency, in particularly in the
goal and objectives for reclamation, drives collaboration in research, and on creating regional best practices
and guidance for reclamation.
Significant advances have been made in the past decade in wetland design and reclamation in the oil sands
region. The timing was right to create a Guidance Document for Wetland Establishment on Reclaimed Oil
Sands Leases (Wetlands Guide) to synthesize current knowledge and inform a new generation of research.
The purpose of this paper is to demonstrate how ecological research and knowledge on natural landforms
and wetlands in the Boreal forests of northern Alberta support mine closure planning and design.

Best practices for wetland reclamation for Alberta oil sand mines T. Charette and J. Martin
The creation of a project team that is comprised of key personnel with a stake in the oil sands mining industry
was critical to deriving a common suite of regional reclamation guidelines. The Aquatic Subgroup of CEMA,
consisting of industry, provincial government, aboriginal groups, and non-government organisations, steered
the project. Members of this steering committee are familiar with regional oil sands industry issues and have
the technical background that enables them to provide relevant input. The steering committee’s involvement
was crucial to ensuring that the needs of the regional industry, government representatives, and
stakeholders were addressed.
The steering committee made decisions on the scope of the project, authors, and target audience. To ensure
maximum usability, the guidance document had to be tailored to its primary users. The target audience is
multidisciplinary closure and landform design teams of oil sands mines; it provides background information
and detailed guidance for all technical disciplines. This document is also intended to serve as a reference for
a broader audience of long-range planners, company managers, regulators overseeing oil sands mining, and
stakeholders with a vested interest in the future of the region.
The design of a wetland is a multi-disciplinary exercise, requiring input from several areas of expertise,
including but not limited to hydrology, hydrogeology, engineering, and aquatic ecology. Thus, a team
approach was adopted. The steering committee proposed a table of contents. Selected authors developed
the table of contents further, reporting back to the steering committee. Authors presented draft chapters at
subsequent workshops with the Steering Committee, which allowed refinement of the product. The entire
document was also subjected to an independent “cold-eye” review by experienced professionals. Table 1
presents the chapters and authors of the Wetland Guide.
Natural wetlands have been studied extensively over the past one hundred years. Some of this knowledge
on natural systems, in particular the drivers of their form and function, can be applied to reclamation with
the goal of maximizing ecological processes and wildlife common to the boreal forest. An understanding of
wetland depth profile — the way the basin of the wetland slopes at the edges, how deep it gets in the middle,
and how it influences the development of emergent plant communities and the kinds of wildlife that will
likely use a site — can guide reclamation. This knowledge also helps set targets in terms of expected species
composition and ecological functions for a reclaimed wetland.
The hydrology, chemistry, and biology of aquatic ecosystems reflect a multitude of landscape features that
affect the design of a wetland, varying from terrestrial to aquatic and from regional to local (Figure 1; Webster
et al., 2006). For instance, the form, or topography, of the land surface and the hydraulic properties of the
geology largely control the hydrology of the landform and quantity of water delivered to a wetland. The
stream network allows connectivity among aquatic ecosystems, which is critical for the movement of water
and species, and for flushing of undesirable constituents. Design of the morphology of the aquatic system
offers an opportunity to create habitat complexity, which increases habitat diversity and long-term
ecosystem stability (Yachi and Loreau, 1999; Drake, 1991). Figure 1 illustrates these and other landscape
elements that affect the design of aquatic ecosystems. The items located on the upper part of the chart are
those over which we have the least degree of control, such as geology and precipitation. Plans and designs
need to take these into account, but design should focus on the items over which we have a greater degree
of control.

Chapter Description Authors

1. Introduction and Introduces the guide and provides Théo Charette, CPP
background context for the remaining chapters Environmental
2. Watershed Describes hydrological and Andrew Baisley, Lindsay Tallon,
hydrology and hydrogeological processes and and Mike O’Kane, O’Kane
geochemistry properties Consultants
3. Natural Wetlands Describes natural wetlands in the Brian Eaton and Jason Fischer,
in the region region including physical, chemical Alberta Innovates Technology
and biological conditions Futures
Lisette Ross and Lynn Dupuis,
Native Plant Solutions
Dale Vitt, Department of Plant
Biology, Southern Illinois
University
Matt Wilson and Théo Charette,
CPP Environmental
4. Lessons learned Describes lessons learned from the David Cooper, Kristen Kaczynski,
from wetland non-oil sands wetland reclamation Andrea Borkenhagen,
reclamation and as well as from the three decades and Stéphanie Gaucherand,
restoration projects of wetlands design and Colorado State University
establishment on oil sands leases Gord McKenna, BGC Engineering
Inc.
5. Wetland design for Wetland design for mine closure Gord McKenna and Vanessa
mine closure plans plans Mann, BGC Engineering Inc.
6. Wetland design at Overview of the design process Gord McKenna, Vanessa Mann,
the landform scale from a landform design perspective and Lisette Ross
7. wetland Overview of the construction and Lisette Ross, Dale Vitt, Gord
construction revegetation process McKenna and Vanessa Mann
8. Operation, Describes operational, Gord McKenna, Vanessa Mann,
maintenance, and maintenance, monitoring and risk and Lisette Ross
monitoring management strategies for
wetlands after their construction
Although hydrology can be difficult to control from a mine reclamation perspective, it is perhaps the most
important factor controlling wetland performance. In Boreal Alberta, annual water level variation in natural
depressional wetlands varies by about 20 cm, on average. However, it is normal for some wetlands to dry up
and return at a later time; some disturbance events (e.g., wet-dry cycles) are normal in wetlands and can
enhance species richness. Certification monitoring programs, and applications for certification, should
recognize this variability. As long as wetland hydro period is within the natural range of variation for the
region, success is attainable.

Wetlands rely on connectivity with uplands and with other aquatic ecosystems to exchange species. This
connectivity creates functional redundancy, and allows biodiversity and ecosystem stability. During times of
stress, some species can find shelter at neighbouring sites as long as they are accessible. Exchange and
dispersal in plankton and plant propagules can occur through stream dispersion, but insects, bird, and wind
dispersal is highly important as well.
In Boreal Alberta, surface water connectivity between wetlands can be limited (Devito et al., 2012). In the
absence of surface hydrological links, physical placement on the landscape becomes critical (Amezaga et al.,
2002). Eaton et al. (2014) recommends a distance between adjacent wetlands of 1 km to accommodate
amphibian dispersal. Adjacency to the natural landscape is particularly important, to allow re-colonization of
species.
A matrix of wetland and terrestrial patches is critical for a high-functioning landscape. Upland areas adjacent
to wetlands (i.e., riparian zones) are highly diverse ecosystems with high species richness, thereby
contributing to overall biodiversity. They are a source of carbon and coarse woody debris to aquatic
ecosystems, they act as corridors for wildlife movement on the landscape, and they provide an erosion
control barrier to aquatic systems. Adjacent terrestrial land is core habitat for many semi-aquatic species that
require both terrestrial and aquatic habitats. For example, many amphibians require aquatic breeding sites
and terrestrial overwintering sites and most species of waterfowl and waterbirds nest in the backshore.
Simple ecosystems are less resilient and therefore more prone to destabilization and collapse. Increasing
diversity can increase functional redundancy, which means that the odds are greater that some species will
be capable of functionally replacing important species. The diversity of many aquatic species, and thus
ecosystem stability, is positively related to increased habitat complexity and heterogeneity. Outside of
landscape-scale diversity, wetland-level biodiversity can be increased by creating a complex wetland shape
that maximizes the length of shoreline per unit area, includes islands and irregularities in the wetland bottom
contours.
Constructing marshes that are similar in structure to natural marshes will increase the likelihood that
constructed marshes become self-sustaining, natural in appearance, and healthy and resilient to
disturbances (Bayley et al., 2014). In Boreal Alberta, depressional wetlands are typically small (5.5 ha on
average), however variability is large. A variety of wetland sizes should be targeted, while focussing on small
wetlands.

The success of a wetland reclamation project is often measured on vegetation performance. Depressional
wetlands (e.g., marshes, open water wetlands) possess a series of vegetative zones that, as a group, are
critical to ecological function (Figure 2). The aquatic zone, defined as those areas flooded long enough to
exclude terrestrial plants, is typically divided into three sub-zones. A wet meadow zone is characterized by
water near the soil surface and a variety of herbaceous species. An emergent zone of cattails and rushes
typically is no deeper than 1 meter. An open water zone may contain submergent plant species and is less
than 2 meters in depth. Research has demonstrated that natural depressional wetlands in Boreal Alberta
typically are composed of one quarter open water and three quarters vegetation (wet meadow + emergent
zones; Figure 3). The presence and proportions of these wetland zones are critical in creating habitat
complexity and supporting a wide array of wildlife with varying habitat needs (Figure 4). A wetland slope of
5% (20:1) or less will create a gentle approach to the open water zone, which allows for development of a
wide wetland plant zone (Bayley et al., 2014). Gentle slopes buffer vegetation against rapid changes in water
depth, thereby increasing resilience against a wide range of precipitation and subsequent water levels
(Forrest, 2010; Wilson and Bayley, 2012).

Shallow open
water
Emergent
Wet meadow
Reclamation can often be described as planning a design around a moving target. An adaptive management
approach recognizes that information and understanding of ecosystems change over time. Adaptive
management, when addressed in the planning stage, provides documentation of alternative strategies that
allows managers the opportunity to change to designs or operations in an efficient and affordable manner.

To assess performance, clear definitions and descriptions of the reclamation objectives, criteria, and
indicators is important. Monitoring and assessment of the identified indicators over time allows for informed
adjustments to be made to the planned reclamation based on data collected on-site and up-to-date technical
information.
Reclamation
Plan
Plan
Construction &
Do Operations
Implementation of
Reclamation Plan
Milestone
or Monitoring
Check Assessment
Met Indicators Missed Indicators
Diagnostic
(understand issue,
Adapt Accept Results
assess risk, develop
options)
Adaptive
Response
Assess Readiness
for Reclamation Decision process
Certificate Adaptive process
Application
The adaptive management process is designed to test predictions and adapt to unexpected results. The
process is:
1. Clear quantifiable goals and objectives,
2. Conceptual designs (outcomes) related to goals and objectives,
3. A monitoring process to measure the success in predicting outcomes,
4. An evaluation process to compare monitoring results and predicted outcomes, and
5. A diagnostic and mitigation process to re-align outcomes to goals and objectives, where necessary.
Figure 5 represents an adaptive management process that can be implemented and applied during all stages
of mining and reclamation, from the planning stages through to the reclamation certification application
(Poscente and Charette, 2012). The purpose of the monitoring plan within the adaptive process is to evaluate
the success of the reclamation prescriptions over time, and adjust or modify (adapt) practices where

necessary to achieve reclamation goals and objectives. If monitoring results are unacceptable, the adaptive
management process can be employed, with changes being made to elements of operations, reclamation
techniques or prescriptions, and monitoring regimes. A variety of responses may be used, depending on
expected outcomes and environmental conditions. Responses may include:
 Continued monitoring – more time may be necessary to properly assess a site
 Reclassification of the ecosite in question – best in the early stages of reclamation
 Determination of indicator inapplicability – chosen indicators may not apply to certain
reclamation areas or older reclamation sites where old approval authorized practices currently
considered outdated.
Once changes have been made, indicators can be re-evaluated and success of reclamation re-assessed. As
monitoring results are accepted, an application for a reclamation certificate may occur.
The goal is to integrate reclaimed wetlands into the natural landscape and provide desirable ecosystem
services. For this to occur, it is important to understand the characteristics and processes that make up a
wetland. Knowledge on the characteristics and drivers of natural ecosystems can be applied to reclamation
with the objectives of maximizing ecological function, integration with the natural landscape, and natural
appearance. Wetlands in the oil sands region are the result of complex interactions among terrestrial and
aquatic features at multiple spatial and temporal scales. The hydrology, chemistry, and biology of wetlands
reflect many landscape features such as climate, geology, aquatic networks, morphology, etc.
Careful placement of the aquatic network is important to allow movement of water and species across the
landscape, which creates resiliency and stability in wetland ecosystems. Stream connections, or in absentia,
the proximity of wetlands to one another (<1 km apart), are necessary considerations to allow dispersal of
plant propagules and invertebrates. A matrix of wetland patches, amidst uplands, allows maximization of
landscape-level diversity.
Diversity in wetland morphology also leads to biological diversity. Complex wetland shapes with crenulated
shorelines, islands, and holes create habitat diversity and the ability of multiple species to co-exist. Another
tool to increase diversity is the creation of a variety of wetland sizes and shapes, although the majority of
wetlands should be of similar size to adjacent natural sites (approximately 5.5 ha in Boreal Alberta). In terms
of slope, a gentle approach toward the middle of the wetland allows the development of wide wetland
vegetation zones, which are critical to wetland function and supporting a wide array of wildlife with varying
habitat needs.
Integration of natural structure in wetland reclamation projects will increase the likelihood that constructed
wetlands become self-sustaining, healthy ecosystems. However, setting specific reclamation objectives,
monitoring, evaluation, and adaptive responses keep wetlands on a trajectory toward certification.
The next generation of wetlands, two fen watersheds (Syncrude and Suncor), are currently being
constructed. Research and monitoring associated with these projects will greatly inform the next version of
the Wetlands Guide, particularly with respect to peatland ecosystems.
We thank members of the Wetland Guide Task Group of CEMA (Rod Hazewinkel, Carol Jones, Ainsley
Campbell, Christine Daly, Chris Powter, Ernst Kerkhoven, Gerri Haekel, Gillian Donald, Richard Cavanagh), the
Wetland Guide authors (see Table 1), and individuals that have reviewed the Wetlands Guide at various
stages (Jonathan Price, Susan Galatowitsch, Jan Ciborowski, Suzanne Bayley, Carl Mendoza, Ann Smreciu,
Line Rochefort, and Norbert Morgenstern). We also thank the CEMA Reclamation Working Group Program
Managers, Kyle Harrietha and Kim Dacyk, and Glen Semenchuk (CEMA Executive Director) for assistance in
the Wetland Guide project.

Suncor Energy Inc. and its affiliates (collectively “Suncor”) do not make any express or implied
representations or warranties as to the accuracy, timeliness or completeness of the statements, information,
data and content contained in this paper and any materials or information (written or otherwise) provided
in conjunction with this paper (collectively, the “Information”). The Information has been prepared solely for
informational purposes only and should not be relied upon. Suncor is not responsible for and is hereby
released from any liabilities whatsoever for any errors or omissions in the Information and/or arising out of
a person’s use of, or reliance on, the Information.
Alberta Energy (2013) Alberta Oil Sands Industry Quarterly Update Spring 2013, viewed 10 May 2013, https://albertacanada.com/
files/albertacanada/AOSID_QuarterlyUpdate_Spring2013.pdf.
Amezaga, J., Santamaria, L. and Green, A.J. (2002) Biotic wetland connectivity – supporting a new approach for wetland policy, Acta
Oecologica, Vol. 23, pp. 213–222.
Bayley, S.E., Wilson, M.J., Rooney, R.C. and Bolding, M.T. (2014) Assessment methods for reclamation of permanent marshes in the
oil sands: handbook and video, prepared by the Bayley Lab, Department of Biological Sciences, University of Alberta.
Edmonton, AB.
CEMA (2014) Guidelines for wetlands establishment on reclaimed oil sands leases third edition, Fort McMurray, Alberta. 494 pp.
Devito, K., Mendoza, C. and Qualizza, C. (2012) Conceptualizing water movement in the Boreal Plains: implications for watershed
reconstruction, Synthesis report prepared for the Canadian Oil Sands Network for Research and Development, Environmental
and Reclamation Research Group. 164 pp.
Drake, J.A. (1991) Community assembly mechanics and the structure of an experimental species ensemble, American Naturalist,
Vol. 137, pp. 1–26.
Eaton, B., Fisher, J., Ross, L., Dupuis, L., Vitt, D., Wilson, M. and Charette, T. (2014) Chapter 3: Natural wetlands in the region, in
Guidelines for wetlands establishment on reclaimed oil sands leases, Cumulative Environmental Management Association.
Fort McMurray, AB. 494 pp.
Forrest, A. (2010) Created stormwater wetlands as wetland compensation and a floristic quality approach to wetland condition
assessment in central Alberta, University of Alberta, Edmonton.
Poscente, M. and Charette, T. (2012) Criteria and indicators framework for oil sands mine reclamation certification, Cumulative
Environmental Management Association, Fort McMurray, AB. 188 pp.
Soranno, P.A., Webster, K.E., Cheruveli, K.S. and Bremigan, M.T. (2009) The lake landscape context framework: linking aquatic
connections, terrestrial features, and human effects at multiple spatial scales, Verhandlungen des Internationalen Verein
Limnologie, Vol. 30, pp. 695–700.
Webster, K.E., Bowser, C.J., Anderson, M.P. and Lenters, J.D. (2006) Understanding the lake-groundwater system: just follow the
water, in J.J. Magnuson, T.K. Kratz and B.J. Benson (eds), Long-term dynamics of lakes in the landscape, New York: Oxford
University Press. 29 pp.
Weller, M.W. and Spatcher, C.E. (1965) Role of habitat in the distribution and abundance of marsh birds, Iowa Agricultural and Home
Economic Experiment Station Special Reports, No. 43. 31 pp.
Wilson, M.J. and Bayley, S.E. (2012) Use of single versus multiple biotic communities as indicators of biological integrity in northern
prairie wetlands, Ecological Indicators, Vol. 20, pp. 187–195.
Yachi, S. and Loreau, M. (1999) Biodiversity and ecosystem productivity in a fluctuating environment: the insurance hypothesis.
Proceedings of the National Academy of Science, Vol. 96, pp. 1463–1468.


S. Costa New Gold Inc. Canada
Planning for mine closure usually focuses on a detailed reclamation plan designed by experts to mitigate
environmental impacts; however, less is known about the social aspects of mine closure and how to mitigate
the socio-economic impacts of closing a mine. Industry best practices highlight the need to engage host
communities in planning for mine closure, still there are limited case studies that show what works best to
face the social challenges of closure. New Gold’s Cerro San Pedro Mine, located in Mexico has taken a
participatory approach to define the company’s Integrated Closure Strategy, engaging with key communities
of interest to understand the social impacts of closure and find feasible economic alternatives for a local and
regional longer-term development.
This paper provides an overview of the engagement process that the Company undertook to discuss aspects
of mine closure and the future without the mine with host communities and employees. The engagement
process is part of a broader approach to integrated closure planning and was led by a reputable non-
governmental organisation specialised in creating spaces of dialogue and consensus-building. The process
used both qualitative and quantitative methods. Findings of this engagement process revealed the specific
challenges and opportunities to strengthen economic activities alternative to mining such as agriculture,
tourism, and small business development; in addition to identifying potential partners. Outcomes of the
engagement process may be summarised as: i) raising awareness and responsibility of mine closure, both
environmental and social impacts, ii) building a common vision of the future without the mine, iii) reaching
consensus among different communities of interest, enabling participation and ownership of a shared
disposition that addresses social impacts of mine closure.
Lessons learned include, among others, the relevance of engaging employees and communities, not only in
implementing mine closure, but also in identifying social impacts of mine closure and building together a
feasible sustainability path forward.
Planning for mine closure usually focuses on a detailed reclamation plan designed by experts to mitigate
environmental impacts; however, less is known about the social aspects of mine closure and how to mitigate
the socio-economic impacts of closing a mine. In Mexico, as in many other countries, the law stipulates the
biophysical closure of the mines, but does not include social impacts of closure. More recently, mining
companies have understood the importance of incorporating social aspects in their closure plans and industry
best practices highlight the need to engage host communities in planning for mine closure.
The purpose of this paper is to illustrate an engagement process with key communities of interest, using the
New Gold’s Cerro San Pedro Mine as case study. This engagement process had the aim of better
understanding the socio-economic impacts of mine closure in order to build a mitigation strategy as part of
the mine closure planning. The paper is divided in five sections; the first section provides an overview of the
local context and the international best practices currently in place to address social aspects of closure. The
second section describes the methodology used by the engagement process and explains its different phases.
The third section of this paper discusses data obtained and the main challenges faced in using qualitative and
quantitative methods. Section four presents the results of such a process and how they were used by New
Gold in planning for mine closure, as well as discussing the main lessons learned. Lastly, section five presents
concluding remarks and suggests further research to better understand social aspects of closure.

Social impacts of mine closure: engaging employees and host communities in planning for closure S. Costa
New Gold is an intermediate gold mining company with assets on three continents, with four producing
mines – the New Afton Mine, the Mesquite Mine, the Peak Mines, and the Cerro San Pedro Mine, and three
significant development projects – the Blackwater project, the Rainy River project and the El Morro project.
Cerro San Pedro is the only New Gold mine in Mexico and will be the first New Gold mine to initiate closure
and, therefore, this process is a significant milestone in the evolution of the Company.
The Cerro San Pedro Mine (Cerro San Pedro), wholly owned by New Gold Inc., has been in production since
2007, and is located in Cerro San Pedro, central Mexico, 20 km northeast of the state capital city of San Luis
Potosí, a city of approximately one million inhabitants. The San Luis Potosí region has a long history of mining
activity, offering a strong pool of local mining suppliers and contractors and skilled workers covering a broad
range of professions and trades. Cerro San Pedro is an open pit mine that produces doré, a mix of gold and
silver, through heap leaching and a Merrill Crowe process plant.
Cerro San Pedro has maintained a close relationship with the residents of the Municipality of Cerro de San
Pedro, which comprises sixteen communities and three ejidos (communally-owned land holdings) within the
mine’s sphere of influence. Cerro San Pedro has, from its inception, provided broad benefits to the local
communities, becoming the main economic activity of the region. Examples of other socio-economic benefits
of the mine have been increased access to medical services, support for the local educational system with
contributions for improving school’s facilities, building new classrooms, providing educational software and
Internet connectivity. The mine has provided local children with transportation to schools, has contributed
to the preservation of local heritage and culture, and has provided financial support for the restoration of
heritage buildings. The mine’s contributions have been recognised in the accreditation as a Socially
Responsible Company (Empresa Socialmente Reponsable - ESR) by the internationally-recognised non-profit
Mexican Centre for Philanthropy (Centro Mexicano para la Filantropía - CEMEFI). While Cerro San Pedro’s
community investments have contributed to the quality of life of the municipality’s residents, as the mine
approaches closure, consideration must be given to the sustainability and continuity of such contributions.
Since it has become clear that the mine life was about to end, management has carried out a strategic review
of community investments and is working to ensure that long term and sustainable benefits from community
investments are achieved.
The history of the Municipality of Cerro de San Pedro has been closely linked to mining. Mining in this area
gave rise to the founding of the city of San Luis Potosí. From pre-Hispanic times to the present, Cerro de San
Pedro has experienced boom and bust times due to mining. Before the arrival of New Gold, the last major
mining project was undertaken by the American ASARCO, which operated in the area during the first half of
the 20th century. After this period, mining activity continued on much smaller scale as mostly artisanal
mining.
One of the main challenges for the area's social and economic development is the lack of social capital. A
Social Baseline Study commissioned by New Gold in 2012 found that there was a “crisis of social organisation”
in the communities around the mine, attributable to changes in land ownership, lack of leadership,
emigration, and social and economic inequality, among other factors. This study also identified areas of
opportunity and challenges for the area's sustainable development, such as agriculture and tourism.
An Economic Impact Study, also commissioned by New Gold, revealed the significance of the Cerro San Pedro
mine to the local and regional economy. For instance, in 2011 Cerro San Pedro represented 33.5% of the
mining sector's GDP in San Luis Potosí, and 0.92% of the state's total GDP. In 2012, taxes paid totalled $761.5
million pesos (about US$ 57.8 million at the time). Total tax expenditure was around $ 2.11 billion pesos from
the start of operations (about US$ 165.1 million at that time). By 2012, Cerro San Pedro had created nearly
860 direct jobs and about 4300 indirect ones. Direct jobs represent about 15% of the workforce of the entire
mining industry in the state of San Luis Potosí and 30.2% of the workforce of the state's metals and minerals
sector.

In the last decade, some toolkits and guidelines have been designed to promote closure best practices –
including the social aspects of the process. As examples, the International Council on Mining and Metals
(ICMM) Planning for Integrated Mine Closure Toolkit was released in 2008, and the Brazilian Institute of
Mining (Ibram) Guide For Mine Closure Planning, was released in 2013. The development of these guidance
materials have been in response to an increase in societal expectations and the fact that national regulations
generally do not include social aspects of mine closure processes, as is the case in Mexico. Issues resulting
from mine closure process, such as the loss of local employment, environmental liabilities and their impacts
on people's health, and the loss of the local economic engine, are addressed in these valuable publications.
Despite the guidance available through organisations such as Ibram and ICMM, there are few documented
mine closure experiences that have been considered exemplary in the industry; and even fewer that have
had a major social component. The majority of cases identified as best practices focus on the biophysical and
technical aspects of remediating the mining site. In order to better understand international benchmarks for
social mine closure, New Gold commissioned a review1 of seven mine closure experiences of several
companies in different parts of the world, including Río Tinto, Newmont and Barrick. The following were
identified as the main characteristics of the leading practices in closure planning:
 responsibility to independent third parties beyond regulatory requirements;
 high levels of transparency;
 ethics in business practices;
 respect for the expectations of communities of interest by including them in the planning and
decision-making processes;
 respect for the rule of law when taking additional measures to satisfy corporate social
commitments;
 respect for international regulations by applying similar business standards at all operations; and
 respect for human rights through their actions in the field.
Aware of the lack of positive benchmarks on social closure yet with leading practices identified, Cerro San
Pedro set out to establish its own Integrated Closure Strategy. New Gold, with the support of external
consultants started a process to identify and assess closure-related risks, identify controls and assign
responsibilities for the implementation and maintenance of controls. An internal committee was set up to
break traditional departmental silos and provide a forum for collaborative and integrated work. The Closure
Committee includes all key departments and regularly meets to review the closure risk registry, discuss
actions and collaboratively identify opportunities and solutions to challenges.
Cerro San Pedro announced its gradual closure of the mine in 2014, to all employees and contractors. The
significance of this process lies in the fact that today one of the most important challenges facing the mining
industry is the planning and implementation of an operation's closure. A responsible mine closure must take
into account social, economic and environmental parameters, seeking to minimise the impacts caused by the
operation and termination of mining activities. Cerro San Pedro adopted an integrated closure strategy that
not only complies with legal laws regarding environmental and biophysical closure, but also includes social
aspects of closure; therefore having a socially responsible and integrated approach through the entire
1
This study was presented by its authors at the 2013 Mine Closure Conference.

process. A responsible mine closure must take into account social, economic and environmental parameters,
seeking to minimise the impacts caused by the operation and termination of mining activities.
Key goals of the Integrated Closure Strategy are the employability of the workforce and a focus on sustainable
economic development of local communities. The Strategy seeks to strengthen the foundations of a
development model focused on different engines for the local economy as alternative to mining, such as
agriculture, tourism, ventures tied to existing productive chains, and direct employment in other industries
of San Luis Potosí. Therefore, Cerro San Pedro’s Integrated Closure Strategy includes two main social
components: 1) to support employees to have a successful employment transition, by strengthening their
employability in the mining industry, in other industries, or through individual entrepreneurship, and 2) to
identify and promote local economic development.
Several actions may allow employees to continue with their professional development after their
employment relationship with the company is over. Cerro San Pedro has offered skills certification and
training programs, for both its employees and for the residents of neighbouring communities. The purpose
of these programs is to provide the tools and skills they require to strengthen their capabilities, and, at the
same time, to promote local development with the necessary social capital to diversify economic activities
of the Cerro de San Pedro communities. This municipality has, besides the mining industry, an economy based
on agricultural and tourist activities.
Cerro San Pedro has taken a participatory approach to define the company’s Integrated Closure Strategy,
engaging with key communities of interest to understand the social impacts of closure and find feasible
economic alternatives for a local and regional longer-term development. The autonomous civil society
organisation Centro de Colaboración Cívica (CCC) (which is a member of the network Partners for Democratic
Change International) was selected for its reputation in Mexico and its alliances with well-respected
organisations worldwide to facilitate an engagement process to involve employees and communities in
discussion about aspects of mine closure and the post-mine future. The decision to involve an independent
third party was made as a means to add credibility and legitimacy to the engagement process. It was
important, for New Gold and CCC, that the process was kept autonomous, transparent and inclusive at all
stages. During most of 2014, CCC carried out a consultation with employees from the different mine
departments, as well as the residents of the various localities of the Municipality of Cerro de San Pedro that
maintain strong connections of financial dependency with the company.
The main objectives of the engagement process were to understand the concerns and expectations of the
company’s workforce, as well as those of the communities surrounding the mine, and to generate inputs for
the development of post-closure economic alternatives. The engagement process lasted about seven months
- from May to November 2014. Even though the planning of certain activities and implementation of some
programs were delayed until results from the engagement were analysed, this did not significantly impact
the general closure schedule. It was important to construct something robust and well informed before Cerro
San Pedro could make key decisions about investments and activities associated with social closure.
The engagement process included information gathering and generation of ideas, proposal development and
discussion, and proposal presentations. A mix of qualitative and quantitative methods was used to engage
with employees and community members. The process designed and facilitated by CCC lasted seven months,
and consisted of three main phases:

New Gold and CCC collaborated in the design of the engagement process, taking into consideration the
company’s Community Engagement Strategy and labour force characteristics. The key objectives of Phase 1
were to:
 introduce the engagement process to management, mine employees and members of direct
impact communities;
 identify concerns and expectations for the future after mine closure; and
 consult with employees and community on their preferred methods of communication and
engagement.
Phase 1 was also an opportunity for the CCC team to familiarise themselves with the community and the
manner in which community members interact with each other and with the company.
The CCC team started out with meetings with key internal communities of interest such as department
managers and 20 in-depth interviews and focus groups with employees at the various departments of the
mine. These interviews were followed by open community meetings in the four direct impact communities:
La Zapatilla, Cerro San Pedro, Monte Caldera and Cuesta de Campa. These communities are considered
‘direct impact’ because they are in close proximity to the site and/or have been identified as most dependent
on mining employment and other support from the mine. La Zapatilla was in fact part of a resettlement
program as a result of the mine development. It was also identified as the community with stronger links and
dependency to the mine. Meetings with the other communities in the municipality took place during the
second CCC visit. Meetings with key community communities of interest identified by the Cerro San Pedro
Community Development Team - such as municipality staff and the local priest - were carried out.
People were invited by the Community Development Team to open community meetings to introduce CCC
and the engagement process. Posters with simple diagrams of the process and information about next
meetings were displayed and distributed in the various communities. While the Cerro San Pedro Community
Development Team participated in first meetings with community members – as a means to introduce the
CCC team and their purpose – CCC facilitated all other meetings and focus groups independently in order to
keep the process autonomous.
Outcomes of this first phase were the revised design of the engagement process, as people were not just
asked about their concerns and aspirations for the future, but also about how they wanted to be engaged.
Inputs from Phase 1 informed the design of the process, and included a preliminary assessment of concerns
and expectations. Phase 1 also built support and buy-in for the consultation and participatory closure
planning program.
Later in the year, a survey was designed specifically for the workforce, with the aim to better understand
their level of comfort and understanding of the closure process at Cerro San Pedro, to document the
workforce’s concerns and their aspirations in regards to Cerro San Pedro programs to support their transition
to other employment. A total of 362 surveys were completed including all site’s departments.
Once engagement channels were defined together with host communities and employees, the next step was
to listen to their concerns and ideas for the future after the mine closure. The key objectives of Phase 2 were
to:
 Carry out workshops with residents of the direct impact communities to identify potential
alternatives for economic diversification in their communities.
 Identify and establish a collaborative relationship with technical experts who could assist with the
assessment of the viability of proposals made by community members and employees.

 Assess proposals based on technical quality and business viability.

Participatory workshops were carried out in La Zapatilla, Monte Caldera and Cuesta de Campa. These
workshops included the participation of technical experts and aimed at the identification and preliminary
assessment of alternatives for the local economy. Consultation with technical experts about ideas generated
by workshop participants was also carried out. All ideas were taken into account, making sure that the views
of everyone that was interested in participating was heard.
Inputs from communities were grouped in similar themes and then assessed by experts to consider their
technical and business feasibilities. This exercise was important for the following phase, where these inputs
were discussed back with communities.
Phase 3 consisted in six days of dialogue tables in the different communities and key stakeholder groups. The
key objective of Phase 3 was to discuss the opportunities and challenges identified with the proposed
projects, which were mostly related to agriculture and tourism. The feedback to communities aimed at
consensus building and reaching agreement about future investment projects.
Detailed reports with proposed agricultural and tourism projects that included feasibility studies by experts,
were presented to Cerro San Pedro and the Municipality. These reports also suggested partnerships and co-
investment opportunities.
The engagement process was agreed with communities and employees about the preferred channels of
communication and discussion. This was key to get process buy-in and participation from communities of
interest.
Data collected about concerns and future after mine closure reflect the following:
 Around 35% of Cerro San Pedro employees are interested in starting their own businesses.
 25% of employees are interested in skills development programs, 40% would prefer to stay in the
mining sector while 50% would prefer to look for employment opportunities in other industries in
San Luis Potosí.
 Communities are concerned about their future without the mine, in particular the future of the
local youth, and would like to see more employment opportunities and local development in the
municipality.
 Past negative experience with failed ventures, particularly with productive projects are reflected
in apathy and lack of confidence in their business competency.
 A key concern within the communities is the weakness of the transportation infrastructure and
services, the lack of roads connecting the communities and the low quality of existing roads.
 Another key concern expressed was in relation to the protection of the environment and human
health as the mine closes.
Inputs from communities and employees collected and analysed in Phase 2 reflect the following:
 There are opportunities for agricultural and tourism development in the region, and there are
synergies between some potential activities and businesses.
 Water is the greater challenge for agriculture in the region.
 Agricultural projects must include consistent technical support for at least three years and should
be well planned and based on a solid understanding of the market.

 Community enterprises must be legally formalised to ensure the adequate level of commitment
from participants.
Data was analysed and presented back to communities and employees for feedback and consensus building.
Agricultural and touristic projects proposals were discussed in the last part of the engagement process. This
was central to raise awareness of co-responsibility of different actors, including communities and employees,
as well as the importance of partnerships with government agencies and other public and private institutions.
Regarding challenges of the engagement process it is important to mention, firstly, finding a reputable and
trustworthy third party to carry the process out. A balance needs to be found between jointly designing and
overseeing the process while keeping its autonomy and independence from the mining company. A strong
collaborative relationship and effective communication between Cerro San Pedro and CCC were central in
achieving goals. During the engagement process, the main challenges were to keep interest and participation
levels throughout the different phases. To overcome this challenge, CCC kept participants well informed
about the next steps so they could better understand and anticipate following steps. It was very important
to manage expectations with this process. Cerro San Pedro was careful to convey the purpose of the
engagement process in order to avoid false expectations about its scope and reach.
Findings of the engagement process revealed the specific challenges and opportunities to strengthen
economic activities alternative to mining. The main challenges identified were water scarcity, gaps in
infrastructure, transportation, gaps in local capacity and in social capital. The main opportunities are the
proximity to a large industrial city, highways and an international airport, cultural and historical richness of
the area, and a recognised tourism potential. The engagement process resulted in proposed agricultural
projects, the strengthening of a proposal for tourism development which had previously been commissioned
by the mine, and a summary of concerns and expectations from mine employees and workers.
Employees were able to share their concerns, their vision for their own future and to articulate how the
company could support their transition to a productive life post mine closure. In regards to opportunities for
economic development, proposals were made by community members with the support of technical experts.
Technical experts contributed with their knowledge to analyse the feasibility of economic development
proposals, and suggested different alternatives to achieve them.
Provided with the findings of this process, the mine was able to review and bring focus to a preliminary
Integrated Closure Strategy, define actions and a budget associated with the social closure aspects.
Other key outcomes of the engagement process have been: raising awareness and responsibility of mine
closure, including both environmental and social aspects, building a common vision of the future without the
mine, reaching consensus among different communities of interest, and enabling participation and
ownership of a shared strategy that addresses social impacts of mine closure. All of these are key components
of a successful community program for closure planning.
Valuable lessons have been learned since Cerro San Pedro started the development of its integrated closure
strategy. What follows is a brief summary of the lessons learned to date.
First and foremost, it is very clear to Cerro San Pedro and to New Gold that engaging employees and
communities in the closure planning is absolutely critical. The mine did not commission or designed a closure
strategy in isolation and based on its own understanding on what would be adequate for the community.
Cerro San Pedro has learned that in order to achieve a shared ownership and responsibility for a prosperous
future the main actors need to be engaged from an early stage. Employees and community members need
to participate in decisions which impact their future and be given all the information they need to be able to

make their own decisions. Their role includes the identification social impacts of mine closure and building
together a feasible path forward.
The benefits of establishing a cross-departmental Closure Committee has also been a very positive lesson.
This committee has resulted in increased understanding of the interdependency of the various departments,
particularly in regards to closure aspects. This understanding has translated into collaboration and innovation
in addressing challenges of both social and technical nature. Regarding community investments, it is clear
that an investment strategy focused on sustainable, long term benefits is critical. The traditional
philanthropic approach to community investments often fail to create sustainable benefits and miss
opportunities to build social capital, which is critical for the social and economic development of local
communities. Cerro San Pedro has re-focused its community investment strategy to support the
development of local social capital and invest in projects which meet a sustainability criteria. The efforts the
Community Development Team has put towards building social capacity and resiliency within community
members have provided very encouraging results in a very short period of time.
Finally, in order to promote the local economy in a meaningful way, the collaboration between public, private
and social actors, is fundamental. Cerro San Pedro has engaged and signed collaboration agreements with
renowned Mexican educational institutions, business incubation centres, international consultants and with
governmental agencies, to promote the development of small businesses and productive projects,
understanding entrepreneurship as a manifestation of the aspirations of self-employment that, in addition,
contribute to the revitalisation of local economies.
New Gold’s Cerro San Pedro Mine is the first New Gold Mine to initiate a closure process and therefore an
important learning experience for New Gold. In order to establish a closure strategy which meets the real
needs of the local communities and addresses their true concerns, the mine launched a participatory process
to help define the company’s social closure strategy. The key outcome of such process was engagement with
key communities of interest to understand the social impacts of closure and find feasible economic
alternatives for a local and regional longer-term development.
The closure process at Cerro San Pedro is it its early stages, yet several benefits have resulted from open and
transparent engagement with communities and employees and from internal collaboration among the
various departments. These lessons will continue to support the closure process at Cerro San Pedro and will
influence how New Gold plans and implements closure in the future.

M.B. Dagva QMC LLC, Mongolia

G.L. Ainsworth QMC LLC, Mongolia
Ts. Davaatseren Erdenet Mining Corporation, Mongolia
Kh. Vladimir Erdenet Mining Corporation, Mongolia
O. Erdenetuya Erdenet Mining Corporation, Mongolia
In 1978 Erdenet Mining Corporation (EMC or Erdenet) was established in accordance with a joint venture
agreement between the governments of Mongolia and the former Soviet Union (51% and 49% respectively).
As one of the largest ore mining and processing operations in the world, the company produces approximately
530,000 tons of copper concentrate and 4,500 tons of molybdenum concentrates annually. Erdenet town was
established in 1974 specifically for the mine and now has a population of around 110,000.
One of the major shortcomings of the literature and legislation on mine closure is “best practice” principles
requires closure planning to commence at the inception of mine feasibility, therefore, essentially developed
for new mining ventures. However, Erdenet has been in operation for 36 years, and understandably, best
practice in 1978 in some instances may not meet the expected standards of today. Over the last decade for
example, advancements in the construction of Tailings Storage Facilities incorporating impermeable geo-
membrane HDPE liners and under drainage collection systems have dramatically reduced seepage and ground
water contamination.
In parallel, Mongolia has evolved from a communist system of government into one of the youngest
democracies and one of the fastest growing market economies in the world. Unfortunately, institutions are
still developing and policies change frequently, resulting in inconsistencies with implementation and
regulation.
This paper discusses the challenges of integrating best practice mine closure into an established mining
operation and its subsequent liabilities, considering the advancements in technical knowledge and mining
practices since the mine’s inception, and the potential influences of changes in mine management, political
parties and policies.
This paper will also briefly address the potential obstacles of securing closure financial provisions for an
established mining operation given the above mentioned challenges.
For Mongolia’s mineral resource sector to take advantage of its resources for economic growth, it needs to
ensure long-term sustainable development. Prior to the transition to a market economy in the 1990s,
Mongolia experienced over 70 years of communist mining control. This new era saw opportunism for local
Mongolian, Chinese and Western mining companies to obtain mining licenses. Unfortunately, some of these
companies did not consider their social and environmental responsibilities. Consequently, these liabilities
have tarnished the mining industry, with debates increasing in recent years around large mining operations
and the actual benefits to Mongolian society.

Challenges of integrating mine closure plans mid-way through the life of mine in Mongolia M.B. Dagva et al.
Mongolia’s population is one of the sparsest in the world, with an average population density of 1.5 persons
per sq km (0.3 persons per sq km in the southern Gobi). Total population recently reached 3,000,000. From
a mining perspective, this sparseness of population can help the development of mining.
On the other hand, Mongolians traditionally lead a pastoral, nomadic lifestyle. Because of the climate and
short growing season, animal husbandry defines the nomadic lifestyle, with agriculture playing a less
important role. Nomads raise five types of livestock – goats, sheep, cattle (including yaks), camels and horses
that provide meat, dairy products, transportation and wool. In this regard, although the herders may not
own the land directly, their livelihood has depended on these grasslands for pasture for many generations.
So naturally, the herders tend to treat exploration and mining activities with great caution.
“Mine closure” is not particularly a new term but historically it was not referred to as often as it is today. The
study “To improve the legal and regulatory framework for Mine closure” (Tsedendor et al., 2010) published
in 2010 regarding mine closure issues, included the following legal and regulatory framework enforced in
Mongolia: Subsoil Law (1989), Minerals Law, Land Law, regulations for either permanent or temporary
closure of mine and the Environmental Impact Assessment Law,
Regulatory framework covered the following:
 Any damage or deterioration to the subsoil during operations is to be restored, rehabilitated and
reported to the local government authorities.
 Notify the Specialised Inspection Agency one year prior to mine closure, describe the
decommissioning process in accordance with the appropriate procedures.
 Determine whether the estimated resource of the ore deposit during mineral exploration is
sufficient enough to recover any potential environmental damage caused by mining.
 Hazardous areas caused by the mining process must be identified with the appropriate warning
signs. These locations are to be surveyed, mapped and presented to all levels of local government
authorities and the Specialised Inspection Agency.
At the time we concluded these issues as unresolved:
 securing funds for mine closure;
 post-closure monitoring plan;
 liabilities of mine closure due to change of ownership;
 unplanned closure; and
 closure rights issues with licenses as collateral
To ensure competitiveness within Mongolia's mineral resource sector, it needs a stable long-term policy
framework. Accordingly, additional provisions in line with international standards for mine reclamation and
closure were added to the Government minerals policy.
The minerals policy is also included in the Minerals Law of Mongolia. Under rights and obligations, an
organisation's mining, processing, rehabilitation and closure procedures shall require approval from the
central government agencies responsible for environmental issues.
The ammendments also iclude a number of provisions relating to the obligations for mining and exploration
license holders. Some of the significant changes include:

 The appointment of a specific person responsible for mine closure.

 Priority rights for Mongolian registered entities to perform contract work, and to procure goods
and services.
 Priority rights for the sale of mined, beneficiated and semi-processed products for Mongolian
industries.
 To notify the Mineral Resources Authority of Mongolia upon discontinuation or termination of the
mine, infrastructure and beneficiation processes.
 The development of deposit resource estimations and reports by approved competent experts.
 To conduct and prepare a feasibility study by competent experts within one year time of receiving
the mining license.
 Product transport, infrastructure construction, mine reclamation and closure funds must be
detailed within the feasibility study.
Mine closure procedures and regulations relating to the management of collateral has yet to be developed
to international standards in Mongolia.
The Erdenet Mine (Mine), operated by the Erdenet Mining Corporation (EMC or Erdenet) is located in the
smallest aimag (province) of Orkhon in the northern part of Mongolia, about 140 km from the Russian border
and about 370 km northwest of Ulaanbaatar capital city of Mongolia by road. Orkhon is situated in the forest-
steppe sub region of the Khangai Mountain region, occupying the midlands of the Buren Mountain range and
the Orkhon and Selenge River Basin. The Selenge River supplies water to the Mine and Erdenet City via a 60-
km long pipeline. The mine area is in a semi-arid region with a typical continental climate of a short summer,
long winter, and a vast temperature fluctuation.
Erdenet mines 25 million tons of ore per year, producing 530,000 tons of copper and 4,500 tons of
molybdenum. Erdenet currently employs about 6,000 workers.
Erdenet’s internal organisational structure includes a number of independent departments:

 The Geology and Prospecting department established in 1981 operates to expand the ore reserve,
conducts geological and hydro geological studies, and also completes wells repairs.
 The Power Energy Unit serves Erdenet City with running water and sewerage services, as well as
supplying recycled water, oxygen and acetylene to the mine.
 The Department of Technological main functions are to automate and control the technical
processes of the plant, to improve and develop the communication and information network
based on latest technical achievements and successes.
 Machinery Repair Plant performs equipment servicing and maintenance; armouring equipment
with rubber or corundum; manufacturing and fabricating nonstandard equipment and spare
parts; diagnosis and repair to lifting machinery; smelting of steel, cast iron and non-ferrous metals
for casting parts.
Erdenet has three subsidiary companies:
 Erdmin Co. Ltd, established in 1997, produces cathode copper, rolled copper, copper wire and
electric copper wire
 The “RSTS” Co. Ltd., established in 1979 as a construction and repair unit, is the chief construction
entity for the company and the city of Erdenet.
 Erdsulj Co. Ltd, established in 1991, is a “sewing workshop” providing working clothes for
company’s workers. Since 2005 the workshop had expanded its activities to supply special
purpose and professional clothes.
Erdenet Mining Corporation also contracts out a number of operations at site:
 leach dump irrigation by two different companies;
 landfill waste is removed by a contractor to the local landfill facility; and
 local police management of mine access and security.
Items 2005 2006 2007 2008 2009 2010 2011 2012

Percentage of mining industry for
21.1 20.5 18.8 16.9 18.2 18.2 21 18.6
Gross Domestic Product
Percentage of EMC for Gross
9.5 10.1 11.2 12.8 11.3 10.1 8.3 7.4
Domestic Product
Percentage of mining for
68.5 75 68.8 60.3 66.4 80.8 89.2 91.3
Mongolian export
Percentage of EMC for Mongolian
36.1 45.4 43.7 30.2 30.4 29.3 20.2 20.1
export
Percentage of EMC on state
10.5 31.7 30 25.6 11.5 11.3 17 13
budget
Table 1 shows that EMC plays a major role within the Mongolian socio-economic framework, producing 20%
of total export products alone and providing 10% of the State budget. EMC also provides 85 to 90% of Erdenet
city’s budget.

Erdenet City is one of the youngest settlements in Mongolia. It was founded in 1974 in an area where large
deposits of copper had been discovered in the 1950s. The initial plan was for 8,000 miners to live in a newly
established town with their families. Assuming 4 to 5 members per family, the town was built to
accommodate 40,000 to 50,000 people. Today it is the third largest city in Mongolia with more than 110,000
inhabitants.
There are more than 1,300 companies currently operating in the city. The Erdenet carpet factory, the largest
carpet manufacturer in Mongolia, employs around 2,200 people. The city is home to a healthy private sector
with a number of foreign invested companies. For example, the American and Mongolian joint venture
“Erdmin” produces cathode copper and copper wire. A single-track railway line (121 km) linking Erdenet to
the Trans-Mongolian Railway was inaugurated in 1977, enabling Erdenet to become an important trading
centre for cashmere, wool, meat, paper manufacturing and metal processing.
Erdenet can be divided into 3 main districts: city, ger district and the industrial district. Approximately half of
the population live in old “Russian” apartments (connected to water, power and sewage) while the other half
live in the ger district (Mongolian traditional nomadic housing) on the outskirts of the city (no utilities). This
situation is progressively worsening as there is a growing inflow of people relocating from other cities (for
higher salaries) and nomads leaving their traditional nomadic herder life for employment. In the mid-1980s,
more than 50% of the inhabitants were Russian working as engineers or miners. After the fall of Soviet
Communism, most Russians left Erdenet. Today only about 8% of the population is Russian.
The EMC executive management team has been discussing the need to introduce a mine closure
management system since 2012. EMC began to cooperate with Qualified Mining Consultants (QMC) on the
matter early 2013. QMC's role is to consult and direct the EMC management team's ideas to international
best practice standards and methodologies. The process is still ongoing, and the strategies and observations
during the process can be classified as below.
International guidelines recommend developing the closure plans in the first stage of the project and refining
the plan as the project progresses. Figure 3 shows the current status of Erdenet within the stages of mining
and mine closure planning.

Stages of Stake
Mining Mine Closure Regulator
Holder
Planning
Exploration
Planning
Feasibility
Planning and
Project Approval Stage Assessment
Design
Incre
il
Construction
deta
Planning
a si n g
a si n g
Stakeholder input
deta
Incre
il
Operation Stage Review

Operation and
Audit
Planning
Erdenet Mining
Corporation
Decommissioning Stage
Decommissioning
Closure
Performance
Completion Stage Review and
Relinquishment
Post-closure
monitoring and
Maintenance

Nearly 40 years of operations since initial project implementation presents several challenges in integrating
a mine closure plan into the project.
Figure 4 shows that potential negative impacts during the study, developmental, operational and closure
stages of new projects can be determined, with pre-emptive measures put in place to avoid these issues.
However EMC may have missed some these opportunities.
Preliminary
Mine Closure Plan
Studies
Preliminary
Mine Closure Plan
Studies
Study and evaluation stage

EMC
Development stage today
Operational stage
Closure and relinquishment
1975 Offsetting 2055

Remaining
negative impacts
Remaining
Potential negative
negative impacts
impacts from Eliminate negative Eliminate negative
mining project impacts impacts
Avoid Avoid Avoid
Nevertheless, preliminary studies of closure planning during the operational stage should look into these
missed opportunities and provide assessment. The baseline environmental and socio-economic evaluations
were not implemented before mining operations began; therefore opportunities to refer back to original
baseline data pre-disturbance are significantly reduced. However, other locations may be identified with
similar biota to help establish a relevant biological baseline.
EMC implements environmental monitoring and mitigation programs as the environmental impacts occur.
For example, there are ongoing issues with windblown dust from the tailings dam and ARD drainage from
waste rock dumps. We have observed that, with the mine closure scenario still unclear and with the main
completion criteria undetermined, the current environmental protection process has been executed without
the appropriate planning.
The final ore reserve depth originally had three planned versions, and together with the economy and market
conditions at the time, the mine margins also need to be determined. There is also ongoing mineral
exploration to expand the reserves. Both of these uncertainties add to the challenges in formulating closure
plans.
The Erdenet settlement, which was established specifically for the mine, has now become a city with over
one hundred thousand residents who are still dependent on the company for revenue and taxes. While the
company has directly created 6,000 jobs, there are no current studies on how many jobs have been created
indirectly, but estimates suggest about 30,000 jobs. Under the companies CSR (Corporate Social
Responsibility) program, EMC has invested in the expansion of the City College, and the construction of the
EMC hospital diagnostic treatment center.

Of late, there has been some confusion with the mine workers and city residents when they first heard
discussions about mine closure. In the Mongolian context “Mine Closure” implies a negative connotation.
EMC management is now careful of using the term outside its meaning.
The above examples show Erdenet’s mining project's environmental, social and community impacts has its
internal challenges and various impacts and complications. QMC’s objective is to determine how best to
integrate mine closure planning with a closure management system that is in line with international best
practices and standards.
In cases where there are too many cumulated technical challenges, the issues need to be addressed and
solved at the management level. Given the current conditions, QMC's first priority is to secure the support,
commitment and investment from all the relevant stakeholders and the EMC management for mine closure
studies. Within this framework, the presentations were proven to be the most effective method to
demonstrate a simplified business case scenario and cashflow models.
No plan for mine closure (similar to unplanned closure)

 retain maximum profits with minimum outlay, seen as profiteering;
 legacies of pollution and contamination to the environment, and the health and safety of the
community;
 liabilities and costs left for community, local and national government;
 irresponsible mining, contradicts Mongolia’s definition of Responsible Mining;
 owners integrity destroyed, bad example for rest of mining industry;
 rest of the mining industry tarnished;
 compromising the environment, health and safety for financial gain;
 international and national worst practice; and
 against international standards and national laws (illegal).
Planning for mine closure at end of mine life
 insufficient funds remaining to complete mine closure;
 owners integrity compromised, bad example for rest of mining industry, both governments will
have to pay for remediation eventually, for a long time;
 legacies of pollution and contamination to the environment, and health and safety of the
community;
 liabilities and costs left for community and local government; and
 irresponsible mining practices.
Integrating mine closure as soon as possible with the required funding provisions
 Integration of closure planning into business plans resulting in reduced actual costs of closure
over the long term;
 Increased community ownership for post operational outcomes;
 international/national best practice;
 bench mark for the Mongolian mining industry;

 rejuvenate foreign trust and investment;

 sustainability for the community;
 recognition and realisation of positive legacies for local communities and lower exposure to
future potential negative legacies; and
 enhanced reputation.
From the above business case options, integrating mine closure into the mine operations as early as possible
is not only a wise business decision, but it is also beneficial to the environment, the community and the
country at large. These comparative business cases were effective in assisting the company to acquire
approval from all levels of corporate management.
In order to acquire acceptance we presented a comparison between the business risks and mine closure
responsibilities using the simplified cashflow models. One unique situation at Erdenet which may be inherited
from the former Soviet era/style of mining is that EMC has been responsible for a significant level of social
responsibility since its foundation. EMC developed departments which are integral to the company but also
have social benefits for example medical and health programs, college scholarships and training programs.
EMC also covers water supply, electricity, waste and other utilities. Costs equal to approximately 7% of OPEX
is spent each year for those social departments. A simplified cashflow model shown in Figures 5 and 6 show
the cost impact during the remaining life of mine, inclusive of copper market price fluctuations and operating
cost variations.

A general mine closure plan was produced for the work package to assist EMC's management decision making
process, with further recommendations of best practices and detailed steps as shown in Figure 7. EMC’s
management decision was to proceed with the development of the concept plans for mine closure and to
integrate them into the company's ten-year business plan. Fortunately, the ten-year business plan is currently
being updated and therefore a timely opportunity to include the mine closure concept plans.
EMC 10 yearly business

GAP Analysis Filling gaps
plan (2015 to 2025)
Conceptual Draft Preliminary

EMC Today closure plan closure plan closure plan
Final closure Post

and post- Closure
closure plan Vision
Social and Economic
Impact Assessment
Environmental Impact
Assessments
Operational Stage (Since 1970s) Closure stage

Currently, the concept plans for mine closure are in progress. The various closure scenarios are being
reviewed and developed, and solutions to the key issues are being resolved. The preferred closure option is
for partial salvage and decommissioning (retaining specific infrastructure for future business opportunities),
to rehabilitate tailings dam, waste dumps, pit, roads, and other disturbed areas, and make voids safe, stable
and secure.
Decisions on whether to retain the power plant, wastewater treatment plant, etc., will be made later in
consultation with the community and interested stakeholders. Erdenet’s current proposition is to utilise the
manufacturing workshop as a continuing business and retain offices for community administration purposes.
Option Outcome
Removal of all infrastructure,
1 Complete decommissioning Successful remediation of contaminated sites
Successful rehabilitation of TSF, WRD and all disturbed areas
Retain specific infrastructure for alternate business’s
Retain some buildings and
2 Successful remediation of contaminated sites
Infrastructure
Successful rehabilitation of TSF, WRD and all disturbed areas
3 Temporary Closure Care and maintenance, safe and stable
4 Unplanned Unforeseen, Liquidation, Government responsibility
Using the domain approach (ICMM, 2008) and completing a gap analysis, the operation was divided into a
number of separate entities (domains) to identify key issues to determine future study directions. Also,
utilising International Council for Mines and Metals (ICMM) Sustainable Development Toolkit for one of the
largest domains - Erdenet city, evaluation and directions for further works are also being carried out.
Two mine cut-off grades scenarios were considered for the entirety of mine life for the waste dump domain.
It was determined that should the cut-off grades be set too high there was the potential risk of ARD with
poor grade ore stockpiles, and so the LOM plan and mine closure plans are being prepared considering this
key criterion.
Preliminary estimates for closure cost are made and assessments are conducted for the expected solutions
from the Mongolian Mining industry and the Government regarding the securing of closure funds by the
EMC' decision making procedure. A decline in copper prices on the world market since 2014, forced EMC to
review their business plans and to re-evaluate its operation costs. Also completing a cash flow analysis and
risk assessments based long-term market trends of copper and mining costs to secure mine closure funding.
Mongolian mine closure regulations and guidelines are not yet completely developed to international best
practice standards. The experience and knowledge acquired during the development of EMC's mine closure
plans and the integration of closure management systems will be put to effect in multi-party work groups for
the development of mine closure procedures and guidelines.
 A long running mine with no closure plans presents challenging issues. In the Erdenet case, the
following were observed: any measures taken regarding environmental and social issues show
weak correlation with each other, and rather than focusing on the mid and post closure stages

where the highest potential risks and costs are estimated, they've tended to focus on domains
that show environmental issues during mine operations.
 Mining commenced almost 40 years ago with a remaining potential resource of another 40 years.
Trying to develop mine closure criteria without pre-mining baseline date is challenging enough
and requires a need to rely on benchmarking more than otherwise used for new projects)
 EMC has cooperated with QMC using a range of methods to assist management in the decision
making process and to pioneer the integration of mine closure planning in Mongolia where mine
closure regulations has yet to be fully developed. Methodologies and guidelines used in
international standards are best suited for these particular conditions. The EMC mine closure plan
has the potential to be used as a case study to help develope the Mongolian mine closure policy’s
and regulatory framework.
 The overlap in concept plans and renewed LOM planning gave us a fresh chance to review the
correlation between the cost and the mine closure results to help resolve the many technical
challenges.
DITR (2006) Mine closure and completion, leading practice sustainable development program fFor the mining industry. Department
of Industry Tourism and Resources, Canberra
ICMM (2008) Planning for integrated mine closure: Toolkit International Council on Mining and Metals.
Mongolian Statistics Bulletin (2013)
Tsedendorj, T., Dagva, M. and Byamba-Yu, J. (2010) To improve the legal and regulatory framework for Mine closure, Report for
policy studies,. Ulaanbaatar, Mongolia, pp. 21–26.

E. Nurtjahya Universitas Bangka Belitung, Indonesia

F. Agustina Universitas Bangka Belitung, Indonesia
Tin mines in the Bangka Belitung Islands have been exploited for about a hundred years. Following the
issuance of a 1999 Ministry of Trade and Industry decree that tin is not an export item to be monitored and
regulated, the Bangka regent issued a decree giving permission for the people to mine tin in 2001.
Consequently, “unconventional mines” (tambang inkonvensional), the term used to describe local small-scale
tin mines, have expanded significantly since 2000.
Bangka Island has a surface of 11,900 km2 and is mainly lowland below 50 m with some hills up to 700 m;
climatic differences within the island are small. Its climate is hot and wet with an average annual rainfall of
approximately 2,400 mm. Mining activities are spread across the island and run by one publicly listed tin
mining company, dozens of private companies and thousands artisanal mining groups.
Tin mining activities increase the wealth of the people, but they decrease environmental stability. Offshore
mining has reduced water quality as total soluble solids have increased and pH decreased; changes in the
seabed have caused changes in benthic flora, fauna and plankton diversity and an increased mortality index
of coral reefs and their associated fish. The number of fish caught in the offshore mining site has decreased.
Inland mining activity has reduced soil fertility and flora and fauna diversity. Inland mining has reduced the
number of individuals, species and plant families. In some areas, illegal mining causes floods in the rainy
season and damages roads and bridges.
Socio-economic secondary data were collected from various sites on Bangka Island through a literature
review. In addition to inadequate commitment and political will on the part of the local and national
governments, a low level of law enforcement seems to be a dominant factor in the low environmental
awareness. These findings may be used to accelerate the mine closure program started by the largest tin
mining company. This paper illustrates some opportunities and alternatives.
Bangka Belitung Islands produced approximately 106,000 t of tin in August 2013, representing more than
one third of global tin supply (IDH, 2014). The majority is exported to Singapore, followed by Malaysia, Japan
and the Netherlands. Tin mining is the most significant activity in the islands, taking place onshore and
offshore, including in protected forests and marine ecosystems. There are between 15,000 and 50,000
artisanal mines and approximately 30 independent smelters (IDH, 2014).
Following the issuance of a 1999 Ministry of Trade and Industry decree (No. 146/MPP/Kep/4/Tahun 1999)
that tin is not an export item to be monitored and regulated, the Bangka regent issued a decree (SK Bupati
Bangka No. 6 Tahun 2001) giving permission for the people to mine tin in 2001. Previously, as a strategic or
group “A” commodity, tin could be mined and marketed on by the public tin company. Consequently,
“unconventional mines” (tambang inkonvensional – TI), the term used to describe artisanal and local small-
scale tin mines, have expanded significantly since 2000. Tin mining has become a dominant economic driver
in the islands, leaving pepper plantations behind (Zulkarnain et al., 2005).
There were 80 dredges and 3,600 floating tin mines off the shore of Bangka Island in 2013 (Bangka Pos, 2013).
Tin production from artisanal mines contributes up to 80% of Indonesian tin exports, and tin mining activities

Managing the socio-economic impact of tin mining on Bangka Island, Indonesia – preparation for closure E. Nurtjahya and F. Agustina
increase the wealth of the local people, but most of the activities neglect good mining practices, safety and
land reclamation (ITRI, 2013). There were 80 mining-related casualties in 2012, and 22 in 2013 (Nurtjahya et
al., 2014). Most accidents at inland mine sites are due to landslides, and non-standard diving devices are
responsible for most deaths under water (Walhi – Friends of the Earth Indonesia, 2013).
The low level of law enforcement seems to be a dominant factor for low environmental awareness among
people, operators and regulators. Without firm and consistent implementation of regulations, reclamation
and revegetation measures and community development activities are meaningless and very inefficient.
People’s involvement and concerns are not genuine. It can be predicted that a mine closure program will fail,
although a lot of money will be spent. Balancing internal and external pressures with socio-economic impacts
would put law enforcement on the right track. The question then is how to manage these pressures to secure
the mine closure process.
Bangka Island, with a population of 991,062, is located off the eastern coast of South Sumatra Island (BPS
2012). Bangka Island has a surface area of 11,900 km2 and is mainly lowland below 50 m with some hills up
to 700 m; climatic differences within the island are small. Its climate is hot and wet, belonging to the Af-type
Köppen-Geiger climate classification (PT Timah Tbk., 1997), with an average temperature of 26.3oC, average
humidity of 61.7% and average annual rainfall of approximately 2,400 mm.
The authors and their students gathered secondary data from sites in all four regencies in Bangka Island:
Bangka, Central Bangka, West Bangka and South Bangka.
Alluvial tin deposits – cassiterite (SnO2) – were revealed after stripping the vegetation above the upper soil
and removing the non-tin deposit overburden. Tin extraction is done by pouring a large volume of highly
pressured water over the sediment, with the heavy tin ore separated from light material such as quartz by
gravity in a series of jigs, shaking tables or traditional jigs (sakan). Non-tin sediment settles in a lower area
with acidic pH that may go below 3.
Dredging is conducted to exploit tin deposits located offshore. Artisanal small-scale miners use small
traditional gravel pumps to pump tin-ore deposits to floating dredge units (TI apung) or modified small fishing
boats. Large dredging ships (kapal keruk) that can excavate tin ore at up to 70 m depth with bucket wheel
dredging (BWD) and offshore suction boats (kapal isap produksi) are used to exploit the tin slurry and pump
it to the concentrating plant. Gravity is used to separate the tin ore from other material. The alluvium is
broken up by a high-pressure jet of water and suctioned to the concentrating plant above the water, changing
the seabed and leaving high turbidity below the water surface.
Inland mining decreases soil properties, with soil texture changing from about 70% to 97% sand faction. The
concentrations of phosphate, potassium and sodium in undisturbed land are higher than in disturbed areas,
and are gradually decreasing as more tin-mined land is abandoned (Nurtjahya et al., 2009b). C-organics are
less than 2%, and the cation-exchange capacity (CEC) of tin-mined lands is very low (0.4–3.9 units) (Nurtjahya
et al., 2009b). The soil temperature may reach 45oC during the day (Nurtjahya et al., 2008c), and evaporation
on sandy tailings may reach 4 L/m2/day or double than of undisturbed soil (Nurtjahya, 2010). In mining
activity, the water and the sediment from the washing process bring acidic material, which makes the
catchment more acidic – with a pH below 3. Together with this material, lower pH negatively affects soil flora
and fauna. The zero-year sandy tailings itself has pH of about 4.5–4.8. The pH is getting higher, up to 5.1 in
an area with 38 years of natural regeneration. As a comparison, the pH of forest and abandoned farmland is
around 4.5–4.7 (Nurtjahya et al., 2009b).
Mining activity changes the vegetation structure and composition. The vegetation structure after 38 years of
natural succession on old tin-mined land was less than 2%, similar to that of a riparian forest on Bangka Island
(Nurtjahya et al., 2009b). In the forest, there are 7,295 individuals per hectare of forest, belonging to 85
species and 44 families. An area with 38 years of natural succession on old tin-mined land has 2,180

individuals per hectare, belonging to 16 species and 13 families; and an area with seven years of natural
succession on old tin-mined land has 890 individuals per hectare, belonging to six species and four families.
After 38 years of natural succession on old tin-mined land, there is no record of poles (trees with diameter
between 10 and 20 cm) or trees with a diameter of 20 cm or more. As a comparison, in an area of abandoned
farmland, there are 7,175 individuals per hectare, belonging to 48 species and 47 families (Nurtjahya et al.,
2009b). The number of arbuscular mycorrhizal fungi (AMF) spores increases with the abandonment of tin-
mined land, and the number of phosphate solubilising bacteria (PSB) shows different readings with the period
of abandonment (Nurtjahya et al. 2009b).
Land recovery and coral reef transplantation are costly. Excluding land function change from pepper
plantation, the revenue from tin through land function change is lower than for non-mining land uses:
protected forest, rubber plantation and beach (Nurtjahya et al., 2014). Cultivation on mined sites requires
large input. The expenditure to convert one hectare of previously tin-mined site into rice field is estimated
at Rp. 31 million, with the land preparation component representing the major portion (about 62%), and
almost 50% of the costs are for physical and chemical soil property improvement (Nurtjahya et al., 2009a).
Offshore tin mining has reduced water quality. This is shown by a 40% total soluble solid (TSS) increase, a
75% sedimentation rate increase, a 25% water pH decrease (i.e. from 8.0 to 6.0, becoming more acidic) and
a 50% dissolved oxygen (DO) increase (Nurtjahya et al., 2014). Wahyuni et al. (2013) reported that, owing to
tin mining, the concentrations of lead (Pb) (0.223 ppm) and TSS in solution offshore at Batu Belubang (705
ppm) were above the ministerial regulations of (Kepmen LH No. 51 tahun 2004) 0.008 ppm and 400 ppm
respectively, and it was normal to find cadmium (Cd) and zinc (Zn) in the water and Pb in sediment.
The increase of TSS affects the aquatic ecosystem, reducing oxygen supply (Mukhtasor, 2007) and changing
photosynthetic processes in phytoplankton and aquatic plants. In another study, offshore mining was found
to cause a 40% reduction of the number plankton species (Nurtjahya et al., 2008a). The number of species of
seagrass in mined water was about 70% of the number in less mined water (Nurtjahya et al., 2008a). The
dominant substrate in mined water was sand and rubble, in contrast to macroalga Halimeda sp. anemone in
less mined water (Nurtjahya et al., 2008a). Seagrass specimens such as Cymodocera rotundata, Cymodocea
serrulata and Thallassia hemprichii were reported in the water at Tukak Beach in South Bangka, whereas the
majority of seagrass was dying in the water at Tanah Merah (Nurtjahya et al., 2014).
The Walhi – an environmental NGO – reported that tin mining was responsible for up to 90% of the damage
to coral reefs and the reduction of coral reef–associated fish (Walhi, 2013). The number of coral reef–
associated fish in mined water was 30% of that in less mined water (Nurtjahya et al. 2008b). Coral reef life
coverage was less than 25% in mined water compared to more than 90% in less mined water (Nurtjahya et
al., 2008a). However, the growth rate of the coral reef species Acropora digitata transplanted to Teluk Limau
Beach, Bangka, was 2.2–2.4 mm/month (Sodikin, 2011).
Pratama (2014) reported that because of floating small-scale artisanal tin mining units (TI apung), the number
of fish caught has decreased, causing some fishermen not to go fishing. In the past, when the condition of
the coral reefs was much better, fishermen in one area could harvest up to five metric tonnes per day, fishing
could be done 300 m to 1 km from the beach, whereas now the fishermen need to go 1.5–5 km with no
guarantee of a good catch.
An agrarian problem has arisen since the start of mining, before the reclamation or revegetation of mined
sites (FEM IPB, 2013). Two problems are identified: (1) the mobility of artisanal small-scale miners (TI and
tambang skala kecil) in reclaimed sites cannot be controlled by the mining company, and (2) ownership
claims related to and occupation of reclaimed sites. These problems may arise at any step of the reclamation
process. During step 1, site determination, artisanal small-scale mining, or huts, or plantation crops suddenly
exist in few days. During step 2, site preparation, artisanal small-scale mining and plantation claims arise.

During step 3, cover crop planting, artisanal small-scale mining, plantation claims and territorial claims arise.
During step 4, revegetation, artisanal small-scale mining and territorial claims exist, and cover crops are
tampered with. During step 5, maintenance, revegetation plants are stolen, cut or claimed, the area is re-
mined or societal conflict exists (FEM IPB 2013).
As one mine site can be re-mined by different group of miners, soil quality decreases before reclamation
(FEM IPB 2013). In the case of reclaimed and revegetated sites, illegal mining has destroyed the reclamation
and revegetation process. Re-mining has financial consequences, as more input is needed for poorer soils. In
many areas, the sites become an open access resource. In one location, the miners have land ownership
issued by the local authority (FEM IPB 2013). From a survey conducted in 2012, only 23.6% of 2,111 hectares
of potential reclamation area is eligible or free of claims by local people (FEM IPB, 2013).
Socio-economic secondary data were collected from various sites on Bangka Island through a literature
review by the authors and some undergraduate students at Universitas Bangka Belitung. A descriptive
qualitative method and purposive sampling were used, with 10-17 interviewees for each study. The
interviewees are artisanal tin miners, artisanal tin mine owners, prominent local people, fishermen and
former fishermen who has changed their profession to be floating TI tin miners.
The positive impact of tin mining is economic (Juniarti, 2014; Indra, 2013; Romeo, 2011). The tin business
brings in money in a short time; local people say the money earned from the tin business is “’hot money.”
Tin mining is considered the largest economic driver, making a significant contribution to the provincial
economy. It is reported that about 10,000 artisanal small-scale miners support more than 50,000 people
(ITRI, 2012). Artisanal small-scale mining contributes up to 80% of Indonesia’s tin exports (ITRI, 2013).
The income percentage for tin miners compared to overall income per month of people in Lubuk Kelik,
Bangka, is 93.4%; for ex-pepper farmers in Silip, Bangka, 95.1%; and for ex-rubber farmers in Bencah, Central
Bangka, 89.1%. Pepper and rubber plantations contribute less than 2.3% each of overall monthly income of
pepper farmers or rubber farmers (Nurtjahya et al., 2008b). The net monthly income of fishermen in Rebo
and Bubus beaches, Bangka, is about one-third of the income of their colleagues working in tin mining
(Nurtjahya et al., 2008a).
The majority of people in Bangka Belitung fall into wealthy categories as of 2011. The increase of income for
the majority people is shown by the number of motorcycles and cars. From 1999 to 2011, motorcycle and
car taxes increased 15-fold. The number of people who perform the hajj pilgrimage increased almost 10%
from 2001 to 2012 (Erman, 2013).
While tin mining increases income for miners or people active in mining, it also causes societal conflicts at
mining sites. Most of the conflict in both inland and offshore mining is between locals and immigrants. Almost
60% of artisanal miners come from adjacent islands, such as mainland Sumatra and Java Island (Erman, 2013).
Walhi – Friends of the Earth Indonesia (2013) reported 12 conflicts between local fishermen and miners
between 2006 and 2011. Attitude changes and conflicts are reported in the hamlets and villages of the
studied area (Table 1).
In terms of education, the drop-out rate from elementary to senior high school has increased. In 2011, the
province of Bangka Belitung had the second-largest student drop-out rate in Indonesia because of children’s
involvement in mining or following their parents when they move to new mining sites (Erman, 2013.

In some areas, fishermen and farmers have changed their professions to become miners. Fishing boats are
modified to become mobile floating dredges in Bangka (Nurtjahya et al., 2008a). Rubber plantations and
pepper plantations have been mined in some areas in Central Bangka and South Bangka (Nurtjahya et al.,
2008b).
Aspect Findings References

Attitude changes  Alcohol drinking, drunkenness, prostitution, Iryanto (2014);
gambling, drug use among male youth Pratama (2014);
 Disobedience to parents Romeo (2011)
 Consumerism
 Neglecting prayer duties
 Profession changes from fishermen to
miners, from farmers to miners, from
labourers to miners
 Less collaboration among villagers
Conflicts  Between café owners and local people, café Anggrewan (2012);
owners and local women Christina (2011);
 Fighting over mining sites between different Juniarti (2014);
dredge-type miners, fishermen and artisanal Bangka Pos (2015)
miners, locals and immigrants
 Between husbands and wives because of
husbands buying the services of café
hostesses
 Between religious local culture and more
secular immigrant culture
 Over financial transparency between head
of village and local people
Offshore mining affects the fish catch. Small pelagic and demersal fish production decreased at three offshore
mined sites in three regencies over the period 2009–2010. At Batu Belubang Beach, Central Bangka, the total
small pelagic fish production decreased more than 50% in the 2009–2010 period (Octarini, 2011). At Rebo
beach, Bangka, small pelagic production decreased 10% and demersal fish catch decreased up to 48% in the
period of 2007–2010 (Sucita, 2011). At Tanjung Ular beach, West Bangka, small pelagic production decreased
24% and demersal fish catch decreased up to 70% in the period of 1998–2008 (Bidayani, 2010). A river that
receives tin sedimentation has 22 species and 10 families of fish, whereas a river free from tin mining has 36
species and 16 families of fish (Muslih et al., 2013).
The habitat changes have caused the benthic mollusc species Laevistrombus canarium L. (locally known as
siput gonggong) of the family Strombidae to be replaced by the bivalve species Anadara granosa (Yulianda
et al., 2009). The mollusc is the raw material of the most expensive seafood cracker in the province, up to
Rp. 250,000 per kg or about US$ 25 per kg.
Flooding in many areas of the province is believed to be caused by tin mining. The original small stream
channels have been changed by the mining activity. The changing channels cause flooding in many areas, and
in one area flooding resulted in the loss of two lives in 2013.

Socio-economic and cultural development is one of four mine closure programs stipulated by article 16 (2) d
of the Ministry of Energy and Mineral Resources regulation for the Republic of Indonesia (Permen ESDM No.
7 Tahun, 2014). A number of social and economic indicators can be used in future surveys to measure the
improvement as perceived by the local community.
Socio-economic impact must be managed to support the mine closure program in the province. To date, the
only mining closure program is being conducted by the only publicly listed tin mining company. Socio-
economic impact should be managed simultaneously by both internal and external stakeholders. Internal
stakeholders implement the laws, socialise the laws to the people and work hand in hand to create
alternative economic drivers after tin mining. External stakeholders such as the Tin Working Group (TWG)
provide clarification of the tin supply chain to support the mining industry in becoming more sustainable. The
IDH Tin Working Group works on mineral certification, although it started voluntarily. The local government
may take the lead to educate and empower the local people by addressing the internal needs and external
pressures in order to achieve sustainable development.
The broader group of stakeholders needs to consider the ecological cost of degraded mined soil for the sake
of sustainable development in the province. Benefits should be passed on to future generations, not held by
the current generation. Notably, more than 50% of mining money is sent outside of the province.
Regulators may play a role in enforcing good mining practices and issue rewards and punishments to miners
as well as providing alternative non-tin economic drivers to the people. Law enforcement should be
implemented fairly, and law the applied to everyone. Leadership, commitment and political will are required
on the part of local and national legislators and executives. Every legislator and executive should have the
same understanding of sustainable development. People should abide by the rule for the sake of their
children and grandchildren. As the regulator, a Bangka Island regent invited a number of undergraduate
students and skilled farmers from Java Island and the Sumatra mainland to change the mindset and educate
local people about agriculture and livestock.
All mining laws (Peraturan Pemerintah Republik Indonesia Nomor 1 Tahun 2014, Undang-undang Nomor 4
Tahun 2009 tentang Pertambangan Mineral dan Batubara, dan PP Nomor 23 Tahun 2010 tentang
Pelaksanaan Kegiatan Usaha Pertambangan Mineral dan Batubara, PP Nomor 24 Tahun 2012 tentang
Perubahan Atas PP Nomor 23 Tahun 2010, PP Nomor 55 Tahun 2010 tentang Pembinaan dan Pengawasan
Penyelenggaraan Usaha Pertambangan Mineral dan Batubara, PP Nomor 78 Tahun 2010 tentang Reklamasi
dan Pasca Tambang) should be implemented. In many cases, especially in Bangka Belitung, the role of
environmental NGOs is effectively to monitor the implementation of the law and trigger discussion among
regulators and between regulators and legislators. Coverage in the local newspaper has given the local
people more knowledge of the pros and cons of tin mining. One example is the issue of radioactivity in the
mining industry. In the last two years, the international networking established by a local NGO has created
pressure on local government to check on the implementation of the laws. Regardless how large the
contribution, local and international NGOs invested in the International Tin Research Institute – ITRI
Indonesia Tin Forum, an international seminar held in Pangkalpinang, Indonesia, in December 2013.
Academics should undertake research and development from a social approach to find economic ways of
engaging in mine site reclamation and revegetation, find alternative economic drivers for the people and
implementing the results of their research and patents. All of the work requires collaboration and integration
from either regulators or business. In a previous study, the percentage of input to prepare mined land before
it was ready to plant was almost 70%, or Rp. 22.8 million per hectare, with compost and mineral soils
representing the two largest expenditures (Nurtjahya et al., 2009a). Corporate social responsibility (CSR) on
the part of mining companies may support research, and at the same time create recognition of the
companies themselves.

Operators, academics and regulators must find the best solution to implement the idea of people-based
reclamation (FEM IPB, 2013) in practice. This is not an easy task. It is also about changing mindsets. However,
involving students of all education levels may create positive results. Weekend camps and mandatory field
involvement (Kuliah Kerja Nyata) for undergraduate students may be alternatives to educate people about
sustainable development and sustainable mining.
Anyone can create environmental awareness. A university lecturer asked for support from the Pantai Rebo
(Bangka Island) local authority and fishermen to conserve the offshore environment from inappropriate
fishing and tin mining. Of 30 respondents, 64% agreed not to use bombs in their fishing, and 30% agreed not
to mine in offshore water, while another 67% somewhat agreed (Umroh, 2015).
An effort should be made to change mindsets, although this requires a very long investment. Education is a
potentially effective way to do this. Environmental awareness comes in many forms. Some local efforts
includes the “Bina Cinta DAS” (loving watershed building) at the province’s Balai Pengelolaan Daerah Aliran
Sungai (BPDAS) for junior high-school students; mangrove planting in Belinyu, Bangka, by multiple
stakeholders, including scouts; and mangrove nursery building at Rebo Beach, Bangka, by university students
as their social project. A legal draft of watershed management principles has been prepared by a
multisectoral forum and is in the process of becoming law.
The international community may play an important role from outside the country. In the last couple of years,
local, national and international environmental organisations have increased their campaigns against
exported tin ingots produced using poor mining practices (Nurtjahya et al., 2014). Some foreign coverage in
newspapers and short films has strengthened the call for good mining practices. A large newspaper in Europe
published a story stating that it is highly likely smartphones contain a few grams of illegal Bangkanese tin
(Hodal, 2012). This pressure led to the December 2013 Indonesia Tin Forum international seminar in
Pangkalpinang, which made more stakeholders aware of the issues. It was also attended by some overseas
smartphone company representatives and local and international consultants.
The TWG, which facilitated the seminar, identified that many of the sustainability issues are derived from the
local and national operating environment and are not directly intrinsic to the business and practices of
mineral exploitation and trade (IDH, 2013). However, the impact of tin mining on the biota is reported to be
a concern for smartphone companies (Nurtjahya et al., 2014).
Thanks to the emerging EU regulation on due diligence in mineral supply chains, including for tin, there is
increasing pressure from national and international NGOs and consumers for producers to supply tin using
standard practices. Responsibly produced raw materials are becoming more important for mineral importers.
Bangka and Belitung Island tin production has been associated with sustainability challenges. These include
critical issues such as mining in protected areas and critical marine ecosystems, the lack of adequate
rehabilitation practices, inadequate occupational health and safety measures and illegal artisanal small-scale
mining (BGR, 2015).
A government body in Germany is concerned about good mining products and supports mineral producers
to improve their social and environmental performance (BGR, 2015). As an advisory institution to the German
government and partner authorities in the mining sector in developing countries, BGR will develop a practical
certification framework to allow semi-industrial operations linked to the artisanal sector in particular to prove
socially and environmentally sound mineral production and encourage their integration into the formal
business sector (BGR, 2015). The framework consists of five basic principles: (1) traceability, (2) fair working
conditions, (3) security and human rights, (4) community development and (5) care for the environment. At
the same time, the framework presents a business opportunity for producers to meet international supply
chain due diligence expectations for traceability and transparency (BGR, 2015). Volunteer tin
producers/smelters could be audited by independent auditors based on the agreed standards (BGR, 2015).

In the last few months, economic growth has slowed. Since 2014, law enforcement related to the tin mining
business has been enhanced. The province should reduce its dependence on tin mining as the first step
towards sustainable development (Tan et al., 2014). The local government may play an important role as a
“conductor” to manage both internal needs and external pressures to hasten the process of achieving good
mining practices and sustainable development. As of 2015, the authority for issuing mining licences is the
governor, not the local government.
The community representation group (LKMTL) withdrew from the Kelian mine closure process in 2003 owing
to some unaccommodated requests from the local community (Nyompe, 2003). Workers and the community
prefer to air their grievances directly with the company because of a lack of confidence in the local
government (McGuire, 2003). The community tried various means to press the company to meet their
demands, including sending their grievances in writing to relevant authorities, the legal aid office and the
National Commission for Human Rights, and holding a series of demonstrations (Nyompe, 2003).
The local government must provide more accurate recommendations to the provincial government for
issuing mining licences. The local and provincial governments together may ask for a higher percentage of
royalties and taxes. Collaborating with other stakeholders such as academics and NGOs, the governments
should enhance efforts to educate people, that is, build awareness of the environment for all ages. Finding
alternative economic drivers is also encouraged. Bringing in skilled agriculture labourers and livestock from
neighbouring islands would accelerate the transfer of technology and agricultural culture to the local people.
Undergraduate students and lecturers from other islands may support mindset changing and transfer some
agricultural knowledge to locals. Tourism investments should be supported, in line with the efforts by the
local Belitung Island government. External influences should be considered motivation. Implementing legal
and sustainable tin supply chains and all aspects of an international audit would increase operators’ and local
people’s compliance with the regulation, and therefore support sustainable mining and sustainable
development. Local and international environmental NGOs may become good partners in monitoring and
evaluating. Their recommendations are needed.
The increase of environmental awareness to empower local people to do business in sectors other than
mining; better cooperation between the local, provincial and central governments on royalties and taxes;
and more contributions from other stakeholders may pave the way to mine closure. Local entrepreneurship
is enhanced in the mine closure process (Permen ESDM No. 7 Tahun, 2014). The local government should
increase the likelihood of establishing sustainable benefits for people beyond mine closure. The only mine
closure program has just been started by a state mine company.
Managing internal needs, external pressures and socio-economic impacts will put law enforcement on the
right track. Correct law enforcement would support and secure the development of socio-economic and
cultural programs. Clearly, socio-economic and cultural development is just one of many mine closure
attributes. Socio-economic and cultural programs, however, would significantly contribute to the success of
mine closure programs in Bangka Belitung.
The local government may play an important role in managing internal needs and external pressures to
hasten the process of implementing good mining practices and sustainable development. The local
government together with the provincial government may ask for a higher percentage of royalties and taxes.
Collaborating with other stakeholders, such as academics and NGOs, education – that is, building awareness
environment – should be enhanced for all ages. Finding alternative economic drivers is also encouraged.
Tourism investment should be supported. External influences should be considered as motivation.
Implementing legal and sustainable tin supply chains and all aspects of international audit would increase
operators’ and local people’s compliance with the regulations, and therefore support sustainable mining and
sustainable development. Local and international environmental NGOs may become good partners for
monitoring and evaluation.

The increase of environmental awareness; empowering local people to do business in sectors other than
mining; better cooperation between local, provincial and central governments on royalties and taxes; and
more contributions from other stakeholders may pave the way to mine closure.
Correct law enforcement would support and secure the development of socio-economic and cultural
programs. The success of that development would provide strong a basis for the success of mine closure in
the province.
The authors gratefully acknowledge the funding of this research by the Directorate General of Higher
Education, Republic of Indonesia (Hibah Bersaing, No. SP2H 092/SP2H/PP/DP2M/III/2007; and No. SP2H
086/SP2H/PP/DP2M/III/2008). Special thanks to the reviewers for their valuable comments on the
manuscript.
Anggrewan, S. (2012) Konflik sosial akibat prostitusi di areal pertambangan timah inkonvensional (TI) Parit 19 Desa Gunung Pelawan
Kecamatan Belinyu, Dissertation, Jurusan Sosiologi, Universitas Bangka Belitung, Pangkalpinang, Indonesia.
Badan Pusat Statistik (BPS) (2012) Bangka Belitung dalam Angka, Pangkalpinang, Indonesia.
Bangka Pos (2013) Setahun, 80 nyawa melayang di lokasi tambang, viewed 27 June 2014, http://bangka.tribunnews.
com/2013/11/20/setahun-80-nyawa-melayang-di-lokasi-tambang.
Bangka Pos (2015) Di Tempilang, Bangka Barat jarang disetor, para ibu serbu kafer remang, 11 March 2015.
Bidayani, E. (2010) Analisis ekonomi dampak kerusakan lingkungan terhadap pendapatan nelayan di pesisir Tanjung Ular Kabupaten
Bangka Barat [tesis], Sekolah Pascasarjana Institut Pertanian Bogor, Bogor, 143 pp.
Bundesanstalt fur Geowissenschaften und Rohstoffe (BGR) (2015) BGR project proposal Bangka Island (Indonesia), 2 p. (unpublished).
Christina, M. (2011) Gerakan sosial terhadap keberadaan kapal isap oleh komunitas nelayan Desa Teluk Limau Kecamatan Parit Tiga
Kabupaten Bangka Barat, Dissertation, Jurusan Sosiologi, Universitas Bangka Belitung, Pangkalpinang, Indonesia.
Erman, E. (2013) Dampak penambangan timah dan respon masyarakat lokal, in Proceeding ITRI Indonesia Tin Forum, 11 December
2013, Pangkalpinang, Indonesia.
Fakultas Ekologi Manusia Institut Pertanian Bogor (FEM IPB) (2013) Laporan Triwulan VII-VIII (August 2012–January 2013) Agraria dan
Reklamasi Berbasis Masyarakat [laporan] Bogor, Indonesia.
Hodal, K. (2012) Death metal: Tin mining in Indonesia, The Guardian, 23 November 2012, viewed 26 February 2014
http://www.theguardian.com/environment/2012/nov/23/tin-mining-indonesia-bangka.
IDH (2013) IDH Tin working group communiqué – To share recent study findings and consult Indonesian stakeholders about their
sustainability views and priorities for action.
Indra, I. (2013) Tambang inkonvensional dan implikasinya terhadap perilaku sosial masyarakat di Desa Tempilang Kabupaten Bangka
Barat, Dissertation, Jurusan Sosiologi, Universitas Bangka Belitung, Pangkalpinang, Indonesia.
International Tin Research Institute (ITRI) (2012), Comments from ITRI: Friends of the Earth campaign “Make it better.”
International Tin Research Institute (ITRI) (2013) Indonesia’s position in the global tin industry, in Proceedings ITRI Indonesia Tin
Forum, 11 December 2013, Pangkalpinang, Indonesia.
Iryanto, R. (2014) Dampak pertambangan inkonvensional terhadap perubahan perilaku pemuda di Desa Bikang, Dissertation, Jurusan
Sosiologi, Universitas Bangka Belitung, Pangkalpinang, Indonesia.
Juniarti, L. (2014) Disharmonisasi keluarga pada pekerja tambang inkonvensional (TI) di Dusun Parit 19 Kecamatan, Dissertation,
Jurusan Sosiologi, Universitas Bangka Belitung, Pangkalpinang, Indonesia.
McGuire, G. (2003) Managing mine closure risks in developing communities – a case study, Kelian Equatorial Mining, Indonesia, in
Proceedings Mining Risk Management Conference, 9-12 September 2003, Sydney, Australia, pp. 1-9.
Mukhtasor, M. (2007) Pencemaran Pesisir dan Laut, Pradnya Paramita, Jakarta, Indonesia.
Muslih, K., Adiwilaga, E.M. and Adiwibowo, S. (2013) Pengaruh penambangan timah terhadap keanekaragaman ikan sungai dan
kearifan local masyarakat di Kabupaten Bangka, Prosiding Pertemuan Ilmiah Tahunan MLI I.
Nurjaman, U. (2013) Pekerja Tambang Inkonvensional (TI) Tertimbun Tanah, viewed 15 April 2013, http://www.babel.polri.go.id/
polres/bangka-selatan/5970.html.
Nurtjahya, E., Agustina, F. and Akbar, A. (2008a) Kajian Manfaat Sosial Ekonomi Penambangan Timah Inkonvensional dan Kerusakan
Lingkungan dan Keanekaragaman Hayati yang Ditimbulkannya di Pulau Bangka, Laporan Penelitian Hibah Bersaing, Januari
2008, Universitas Bangka Belitung, 70 pp.
Nurtjahya, E., Agustina, F. and Putri, W.A.E. (2008b) Neraca Ekologi Penambangan Timah di Pulau Bangka Studi Kasus Pengalihan
Lahan di Ekosistem Darat, Berkala Penelitian Hayati, Vol. 14(1), pp. 29–38.
Nurtjahya, E., Nur, M.M. and Mulyono, E. (2009a) Rice field cultivation on tin-mined land in Bangka Island, Indonesia, in Proceedings
Mine Closure 2009, A.B. Fourie and M. Tibbett (eds), 9-11 September 2009, Perth, Australia, Austra-
lian Centre for Geomechanics, Perth, pp. 549–560.

Nurtjahya, E., Setiadi, D., Guhardja, E., Muhadiono, M. and Setiadi, Y. (2008c) Establishment of four local tree species for potential
revegetating of tin-mined land in Bangka Island, Indonesia, in Proceedings Third International Seminar on Mine Closure, A.
Fourie, M. Tibbett, I. Weiersbye and P. Dye (eds), 14-17 October 2008, Johannesburg, South Africa, Australian Centre for
Geomechanics, Perth, pp. 751–758.
Nurtjahya, E., Setiadi, D., Guhardja, E., Muhadiono, M. and Setiadi, Y. (2009b) Succession on tin-mined land in Bangka Island, Blumea,
Vol. 54(1–3), pp. 131–138.
Nurtjahya, E., Umroh, U. and Agustina, F. (2014) Impact of tin mining on the biota of Bangka Island, Indonesia – a proof to convince
the tin supply chain of smartphone companies, in Proceedings Ninth International Seminar on Mine Closure, 1-3 October
2014, Johannesburg, South Africa, Australian Centre for Geomechanics, Perth.
Nyompe, P.E. (2003) The closure of the Kelian gold mine and the role of the business partnership for development/World Bank, paper
presented at the meeting on Indigenous Peoples, Extractive Industries and the World Bank, 15 April 2003, Oxford, England.
Octarini, S. (2011) Analisis bioekonomi dampak aktivitas tambang inkonvensional (TI) apung terhadap produksi tangkapan nelayan
Batu Belubang Bangka Tengah, Dissertation, Jurusan Perikanan Universitas Bangka Belitung, Pangkalpinang, Indonesia.
Permen ESDM (2014) Peraturan Menteri Energi dan Sumber Daya Mineral Republik Indonesia Nomor 07 Tahun 2014.
Pratama, P. (2014) Alih profesi nelayan menjadi penambang timah di Lingkungan Nelayan 2 Sungailiat, Dissertation, Jurusan Sosiologi,
Universitas Bangka Belitung, Pangkalpinang, Indonesia.
PT Timah Tbk. (1997) ANDAL, RKL, dan RPL. Kegiatan Penambangan Timah dan Pasir Laut di Perairan P. Bangka Kabupaten Bangka,
Propinsi Sumatera Selatan.
Romeo, B. (2011) Tambang inkonvensional dan perubahan perilaku ekonomi masyarakat Desa Semulut Kecamatan Jebus,
Dissertation, Jurusan Sosiologi, Universitas Bangka Belitung, Pangkalpinang, Indonesia.
Sodikin, H. (2011) Pertumbuhan Acropora digitata pada transplantasi karang di pantai Teluk Limau Sungailiat Provinsi Kepulauan
Bangka Belitung, Dissertation, Jurusan Perikanan Universitas Bangka Belitung, Pangkalpinang, Indonesia.
Sucita, E.Y. (2011) Analisis bioekonomi dampak aktivitas penambangan timah terhadap produksi hasil tangkapan nelayan di Desa
Rebo Kabupaten Bangka, Dissertation, Jurusan Perikanan Universitas Bangka Belitung, Pangkalpinang, Indonesia.
Tan, K.G., Amri, M., Low, L., Tan, K.Y., Merdikawati, N., Annisa, R., Nugroho, B.A., Eman, E.F., Pratama, F.S.B., Kusuma, R.R., Novrianto,
R., Rachman, F., Rianto, F. and Siregar, F.K. (2014) Analisis daya saing dan strategi pembangunan untuk 33 provinsi Indonesia,
Makalah Kepulauan Bangka Belitung, ACI Policy Paper Series No. 16/2014, National University of Singapore, Singapore.
Wahyuni, H., Sasongko, S.B. and Sasongko, D.P. (2013) Kandungan logam berat pada air, sedimen dan plankton di daerah
penambangan masyarakat Desa Batu Belubang Kabupaten Bangka Tengah, in Proceedings Seminar Nasional Pengelolaan
Sumberdaya Alam dan Lingkungan 2013, pp. 489–494.
Walhi (2013) 50 persen terumbu karang Babel rusak, viewed 24 June 2014, http://www.antaranews.com/berita/396700/50-persen-
terumbu-karang-babel-rusak.
Walhi – Friends of the Earth Indonesia (2013) Mari buat Bangka Belitung lebih baik, in Proceedings ITRI Indonesia Tin Forum, 11
December 2013, Pangkalpinang, Indonesia.
Yulianda, F., Achmad, F., Armin, A.H., Sri, H. and Kusharjani, K. (2009) Ekologi Ekosistem Perairan Laut Tropis, PUSDIKLAT Kehutanan
– Departemen Kehutanan RI, SECEM – Korea International Cooperation Agency, Bogor, Indonesia.
Zulkarnain, I., Erman, E., Pudjiastuti, T.N. and Mulyaningsih, Y. (2005) Konflik di Kawasan Pertambangan Timah Bangka Belitung:
Persoalan dan Alternatif Solusi, LIPI Press, Jakarta, Indonesia, 193 pp.

M.P. Dobrowolski Iluka Resources Ltd and The University of Western Australia, Australia
S.A. Rowlands Iluka Resources Ltd, Australia
R. Hattingh Iluka Resources Ltd, Australia
Tailings produced from the density separation of mineral sands from clay, silt, and sand materials can remain
geotechnically unstable for some time due to their high water content. This is especially true when tailings
are disposed in mining voids that restrict free drainage, since evaporative drying is limited to the top layer of
the tailings. Solar/evaporative drying of the saturated tailings can be effective, depending on the climate, but
this requires costly earthworks and disturbs a larger area. Using deep-rooted vegetation in this biologically
benign material may offer a cost-effective alternative to accelerate drying tailings at depth to achieve the
desired geotechnical stability.
Two trials are being undertaken in historic tailings disposal areas on mineral sands mines in Southwest
Australia, a region with high potential evapotranspiration during summer. The tailings disposal methods at
the two sites differed: one was mainly fines (clay and silt-sized particles), and the other was a mixture of fines
and sand tailings stabilised by flocculant. Both disposal methods have resulted in saturated tailings at depth,
which may be geotechnically unstable in the medium term. The local groundwater level at both sites remains
below the saturated tailings, making it possible for deep-rooted vegetation to dry a substantial depth of
tailings without the tailings being recharged from surrounding groundwater. At the first trial, where the
tailings have been covered with overburden and topsoil, trees chosen for their waterlogging tolerance,
Eucalyptus camaldulensis, Melaleuca rhaphiophylla, and Casuarina obesa, were densely planted in plots. The
soil profile within and outside of the plots is being monitored for changes in soil water content by measuring
neutron thermalisation. Water content acts as a proxy for geotechnical stability, and given the accuracy of
this method, the long-term water balance of the tailings can be assessed separately from the seasonal
changes expected with predictable winter rainfall recharge and summer drought. After 9 months of growth,
trees at this trial are drawing water nearly 2 m deep at the interface of the overburden and tailings material.
The second trial, which is being established, will monitor changes in geotechnical stability using direct cone
penetrometry, allowing greater spatial assessment of stability changes than the fixed access points used with
neutron thermalisation. This trial will test similar species’ ability to dewater tailings of higher fines content at
a site with a similar Mediterranean climate but lower rainfall; the aim here is to test the general applicability
of vegetation to penetrate roots into tailings for effective dewatering. This paper presents a detailed
rationale, design, and preliminary results of these trials aiming to improve the geotechnical stability of
mineral sands tailings.
The mining of mineral sands produces tailings from the wet concentration step: the separation of the heavy
mineral sands from lighter quartz and clays. These lighter quartz and clay materials represent the bulk of the
mined ore body. They are often disposed in the mining void as separate streams of sand tails and clay fines.
Alternatively, the fines are mixed with flocculant, thickened, and then remixed with sand tails for disposal in
the void. Clay fines retain more water and drain slower than sand tails, especially when their drainage is
restricted in mine voids. This high water content can cause geotechnical instability, which is problematic for
both capping of the material and future land use.

Improving geotechnical stability of saturated mine tailings by dewatering with vegetation M.P. Dobrowolski, S.A. Rowlands and R. Hattingh
There are a number of ways to address geotechnical instability caused by the high water content of tailings.
Over time, natural consolidation will occur with overburden pressure and solar/evaporative drying in
climates with net annual evaporation. Rapid drying is usually achieved by using earthworks, such as
excavation and solar drying of thin layers of tailings, or by applying earth to increase overburden pressure,
which in turn expresses the water. Engineering methods such as wicking may also be applicable. Earthwork
and engineering methods, however, are costly. Taking advantage of deep-rooted vegetation to act as
transpiration pumps may offer a cost-effective alternative to more rapidly dry tailings at depth to achieve
the desired geotechnical stability.
To achieve drying of the profile with deep-rooted vegetation may require a combination of well adapted
plant species, suitable physical and chemical tailings characteristics, and a favourable climate. The species
should ideally be tolerant of waterlogging to the initial saturated condition and/or capable of rooting into
the phreatic zone and should have a high water use. The tailings should hold sufficient water available to
plants, generally considered in the range from −10 kPa to −1,500 kPa, and should achieve sufficient strength
at the residual (−1,500 kPa) water content, once plant roots have removed that available water. Chemical
characteristics, such as pH, salinity, and plant nutrient content should not limit plant growth. Finally, a
subhumid climate, especially one where the season of maximum precipitation is out of phase with the season
of highest evaporative demand, will allow net depletion of water from the tailings profile.
Examples of revegetation of mine tailings are more commonly termed “phytoremediation” and are
concerned with overcoming or treating chemical limitations or contaminants, or with capping and covering
tailings to prevent downward infiltration and leaching of the material (e.g., Robinson and Anderson, 2007).
Examples of revegetation of tailings exclusively for the purpose of dewatering to achieve geotechnical
stability, however, are not unknown (e.g., Silva, 1999; Wu, 2009).
The trials described in this paper aim to test the utility of using deep-rooted vegetation to dewater mineral
sands tailings sufficiently to achieve geotechnical stability. Early indications suggest that, at one site, the trees
established on a 1.5–2 m soil cover above wet codisposed tailings have roots that are drawing water from
the saturated layer. At a second site with tailings consisting of clay fines, soil strength is likely to increase
through similar plant root penetration and transpirational drying to sufficient depth to allow trafficability.
The magnitude and/or rate of these effects cannot yet be estimated.
Located in Southwest Australia, the two mine sites at which the trials have been established are at Gingin
and Eneabba, 70 km and 240 km north of Perth, Western Australia, respectively. These sites experience a
subhumid, Mediterranean climate with mild wet winters and dry hot summers (Figure 1). Annual
evaporation, in excess of 2,000 mm for both sites, far exceeds the annual precipitation of 650 mm at Gingin
and 490 mm at Eneabba. This climate makes drying of tailings possible given favourable site conditions.

Vegetation
These two Iluka Resources mines are currently being rehabilitated. Their wet tailings differed substantially,
which offered a test of the general applicability of using deep-rooted vegetation to dry wet tailings to achieve
geotechnical stability.
At Gingin, the tailings consisted of a mixture of clay fines (clay- and silt-sized particles), sand tails, and
flocculant that were placed in mine pits and covered with a total depth of 1.5–2 m of overburden (deeper
soil material), subsoil, and topsoil. Observations of bow waves in the tailings material while it was being
covered by heavy machinery and the expression of water to the surface through the cover layers indicated
the tailings remained wet. A hydrological model of the mine site based on local piezometers indicated that
the tailings deposited in pits were likely to sit above the local groundwater table in most locations; therefore,
water within the tailings would not be replenished by groundwater if it was removed by deep-rooted
vegetation.
At Eneabba, the tailings were unmixed clay fines that were placed in a mine pit and remained in an uncovered
tailings dam for solar drying. A particle size analysis indicated the tailings’ properties were dominated by clay-
sized particles: 55% clay (< 2 μm), 6% silt (2–20 μm), and 39% sand (> 20 μm). Large polygonal cracks had
formed in the tailings surface, which filled with water over winter, recharging the tailings profile (Figure 2).
They were not trafficable by heavy earthmoving equipment for further rework. As at Gingin, the base of the
tailings dam was above the local groundwater of the region.

At Gingin, three locations (replicates) over former mine pits were chosen for the establishment of tree plots.
At each location, three circular plots, 8 m in radius, were densely planted with seedlings of Eucalyptus
camaldulensis (river red gum), Melaleuca rhaphiophylla (swamp paperbark), or Casuarina obesa (swamp she-
oak) in June 2013 (Figure 2). The choice of these three plant species was driven by their known waterlogging
tolerance and local adaptation as well as the desire to test the species’ characteristics in practice: their
establishment success, growth rate, ability to root at depth, and water use. Plants were supplemented with
fertiliser tablets to ensure growth was not limited by poor soil nutrient conditions.
Similarly at Eneabba, two replicates of 20 m square plots of each of E. camaldulensis and M. rhaphiophylla
were densely planted on the clay fines tailings dam, directly in the tailings material, with fertiliser
supplementation, in July 2014. Since no topsoil cover had been applied at this site, the soil was tested prior
to planting to identify likely plant nutritional limitations of the tailings (clay fines) material. This testing
indicated N, P, Mn, Cu, and Mo would be limiting for plant growth but could be supplied with a general
fertiliser application.
At Gingin, dewatering of the tailings was measured using neutron thermalisation because of this method’s
accuracy in measuring changes in water content within a soil profile (Gardner, 1986). Aluminium access
tubes, 3 m long and sealed at the base, were inserted into spiral augered holes in the soil profile. They were
located inside, on the edge, and outside the circular plots of trees, as illustrated in Figure 3. The spatial
arrangement of access tubes was designed to measure soil profile drying in the presence and absence of tree
growth, allowing for expected spatial variation in the soil profile. A total of 16 access tubes was inserted at
each of the three locations over former mine pits. Neutron thermalisation was monitored at one- to two-
month intervals once the plants were established, using a Hydroprobe® 503DR-1.5 Direct Reading Neutron
Moisture Probe (from CPN International Inc. in Martinez CA, USA). Counts were recorded at 30 cm intervals
for 30 s each. No calibration for volumetric water content was done because the changes in count ratio
accurately describe changes in water content within the profile. All access tubes were surveyed to an
accuracy of ± 30 mm to allow settling/consolidation of the profile to be monitored over time.
At Eneabba, rather than assessing dewatering, researchers measured soil strength by taking soil strength
soundings to a depth of 4 m into the soil profile, using a PANDA® (from Sol Solution in Riom, France), a
variable energy lightweight dynamic cone penetrometer (DCP) (see Figure 2). Cone resistance measured by
a DCP is correlated with water content; more importantly, it is a direct measure of geotechnical stability, the
ultimate aim.

Vegetation
Although the aims were to test the possibility that deep-rooted vegetation could dry wet tailings produced
from mineral sands mining, and if possible, to estimate the rate at which this could be achieved, it was clear
from initial observations that plant establishment was an obstacle. During the first summer at Gingin, almost
all trees died from drought at two locations, probably because they were unable to root sufficiently deep
through the soil cover above the wet tailings. At the third location at Gingin, also the wettest, most trees
survived or thrived; the E. camaldulensis plot at this location grew over 2 m tall in the first 12 months. In
contrast, at the hotter and drier Eneabba site, E. camaldulensis seedlings died and M. rhaphiophylla seedlings
survived in the clay fines tailings, which was initially more waterlogged than the Gingin soil profile.
Regardless of plant establishment difficulties at Gingin, observations of the soil profile water content
indicated that the roots of the trees were penetrating the tailings material within one or two years and
drawing on the entrained water. Substantial drying of the profile to almost 2 m deep was observed in year 1;
spring 2013 to autumn 2014 neutron thermalisation measurements are illustrated in Figure 4 by arrows
indicating the E. camaldulensis plot in which trees grew over 2 m tall. Comparing these access tubes inside
and on the edge of a tree plot to those outside the plot indicates that soil drying due to the grass and
herbaceous cover only occurs to a depth of 1 m. In the neutron thermalisation measurements at this location
in year 2 (spring 2014 to summer 2015), the soil profiles were completely recharged with winter rainfall, but
drying to greater than 2 m depth was apparent, and to a greater degree, at the same access tubes as year 1.
Additional access tubes showed drying at depth both in the E. camaldulensis plot and to a small degree in
the M. rhaphiophylla plot, as indicated by arrows in Figure 4. This location is being monitored, and trees at
the other two locations are being re-established to determine whether long-term drying at depth can be
achieved in this soil/tailings profile.
At the Eneabba site plant, growth in the first year was too small to expect DCP measurement to detect
differences in cone resistance between vegetated and unvegetated areas on the tailings dam. Initial
soundings were made 4 m deep using the DCP; these soundings indicated spatial variability in soil strength
related to the large polygonal cracking occurring on the surface of the tailings (data not shown). Changes in
soil strength down the tailings profile will be monitored over time as the tree canopy grows and transpires
the tailings water.
A single, small (1.5 m) Melaleuca tree had naturally established on the tailings dam some distance from the
trial plots and offered an opportunity to measure soil strength adjacent to and further away from it. Cone
resistances in soundings taken at 1–3.5 m on either side of the established tree were in the range of 4–
10 MPa at least 2 m deep in the tailings and up to 4 m deep, whereas soundings taken on either side at 7.5 m
and 8 m away were much weaker, dropping and staying below 3 MPa at depths > 0.3 m (Figure 5). Although
this is only associative data and not experimental observation, the increase in soil strength close to an
established tree in the tailings would be expected if the roots were penetrating and drawing water at depth.
If this result is repeated in the experimental plots, strengthening of the tailings to considerable depth (> 4 m)
with tree dewatering would appear to be possible at this site.


Vegetation
The DCP measurements of soil strength also indicate another factor critical to the success of dewatering to
depth: the penetrability of the tailings by plant roots. Generally, soils with a penetrometer resistance of
2 MPa will limit root growth by mechanical impedance unless a network of fissures exists for roots to exploit
(Bengough et al., 2006). This can occur in relatively moist soils with matric potentials of −100 kPa to −200 kPa
(Bengough et al., 2011). The DCP measurements taken in these tailings show that potentially root-limiting
soil strength (> 2 MPa) is commonly encountered in relatively wet material, so root penetration could well
depend on initial drying and cracking of the tailings.
Once the initial hurdle of establishing trees on tailings is overcome, monitoring of the soil water content
through a covered deposit of mineral sands tailings indicates that water removal is being achieved to a depth
of 2 m, although this is fully recharged by winter rainfall. The rate of dewatering using deep-rooted vegetation
cannot yet be estimated from these initial results. At a second uncovered tailings dam, trees have not grown
sufficiently to lead to any detectable increase in soil strength as a result of transpirational dewatering of the
tailings; however, observations on a naturally established tree indicate that improved geotechnical stability
to a considerable depth (> 4 m) may be possible.
We wish to acknowledge the assistance of Hughan McVee and Cameron Payne in establishing and
maintaining the field trials, Andrew Nelson for hydrological interpretation, Daniel Byrne for mine pit and site
survey interpretation, and Kerri Conder for ongoing monitoring.

Australian Government (2015) Bureau of Meteorology, viewed 14 January 2015, www.bom.gov.au.

Bengough, A.G., Bransby, M.F., Hans, J., McKenna, S.J., Roberts, T.J., and Valentine, T.A. (2006) Root responses to soil physical
conditions: growth dynamics from field to cell, Journal of Experimental Botany, Vol. 57, pp. 437–447.
Bengough, A.G., McKenzie, B.M., Hallett, P.D., and Valentine, T.A. (2011) Root elongation, water stress, and mechanical impedance:
a review of limiting stresses and beneficial root tip traits, Journal of Experimental Botany, Vol. 62(1), pp. 59–68.
Gardner, W.H. (1986) Water content, in Methods of soil analysis, part 1 – physical and mineralogical methods, 2nd edition, A. Klute
(ed), American Society of Agronomy Inc. and Soil Science Society of America Inc., Madison, Wisconsin, USA, pp. 493–544.
Robinson, B. and Anderson C. (2007) Phytoremediation in New Zealand and Australia, Methods in Biotechnology, Vol. 23,
Phytoremediation: methods and reviews, N. Willey (ed), Humana Press Inc., Totowa, NJ, USA, pp. 455–468.
Silva, M.J. (1999) Plant dewatering and strengthening of mine waste tailings, PhD thesis, University of Alberta, Edmonton, Alberta,
Canada, 293 pp.
Wu, S. (2009) A greenhouse study of selected native plant species for dewatering CT, Master of Science thesis, University of Alberta,
Edmonton, Alberta, Canada, 135 pp.

S. Ebbs Southern Illinois University, USA

L. Glaeser Southern Illinois University, USA
M. House Southern Illinois University, USA
D. Vitt Southern Illinois University, USA
The Canadian oil sands industry is actively engaged in the revegetation and reclamation of sites associated
with bitumen mining. One current reclamation research effort sponsored by Syncrude Canada Ltd., referred
to as the Sandhill Fen Research Watershed (SFRW) project, is taking place on a former open pit mine in
northern Alberta. One aspect of this project involves the establishment of a wetland on the former mine pit,
which has since been back filled with soft tailings. A component of the active research taking place at the
SFRW is the evaluation of using plant ecophysiological parameters as markers of success. The parameters
being used include rates of photosynthesis, transpiration, stomatal conductance, and additional derived
parameters (e.g., water use efficiency). These ecophysiological measures provide quantitative information on
the physiological status of the plants and a potential assessment tool for plant performance during the early
stages of reclamation. Additionally, this approach could provide insight into the trajectory of the plants in the
reclaimed vegetation community, and the long term potential success/stability of the reclamation.
Comparable data from the same plant species growing on disturbed and undisturbed benchmark sites is being
collected. These sites can provide a frame of reference for predicting reclamation trajectories, as well as
expectations for plant physiological performance as a function of time. Ecophysiological data collected one
year after the initiation of the project demonstrated that plants in the wetland were showing similar
performance to plants on the benchmark sites, suggesting that the reclamation was on a comparable
ecological trajectory to those reference sites. For example, an array of ecophysiological parameters was
monitored for water sedge (Carex aquatilis), a key species planted in the pilot wetland. Photosynthesis rates
for water sedge in the Sandhill Fen pilot ranged from 2.2 – 20.5 moles CO2 m-2 min-1 as compared to values
ranging from 1.4 – 14.2 moles CO2 m-2 min-1 for plants on the benchmark sites. Water use efficiency values
ranged from 0.9 to 6.1 for water sedge in the pilot site as compared to 0.6 to 4.8 on the benchmark sites.
Across the SFRW, there is a gradient of soil moisture from dry to inundated. Plant performance relative to this
moisture was monitored closely to offer guidance on how water levels influence plant establishment. The
results thus far illustrate the value of this ecophysiological approach in providing markers of success for
wetland reclamation efforts. The results also suggest that this same approach might be applicable to other
mine closure scenarios, providing a means to relate the ecological trajectory of reclamation to relevant
benchmark sites.
During mine site reclamation, natural vegetation communities are commonly used as references.
Appropriate reference communities can represent a general or specific assemblage of plant species or a
community that provides one or more desired ecosystem functions. The progress of the reclaimed vegetation
community over time can be compared to the natural reference community to assess whether the
reclamation is moving on a trajectory towards the reference community structure or on an altered trajectory
(White and Walker, 1997). Reference communities have been used in a variety of ecological contexts,
including forest, grassland, and wetland restoration. Reference communities have also been used to assist in

A plant ecophysiological approach to assess the performance and potential success of mine revegetation S. Ebbs et al.
recent mine closure and/or revegetation efforts (e.g., Gravina et al., 2011; Rooney and Bayley, 2011; Vickers
et al., 2012). The progress and potential success of the reclamation relative to the reference community has
been tracked by monitoring parameters such as species richness, species density, or net primary production.
Often the potential stability and trajectory of the restored community does not become evident for several
years. Early markers of success may be measureable in terms of structural aspects of the plant community,
such as the recruitment of foundation species, survival of naturally recruited or introduced plants, percent
cover achieved by the community or indices such as the community structure integrity or higher abundance
indices (Jaunatre et al., 2013). Another approach to evaluate the early progress of a reclamation effort is to
take a functional approach by measuring ecophysiological parameters of target or representative plant
species to assess their performance relative to the same species in a reference community. This
ecophysiological approach has been used to monitor establishment and performance of native, introduced,
or exotic plants is some ecological restorations (e.g., Lambert et al., 2011; Ostertag et al., 2008; Rovai et al.,
2013), but has not been applied as widely during mine closure reclamation scenarios.
The extraction of mineable surface deposits of bitumen from the oil sands in Alberta can take place through
open pit mining. When the bitumen deposit is exhausted, the resulting pit (referred to as an in-pit) is often
utilised for tailings storage and backfilled with mine waste materials such as salty sand (Wytrykush et al.,
2012). The current government mandate to the oil sands industry (Alberta Environment, 1999) is that the
mined areas must be replaced with self-sustaining and maintenance-free ecosystems that provide land
capabilities equivalent or superior to the pre-disturbance conditions. Those ecosystems should consist of an
integrated landscape of uplands and wetlands. Pre-disturbance land capabilities include ecosystems services
such as wildlife habitat, wetland and watershed functions, sources of timber, traditional foods, medicinal
plants, or recreational use (Oil Sands Reclamation Committee, 1998). Achieving these endpoints have been
described as the engineering of new ecosystems over a time scale of >100 years (Johnson and Myanishi,
2008). There are still many years of oil sands extraction planned for most mining leases, and therefore mine
closure has not been completed to date in Alberta’s oil sands industry. The industry is, however, executing
progressive reclamation. Many operational scale revegetation efforts have been completed, for example
Syncrude Canada Ltd has over 3000 hectares of permanently reclaimed uplands. Wetland reclamation is
relatively new and has only been attempted at the pilot or small scale.
Syncrude Canada, Ltd. initiated an experimental watershed in 2008, referred to as the “Sandhill Fen Research
Watershed”, to establish an integrated upland-wetland landscape on a former open pit mine. The total area
of this watershed is 52 ha, 17 ha of which is a central wetland (Sandhill Fen). The project aimed to understand
the initial conditions needed for a stable wetland to develop over time, particularly if the trajectory of
establishment leads to a boreal rich fen. Rich fens have a relatively high abundance in the natural northern
Alberta boreal region were Syncrude operations are located. Rich fens are dominated by true mosses and
include a diversity of vascular plant species (Vitt and Chee, 1990). Establishment of a fen represents a
particular challenge as there have been no prior efforts to create a peat-forming wetland. Examination of
cores from existing fens in the region indicate that the presence of vascular plants are critical to the early
establishment of these peat-forming wetlands, most likely as nurse plants to facilitate the establishment of
the underlying moss layers. To aid in the development of future revegetation plans for soft tailings
reclamation, experimental plots were established in the experimental wetland in order to evaluate the
growth and performance of an array of fen vascular plant species. One year after the plots were established,
the performances of three species were evaluated across the wetland area using ecophysiological
parameters. Reference communities in northern Alberta for each plant species were also identified and
included sites ranging from either recently disturbed or have been undisturbed for millennia. The same
ecophysiological parameters were measured to assess performance of plants in the experimental watershed
relative to those from the reference sites. Another reference site used was a nearby, smaller scale
reclamation effort.
The first objective of the research was to establish the range of values for each ecophysiological parameter
across the reference sites and the range of values across a moisture gradient that developed on Sandhill Fen.

Vegetation
The second objective was to compare the ranges of values obtained to assess the physiological performance
of the species in the experimental watershed to the reference sites. The final objective was to demonstrate
the utility of these ecophysiological parameters for assessing the early performance of plants in a wetland
reclamation scenario. The more general application of this monitoring approach to the success of mine
closure and revegetation efforts is discussed.
A complete description of the construction of the Sandhill Fen Research Watershed (SFRW) can be found in
Wytrykush et al. (2012). The Sandhill Fen Research watershed project is located in the north section of East
In-pit on Syncrude Canada's oil sands lease. The former mine was used as a tailings storage facility and filled
with composite tailings. Before construction of the Sandhill Fen research project the site location was a
relatively flat, soft tailings sand feature. Above-grade landform construction components of the watershed
(hummocks) were mechanically constructed with heavy equipment. Construction started in the winter of
2009 and was completed in the fall of 2012 and summer of 2013. The watershed consists of seven small
upland hills on the northern and southern borders with water flowing into a central wetland (i.e., “Sandhill
Fen”) approximately 800 m in length and 350 m in width at its widest point (Figure 1). For the central wetland,
approximately 0.5 m of peat was spread over the 17 ha area. The peat was collected from a peatland complex
on the oil sands lease in the fall/winter of 2010 and stored in piles until fall/winter2011, at which point it was
spread across the wetland area. Mixed with the peat were small amounts (<5% w/w) of the underlying clay
mineral soil beneath the peat deposit. Some physicochemical parameters for the peat are shown in Table 1.
Some initial water for the wetland was pumped from a nearby natural lake (Mildred Lake) into a small holding
pond separated from the wetland by a rocky dam. Water was released into the wetland area through the
dam throughout the summer and fall of 2012 and 2013.

Parameter
Bulk density (g cm-3) 0.25 ± 0.06

Organic matter (%) 36.5 ± 6.4
Moisture content (%) 69.0 ± 10.1
pH (substrate) 5.3 ± 0.3
pH (pore water) 7.0 ± 0.4
Carbon (%) 28.5 ± 9.3
Nitrogen (%) 1.4 ± 0.4
C:N ratio 20.5 ± 2.8
Extractable NO3--N (g g-1, 1 M KCl extract) 4.2 ± 0.9
Extractable NH4+-N (g g-1, 1 M KCl extract) 39.3 ± 5.8
Extractable PO43--P (g g-1, Bray extractable P1) 33.8 ± 3.1
During spring 2012, experimental plots were established within the wetland area (Figure 1). Twenty research
plots were positioned across the fen; 16 are located in the main fen watershed and four plots (referred to
hereafter as the perched plots) were positioned on one of the southern upland hills approximately 25 m
above the wetland. All 20 plots were positioned to span the environmental gradients of moisture that were
expected in the watershed. The perched plots were isolated from the fen watershed to create a separate
hydrology and moisture regime. The 20 research plots are 10 m x 10 m and divided into four equal quadrants.
Within one of the quadrants, monocultures of native wetland species were planted in 1 m x 1 m subplots.
Three of the species established in those subplots were examined here: water sedge (Carex aquatilis), black-
girdle bulrush (Scirpus atrocinctus), and seaside arrowgrass (Triglochin maritima). The first two species are
monocotyledonous sedges while seaside arrowgrass is also monocotyledonous but a member of its own
taxonomic family (Juncaginaceae), closely related to the Liliaceae. All three species are indigenous to fens
and marshes of Alberta. All three plant species utilise the C3 photosynthetic pathway (Bruhl and Wilson,
2007; Wang et al., 2006). The plants were grown in 2011 at the Smoky Lake Forest Nursery (Smoky Lake, AB)
using seed collected from fens across northern Alberta. Seed was germinated in standard nursery trays in a
peat-mineral soil mix watered as needed. The plants were grown under ambient greenhouse conditions and
then transplanted in spring 2012 as plugs into the subplots at a density of twenty-four plants m-2 for Scirpus
atrocinctus and Triglochin maritima and seven plants m-2 for Carex aquatilis seedlings. The plugs were planted
directly into the peat layer.
By summer 2013, at the initiation of the study reported here, there was 100% survival of each of the three
species across the 20 plots. The wetting of the fen was not uniform or continuous, but four discrete categories
of soil moisture could be distinguished. There were five plots that had been under continuous standing water
of at least 10 cm (inundated plots). An additional five plots were continually saturated with water at or within
10 cm below the substrate surface but had not been inundated (saturated plots). A total of six plots were
well hydrated with water due to past inundation but at the time the measurements were made were not
fully saturated (soil moisture >50%, unsaturated plots). The four perched fens received water predominantly
from snow melt, but augmented with some water pumped to these areas. These plots (perched plots) also

Vegetation
experienced variable levels of soil hydration as for the unsaturated plots but at the time of measurement
were more similar to upland conditions than the wetland conditions in the watershed.
The plants in plots from the Sandhill Fen were compared to plants growing on naturally occurring reference
communities in northern Alberta, each with varying ages of disturbance.
The Anzac sites include three sites at a roadside locality disturbed in 2006 owing to road expansion. Anzac
Roadside 1 (“Anzac ditch”): A poorly drained ditch on organic soil with abundant sedges (56°22.56’N,
111°01.32’W). Sampled: Water sedge (C. aquatilis).
Anzac Roadside 2 (“Anzac mineral”): A site with wet mineral soil and abundant mosses and graminoids
(56°22.59’N, 111°01.17’W). Sampled: Water sedge (C. aquatilis).
Anzac Roadside 3 (“Anzac fen”): A fen with populations of Sphagnum angustifolium and various sedges
(56°22.57’N, 111°01.31’W). Sampled: Water sedge (C. aquatilis).
Along the access road to Engstrom Lake is a poorly drained wetland with sedges. The most recent disturbance
to this area is unknown, but it must have occurred within the past 30 years when construction began for
access roads to nearby recreational areas (56°12.02’N, 110°53.47’W). Sampled: black-girdle bulrush (S.
atrocinctus).
An undisturbed minerotrophic wetland is located near an access road to the east of Mariana Lake. No
disturbance to this area is evident (55°54.01’N, 112°03.88’W). Sampled: black-girdle bulrush (S. atrocinctus).
This reference site is a mature open sedge/moss-dominated rich fen with abundant scorpidium moss
(Scorpidium scorpioides). No disturbance to this area is evident (56°57.04’N, 111°31.82’W). Sampled: Seaside
arrowgrass (T. maritima).
The U-cell Experimental Area is located on the Syncrude Canada Ltd. production facility and served in this
study as reference reclamation plots. U-cell is a set of 28 cells (10 x 20 m) with different substrate and water
chemistry treatments (Wytrykush et al., 2012). In 2008, a variety of water and substrate treatments were
established. These included a set of 14 cells filled with stockpiled peat at 15, 50, and 100 cm depth. Seedlings
of a variety of wetland plants propagated as described above were established in the cells and their survival
and growth monitored. Some cells received fresh water while other cells received mine process water. In
2010, process waters in U-cell treatments had electrical conductivities ranging from 388 to 1,360 µS cm-1
(Wytrykush et al., 2012). Individuals of water sedge, black-girdle bulrush, and seaside arrowgrass were
among the species planted in the cells. Measurements were taken from cells with large populations of these
species. (57° 03'01.00" N, 111°39'20.33"W). Sampled: Water sedge (C. aquatilis), on peat substrate with fresh
water from 2009-2011, mine process water in 2012, and rain-fed in 2013-2014 (Cell 8); Black-girdle bulrush
(S. atrocinctus) on peat substrate receiving fresh water until it was rain fed in 2013 to 2014 (Cell 6): Seaside
arrowgrass (T. maritima) on mineral substrate receiving fresh water from 2009 to 2010, process water from
2011-2012, and rain fed in 2013-2014 (Cell 16). U-Cell was included as a reference site to illustrate a recent
reclamation scenario with highly variable conditions over the years since establishment.

Ecophysiological parameters were measured for the three plant species using a CI-340 portable
photosynthesis system (CID Bioscience, Camas, WA) under ambient humidity, temperature, and light
intensity using an open system configuration with a 6.25 cm2 flat rectangular chamber. All measurements of
plants on the Sandhill watershed and the reference sites were made in summer 2013 between 9 AM and
noon from mid-June to the end of July (18 June – 31 July). Leaves selected for measurement were fully
expanded and free from damage. Once enclosed in the chamber and after an acclimation period of 5 min,
photosynthetic rate, evapotranspiration, stomatal conductance, and photosynthetically active radiation
(PAR) were measured. Measurements were taken on five plants from each plot on Sandhill Fen and from
each reference site. All three parameters are common indicators of plant physiological status. Photosynthesis
and evapotranspiration are highly responsive parameters and their rate drops quickly if plants are
experiencing stress. Stomatal conductance is an indicator of the degree to which the stomatal pores on the
leaves are open to allow for gas exchange and evapotranspiration. The diameter of the pore can be
biologically controlled to balance the entry of CO2 for photosynthesis against the loss of water. Stomatal
conductance is highest when plants have fully open stomates and an adequate supply of water and decreases
under water-limiting conditions or in response to other stresses.
The data were subjected to a one-way ANOVA analysis with Tukey’s HSD used to provide post hoc
comparisons. In addition to the ecophysiological parameters measured directly, two additional parameters
were derived from the data recorded. Radiation use efficiency (RUE), a representation of the efficiency to
which the light available was used to drive photosynthesis, was calculated as the ratio of photosynthetic rate
to photosynthetically active radiation (Sinclair and Horie, 1989). Water use efficiency (WUE) is a common
parameter used to relate the amount of water expended per unit of photosynthesis achieved and was
calculated as the ratio of photosynthetic rate to evapotranspiration.
Water sedge (Carex aquatilis) has a wide distribution across Canada and the United States (south to Arizona,
Missouri, and Virginia) and has been classified as a native and obligate wetland plant (FNA, 2002). This species
can be found throughout the boreal wetlands of Canada. Water sedge is also resilient to disturbance and can
naturally colonise disturbed soils, road side ditches, and mining pads (Koropchak et al., 2012). Water sedge
has been included in a variety of reclamation and revegetation efforts, including revegetation efforts on oil
pads and mined peatlands (Vitt et al., 2011). The plants are capable of growing on mineral soils although the
species shows more robust growth on organic soils and under wetland conditions (Koropchak et al., 2012).
In the study here, the rate of photosynthesis by water sedge was significantly lower on the mineral soil from
the Anzac site as compared to the undisturbed Mariana Lake wetland site, but did not differ significantly from
any of the other reference sites (Figure 2). The rate of evapotranspiration on the mineral soil differed
significantly only from the U-cell site and there was no significant difference across the reference sites for
stomatal conductance. Aside from the differences displayed by plants from the mineral soil, there was no
significant difference in any of these parameters as a function of time from the disturbed sites to the
undisturbed Mariana Lake site. The range of values for the three ecophysiological parameters across all the
reference sites was no more than three-fold. The water sedge plants in the Sandhill Fen experienced a range
of soil moisture regimes from inundated plots to the more mesic terrestrial conditions of the perched fens
and the unsaturated plots, but did not show any significant differences in the parameters as a function of soil
moisture. More importantly, after only one year of growth in the Sandhill Fen plots, the only significant
differences observed were between the Sandhill Fen plots and the Anzac mineral soil and this was limited to
rates of photosynthesis and evapotranspiration for the wetter Sandhill Fen plots. The values for three

Vegetation
parameters across this moisture gradient did not differ significantly from the organic substrates from any of
the reference sites. This demonstrates that within one year of planting, the water sedge plants were
performing as well as plants on the other reference sites with organic substrates, including the undisturbed
Mariana Lake site. This was also evident when the individual ecophysiological parameters were integrated
into either the radiation use efficiency (RUE) or water use efficiency (WUE). The range of RUE values across
the references sites was from 1.2 to 9.2 and for WUE was from 0.58 to 4.8. By comparison, the corresponding
ranges across all the plots from the Sandhill Fen were 1.2 to 13.2 and 0.86 to 6.1, respectively. The ranges
overlap with one another, with some plants in the Sandhill Fen demonstrating even greater efficiency than
plants on the reference sites.

Black-girdle bulrush (Scirpus atrocinctus) is distributed in the United States in New England, West Virginia,
and the states bordering the upper Mississippi river valley and Great Lakes (e.g., Iowa, Wisconsin, Michigan,
Illinois) and across Canada (FNA, 2002). The species is classified as a native species and as either an obligate
or facultative wetland plant. Black-girdle bulrush is most often associated with wetlands or wet soils and has
been found in abundance on some disturbed peatland sites (Mahmood and Strack, 2011; Sottocornola et al.,
2007). Across the three reference sites utilised for this species, significant differences were observed only for
the rates of photosynthesis and evapotranspiration between the more recently disturbed plots (i.e., U-cell
and/or Engstrom Lake) and the undisturbed Mariana Lake site (Figure 3). The range in the data was less than
two-fold for the two parameters. The only significant differences observed for this plant species on the
Sandhill Fen plots was that the rate of stomatal conductance was significantly higher on the two groups of
wetter plots as compared to the two groups drier plots. This rate was significantly higher than what was
observed for any of the reference sites. The rates of photosynthesis and evapotranspiration were not
significantly different between the Sandhill Fen plots and all the reference sites with the values for each
parameter falling within the narrow range of variability between the reference sites. The same trends were
observed for the RUE and WUE. The range of RUE values on the references sites was 0.26 to 10.0 as compared
to a range of 0.68 to 12.4 for the Sandhill Fen plots. The ranges for WUE were 0.06 to 4.2 and 0.71 to 3.8,
respectively. The ranges for both parameters correspond well between the reference sites and the Sandhill
Fen sites.
Seaside arrowgrass (Triglochin maritima) has a distribution similar to water sedge and is a native and obligate
wetland species. This species frequently co-occurs with water sedge in extreme rich fens. Because arrowgrass
shows some tolerance to disturbance and is a halophyte (Cody, 1996), this species has also been considered
for revegetation scenarios, including those associated with the oil sands industry. The only parameter that
differed significantly between the two reference sites was the rate of evapotranspiration (Figure 4), which
was higher for plants at the U-cell site as compared to the rich fen site. The U-cell site is situated on a hilltop
above the refinery complex and is considerably windier than the other sites examined. This greater wind
velocity may be partially responsible for this difference, and also the higher rates observed for the other two
species. Across the moisture gradients identified in the Sandhill Fen, the pattern in the ecophysiological data
was the same as that observed for black-girdle bulrush. The only significant difference observed was for
stomatal conductance between the wetter and drier groups of plots, which likely represents a response of
this wetland species to the mesic conditions in these plots. The range observed in the values for each
parameter for the plants on the Sandhill Fen, particularly for the wetter groups of plots, corresponds closely
to both the recent disturbance at the U-Cell site and the undisturbed rich fen. As for the other two species,
there was clear overlap in the range of values for RUE (5.0 to 19.6 reference sites; 0.26 to 43.1 Sandhill Fen)
and WUE (3.3 to 26.0 reference sites; 0.67 to 25 Sandhill Fen) although the ranges were wider for plants from
the Sandhill Fen.
Mine closure and reclamation are ongoing endeavours for the oil sands industry in Alberta. The prescriptions
needed to guide those efforts are continually evolving. Incorporating large scale wetland complexes into the
reclamation landscape is fundamental to Syncrude’s future reclamation efforts. The Sandhill Fen project is a
wetland restoration but is also exploring opportunities to create a wetland with healthy and diverse set of
plant species. The terrain in northern Alberta has extensive networks of peat-forming wetlands (i.e., fens and
bogs) which could be impacted by oil sands mining. Successful wetland reclamation could return the
landscape function provided by natural peat-forming wetlands, thereby fulfilling the mandate established by
Alberta Environment to restore the ecosystem with equivalent ecological capacity (Alberta Environment,
1999). Peat-forming fens and bogs are highly effective carbon sinks, and globally have stored an estimated
436 Gt of carbon (Loisel et al., 2014).

Vegetation


Vegetation
Successful reclamation of wetlands, including those that may form peat, is a priority for the oil sands industry
and could have the benefit of reducing the carbon footprint of the industry. There has been some success
with peatland restoration in Europe (Timmermann et al., 2006; Vasander et al., 2003; Zerbe et al., 2013) and
Canada (Rochefort and Lode, 2006). Peatland reclamation had not been attempted at this scale prior to the
Sandhill Fen research project, yet represents one of the more likely mine closure scenarios site managers and
regulatory agencies are interested in early markers of success for vegetation establishment during
reclamation efforts in order to evaluate reclamation progress and suitability for certification. The
ecophysiological approach demonstrated here offers a set of markers that can provide such information.
The selection of reference communities for this study included both recently disturbed and undisturbed sites
with the intention of bracketing the most likely range of values. Establishing these ranges was part of the first
objective of this study. This strategy provided at least two points along a vector describing the performance
of these vascular plants between the sites. As expected, more recent disturbance generally resulted in lower
values for the ecophysiological parameters as compared to the undisturbed sites. For water sedge, the results
from the Anzac sites illustrated that substrate type was also a determining factor in the performance of this
plant species. These results for water sedge underscore prior research (Mollard et al., 2012; Raab and Bayley,
2013, Vitt et al., 2011) that has demonstrated the robustness of this species for reclamation on disturbed
sites, including mineral soils. The constraints that substrate type places on physiological performance of
water sedge, and potentially early establishment of this species during reclamation efforts, are also evident.
Nonetheless, the range of values observed for this species varied by no more than three-fold for the
parameters measured. A similar range of values was also observed for the parameters measured for the
other two species. The U-cell site provided another type of reference community for this study, one that is
highly disturbed, but also an experimental reclamation with some similarity to the reclamation projected for
Sandhill Fen. The results for all three species from the U-Cell site consistently fell within the observed ranges,
providing additional evidence that the ranges established are representative for each ecophysiological
parameter and for each plant species.
The topography of Sandhill Fen resulted in a variable distribution of water, creating the four distinct soil
moisture conditions described here. The advantage that arose from this soil moisture variability is that it
allowed for the range of values for each parameter to be established (an aspect of the first objective of this
research) and for that factor to be considered in evaluating the performance of these three species one year
after establishment. In any revegetation effort, the availability of water is critical to initial plant
establishment, particularly for wetland plants. Wetland plant species able to withstand periods of water
deficit while maintaining their physiological performance would be highly desirable for reclamation efforts
in general, as well as for the specific reclamation efforts that would take place on disturbed sites associated
with oil sands mining. The three plant species examined in this study are obligate wetland plants and their
physiological performance clearly declined on the drier plots associated with Sandhill Fen, but all three
species were able to survive these conditions and sustain their physiological performance. The range
observed in the parameters across this moisture gradient aligned well with the range of values associated
with the corresponding reference sites. This specific comparison was the second objective of this research.
Each plant species after one year of growth was performing as well as plants on the undisturbed sites. Such
information provided a clear indication of successful initial establishment of the plants across the Sandhill
Fen plots. It is particularly remarkable that these wetland plants were able to establish themselves in less
than ideal areas of the fen watershed under drier conditions than would normally be considered unsuitable
for these species. The planning of a reclamation effort might presume that these plants could not be
established successfully without the necessary wetland conditions. The evidence here suggests that
establishment is possible without such conditions, but that physiological performance would be less than
optimal. This sub-optimal performance still fell within the ranges that were established from the reference
communities, showing that the dry conditions in areas of the Sandhill Fen represented in the first year of this
project a disturbance no more stressful than that in the disturbed reference sites.

The use of the ecophysiological parameters to assess physiological performance and early establishment of
plants in the Sandhill Fen provides industry with a potential set of markers to evaluate the initial success of
the planned reclamation restoration efforts as compared to benchmark performance of each species. Here
we demonstrate that performance of individual species differs and species must be treated individually. The
ecophysiological parameters described here have been used in subsequent years at Sandhill Fen on the same
plots and reference communities to extend the monitoring of these plant species. The results (data not
shown) are the same as what has been reported here. The plants established in the plots in 2012 continue
to thrive and consistently produce biomass each year as indicated by the appearance of litter. There is also
evidence from other areas within Sandhill Fen that multiple species of bryophytes associated with peat-
forming wetlands have become established across the fen (Vitt and House, 2015) with the plant species
associated with this study serving as nurse plants to create the necessary microclimate for moss
establishment. The first year results presented here offered early insight into the trajectory of establishment,
which aligned with the later successful performance and establishment that was observed. While not
observed here, a scenario where there was poor ecophysiological performance outside the range of values
early in a reclamation effort, relative to the range of values from reference sites, would indicate conditions
where establishment could be in jeopardy or might proceed more slowly than expected. Such information
would allow site managers the opportunity to intervene earlier to correct the conditions that are limiting
performance. For example, in the case of Sandhill Fen, the non-uniform wetting was an initial cause for
concern as it was expected that the drier conditions would limit plant performance. The first year data
reported here reduced that concern, contributed to later discussions about the design of wetland
reclamations, and provided insight into the type of wetland that would likely develop.
By extension, the same framework could be applied to mine closure scenarios. Once an array of foundation
species suitable for reclamation have been selected, the next step would be the identification of an array of
regional reference communities to establish the range of values for these parameters exhibited by these
plant species. Any threshold values below which physiological performance of reference plants decreased
could be noted, particularly if an association with a particular disturbance or stress could be assigned.
Empirical studies could also be conducted with those species to determine the limits of tolerance to a
disturbance or stress expected during the reclamation. For example, given that elevated concentrations of
sodium might intrude into the growth substrate during reclamation efforts above in-pit tailings, greenhouse
studies with water sedge and American sloughgrass (Beckmania syzigachne) were performed to identify the
threshold for sodium toxicity to these two species (Glaeser et al., 2015; Koropchak, 2010). The ranges of
values then form a rubric against which performance of the plants in the reclamation could be compared.
Alignment with, or deviation outside, this range could be an early signal to industry that could guide
subsequent management of the reclamation or inform future reclamation plans. The decision structure could
also integrate any other site-specific factors that might influence plant establishment. An additional value of
these measurements is that it could allow industry to provide stakeholders and regulatory agencies with early
results on the progress of plant establishment. These results would provide functional information on the
status of the foundation species that could complement any additional descriptive data that was collected.
As the reclamation progresses, and even after certification, the continued use of these ecophysiological
parameters offers a strategy to monitor key plant species in the reclaimed area and assess its long-term
stability.
This research was supported by a grant from Syncrude Canada Ltd. The authors would like to offer special
thanks to Jessica Piercey and Carla Wytrykush of Syncrude Canada Ltd. for their assistance with the
development and implementation of this research in the Sandhill Watershed. The authors would also like to
thank Alice Grgicak-Mannion for providing the schematic diagram of the Sandhill Fen Research Watershed.
Jessica Piercey provided valuable revisionary suggestions for the manuscript for which we are grateful.

Vegetation
Alberta Environment (1999) Regional Sustainable Development Strategy for the Athabasca Oil Sands Area, Alberta Environment,
Government of Alberta, Report no. 1/754.
Bruhl, J.J. and Wilson, K.L. (2007) Towards a comprehensive survey of C3 and C4 photosynthetic pathways in Cyperaceae, Aliso: A
Journal of Systematic and Evolutionary Botany, Vol. 23(1), pp. 99−148.
Cody, W.J. (1996) Flora of the Yukon Territory, 2nd ed, Ottawa: NRC Research Press.
FNA (2002) Magnoliophyta: Commelinidae (in part): Cyperaceae, New York: Oxford University Press.
Glaeser, L.C., Vitt, D.H. and Ebbs, S.D. (2015) Responses of the wetland grass, Beckmannia syzigachne, to salinity and soil wetness:
Consequences for wetland reclamation in the oil sands area of Alberta, Canada Ecological Engineering (in review).
Gravina, A., McKenna, P. and Glenn, V. (2011) Evaluating the success of mineral sand mine rehabilitation on North Stradbroke Island,
Queensland: comparisons with reference eucalypt communities, Proceedings of the Royal Society of Queensland, Vol. 117,
pp. 419−435.
Jaunatre, R., Buisson, E., Muller, I., Morlon, H., Mesléard, F. and Dutoit, T. (2013) New synthetic indicators to assess community
resilience and restoration success, Ecological Indicators, Vol. 29(0), pp. 468−477.
Johnson, E.A. and Myanishi, K. (2008) Creating new landscapes and ecosystems: the Alberta oil sands, Annals of the New York
Academic of Sciences, Vol. 1134, pp. 120−145.
Koropchak, S. (2010) Carex aquatilis as a pioneer species for boreal wetland reclamation in northern Alberta, Southern Illinois
Unviersity, Carbondale, IL.
Koropchak, S., Vitt, D.H., Bloise, R. and Wieder, R.K. (2012) Fundamental paradigms, foundation species selection, and early plant
responses to peatland initiation on mineral soils, In Viit, D. and Bhatti, J. (Eds.), Restoration and Reclamation of Boreal
Ecosystems, Camridge University Press, Cambridge, pp. 76−100.
Lambert, A.M., Baer, S.G. and Gibson, D.J. (2011) Intraspecific variation in ecophysiology of three dominant prairie grasses used in
restoration: cultivar versus non-cultivar population sources, Restoration Ecology, Vol. 19(101), pp. 43-52.
Loisel, J., Yu, Z., Beilman, D.W., Camill, P., Alm, J., Amesbury, M.J., Anderson, D., Andersson, S., Bochicchio, C. and Barber, K. (2014)
A database and synthesis of northern peatland soil properties and Holocene carbon and nitrogen accumulation, the Holocene,
Vol. 24(9), pp. 1028−1042.
Mahmood, M.S. and Strack, M. (2011) Methane dynamics of recolonized cutover minerotrophic peatland: implications for
restoration, Ecological Engineering, Vol. 37(11), pp. 185−1868.
Mollard, F.P.O., Roy, M.-C., Frederick, K. and Foote, L. (2012) Growth of the dominant macrophyte Carex aquatilis is inhibited in oil
sands affected wetlands in Northern Alberta, Canada, Ecological Engineering, Vol. 38(1), pp. 11−19.
Oil Sands Reclamation Committee (1998) Guidelines for Reclamation to Forest Vegetation in the Athabasca Oil Sands Region, Alberta
Environmental Protection, Report no. ESD/LM/99-1.
Ostertag, R., Giardina, C.P. and Cordell, S. (2008) Understory colonization of Eucalyptus plantations in Hawaii in relation to light and
nutrient levels, Restoration Ecology, Vol. 16(3), pp. 475−485.
Raab, D. and Bayley, S.E. (2013) A Carex species-dominated marsh community represents the best short-term target for reclaiming
wet meadow habitat following oil sands mining in Alberta, Canada, Ecological Engineering, Vol. 54(0), pp. 97−106.
Rochefort, L. and Lode, E. (2006) Restoration of degraded boreal peatlands, In Wieder, R.K. and Vitt, D.H. (Eds.), Boreal Peatland
Ecosystems, Springer Berlin Heidelberg, pp. 381−423.
Rooney, R.C. and Bayley, S.E. (2011) Setting reclamation targets and evaluating progress: submersed aquatic vegetation in natural
and post-oil sands mining wetlands in Alberta, Canada, Ecological Engineering, Vol. 37(4), pp. 569−579.
Rovai, A.S., Barufi, J.B., Pagliosa, P.R., Scherner, F., Torres, M.A., Horta, P.A., Simonassi, J.C., Quadros, D.P.C., Borges, D.L.G. and
Soriano-Sierra, E.J. (2013) Photosynthetic performance of restored and natural mangroves under different environmental
constraints, Environmental Pollution, Vol. 181(0), pp. 233−241.
Sinclair, T.R. and Horie, T. (1989) Leaf nitrogen, photosynthesis, and crop radiation use efficiency: a review, Crop Science, Vol. 29(1),
pp. 90−98.
Sottocornola, M., Boudreau, S. and Rochefort, L. (2007) Peat bog restoration: effect of phosphorus on plant re-establishment,
Ecological Engineering, Vol. 31(1), pp. 29–40.
Timmermann, T., Margóczi, K., Takács, G. and Vegelin, K. (2006) Restoration of peat-forming vegetation by rewetting species-poor
fen grasslands, Applied Vegetation Science, Vol. 9(2), pp. 241–250.
Vasander, H., Tuittila, E.S., Lode, E., Lundin, L., Ilomets, M., Sallantaus, T., Heikkilä, R., Pitkänen, M.L. and Laine, J. (2003) Status and
restoration of peatlands in northern Europe, Wetlands Ecology and Management, Vol. 11(1-2), pp. 51–63.
Vickers, H., Gillespie, M. and Gravina, A. (2012) Assessing the development of rehabilitated grasslands on post-mined landforms in
north west Queensland, Australia, Agriculture, Ecosystems & Environment, Vol. 163(0), pp. 72–84.
Vitt, D.H. and Chee, W.-L. (1990) The relationships of vegetation to surface water chemistry and peat chemistry in fens of Alberta,
Canada, Vegetatio, Vol. 89, pp. 87–106.
Vitt, D.H., Wieder, R.K., Xu, B., Kaskie, M. and Koropchak, S. (2011) Peatland establishment on mineral soils: effects of water level,
amendments, and species after two growing seasons, Ecological Engineering, Vol. 37(2), pp. 354–363.
Vitt, D.H. and House, M. (2015) Reclaiming peatlands after disturbance from oil sands mining: establishment of bryophytes from
indigenous sources, The Bryologist, Vol. In review, pp.
Wang, R.Z., Liu, X.Q. and Bai, Y. (2006) Photosynthetic and morphological functional types for native species from mixed prairie in
Southern Saskatchewan, Canada, Photosynthetica, Vol. 44(1), pp. 17–25.

White, P.S. and Walker, J.L. (1997) Approximating nature's variation: selecting and using reference information in restoration ecology,
Restoration Ecology, Vol. 5(4), pp. 338–349.
Wytrykush, C., Vitt, D.H., McKenna, G. and Vassov, R. (2012) Designing landscapes to support peatland development on soft tailings
deposits, In Vitt, D.H. and Bhatti, J. (Eds.), Restoration and Reclamation of Boreal Ecosystems, Cambridge University Press,
Cambridge, pp. 161–178.
Zerbe, S., Steffenhagen, P., Parakenings, K., Timmermann, T., Frick, A., Gelbrecht, J. and Zak, D. (2013) Ecosystem service restoration
after 10 years of rewetting peatlands in NE Germany, Environmental Management, Vol. 51(6), pp. 1194–1209.

M. Guittonny-Larchevêque Université du Québec en Abitibi-Témiscamingue (UQAT), Canada
In forested regions, the revegetation of milled mine tailings should aim at tree establishment to decrease the
environmental impacts of mining and foster social acceptability. However, tree establishment on fine grain-
sized mine tailings is difficult due to physical restrictions, in particular low macroporosity and saturated
hydraulic conductivity, which decrease substrate aeration and may impede root growth. Compared to trees,
graminoids, which are routinely seeded on fine-grain mine tailings to stabilise the surface and decrease
erosion, can better tolerate physical restrictions associated with mine tailings. This study aimed to identify
which graminoids could be used to improve the structural properties of tailings through root development
and further facilitate tree root development. We also examined whether water stress exposure modifies
graminoid root effects on the tailings structure.
Five graminoid species with differing root system morphology were selected for a pot experiment under
greenhouse conditions: Lolium perenne L., Avena sativa L., Festuca rubra L., Bromus inermis Leyss., and
Phalaris arundinacea L. The grasses were seeded on low-sulphur tailings from a gold mine (100 pots = 10
repetitions x 5 species x 2 watering regimes) and on conventional growing substrate as control (i.e., peat). All
pots were NPK-fertilized. Germination and survival rates were monitored. Two months after seeding, plant
aerial and root development were measured and tailings macroporosity (i.e., air-filled porosity, pore diameter
> 50µm) and density were determined. All graminoids increased the tailings’ macroporosity, with a positive
linear relationship to root biomass. B. inermis showed the best germination and survival rates and the
greatest increase in substrate macroporosity. Its root system showed the greatest mean root diameter and
volume of the tested species. Reduced watering increased specific root length, resulting in similar
macroporosity in tailings compared to normal watering, despite lower root biomass production.
In Canada, 50% of the boreal region is covered by forest (Environment Canada, 2010). Mines operating in this
region must include in their reclamation plan revegetation strategies that encourage tree establishment in
an attempt to restore disturbed habitats and to increase the social acceptability of mining operations.
Like other mine wastes, metalliferous mine tailings generally cover large surface areas. They consist of the
finely crushed ore (70 to 80% of particles range from 2 to 80 µm, Aubertin et al., 2002) that remains after
valuable metals have been removed. They are transported from the mine plant (usually pumped) as aqueous
slurries and deposited into tailings storage facilities. These substrates are among the most difficult to
revegetate. First, they lack organic matter, nutrients, and soil organisms (Burger and Zipper, 2002). Second,
they lack physical structure, may have extreme pH values, and contain potentially phytotoxic levels of salts
and heavy metals (Tordoff et al., 2000). In metalliferous mine tailings, root development may be restricted
to cover soils (Borgegard and Rydin, 1989; Larchevêque et al., 2013), probably due to unsuitable properties
of the underlying tailings for root growth (Evanylo et al., 2005), in particular, low air-filled porosity
(Larchevêque et al., 2013).
Because non-acid-generating tailings may be less toxic to plant roots (Bagatto and Shorthouse, 1999), direct
herbaceous seeding is used for revegetation. Herbaceous species are usually sown in mixtures of grasses
and legumes (nitrogen-fixing). Graminoids in particular can rapidly cover the tailings and protect them from
erosion (Burger and Zipper, 2002). Unlike herbaceous species, tree development is restricted in the
compacted, low aeration, high pH, and dissolved salt conditions that are associated with mill tailings

Selecting graminoids to increase substrate macroporosity for the revegetation of milled tailings of a gold mine M. Guittonny-Larchevêque
(Emerson et al., 2009). Rapid and adequate root development of young trees appears to be particularly
critical for promoting water access, regardless of the tree regeneration strategy (i.e., tree planting or natural
colonisation). At planting, trees typically undergo a planting shock due to moisture or nutrient stresses, which
can decrease plantation productivity for several years (Grossnickle, 2005). The resulting growth stagnation
corresponds to the time the trees need to grow an adequate root system for the planting site (Grossnickle,
2005). Moreover, the regeneration of boreal tree species in disturbed areas is controlled by water availability,
because seedling mortality in the first growing season is due mainly to desiccation (Greene and Johnson,
1998).
Mining studies have focused mainly on improving the structure and nutrient content of tailings and
limiting toxicity in order to improve plant development (Abrutat et al., 1994; Boyter et al., 2009; Emerson et
al., 2009; Lunt and Hedger, 2003; Whitbread-Abrutat, 1997; Larchevêque et al., 2013 and 2015). However,
herbaceous roots (dead or alive) create continuous biopores, which could improve tailings macroporosity,
aeration, and water circulation (Grevers and De Jonc, 1990; Angers and Caron, 1998), thus encouraging the
development of tree roots (Dexter, 1991). Because over 50% of root biomass can die each year (Wick et al.,
2007), roots may make a significant contribution to tailings structuration.
The beneficial changes in the substrate due to herbaceous species and their facilitating effect on the
rooting of other species have been studied primarily in agricultural settings (Wang et al. 1986; Angers and
Caron, 1998; Dexter, 1991). Root diameter in particular directly influences macropore formation and soil
hydraulic conductivity (Bodner et al., 2014). When stressed, plants adapt through root morphology changes
that can modify their effect on substrate structuration. For example, root diameter decreases and specific
root length (SRL: root length per dry mass) or root length density (RLD: root length by unit of soil volume)
increases in response to water stress (McCully, 1999).
This study aimed to identify which graminoids could improve tailings structural properties (i.e. macroporosity
and density) through root development and hence facilitate tree root development. We also examined
whether water stress exposure modifies the graminoid root effects on tailings structure. Five graminoid
species with differing root system morphology were selected for a pot experiment under greenhouse
conditions: Lolium perenne L., Avena sativa L., Festuca rubra L., Bromus inermis Leyss., and Phalaris
arundinacea L. The grasses were seeded on non-acid-generating tailings from a gold mine and on a peat
control. Germination and survival rates were monitored. Two months after seeding, plant aerial and root
development were measured and tailings macroporosity and density were determined.
The tailings were obtained from the Canadian Malartic gold mine (property of the Canadian Malartic
Partnership, Malartic, northwestern Quebec, Canada, 48°13'N, 78°12'W). This large, open-pit mine began
production in 2011. The Canadian Malartic ore is mineralised greywacke. The tailings have low sulphur
content (around 1% S) and contain calcite, which can neutralise acidity. They consist of finely milled wastes
from the gold extraction process (cyanide leaching) with 86% particles < 80 µm and a uniformity coefficient
(CU = D60/D10) of 11.1. (D10: 0.003 mm and D60: 0.034 mm, D10 and D60 being the size of the sieve through
which 10% and 60% of the material passes, respectively). They were deposited in the storage facility as
thickened tailings (around 60% solids by mass) less than six months before the experiment. The tailings were
treated by cyanide detoxification (SO2/H2O2 technology), which left free CN- concentrations lower than 20
mg kg-1. As expected, the tailings had low macroporosity and high density (Table 1) that approached the
thresholds where root growth is restricted (Archer and Smith, 1972; Schuurman, 1965). Chemical
characteristics of the tailings are summarised in Table 1. Trace metal concentrations are below Quebec’s
regulatory thresholds for residential lands (Government of Quebec, 2014), except for total cadmium (Cd),
which appears to exceed the threshold slightly (see Table 1). Frozen tailings were sampled with a mechanical
shovel at the surface (0 to 1 m) of the tailings facility in February 2014 and transported to the greenhouse
for the pot experiment.

Vegetation
Regulatory
Tailings threshold**
(residential lands)
Macroporosity % 6 (1)
Density % 1.3 (0.1)
pH 7.9 (0.1)
OM* % 0.1 (2)
EC* cS.m-1 10 (1)
Total N g.kg-1 0.1 (0.5)
Total P g.kg-1 0.7 (0.01)
Olsen P mg.kg-1 8 (5)
-1
Total K g.kg 9 (0.4)
Total Ca g.kg-1 17 (2)
Total Mg g.kg-1 14 (0.6)
Total Na g.kg-1 0.5 (0.05)
Total Al g.kg-1 14 (1)
Total Fe g.kg-1 34 (2)
Total B mg.kg-1 0.8 (0.5)
-1
Total Cd mg.kg 6 (0.2) 5
Total Cr mg.kg-1 168 (10) 250
Total Cu mg.kg-1 52 (2) 100
Total Mn mg.kg-1 441 (16) 1000
Total Ni mg.kg-1 69 (9) 100
Total Pb mg.kg-1 18 (100) 500
Total Zn mg.kg-1 73 (4) 500
* OM: organic matter; EC: electrical conductivity

** Government of Quebec (2014)
Five graminoid species with differing root system morphology were selected. L. perenne has a dense and
fibrous root system that can extend deep into the soil (De Baets et al., 2011). A. sativa also has a fibrous root
system, but with thicker roots that extend superficially (De Baets et al., 2011). B. inermis and P. arundinacea
have a fibrous and rhizomatous root system that extends deep into the soil. B. inermis roots can penetrate
soils to 2 m depth and more (Sahramaa and Jauhiainen, 2003). F. rubra has a tracing root system that spreads
by rhizomes (Green and Renault, 2008). Seeds of each species were obtained from Coop Val Nord (Amos,
QC). The selected species are among the herbaceous species used in agriculture and mine revegetation in
northwestern Quebec.

In mid-March 2014, plastic pots (3.6 L) were filled with 1 kg (770 cm3) of air-dried tailings in a greenhouse. In
each pot, 5 seeds were then planted at 3 cm intervals, according to recommended seeding methods in mine
revegetation studies (Burger and Zipper, 2002). Seeds were buried at depths of 0 to 2 cm, depending on the
species, to promote germination.
A split-plot design was used: 100 pots = 10 blocks (replicates) x 2 watering regimes (whole plot factor: normal
versus reduced watering) x 5 graminoid species (sub-plot factor: L. perenne, A. sativa, B. inermis, P.
arundinacea, F. rubra). Additional pots filled with commercial PRO-MIX® (Premier Horticulture Ltd, Dorval,
Quebec, Canada) peat were also seeded with the five graminoid species (4 repetitions x 5 species). This
standard substrate with optimised effect on plant development was used to get ideal germination rates and
root development of plants, acting as positive control.
The amount of water corresponding to field capacity (534 g) was calculated by measuring the mean weight
difference for four pots watered at field capacity before and after oven drying for 48 hours (100 °C). Pots
were then watered at field capacity and allowed to dry by evaporation. The mass loss was measured twice
daily to establish the water loss curve with time and to calculate the mean water loss per day, which was
then used for the normal daily watering regime (160 mL.day-1). For the reduced watering regime, watering
was reduced to 100 mL.day-1.
All pots were fertilised with 1 g of granular mineral fertiliser (9% nitrogen (N) – 28% phosphorus (P) – 9%
potassium (K), starter fertiliser, Coop agricole Amos, QC), which was mixed with the tailings before seeding
according to recommended concentrations for mine tailings revegetation (Burger and Zipper, 2002). Plants
grew for two months, from March 13 to May 23, 2014, under natural daylight conditions complemented by
high pressure sodium lamps under 16-h photoperiod at 25 °C. The watering regimes were applied two weeks
after seeding in order to remove water as a limiting factor while the seeds germinated. All pots were watered
in the morning with 80 mL of water containing 0.3 g.L-1 of liquid fertiliser (30% N – 10% P – 10% K, Nutrite
tree and shrub fertiliser with micronutrients, St-Michel, QC). In the afternoon, pots under the normal
watering regime received 80 mL of water, and pots under the reduced watering regime received only 20 mL
of water.
To determine whether real water loss accounted for the transpiration of developing plants and to
consequently adjust the watering regimes, pots were weighed each day before the first watering for two
weeks and volumetric water content was measured with an ECH2O-EC5 (Decagon Devices, Inc., Pullman, WA)
probe. The results revealed no increase in volumetric water content for pots under the reduced watering
regime. Consequently, the reduced watering regime was changed to 80 mL of water with liquid fertiliser in
the morning and no watering in the afternoon.
For tailings characterisation, nutrient and trace metal analyses were conducted on three sieved (2 mm mesh),
finely ground, oven-dried samples (50°C) (Lakehead University Centre for Analytical Services, Thunder Bay,
ON, Canada). Total N was analysed by the Dumas combustion method (LECO CNS 2000, Mississauga, ON) and
organic carbon (C) by the thermogravimetric method (LECO-TGA, Mississauga, ON). Organic matter
concentrations were calculated as 1.72 × organic carbon (C). Following HNO3-HCl digestion, sample
concentrations of total P, K, Ca, Mg, Na, Al, As, B, Ba, Be, Cd, Co, Cr, Cu, Fe, Mn, Mo, Ni, Pb, S, Se, Sr, and Tl
were determined by inductively coupled plasma-atomic emission spectrometry (ICP-AES, Vista PRO, Varian
Canada, Mississauga, ON). Available P was determined in a sodium bicarbonate solution using
spectrophotometry (Olsen et al., 1954). pH was determined in a saturated paste extract and electrical
conductivity in a 1:2 water solution.

Vegetation
Germination rates (%) were assessed for each pot (100 pots) two weeks after seeding as number of
germinated seeds / 5 x 100. Survival rates (%) were calculated for each pot at the end of the experiment as
number of alive plants / number of germinated seeds x 100. Success rate (%) was calculated as germination
rate x survival rate.
All the following measures were performed on the five pots by the graminoid species and watering regime
combination (50 pots) that showed the best aerial development of the ten pots available per combination.
Plant aerial parts were cut at ground level 24 hours after the last watering. Aerial parts were weighed after
oven drying at 70 °C for 48 hours (dry mass). Aerial biomass was calculated by pot and alive plant.
Undisturbed 100 cm3 soil samples were taken at 0 to 10 cm depth with a double cylinder soil sampler at the
centre of each pot (50 cores). Bulk density and macroporosity (air-filled porosity, % of pores > 50 μm) were
determined following procedures described by Cassel and Nielsen (1986). Cores were brought to saturation
under vacuum and weighed (W1). They were then set on the porous surface of a sand box apparatus
(Eijkelkamp Agrisearch Equipment) and brought to equilibrium at a tension of −10 kPa (field capacity). Cores
were weighed (W2) and oven dried (105°C, 48 hours) prior to final weighing (W3). Macroporosity or air-filled
porosity (%) and bulk density were estimated with the following formulas:
Macroporosity = (W1 − W2/100 cm3) × 100
Bulk density = W3/100 cm3
All remaining tailings in the pots after coring (670 cm3 = 770 cm3 – 100 cm3 of the extracted core) were used
for root morphology analysis. At the laboratory, graminoid roots were washed from the tailings above a fine
metallic grid. Scanned images of fresh roots were analysed with WinRHIZO software (regular version, Regent
Instruments Inc., Sainte-Foy, QC) for root length, surface, diameter, volume, and tip number in relation to
the remaining tailings volume. Root length density (RLD, cm.cm-3 substrate), root surface density (cm2.cm-3
substrate), root volume density (cm3.cm-3 substrate) and number of tips density (nb.cm-3 substrate) were
calculated. Root samples were then oven-dried at 70°C for 48 hours and weighed to measure root biomass
(by pot and plant), root soil density (g.cm-3 substrate), specific root area (SRA, m2.kg-1), and specific root
length (SRL, m.g-1). Roots of all samples had a diameter of 2 mm or less.
Germination, survival, and success rates were compared by species and watering regime using the χ 2 test
(PROC FREQ, SAS V.9.2, SAS Institute Inc., Cary, NC). Data on root parameters, biomasses, substrate
macroporosity, and density were subjected to two-way analyses of variance (watering regime, species) (SAS,
PROC MIXED). Because the interaction between the two tested factors (watering regime and species) was
not significant for all considered variables, the data were presented by species (combined watering regimes)
or by watering regime (combined species). Tested factors were fixed effects, and the block factor was
considered a random effect. When effects were significant for a given variable, least-square means were
estimated (SAS, LS MEANS statement), and Tukey’s tests were conducted to separate the means. Pearson
correlations between tailings macroporosity at the end of the experiment and root parameters were
analysed (SAS, PROC CORR). Overall significance for analyses was set at alpha = 0.05.

Germination, survival, and success rates differed significantly among the five tested species, irrespective of
watering regime (Table 2). On tailings, B. inermis and F. rubra had the best germination rates, whereas A.
sativa and B. inermis had the best survival rates. Consequently, B. inermis showed the best overall success
rate of the five species, with a one in two chance for a seed to produce an alive plant. B. inermis and P.
arundinacea had lower germination rates on tailings compared to peat. Tailings also decreased L. perenne,
P. arundinacea, and F. rubra survival rates compared to peat. The watering regimes had no effect on any
species’ survival after germination (results not shown), demonstrating that the applied water stress was of
moderate intensity.
Germination Survival Success

% % %
Species Tailings Peat Tailings Peat Tailings
A. sativa 31 33 91 100 28
B. inermis 56 100 91 100 49
L. perenne 46 50 61 100 27
P. arundinacea 30 90 65 100 19
F. rubra 57 55 56 100 33
2
χ test * * *
On peat, all species showed greater aerial and root biomass than on tailings, especially B. inermis and P.
arundinacea, except for A. sativa, with similar biomass on both substrates (results not shown). B. inermis
showed the greatest root biomass production by plants seeded on peat (1.4 g), with much lower biomass
when seeded on tailings. On tailings, A. sativa produced the greatest root biomass by plant, closely followed
by B. inermis (Table 3). B. inermis’s success was nearly twice that for A. sativa, and B. inermis produced the
greatest root biomass of all the species seeded in tailings pots. However, of all the species, A. sativa produced
the greatest aerial biomass by plant and by pot (Table 3). Reduced watering decreased the production of
aerial and root biomass by all plants (Table 3).

Vegetation
Root Root Aerial Aerial

biomass by biomass by biomass by biomass by Bulk
pot plant pot plant Macroporosity density
Species g g g g % g.cm-3
0.36 (0.09) 0.26 (0.05)
A. sativa 4.8 (0.31) b 3.2 (0.25) b 13.8 (0.7) a 1.3 (0.04) a
a c
0.53 (0.08) 0.23 (0.05)
B. inermis 1.1 (0.30) a 0.5 (0.24) a 14.4 (0.6) b 1.3 (0.04) a
b bc
0.15 (0.09) 0.06 (0.05)
L. perenne 0.4 (0.31) a 0.1 (0.25) a 11.6 (0.7) a 1.4 (0.04) a
a ab
0.09 (0.10) 0.04 (0.06)
P. arundinacea 0.4 (0.31) a 0.2 (0.25) a 12.4 (0.7) a 1.3 (0.04) a
a ab
0.08 (0,09) 0.02 (0.07)
F. rubra 0.4 (0.31) a 0.1 (0.25) a 13.1 (0.7) a 1.3 (0.04) a
a a
Watering regime
0.31 (0.06) 0.18 (0.04)
Normal 1.7 (0.20) b 1.1 (0.16) b 12.7 (0.4) a 1.3 (0.03) a
a b
0.18 (0.06) 0.07 (0.03)
Reduced 1.1 (0.18) a 0.5 (0.15) a 13.4 (0.4) a 1.3 (0.03) a
a a
Of the tested plants, A. sativa grew roots with the longest length and the largest number of tips by plant,
whereas P. arundinacea showed the lowest root length density and the greatest root density (Table 4). Along
with the widest root diameter, B. inermis showed the greatest root volume by pot and by plant. Reduced
watering generally decreased root length, number of tips, volume, and density, while it increased specific
root length (Table 4). Root diameter was not affected by the watering regime.
After two months of plant growth, tailings macroporosity doubled compared to before plant growth.
Macroporosity was the greatest in tailings with B. inermis compared to tailings with all other tested species
(Table 3). The watering regime did not influence tailings macroporosity. However, tailings macroporosity at
the end of the experiment was positively correlated with plant root biomass (r = 0.58, p = 0.099), and this
relationship was more significant when plants were segregated by watering regime: reduced (r = 0.69,
p = 0.0006) versus normal watering (r = 0.44, p = 0.05). Tailings density was not affected by either plant
development or watering regime.
Our results demonstrated that all the tested graminoids could improve tailings macroporosity to proportions
suitable for plant root development after only two months of growth. Thus, through graminoid root
development, tailings macroporosity exceeded the 10% threshold reported in the literature to impede plant

root growth (Archer and Smith, 1972). In our study, macroporosity improvement was strongly related to root
biomass production by graminoids, as the two variables were positively and significantly correlated. In
addition to a direct effect of plant roots on macroporosity through biopore creation,
root length and biomass were positively correlated to soil aggregation (Angers and Caron, 1998), which can
participate in macropore formation. The positive effect on tailings structure was observed despite lower root
development in tailings compared to a standard growing substrate. Hence, direct graminoid seeding appears
to be a promising method to improve tailings macroporosity before tree establishment. However, no positive
effect of graminoid roots on tailings density was found in our study.
Root mean Root length Root tips Root volume Specific root
diameter density number by pot by pot length
Species mm cm.cm-3 nb.cm-3 cm3.dm-3 cm.g-1
A. sativa 0.45 (0.04) a 0.8 (0.06) b 13 (1.4) b 1.9 (0.9) a 5,575 (3016) a
B. inermis 0.56 (0.03) b 0.7 (0.06) b 6 (1.3) a 6.2 (0.9) b 4,450 (2844) a
L. perenne 0.39 (0.04) a 0.7 (0.07) b 5 (1.5) a 0.8 (1) a 6,125 (3284) a
P. arundinacea 0.32 (0.04) a 0.3 (0.07) a 3 (1.6) a 0.2 (1) a 13,884 (3434) a
F. rubra 0.40 (0.04) a 0.5 (0.06) b 2 (1.5) a 0.6 (0.9) a 7,047 (3179) a
Watering regime
Normal 0.43 (0.03) a 0.7 (0.04) b 7 (1) b 2.3 (0.6) a 4,013 (2103) a
Reduced 0.42 (0.02) a 0.5 (0.04) a 5 (0.9) a 1.6 (0.6) a 10,819 (1886) b
Root tips
Root length by Root volume
number by
plant by plant
Root density plant
Species g.cm-3 cm nb cm3
A. sativa 0.4 (0.05) a 340 (39) b 6,472 (770) b 0.9 (0.38) a
B. inermis 0.1 (0.05) a 214 (37) a 1,683 (725) a 2.0 (0.36) b
L. perenne 0.3 (0.06) a 159 (43) a 1,270 (838) a 0.2 (0.41) a
P. arundinacea 0.6 (0.06) b 82 (44) a 778 (876) a 0.1 (0.43) a
F. rubra 0.2 (0.06) a 94 (41) a 419 (811) a 0.1 (0.40) a
Watering regime
Normal 0.4 (0.04) b 217 (27.2) b 2,865 (536) b 0.9 (0.26) b
Reduced 0.3 (0.03) a 138 (24.4) a 1,384 (481) a 0.4 (0.24) a

Vegetation
20
Tailings macroporosity (%)

15
10 y = 3.2531x + 12.265
R² = 0.208
0
0,0 0,5 1,0 1,5
Root biomass by pot (g)
20
15
10
y = 3.429x + 12.735
5 R² = 0.4762
0
0,0 0,5 1,0 1,5
Root biomass by pot under reduced watering (g)
25
20
15
10
y = 3.9772x + 11.447
R² = 0.1922
5
0
0,0 0,5 1,0 1,5
Root biomass by pot under normal watering (g)

B. inermis was the most efficient graminoid to improve tailings macroporosity, probably due to its wider root
diameter, greater root volume, and better success rate (germination x survival) on tailings. Our results
confirmed that the wider the root diameter, the greater the influence on substrate macroporosity (Bodner
et al., 2014). B. inermis could therefore be particularly useful in seeding mixes that aim to improve tailings
structure before tree establishment. However, other characteristics of herbaceous species should be
considered in order to foster tree establishment after seeding, in particular, their competitive ability to
absorb water, nutrients, and light. The negative competitive effect of herbaceous species vis-à-vis trees has
been highlighted in studies of tree planting on mine substrates (Kost and Vimmerstedt, 1994; Skousen et al.,
2006; Halofsky and McCormick, 2005). For example, B. inermis and F. rubra are potential candidates because,
being low growing grasses, they do not compete for light. On the other hand, L. perenne and P. arundinacea
grow to a higher maximal height and may be highly light-competitive with young seedlings.
A. sativa, an annual graminoid, was able to efficiently increase macroporosity in just one growing season
through maximal root biomass and length production. This species is often used as a nurse plant for tailings
revegetation. The fact that it dies after one season may limit its competitiveness with tree seedlings.
However, perennial graminoids are reported to be more efficient than annuals for soil structure
improvement (Angers and Caron, 1998). In our study, A. sativa achieved similar macroporosity improvement
to perennial graminoids, but perennials may show better performance over a time range longer than two
months.
Under limited water conditions, the root biomass of graminoids decreased along with their aerial biomass
production, but the improving effect on tailings macroporosity remained. The linear relationship between
root biomass and tailings macroporosity changed according to water stress level. Interestingly, graminoid
roots became more efficient at improving macroporosity at low root biomass rates associated with limited
water conditions (greater y-intercept of the regression line). Thus, water-stressed graminoids increased their
specific root length, producing longer root length for a given root biomass compared to non-stressed
graminoids. This could explain why, despite lower root biomass under water-stressed conditions, the tailings
macroporosity remained similar to that for non-water-stressed plants. Fine and extended roots are
particularly effective in forming stable aggregates (Oades, 1984; Tisdall, 1991), which may have participated
in macropore formation. Increased length by unit mass of root (i.e., increased specific root length) is one of
the most common responses of plants exposed to water stress (Comas et al., 2013). This response allows
plants to increase the explored volume of the substrate. These results suggest that graminoids should
maintain their ability to improve tailings macroporosity under varying water availability. This is important
regarding the applicability of our results to field conditions where water availability is more variable than in
a controlled experiment.
In this controlled-environment study, several root parameters were related to tailings macroporosity
improvement, highlighting that plants can modify a tailings structure through different strategies. Root
biomass was directly related to improved macroporosity for all graminoid species studied. Wider root
diameter for B. inermis resulted in maximal improvement through greater root volume. Longer specific root
length under water stress maintained the macroporosity improvement despite lower root biomass. These
promising results should be confirmed in field experiments.
Our results highlighted that graminoids could be added to seeding mixes for two-step revegetation programs
aiming to establish trees on tailings facilities (first step: herbaceous seeding; second step: tree planting). Root
development is rarely considered in assessments of tree plantation success, despite the fact that tree water
and nutrient acquisition as well as tree anchorage require well-developed root systems. Tree root
development could be improved in tailings due to greater macroporosity after graminoid seeding. Tree
plantations could be established soon after herbaceous seeding, because structural improvement by
graminoid roots is rapid. In any case, large-dimension tree seedlings should be planted to limit competition
from graminoids (Jobidon et al., 1998), and graminoids with low competitive ability should be selected.

Vegetation
We thank Mine Canadian Malartic for providing the tailings for the experiment, as well as the Research
Institute in Mines and the Environment and the Commission Scolaire Harricana, Amos, Quebec for providing
laboratory and technical equipment. This research was funded by a grant to MGL from the Foundation of
Université du Québec en Abitibi-Témiscamingue. This research was made possible through student hosting
from Université de Montpellier 1 (France). We also thank Margaret McKyes for English correction of the
manuscript.
Abrutat, P.H., Edwards, R.P. and Wilkins, C. (1994) The potential of in situ amendments to improve alder establishment in pots of
metalliferous mine waste, in Proceedings International Land Reclamation and Mine Drainage Conference and the 3rd
International Conference on the Abatement of Acidic Drainage, 24–29 Avril 1994, Pittsburgh, PA, USA, pp. 305–312.
Angers, D.A. and Caron, J. (1998) Plant-induced changes in soil structure: processes and feedbacks, Biogeochemistry, Vol. 42(1/2),
pp. 55–72.
Archer, J.R. and Smith, P.D. (1972) The relation between bulk density, available water capacity, and air capacity of soils, Journal of
Soil Science, Vol. 23(4), pp. 475–480.
Aubertin, M., Bussière, B. and Bernier, L. (2002) Environnement et gestion des résidus miniers, CD-Rom, Les Éditions de l'École
Polytechnique de Montréal, Montréal, QC, Canada.
Bagatto, G. and Shorthouse, J.D. (1999) Biotic and abiotic characteristics of ecosystems on acid metalliferous mine tailings near
Sudbury, Ontario, Canadian Journal of Botany, Vol. 77, pp. 410–425.
Bodner, G., Leitner, D. and Kaul, H.-P. (2014) Coarse and fine root plants affect pore size distributions differently, Plant and Soil, Vol.
380, pp. 133–151.
Borgegard, S.-O. and Rydin, H. (1989) Biomass, root penetration and heavy metal uptake in birch in a soil cover over copper tailings,
J. Applied Ecology, Vol. 26(2), pp. 585–595.
Boyter, M.J, Brummer, J.E. and Leininger, W.C. (2009) Growth and Metal Accumulation of geyer and mountain willow grown in topsoil
versus amended mine tailings, Water Air Soil Pollut., Vol. 198, pp. 17–29.
Burger, J.A. and Zipper, C.E. (2002) How to restore forests on surface-mined land, Reclamation guidelines for surface mined land in
Southwest Virginia, Virginia Cooperative Extension, Powell River Series, Publication 460–123.
Cassel, D.K. and Nielsen, D.R. (1986) Field capacity and available water capacity, Methods of soil analysis. Part 1 – Physical and
mineralogical methods, Agronomy Series No 9, ASA, Inc., SSSA, Inc., Madison, WO, USA, pp. 901–926.
Comas, L.H., Becker, S.R. Cruz, V.M.V. Byrne, P.F. and Dierig, D.A. (2013) Root traits contributing to plant productivity under drought,
Frontiers in plant science, Vol. 4(442), pp. 1–16.
De Baets, S., Poesen, J., Meersmans, J. and Serlet L. (2011) Cover crops and their erosion-reducing effects during concentrated flow
erosion, Catena, Vol. 85(3), pp. 237–244.
Dexter A.R. (1991) Amelioration of soil by natural processes, Soil & Tillage Research, Vol. 20(1), pp. 87–100.
Emerson, P., Skousen, J. and Ziemkiewicz, P. (2009) Survival and growth of hardwoods in brown versus gray sandstone on a surface
mine in West Virginia, J. Environ. Qual., Vol. 38, pp. 1821–1829.
Evanylo, G.K., Abaye, A.O. Dundas, C. Zipper, C.E. Lemus, R. Sukkariyah, B. and Rockett, J. (2005) Herbaceous vegetation productivity,
persistence, and metals uptake on a biosolids-amended mine soil, Journal of Environmental Quality, Vol. 34, pp. 1811–1819.
Government of Quebec (2014) Annexe I – Règlement sur la protection et la réhabilitation des terrains, Loi sur la qualité de
l’environnement, Chapitre Q2 r.37. , viewed 30 Sept. 2014, http://www2.publicationsduquebec.gouv.qc.ca/dynamicSearch/
telecharge.php?type=3&file=/Q_2/Q2R37.htm
Green, S. and Renault, S. (2008) Influence of papermill sludge on growth of Medicago sativa, Festuca rubra and Agropyron
trachycaulum in gold mine tailings: A greenhouse study, Environmental Pollution, Vol. 151, pp. 524–531.
Greene, D. and Johnson, E. (1998) Seed mass and early survivorship of tree species in upland clearings and shelterwoods, Canadian
journal of forest research, Vol. 28, pp. 1307–1316.
Grevers, M.C.J. and De Jonc, E. (1990) The characterization of soil macroporosity of a clay soil under ten grasses using image analysis,
Can. J. Soil Sci., Vol. 70, pp. 93–103.
Grossnickle, S.C. (2005) Importance of root growth in overcoming planting stress, New For., Vol. 30, pp. 273–294.
Halofsky, J.E. and McCormick, L.H. (2005) Effects of unseeded areas on species richness of coal mines reclaimed with municipal
biosolids, Restoration Ecology, Vol. 13(4), pp. 630–638.
Jobidon, R., Charette L. and Bernier P.Y. (1998) Initial size and competing vegetation effects on water stress and growth of Picea
mariana (Mill.) BSP seedlings planted in three different environments, For. Ecol. Manage., Vol. 103, pp. 293–305.
Kost, D.A. and Vimmerstedt, J.P. (1994) Ground cover and tree growth on calcareous minesoils : greater influence of soil surface than
nitrogen rate or seed mix, Proceedings International Land Reclamation and Mine Drainage Conference and the 3rd
International Conference on the Abatement of Acidic Drainage, 24–29 Avril 1994, Pittsburgh, PA, USA, pp. 295–304.
Larchevêque, M., Desrochers, A. Bussière, B. Cartier, H. and David, J.-S. (2013) Revegetation of non-acid-generating, thickened tailings
with boreal trees: a greenhouse study, Journal of Environmental Quality, Vol. 42, pp. 351–360.
Larchevêque, M., Desrochers, A. Bussière, B. and Cimon, D. (2015) Planting trees in soils above non-acid-generating wastes of a boreal
gold mine, Ecoscience (accepted).

Lunt, P.H. and Hedger, J.N. (2003) Effects of organic enrichment of mine spoil on growth and nutrient uptake in oak seedlings
inoculated with selected ectomycorrhizal fungi, Restoration Ecology, Vol. 11(2), pp. 125–130.
McCully, M.E. (1999) Roots in soil: unearthing the complexities of roots and their rhizospheres, Annu. Rev. Plant Physiol. Plant Mol.
Biol., Vol. 50, pp. 695–718.
Oades, J.M. (1984) Soil organic matter and structural stability: mechanisms and implications for management, Plant and Soil, Vol. 76,
pp. 319–337.
Olsen, S.R., Cole, C.V. Watanave, F.S. and Dean, L.A. (1954), Estimation of available phosphorus in soils by extraction with sodium
bicarbonate, USDA Circ. 939., pp. 1–19.
Sahramaa, M. and Jauhiainen, L. (2003) Characterization of development and stem elongation of reed canary grass under northern
conditions, Industrial Crops and Products, Vol. 18(2), pp.155–169.
Schuurman, J.J. (1965) Influence of soil density on root development and growth of oats, Plant and Soil, Vol. 22, pp. 352–374.
Skousen, J., Ziemkiewicz, P. and Venable, C. (2006) Tree recruitment and growth on 20-year-old, unreclaimed surface mined lands in
West Virginia, International Journal of Mining, Reclamation and Environment, Vol. 20(2), pp. 142–154.
Tisdall, J.M. (1991) Fungal hyphae and structural stability of soil, Aust. J. Soil Res., Vol. 29, pp. 729–743.
Tordoff, G.M., Baker, A.J.M. and Willis, A.J. (2000) Current approaches to the revegetation and reclamation of metalliferous mine
wastes, Chemosphere, Vol. 41, pp. 219–228.
Wang, J., Hesketh, J.D. and Woolley, J.T. (1986) Preexisting channels and soybean rooting patterns, Soil science, Vol. 141, pp. 432–
437.
Wick, A.F., Stahl, P.D., Rana, S., and Ingram, L.J. (2007) Recovery of reclaimed soil structure and function in relation to plant
community composition, Proceedings National Meeting of the American Society of Mining and Reclamation, Gillette, WY,
USA, pp. 941–957.
Whitbread-Abrutat, P.H. (1997) The potential of some soil amendments to improve tree growth on metalliferous mine wastes, Plant
and Soil, Vol. 192, pp. 199–217.

S. Hilts Teck Resources Limited, Canada

W. Franklin Teck Coal Limited, Canada
T. Gullison Hardner & Gullison Associates, Canada
In 2011, Teck adopted an aspirational, long-term (2030) vision to achieve a net positive impact (NPI) on
biodiversity by maintaining or re-establishing self-sustaining landscapes and ecosystems that lead to an
agreed set of viable, appropriate and diverse long-term land uses in the areas where Teck operates. In the
same year, Teck completed a high-level internal guidance manual, which described emerging trends in
biodiversity management and provided a tool kit to help sites with initial stages of biodiversity planning.
In 2012, Teck conducted a company-wide scan to prioritise biodiversity issues at its operating and advance
project sites. This exercise helped the company select two sites for pilot testing its biodiversity planning
workbook and aided in prioritising the remaining sites for corporate and technical support.
Teck next engaged its personnel from operating sites and other corporate functional units to develop a draft
NPI strategy to direct the scope of Teck’s NPI vision and to explain how sites should use the mitigation
hierarchy to pursue NPI and which biodiversity targets a site should seek to achieve. A key aspect of Teck’s
strategy is the use of quantitative metrics to demonstrate NPI on beneficial, valued and sustaining ecosystem
and biodiversity elements that are relevant to the company’s operations and activities. These elements
include natural habitats and ecosystems; critical landscape functions (e.g., connectivity); highly threatened
and/or vulnerable populations and species of plants and animals; and ecosystem services.
A draft biodiversity management planning workbook was piloted in 2013 at the two sites selected in 2012. In
2014, the spreadsheet-based workbook and a guidance document for developing site-specific biodiversity
management plans were rolled out to all sites. By the end of 2015, Teck is committed to developing
biodiversity management plans that set out how NPI may be achieved on the basis of information that has
been gathered and assessed to date.
Experience with applying Teck’s biodiversity management vision, targets and tools has produced several key
learnings:
 Teck’s NPI-based approach is often more in line with the expectations of communities of interest
than other approaches prescribed by environmental assessment agency policies.
 The NPI-based approach can motivate improved efforts at the avoidance, minimisation and
rehabilitation steps of the mitigation hierarchy.
 Carefully selected and managed offsets will often be required to replace ecosystem and
biodiversity elements that cannot be restored on site in a meaningful time frame.
This same experience has revealed some challenges, including developing metrics for NPI planning and
performance monitoring that achieve credibility and comparability, demonstrating additionality of
conservation actions, ensuring the longevity of offsets and identifying available offset opportunities.
Biodiversity conservation is the practice of sustaining the abundance and variety of living organisms on our
planet through actions ranging from protecting animal or plant populations and habitats to rehabilitating
impacted ecosystems. Biodiversity is also the basis of many ecosystem services that sustain people and the

Achieving biodiversity conservation goals in mine development, operation and closure S. Hilts, W. Franklin and T. Gullison
natural environment — from providing clean water to recycling nutrients and pollinating. Increasingly
common pressures on and related losses of biodiversity threaten many of the ecosystem services on which
people and nature depend. In recent decades, conservation scientists, advocates, governments and industries
have become aware of the need to assess and address potential impacts on plants, wildlife, fish and their
habitats on multiple scales and, in many cases, with a greater sense of urgency.
Mining can impact plants, animals and their habitats both directly and indirectly. Direct impacts can result
from any mining activity that involves land clearing (such as road construction, exploration drilling and
overburden stripping or tailings impoundment construction) or discharges to water bodies or the air. Indirect
impacts can result from social or environmental changes that are induced by mining operations, particularly
when mining opens up an area for other economic activities and increased habitation by people. In cases
where mines are developed in landscapes and other pressures on biodiversity are present, the potential for
cumulative impacts must be considered.
Biodiversity is a key concern for many of Teck’s communities of interest such as investors, Indigenous
peoples, governments, local communities and conservation organisations. Impacts to biodiversity judged
unacceptable by communities of interest can potentially slow or even halt the approval process for mine-life
extensions and new projects, while exemplary biodiversity stewardship has the potential to strengthen
relations with and support from communities of interest.
In mid-2011, Teck published a Sustainability Review report (Teck Resources Limited, 2011), which shared our
long-term (2030) vision and goals for performance in six focus areas: communities, water, energy,
biodiversity, materials stewardship and our people. These focus areas and the strategies for each were
developed by the company from late 2009 through mid-2011 and involved extensive research and dialogue
with senior management of the company; identified emerging leaders of the company; and a cross-functional
working group of Teck employees, sustainability consultants and external communities of interest.
Teck is committed to having a net positive impact (NPI) on biodiversity in the regions where we operate. The
company’s intention is to achieve NPI on biodiversity by maintaining or re-establishing self-sustaining
landscapes and ecosystems that lead to an agreed set of viable, appropriate and diverse long-term land uses
in the areas where Teck operates. Our aim is to leave key biodiversity elements — consisting primarily of
ecosystems and the species that depend on them — better off overall as a result of our presence in a region
than they would be if we were never there. Teck’s portfolio of operating sites consists of six open-pit
steelmaking coal mines in western Canada, three open-pit copper mines (two in Chile and one in western
Canada), an underground copper mine in eastern Canada, an open-pit zinc mine in Alaska, U.S.A., an
underground zinc mine in Washington, U.S.A. and an zinc refinery in western Canada.
In late 2011, we completed a high-level internal guidance manual, which described emerging trends in
corporate biodiversity management and provided a tool kit to help sites with initial stages of biodiversity
planning. It reflects the general principles of the Good Practice Guidance for Mining and Biodiversity of the
International Council on Mining and Metals (ICMM) (ICMM, 2006) and draws from many of the tools
contained in this publication. The Teck guidance manual also describes key elements of relevant international
frameworks for biodiversity management, particularly the International Finance Corporation’s Performance
Standard 6 for Biodiversity (IFC, 2012) and the Business and Biodiversity Offsets Programme principles for
offsets design (BBOP, 2012).
The Teck sustainability strategy includes long-term goals (for achievement by 2030) and short-term goals (for
achievement by 2015). This paper describes our approach to achieve our key operational 2015 biodiversity
goal:
Develop comprehensive biodiversity management plans including targets and actions, to minimize
impacts at all operations and advanced projects, in accordance with our Biodiversity Guidance
Manual and corporate standards. (Teck Resources Limited, 2011)
An overview of the steps we have taken to achieve our short-term goal is shown in Figure 1. The next section,
describes our approach to (i) identifying pilot sites for developing biodiversity management plans, (ii) working

Vegetation
with our personnel and communities of interest to develop a strategy and targets for achieving NPI at our
sites, (iii) developing and testing a biodiversity management plan workbook and guidance manual at two pilot
sites and (iv) rolling out the development of biodiversity management plans to the remainder of our sites
and developing additional guidance. Section 3 (Results and discussion) presents the results of our efforts
including our detailed NPI strategy and goals and lessons from our pilot project and broader roll-out.
We conducted a high-level scan of biodiversity considerations across Teck’s portfolio of sites in order to select
two pilot sites and prioritise the remaining sites for corporate and technical support once the pilots were
complete. The two sites selected as pilot locations would receive corporate and technical expert support to
develop the first iteration of a biodiversity management plan using a draft Microsoft Excel TM-based
workbook. This exercise included 13 operating sites (12 mines and one smelter/refinery), two development
projects in the feasibility stage and one project in the pre-feasibility stage.
First, we developed criteria to support the prioritisation exercise (Table 1), which reflected factors such as
biodiversity risks and opportunities and legal requirements.
The next step was to collect sufficient information to assess the 16 sites against the criteria that had been
developed. The information sources consulted in the assessment are listed in Table 2. With a few exceptions,
site managers had sufficient information to be well informed about the biodiversity issues that were relevant
to their sites. Therefore, it was not necessary to rely heavily on desktop searches of online biodiversity
databases (e.g., NatureServe). Each site received a score from 0 to 2 for each criterion, and then these
individual scores were combined and normalised to create an overall site score that could range from 0% to
100%. Sites with the highest scores were the leading contenders to be selected as pilot sites.

Criterion Rationale for collection/inclusion

Risk to biodiversity posed by An expectation of risks to high-value biodiversity elements would
site suggest additional mitigation actions might be needed to achieve net
positive impact.
Impacts to biodiversity created Historical impacts to high-value biodiversity features would suggest
by site additional work might be needed to achieve net positive impact.
Timing of biodiversity Mines in the process of permitting might be required to have plans
management plan requirement sooner than Teck’s own target date of 2015.
Stakeholder risks Strong negative stakeholder positions concerning actual or perceived
impacts or risks to biodiversity might create reputational or regulatory
risk.
Stakeholder opportunities Existence of compelling, time-sensitive opportunities to implement
conservation projects with stakeholders might help build social
license.
Maturity of current biodiversity Readiness at sites with respect to biodiversity stewardship would
management at site influence the amount of effort and resources required to support
biodiversity planning.
Stage of mine Certain stages of the mine life cycle may offer time-sensitive
opportunities to implement biodiversity mitigation actions.
Representativeness Sites that are representative of the conditions and challenges facing
other sites may be more useful as pilots than sites that are unique.
Source Description
Media search Google News archives were searched for articles on all Teck sites published
from 2007 to 2012. These articles were then reviewed for mention of
biodiversity issues. The issue and website link were recorded for each issue
mentioned.
Non-governmental Websites of 50–100 higher profile NGOs were searched for reference to
organisations (NGO) biodiversity issues (either positive or negative) for each of the Teck sites over
website search the last five years. Information on the biodiversity issue, year, NGO and its
website were noted.
Site manager An interview template was designed that addressed most or all of the
interviews prioritisation criteria. Interviews were conducted with
environmental/biodiversity managers at each site; these took from 1 to 1.5
hours to complete.
Biodiversity documents For some sites, considerable biodiversity information was available from Teck
from site or regulator websites or was provided by site managers after the interviews.

Vegetation
In order to develop detailed NPI strategy and targets, in late 2012 Teck engaged its operations and corporate
functional personnel (e.g., Exploration, Project Development) to provide direction on the scope of Teck’s NPI
vision, the methods for applying the mitigation hierarchy in pursuing NPI and the biodiversity targets a site
should seek to achieve. A custom online survey was used to survey personnel at their convenience, and this
was followed by a facilitated workshop.
The survey offered respondents a range of possible targets for various aspects of a corporate NPI strategy.
For example, with respect to the baseline against which sites would measure their net biodiversity impacts,
respondents were offered the following choices: (i) the premine condition; (ii) 2011, the year Teck made its
NPI commitment; (iii) the year that Teck acquired the mine; and (iv) other (respondents who chose this option
were asked to provide details). The consensus or majority choices would form the basis of an initial draft of
NPI Strategy and Targets, which could then be subjected to additional review.
Once the NPI strategy and targets were developed, we developed a workbook that took sites through a
systematic process in developing their biodiversity management plans. We define a biodiversity management
plan as a high-level document that describes an operation’s plan for achieving NPI, as defined by the NPI
strategy and targets.
In general, the biodiversity management plan should include such things as the following:
 a list of ecosystem and biodiversity elements at the site that fall within the scope of Teck’s NPI
strategy and targets;
 a summary of the risks and impacts that the site and its activities pose to the identified ecosystem
and biodiversity elements (both risks and impacts should be calculated in reference to the
relevant baseline as defined by Teck’s NPI targets and strategy);
 a mitigation plan, developed using the biodiversity mitigation hierarchy, that demonstrates how
the site will manage its impacts (avoid, minimise, rehabilitate, offset and undertake additional
conservation actions) to achieve net positive impact for each ecosystem and biodiversity element;
and
 a list of activities required to implement the mitigation plan, including monitoring of plan
execution and effectiveness.
The draft Biodiversity Management Plan Workbook was tested and refined on the pilot sites identified above
before being rolled out to a broader audience.
Personnel from all 16 operating sites and advanced projects participated in interviews and provided data and
other information for this assessment. Personnel were well informed about biodiversity issues at their sites,
and the 100% participation rate helped to ensure the accuracy of the results and provided an opportunity to
engage sites regarding the biodiversity program.
The two criteria that made the largest contribution to site prioritisation scores were the residual risks to
ecosystem and biodiversity elements posed by Teck’s sites and the risks to Teck posed by community of
interest positions around biodiversity issues (Figure 2). The majority of sites scored high against both criteria,
so these two criteria made the largest contribution to overall site scores. The two criteria are clearly linked,
and they foreshadow the benefits in risk reduction and increased social license that successful
implementation of our NPI strategy may deliver.

The two criteria that made the smallest contribution to prioritisation scores were representativeness and
timing of the biodiversity management plan requirement. Only the individual steelmaking coal mines in the
Elk River Valley in southeastern British Columbia, Canada, were found to be representative of many other
Teck sites; and only Line Creek had an external requirement for a plan that was significantly in advance of
Teck’s internal deadline.
Line Creek, an open-pit steelmaking coal mine, emerged as the top priority for the pilot testing, owing to
clear and imminent regulatory and community of interest expectations for development of a stand-alone
biodiversity management plan as part of its Environmental Assessment Certificate for a mine-life extension
project. In the end, Quebrada Blanca, an open-pit copper mine in Chile, was selected as the second pilot site
in order to include one pilot each from the Northern and Southern Hemispheres and from two of Teck’s three
main commodity-based business units. The remaining sites were prioritised for support going forward,
primarily on the basis of the timing of regulator and/or community of interest expectations.
The two pilot sites received corporate and expert technical support to develop first-iteration biodiversity
management plans. External technical consultants worked with site environmental managers and their
regional consultants to populate the Biodiversity Management Planning Workbook with baseline information
on relevant ecosystem and biodiversity elements and existing or planned mitigation actions. Consultants and
managers then worked together to evaluate impacts and risks and identify where anticipated residual risks
would likely require reclamation and closure plans and/or the design of appropriate biodiversity offsets to
be revisited.
As in the overall sustainability strategy development process, the development of our NPI strategy revealed
that our personnel are very interested in participating as the company sets direction for this key area of

Vegetation
environmental stewardship. Participation in the online survey and workshop to develop Teck’s NPI strategy
was very strong, and several participants expressed appreciation at the opportunity to be significantly
involved in a manner that fit with their workload and schedule.
The results of the online survey were presented and discussed at the workshop (an example of survey results
is shown in Figure 3). Presenting the results in this format facilitated a rich discussion around the different
views that people held; during the course of this discussion, it was clear that some people modified their
positions.
A draft of NPI Strategy and Targets, which used the majority responses to the survey, was then presented
and discussed. Together the workshop participants worked through their differences and towards a revised
draft that met most people’s expectations. The active input and participation from the executive level of the
company was instrumental in helping the group reach consensus on the content of the strategy.
Some of the key elements of Teck’s NPI strategy and targets emerged from the workshop:
 The scope of the NPI commitment applies to natural habitats and ecosystems; critical landscape
functions (e.g., connectivity); and highly threatened and/or vulnerable populations and species of
plants and animals, sites and ecosystem services.
 Teck sites will use the biodiversity mitigation hierarchy to manage their biodiversity impacts and
develop mitigation strategies to achieve NPI (Table 3). Figure 4 shows how the mitigation
hierarchy can be applied during the various stages of the mining life cycle to move from an
unmitigated impact level to an NPI by mine closure or sooner.
 The baseline against which NPI will be measured is the condition of the site prior to mining.
 Quantitative metrics will be used to demonstrate NPI on beneficial, valued and sustaining
ecosystem and biodiversity elements that are relevant to Teck’s operations and activities.
 For those biodiversity elements that can tolerate temporal losses without losing significant
viability or function, NPI should be achieved by the time of mine closure. Otherwise, NPI should
be achieved as soon as practicable without jeopardising the viability or function of those
elements.

Mitigation type Description

Whenever possible, avoid impacts. Some biodiversity features are sufficiently
Avoid
important and/or threatened to require a change in plans to protect critical areas.
Minimise impacts that are unavoidable, adopting best practices in mine
Minimise
operations in order to reduce the severity of impacts.
Rehabilitate areas to re-establish ecosystem and biodiversity elements.
Rehabilitate Reclamation practices can replace much or most of the diversity of natural
habitats that existed before mining.
It may not be feasible to fully rehabilitate all ecosystem and biodiversity elements
Offset that are impacted. For these elements, design and implement biodiversity offsets
to achieve a net positive impact on biodiversity.
Certain types of conservation actions are important, but they may produce
Other conservation benefits that are measured in different ways than our impacts are measured, and
actions so they do not contribute quantitatively to our NPI calculations (e.g.,
environmental education).
A key lesson from Teck’s biodiversity strategy development and communication process has been that many
employees from various levels and locations in our company are interested in participating in efforts to define
aspirational goals that will drive continued performance improvements that will ultimately be needed to
sustain the longevity of our business and the positive contributions we make to society. Packaging our bold
vision and goals in a clear report and other communication tools has also been effective in gaining increased
support and interest from investor groups, potential partner organisations and our communities of interest.
One ongoing challenge has been to effectively reach all segments of our target audiences with our key
messages. “Biodiversity” is a term with a very broad meaning, and it is not simple to concisely convey what
our strategy means in terms of concrete differences in practical approach and actions. This will require

Vegetation
continued efforts in expertly crafting simple messages and case studies that resonate with people, especially
the general public and our employees in communities where we operate.
We developed a draft workbook-based approach to developing biodiversity management plans and tested
the workbook at the two pilot sites. The process was broken down into nine steps (Figure 5). Steps 1 and 2
scope the assessment both spatially and with regard to the ecosystems and biodiversity elements that should
be included. In Steps 3 and 4, impact and risk assessments determine current net impact and risk rating (i.e.,
including consideration of planned mitigation) on ecosystems and biodiversity elements. For sites that
currently expect a net negative impact on all or a subset of elements, or that pose high or critical risks, Steps
5 through 7 involve the development of additional mitigation actions following the biodiversity mitigation
hierarchy to achieve NPI. In Step 8, sites develop and schedule actions to implement the plan. In Step 9, sites
implement the action plan, monitor their progress towards NPI and use this information to adaptively
manage the plan as necessary.
The process is designed so that a biodiversity management plan can be produced at any stage in the life of a
mine, although its completeness may vary depending on what information is available for the site at that
time. Older operations may find that developing the first iteration of their plan identifies significant gaps in
Step 2. For example, baseline studies may not be available for all types of potential ecosystem and
biodiversity elements. Accordingly, the priority actions described in the first iteration of the plan will include
conducting additional baseline studies. It may take several more iterations of the biodiversity management
plan before the site has a fully developed mitigation plan. In contrast, recently permitted sites may find that
they have all the necessary information to develop a comprehensive mitigation plan in their first iteration.
Teck’s biodiversity management plans are intended to be “live” documents, updated periodically to reflect
the results of the biological monitoring of the effectiveness of management, changing regulations and
changes in the conservation status of different types of biodiversity elements.
Important lessons were learned at the pilot sites:
 Teck’s NPI strategy and targets align more closely with the expectations of communities of
interest than approaches often prescribed by environmental assessment agency policies.
 The process of developing a biodiversity management plan generated considerable creativity and
enthusiasm on the part of site personnel to identify additional avoidance, minimisation and
rehabilitation measures. For example, personnel at the Line Creek Operations pilot site discovered
that they could revisit and revise the site’s reclamation and closure plan to increase its
contribution to achieving NPI. Personnel also realised that Teck could design and implement
possible biodiversity offsets using its own considerable land holdings.
 Most Teck sites are likely to require carefully selected and managed biodiversity offsets to
mitigate for ecosystem and biodiversity elements that cannot be restored on site in a meaningful
time frame.
 The development of biodiversity management plans will need to overcome some technical
challenges such as reconstructing credible premine biodiversity baselines for older sites,
understanding a site’s impacts to species that are poorly studied (e.g., rare or listed, irregularly
distributed plant species) and identifying high-quality benchmark vegetation that can be used to
develop a measurement framework to assess the success of our rehabilitation efforts.
Experience at the pilot sites also indicated that some improvements to the workbook were needed and that
sites would benefit from a guidance document, which was subsequently developed.

Steps in Developing a Biodiversity Management Plan
1. Area
delineation
Decision Points
No additional controls and
2. Elements A mitigation required to
identification & achieve NPI
Updating and
adaptive amalgamation Additional controls and
management B mitigation required to
achieve NPI
3. Identification
of potential Biodiversity offsets
effects C required to achieve
NPI
9. Implementation 4. Risk and

and Monitoring A impact
assessment
B 6. Opportunity
5. Additional identification
mitigation planning C
and updated risk and
impact assessment
A 7. Offset
calculations
8. Develop
actions
In 2014, the workbook and guidance document for developing site-specific biodiversity management plans,
revised based on the pilot site experiences of 2013, were rolled out to the remaining sites and projects.
Corporate personnel and consultants continued to be available through 2014 to assist sites in applying the
guidance to build their biodiversity management plans. The roll-out was successful and the majority of sites
have developed at least a first iteration of their biodiversity management plans. Several factors have
contributed to this success, including a strong and vocal commitment to biodiversity stewardship from our
senior management, inclusion of progress made on developing biodiversity management plans as a key
operational performance measure, and the interest and dedication of our personnel. Once planning is
completed at our sites, the focus will shift to implementation.
This paper describes how Teck corporate and operations/project personnel have worked together to
carefully define the scope and details of our biodiversity commitment. Our NPI strategy defines clear targets

Vegetation
that drive biodiversity actions that can be embraced by employees and communities of interest. We have
also described how we identified pilot sites to test our tools and approaches and then rolled out the revised
tools across the entire company. Use of these planning tools assists operations to approach their plan
development and implementation in a similar manner. A consistent approach is valuable in enabling the
effects of our activities to be compared and evaluated cumulatively across a region. In addition to meeting
the commitments in our corporate sustainability strategy, Teck biodiversity management plans also serve to
meet aspects of our internal Health, Safety, Environment and Community (HSEC) Management Standards
and the Mining Association of Canada’s Towards Sustainable Mining Biodiversity Conservation Management
Protocol (Mining Association of Canada, 2013).
A key reason for our progress to date is the degree of commitment at all levels of our company. Our senior
management has advocated strongly for Teck’s need to build from its experiences to further improve its
biodiversity stewardship. We have built a strong framework for our future advances with a process that has
benefitted from the active participation of our sites, ensuring that they support and inform what Teck is
trying to achieve.
Business and Biodiversity Offsets Programme (BBOP) (2012) Standard on biodiversity offsets, Business and Biodiversity Offsets
Programme, Washington, D.C.
International Council on Mining and Metals (ICMM)(2006) Good practice guidance for mining and biodiversity, International Council
on Mining and Metals, London, UK.
International Finance Corporation (IFC) (2012) Performance standard 6: Biodiversity conservation and sustainable management of
living natural resources, IFC performance standards on environmental and social sustainability, International Finance
Corporation, Washington, DC.
Mining Association of Canada (2013) TSM assessment protocol: a tool for assessing biodiversity conservation management
performance, The Mining Association of Canada, Ottawa, Canada.
Teck Resources Limited (2011) 2010 Sustainability Review, viewed 14 February 2015, http://www.teck.com/DocumentViewer.
aspx?elementId=197070&portalName=tc.


F. Le Loher French Geological Survey (BRGM), France

F. Cottard French Geological Survey (BRGM), France
P. Chartier Languedoc-Roussillon Division, French Ministry of Environment, France
The Salsigne mine in southern France was one of the largest gold and arsenic mines in Europe, producing 105
t of gold and almost 200,000 t of arsenic compounds over the period of a century. The site is characterised by
a complex geological context associating gold-bearing sulphide deposits hosted by carbonate meta-
sedimentary rocks and thrust-controlled ore emplacement. Different types of sulphidic ores have been
processed using several pyro-metallurgic and hydrometallurgical methods, one after the other or combined
over time. These have generated significant quantities of mining waste, mainly in the form of tailings, which
was the source of a serious arsenic contamination in the surrounding soils, surface and groundwater. The
operator closed the mine in 2004, and the French government rehabilitated highly contaminated areas at the
same time.
Since 2006, on behalf of the French State, BRGM has run a comprehensive program to monitor the residual
impacts related to mining and metallurgical waste disposals, as well as the flooding of underground workings.
A geo-environmental model has been designed to help understand the natural processes taking place at
different sites — in particular, the La Combe du Saut former mineral processing site. This required the
integration of current monitoring data with results from specific investigations conducted on groundwater
geochemistry. The model also takes into account other data, such as ore deposit geology, watershed
characteristics and information related to historical operations.
The use of a geo-environmental model is a practical approach for integrating past and current monitoring
data within the complex geological and hydrogeological context. During the post-closure phase, it provides
the authorities with a low-cost means to help control long-term residual risks and prioritize effective remedial
actions. It can also be used to review solid waste management practices applied in the 1980s and 1990s with
the objective of proposing new guidelines for operational mines on aspects like waste storage design, location,
construction and closure options.
The Salsigne mining district is located 14 km north of the city of Carcassonne in southern France, as illustrated
in Figure 1. It comprised one of the largest gold and arsenic mines in Europe for almost a century. From 1924
to 2004, the main extraction site produced a total of 105 t of gold and almost 200,000 t of arsenic trioxide
and other by-products: mainly silver (250 t), copper, bismuth and sulphuric acid. Historically, the area has
been the site of mine workings and metallurgical activities (mainly for iron, lead and silver) since the Bronze
Age. Gold was discovered at the end of the nineteenth century as inclusions within sulphides (e.g.
arsenopyrite), giving rise to important local industrial development. Since then, different operators and
mining companies have worked the mine for about a century through a series of complex mergers and
acquisitions. For instance, from 1989 to 1997, the SNC Lastours Company exploited the residual tailings
disposed on the site since 1924; the Mine d’or de Salsigne (MOS) Company, the last operator, started mining
in 1992 and did so until the end of 2004.

A geo-environmental model for long-term post-mining management: The former Salsigne gold mine area F. Le Loher, F. Cottard and P. Chartier
Industrial activities were developed in different places in the upstream part of the Orbiel River, where ore
processing plants were successively constructed from 1924 to 1989 at the La Combe du Saut site located
four kilometres southeast of the main Salsigne mine site.
This paper aims to present a learning experience from the long-term monitoring program implemented on
this large-scale mining site at the end of the rehabilitation programs in 2006. It is focused on describing the
site through its geological and industrial contexts as well as its waste management practices, integrated in a
geo-environmental site model that helps in understanding the monitored residual environmental concerns
that still exist at the site.
The Salsigne area is characterised by a complex geological setting and a wide variety of mineralised zones
exploited during a long period using continuously evolving methods and processes that should be well
described and understood when tackling mining environmental problems.
The Salsigne deposits are similar to those reported for many mesothermal gold deposits in metasedimentary
belts. The main deposit is composed of bedded ore (generally enriched sandstone), massive sulphide layers
and quartz lodes as well as sulphide-poor disseminations hosted by metasedimentary rocks of the Early
Cambrian to Devonian age. Deposits are restricted to medium-grade, generally greenschist facies
metamorphic rocks. Successive ore emplacements were controlled by a complicated tectonic regional setting
associating highly deformed shear zone and major folding and thrusting (Lescuyer et al., 1993). These host
rocks are dominantly composed of metasandstone and metasiltstone progressively grading to limestone and
dolomite. Distinct massive sulphidic layers (called 2x and 3a2x ores) and enriched sandstones are found in
the interbedded siltstone-sandstone-carbonates of the northern limb of the Salsigne synform (Figure 2).
Within this diversified geologic terrane, natural groundwater that flows in the vicinity and within the deposits
is dominantly along fractures and through a developed karst system in carbonate rocks.
As illustrated in the cross section of Figure 2, the two major quartz lode systems, the Fontaine de Santé and
Ramèle lodes, which cross-cut the thrusted metasedimentary pile, are between 600 and 800 m long and 300
m apart with an average dip of 55° E. They are mainly composed of interlaced mineralised quartz and sulphide
veins that can be up to four metres thick. The different types of sulphidic ores are characterised by a high
content of iron sulphides (pyrite and pyrrhotite) and, in arsenic-bearing sulphides, mainly arsenian pyrite and
arsenopyrite associated with some chalcopyrite (0.05–1.3%) and traces of bismuthinite. Ore arsenic
concentrations vary from 1% in the massive iron sulphide mineralisation up to 15% in the massive arsenic

Modelling
sulphide layers, the famous 2x ore discovered at depth in 1973 (Tollon, 1970) (Table 1). Gangue minerals
include quartz, feldspars, some mica and abundant carbonates in the form of calcite, ankerite, siderite and
dolomite that will largely contribute to acid rock drainage buffering. In near surface parts of the deposit, iron-
and arsenic-bearing sulphides are exposed to weathering, which resulted in limonite and other arsenic-iron
oxide minerals. Finally, the deposit exhibits a characteristic suite of trace elements, including arsenic, silver,
copper and bismuth, whose abundances are elevated.
Massive Massive arsenic Iron sulphide-rich

Type of ore and Quartz Enriched sulphide disseminated
iron
composition veins sandstone
sulphide ore (2x) ore (3a2x) ore
Mineralogical ore
Silicates (%) 18.70 15.30 12.00 11.00 14.90
FeAsS (%) 15.80 10.10 2.20 31.50 2.80
FeS (%) 19.50 3.10 48.90 16.50 39.10
CuFeS2 (%) 1.30 0.05 0.60 0.30 0.40
Trace element
S (%) 10.70 3.100 18.500 12.300 15.000
As (%) 7.30 4.600 1.000 14.500 1.300
Cu (%) 0.40 0.020 0.200 0.110 0.100
Fe (sulphide) (%) 18.30 5.500 32.000 21.400 26.000
Au (ppm) 13.40 121.000 14.000 8.200 4.000
Ag (ppm) 72.50 19.200 2.000 50.200 5.000
Bi (%) 0.13 0.230 0.040 0.280 0.010
Sb (%) 0.03 0.010 0.001 0.005 0.002
Pb (%) 0.07 0.030 0.002 0.030 0.005
Zn (%) 0.02 0.006 0.001 0.004 0.003

As illustrated in Figure 3, the site is located at a major regional unconformity corresponding to the contact
between the Palaeozoic metasedimentary sequences already described above and a much younger and
gently dipping sedimentary formation dated early Palaeocene, mainly composed of thin layers of sandstone
and marl at the bottom and thicker limestone horizons in the upper part. The lower horizon within these
tertiary formations is of higher permeability and constitutes a small ground and surficial water conduit that
has great importance for site scale in that it imposes a significant control on the local flow regime, which has
a broadly north-south direction.
During the last century, two mining techniques and a large variety of ore processing methods were used to
extract and process the different types of ore.
Underground mining with two successive shafts and, more recently, a three-kilometre-long decline was the
principal mining method used, but this was complemented by open pit mining from 1981 to 2004. A total of
100 km of galleries and drifts had been worked through 16 levels and over a depth of 450 m. Slightly dipping
ore bodies (enriched sandstone and sulphide-rich layers) had been exploited by room and pillar caving,
whereas more steeply quartz lode systems were worked by cut and fill methods (Trueb, 1996).
The open pit mine was 170 m deep with north-south oriented workings conducted on a surface area of
450,000 m2. Today, three large waste rock deposits are located around the former excavation and cover a
total area of 1,300,000 m2.
In total, the Salsigne underground mine produced about 10 million tonnes of ore grading 9 g/t of gold that
were complemented by five million tonnes of excavated surficial ore with an average gold grade of 7 g/t.
The ore mineralogical characteristics with gold particles locked in sulphide minerals and, in particular, pyrite
and arsenopyrite, resulted in the introduction of pyrometallurgy from the beginning of 1924. At that time,
roasting was the only process available to metallurgists for treatment of refractory gold ores, but in 1928, a
pre-concentration step of sulphide mineral flotation was set up with the view of improving costs by only
roasting the concentrates (20 to 30% of ore weight) instead of treating the whole ore. This process generated
about seven million tonnes of tailings, mainly composed of gangue minerals and residual sulphides, which
were accumulated and stored in tailings ponds with an area of 350,000 m2 on the La Combe du Saut site until
1989.

Modelling
Meanwhile, hydrometallurgy was introduced in 1938 with a cyanide and Merril-Crowe process applied to the
roasted concentrates for several decades. MOS Company modified this in 1992 to a carbon in leach (CIL)
circuit applied to the concentrates following a very fine regrinding step (< 25 µm), as illustrated in Figure 4.
A carbon in pulp (CIP) cyanidation process was implemented by SNC Lastours Company to re-treat the large
and old tailings dumps, which contained a grade of 2 g/t of gold. Those tailings materials were mined using
hydraulic methods (hydraulic mining monitors), reprocessed and stored in an engineered impoundment
called the Artus tailings dam, located east of the ore processing plant.
The successive use of various ore processing techniques resulted in the production of huge quantities of
different types of waste, mainly in the form of metallurgical slags and tailings.
A total production of about 900,000 t of slags was estimated, corresponding to the use of the
pyrometallurgical process that prevailed in Salsigne over a period of 70 years. A large quantity of the slags
produced (300,000 t) had been reused on the site as drainage materials within waste facilities or directly sold
outside as a by-product for its abrasive properties. At the end of activities in 2005, about 600,000 t remained
in the form of several dumps scattered throughout the site.
The subsequent flotation process generated the most significant quantity of waste in the form of fine-grained
tailings. Apart from the already-cited materials temporarily stored on site and reprocessed, two conventional
tailings dams were constructed from the end of the 1980s to the beginning of the 1990s in order to receive
waste from two separate cyanidation plants belonging to two different operators (MOS Company and SNC
Lastours Company).
The largest facility, the Artus tailings storage facility (TSF), was designed to store about seven million tonnes
of flotation tailings generated by reprocessing the old tailings with cyanide solutions. It was located in a small
valley in the form of two successive ponds, each constructed with the upstream method using materials
hydraulically separated by cycloning. These methods were, and are still today, the most economical methods
of tailings dam construction. The site was selected to be as close as possible to the cyanidation plant in a

place where the expected volume of storage was sufficient and the geological setting considered appropriate,
at that time, to receive the tailings materials and process water.
The selected area consisted of a 500,000 m2 natural basin bordered to the east by a continuous ridge of low
hills corresponding to the early tertiary geological formation, while the site’s underlying bedrock was
composed of highly deformed Palaeozoic schists (early Cambrian). As a consequence, and as shown in Figure
5, the facility was constructed astride two host geological formations characterised by contrasting
hydrogeological characteristics. While the Palaeozoic schists had quite low permeability, the overlying
Eocene sedimentary sequence was known to contain several small aquifers with springs occurring within the
sandy and marl horizon at the bottom, and karstic characteristics within the overlying limestone formation.
The construction started in 1988 on the left bank of the Orbiel River with a complex subsurface drainage
system designed to both remove pore water from the tailings impoundment and capture groundwater
coming from the catchment area and, particularly, the host tertiary formation. This drainage system
consisted of different types of drains, as illustrated in Figure 6. Blanket toe drains were first constructed
against the starter dams, along with three penstocks connected to internal finger drains and completed,
when the dam was raised, by a system of elevated drains located around the impoundment and at medium
depth within the embankment. In addition, a system of trenches was constructed in the upstream part of the
foundations in order to drain and collect water flowing from the tertiary formations. The whole drainage
system was connected to six reclaim ponds in a cascade series arrangement situated outside the
impoundment with the aim of enhancing evaporation and partly reclaiming the water for use in the
processing plant, as required.
At the end of the operation in 2004, rehabilitation works performed by the mine operator consisted of
covering beaches and dams with 0.5 m of marl and limestone with the aim of stabilising soil slopes and
improving revegetation. A surface water runoff system was also constructed around the disposal site in order
to manage heavy rainfall and storm waters.

Modelling
The second tailings dam, the Montredon TSF, was constructed in 1994 in the northern part of the site to store
about 500,000 t of new tailings produced by the flotation and cyanidation plant owned by MOS Company
(Figure 7).
This facility is located within the tertiary sedimentary formation, and, more specifically, hosted by a 10–15
metre-thick limestone upper horizon that had been partly excavated to ensure structural stability for the
basin’s foundations, upon which the dam was constructed. The bedrock was then covered by a five-metre-
thick layer of compacted clay, and an embankment underdrainage system was installed in the lower part of
the west wall. The dam was raised following the upstream method, with clay (5 m thick) used to build the
retention dykes while tailings (average grain size < 25µm) were deposited using a spigotting technique from
the embankment crest. Part of the process water was reclaimed for use in the plant water circuit through a
water pond.
The facility was designed for an initial capacity of 400,000 m3 (560,000 t) tailings and a total height of 20 m.
It closed in 2004 following four successive raises to reach an ultimate height of 28 m. This expanded capacity
was constructed in order to store the final tailings production together with the content of former ponds
located adjacent to the TSF, which contained about 20,000 m3 of calcium arsenates. The final capacity
reached one million tonnes of tailings and other types of waste. Following the successive dam raises, the
disposal was enlarged and its stability reinforced with the deposition of compacted rockfill materials on the
western and outer slope of the dam to form a buttress berm. This retaining slope weighting structure was
constructed during the final stage of the dam construction.
The closure works included capping of the facility with the aim of limiting infiltration and allowing vegetation
to grow. At the top, the cover system comprised the following layers, listed from bottom to top: rock-fill
subgrade (schists materials), a five-millimetre-thick geosynthetic clay liner (GCL) protected by a separate
geotextile, a layer of fine-grained slags used as drainage material and a final 0.5 metre-thick marl-limestone
cover over the whole facility. A central piezometer equipped with a pump was installed in order to withdraw
residual water from the entire disposal facility.

Beginning in 1970, effluents from the different plants were treated in a water treatment facility using lime to
remove most arsenic from hydrometallurgical process solutions and waste waters by precipitation of alkaline
calcium arsenates Ca3(AsO4)2 and Ca(OH)2. Over three decades, about 40,000 m3 of such precipitates were
produced and stored in different ponds scattered throughout the site. Today, the facility is still in operation
for the treatment of an average volume of 100,000 m 3 per year of contaminated water, with a yearly
production of about 400 t of calcium arsenates, which are stored on site in a specific confined cell.
Geo-environmental models have proven to be powerful tools for understanding the environmental processes
controlling the release, transport and fate of contaminants in mineralised areas (Plumlee, 1999; Plumlee and
Nash, 1995). They are also effective in the process of predicting and anticipating potential environmental
impacts that will occur within the mine life cycle and, more specifically, for taking the appropriate remedial
actions necessary during the post-mining phase (Wanty et al., 2002). Geo-environmental models seek to build
a holistic picture of an ore deposit and the mining/processing operation so that the valuable information and
insights gained from other disciplines are not lost in the design of environmental programs.
By integrating data from multiple origins — such as ore body geology and mineralogy, mining/ore processing
operations and resulting waste management, and hydrologic characteristics — geo-environmental models
facilitate the global interpretation of data in a given complicated mining site and provide the basis for
optimising the post-closure monitoring program.
In the Salsigne area, the orebody’s highly sulphidic mineralisation, expected to be the cause of significant
acid rock drainage, is largely buffered by the carbonate host rocks and the presence of carbonates in the
gangue mineralogy. Consequently, drainage waters and groundwater have near-neutral pH values, generally
between 7 and 8, and the arsenic-rich nature of the ores is manifested in the arsenic-rich water compositions,
which sometimes exhibit concentrations as high as several mg/L. Tailings generated through a flotation
process were enriched in gangue carbonate minerals and show water compositions with similar near-neutral
pH values and much higher iron concentrations — as high as several mg/L. Owing to the abundance of
arsenopyrite, arsenic is present in tailings drainage waters, typically in elevated concentrations ranging from
3 to 16 mg/L. This likely reflects the decomposition of iron arsenate species at higher pH and the high mobility
of reduced arsenite species in reduced drainage waters. Other dissolved base metals, such as Pb, Cu and Zn
are conspicuously low or absent in all types of mine waters. High concentrations of arsenic with low

Modelling
abundance of the base metals are typically characteristic of carbonated sediment-hosted gold deposit mine
waters (Plumlee et al., 1999).
Among the other potential environmental effects of the Salsigne mine, the hydrogeological features of the
limestone host rocks include karstic water flows, which lead to rapid exchanges between the mine waters
and local streams and the regional Orbiel River. This necessitated the prediction of the quality of the
reactivated natural discharge seeps following the mine shutdown.
Owing to the complexity of the historical site ownerships over a century at the La Combe du Saut site, the
French State decided in 2000 to develop a site management strategy by undertaking an extensive
rehabilitation and clean-up program. This project was carried out under the supervision of the French
Environmental Agency (ADEME) on behalf of the state. It started with a detailed site environmental and
human health risk assessment coupled with a comprehensive inventory of historical waste deposits. This
resulted in the identification of 42 disposal locations totalling 2.6 million m3 of waste and contaminated soils
containing approximately 46,000 t of arsenic over an area of 530,000 m2.
The principal pollution flows from the site into the environment were caused at that time by:
 soil erosion by water runoff and direct flow migration of arsenic into the Orbiel River during heavy
rainfall (estimated 1,300 kg/year );
 slow infiltration of water through the waste from the site towards the local surficial aquifers and
finally to the Orbiel River (estimated 300 kg/year);
 transportation of pollution by dust in a particularly windy area.
After demolition of the vast industrial complex (Figure 8), the remediation project aimed to reduce pollutant
flows through water and the air; for this purpose, only the most polluted wastes and soils (about 475,000 m3
in total) were excavated and confined in situ under a bituminous cover. This waste storage was constructed
within a small valley or thalweg and consisted of a channel 300 m long and 20 m deep. The confined wastes
mostly comprised, among other components, some residual tailings and sulphide concentrate, slag materials,
Ca arsenates, 35,000 m3 of demolition waste and 400,000 m3 of removed contaminated soils (> 3,000 ppm
of As).
In addition to the waste containment of highly contaminated materials, there was a need to manage large
areas of soil with a lesser but significant residual arsenic content. A solution consisting of an earlier tested
phytostabilisation technique was implemented in order to minimise the transfer of residual arsenic, mainly
in the form of particulate matter, via surface runoff to the nearby Orbiel River. With this technique, arsenic
was immobilised by the spreading of steel shot and through revegetation using a mixture of selected
indigenous plant seeds. In this way, the adapted vegetation cover provides stability to the soil, wind erosion
is largely prevented and metal leaching to the groundwater can be highly reduced. The project was
successfully carried out within a European-funded LIFE project called the Difpolmine project from 2002 to
2006 (Jacquemin, 2006).

In 2004, MOS Company, the last mine operator, closed the mine site and the associated waste rock dumps,
representing several million tonnes of waste. Specific measures included partly backfilling the open pit,
terracing and vegetating the large waste dumps and, finally, installing gates and fencing. Furthermore, at the
La Combe du Saut site, the processing plant still under MOS responsibility was dismantled, and the two TSFs
were closed and rehabilitated as described in the previous section.
Following the extensive remediation programs completed in 2006 by both the mine operator and the French
State, a site environmental management strategy was developed with the objectives of controlling the
potentially adverse effects of closure and clean-up projects and managing residual risks. This task was
delegated to BRGM, the French Geological Survey, which has acted as the post-mining management agency
on behalf of the state since 2006.
The management system was set up in 2007 and has been continuously upgraded to include the following
tasks:
 survey of post-mining hydrogeochemical conditions at the mine site and, in particular, the deep
mine flooding process;

Modelling
 survey, control and maintenance of closed waste management facilities, including both TSFs and
the confined waste disposal constructed by ADEME with the objective of controlling their long-
term physical integrity and the chemical evolution of drainage water seepages;
 monitoring of the surface water quality within the watershed through a network of six water
monitoring stations on the Orbiel River and five piezometers located in the river alluvial
formations;
 management and maintenance of the water treatment scheme still in operation, including
management of calcium arsenates produced daily;
 monitoring of both Artus and Montredon TSFs through a comprehensive network of regular
topographic surveys of the tailings pond surfaces and seepage monitoring.
Also included are piezometer readings and inclinometer measurements as well as pore pressure probe
measurement.
At the mine site level, few significant residual impacts remain to date at the surface. Following the end of
dewatering in 2001, the deep mine flooding was not yet completed at the time of this writing. As shown in
Figure 9, the deep water table is still slowly rising (about a few centimetres per year), with high water table
fluctuations of between five and 10 metres (lowest and highest water levels) each year. Apart from mining
voids, which can act as preferential groundwater flow pathways, the fluctuating system is highly controlled
by meteoric water inflow and by karstic outflow conduits whose detailed functioning at the regional scale is
not yet completely understood. New piezometers are planned in the near future to better monitor the rising
water table, check groundwater quality and deal with the expected overflow issue that will occur in the
future.
Once the clean-up program was completed at the La Combe du Saut site, a few natural springs were identified
along the tertiary formations’ southern embankments, where some groundwater is channelised. The Point V

spring, identified in the mid-2000s, is located in the southwest of the La Combe du Saut site and is
characterised by elevated concentrations of arsenic (yearly average concentration of 40 mg/L) and a flow
fluctuating between 2 and 12 m3 per hour, which is strongly related to rainfall events. In addition, this spring
provides a water flow to the water treatment system of about 20,000 m3 per year. This represents the major
load of arsenic (0.8 t/year or about 65% of the total) to treatment that is precipitated in the form of calcium
arsenates. This represents one of the major remaining environmental concerns in the area. The main treated
inflow comes from the Artus TSF drainage and the Point V spring, as illustrated in Figure 10.
Process improvements are currently being assessed with a view to replacing the current water treatment
facility with new cost-effective methods, including passive treatment processes.
The stability of the Artus TSF has increased considerably over time since operations ended. A 20 cm
settlement has been observed, but one of the major problems encountered is related to the hydrologic
characteristics of the site. Surface and groundwater have not been diverted around the impoundment nor
stored behind it and are simply drained beneath the impoundment itself. Consequently, a significant quantity
of contaminated drainage water is recovered in the downstream water pond no. 6 (see Figure 6) and is
subsequently pumped to the water treatment system. In 2014, the total volume of effluent treated reached
70,000 m3 at a rate of approximately 190 m3per day and an average arsenic concentration of 8 mg/L. This
contaminated water drainage is expected to last for decades, requiring constant treatment and expensive
maintenance.
Once the Montredon TSF was closed and covered, early signs of instability occurred in the upper part of the
western embankment in relation to the upper dam structure. However, these defects were due to undrained
water that locally accumulated at the base of the cover system and were rapidly repaired. During a drilling
campaign in 2008, it was observed that most of the confined materials were still in a semi-fluid state with
locally high water content. Nevertheless, because of the pumping and drainage realised since closure, an
average settlement of 50 cm has been measured at the surface. This vertical movement corresponds to a
partial consolidation of materials, which slightly improves compaction and geotechnical behaviour. Because
some excess water is still present within the tailings, the installation of new piezometers is planned in 2015
with the aim of determining the current physico-chemical properties of tailings and, in particular, their water
content.

Modelling
The hydrogeology of the La Combe du Saut site is not yet completely understood despite the extensive
groundwater and surface water monitoring program currently in place. However, recent improvements have
been made in the characterisation of the different water flows and the understanding of exchanges between
the various aquifers involved and their relationships with the Orbiel River and its alluvial water body.
The Orbiel water monitoring network set up during the site rehabilitation program is still in operation today.
This makes it possible to monitor surface water quality over time within the watershed and rapidly detect
any unfavourable trend. Figure 11 illustrates arsenic concentrations measured at the La Combe du Saut
downstream monitoring station on the Orbiel River. It shows that, following a peak of concentration
corresponding to the end of the rehabilitation project in 2006, a regular decrease of contamination has been
measured in the river. Since then, slight and persistent water contamination has been observed, particularly
during the summer season in relation to low water levels. This residual impact is due to the site’s shallow
groundwater flows, including inflows from the Artus TSF adjacent to the river.
To address this concern, different studies are currently being conducted at the La Combe du Saut site to
improve the hydrogeological site conceptual model and to accurately investigate the origins of groundwater
contamination. Provisional results of hydro-geochemical studies based on major and trace elements analysis
show that elevated arsenic concentrations at the Point V spring, for instance, are the result of the dissolution
of secondary oxidation products such as Ca arsenates, with no net acidification, likely due to buffering by
tertiary sedimentary host rocks. Other investigations involving ground geophysics have also been conducted
recently to locate residual and hidden arsenate disposal sites that could be the sources of contamination in
the area.
Lessons learned from the understanding of historical data and monitoring results gathered during the
10 years following mine closure and site rehabilitation have led to a better appreciation of the environmental
issues still not fully resolved during the post-mining phase. Most are related to the waste management

strategy adopted at the time of the operation, when the choice of a site was usually controlled by the location
and type of ore processing plant the impoundment was to serve.
When following best practices, however, the choice of a mine waste disposal site during siting studies
depends on both the properties and the potential environmental impacts of the tailings as well as
topographic, hydrologic and geological conditions of the selected site. In the La Combe du Saut site context,
natural groundwater drainage conditions, mainly controlled by the tertiary geological formations, would
today constitute a crucial factor to be taken into consideration in the choice of any waste management
methods. The objective of not interrupting significant natural drainage pathways in the design of the Artus
TSF, for instance, would also be a determining management factor. This large waste disposal site is at the
source of large volumes of contaminated mining waters that need to be treated for the very long term.
Following cessation of mining activities in the Salsigne area in 2004, extensive closure and site rehabilitation
program have considerably improved the environmental situation by decreasing the levels of arsenic
contamination, particularly in the watershed surface waters. Nevertheless, some concerns in relation to
groundwater contamination remain locally at the La Combe du Saut site, which is characterised by a long
history of successive ore processing techniques and diversified waste management strategies.
In the context of the post-mining phase managed by BRGM on behalf of the French State, all the necessary
information provided by the geo-environmental model of the Salsigne gold deposit should be considered
when selecting adaptive remedial solutions. Such a model based on the geology of mesothermal gold
deposits in metasedimentary belts helps integrate geological and industrial-related controls on the
environmental behaviour of the contaminated site. It also facilitates the understanding of such controls on
ground water flows and on arsenic mobility in surficial mining-related environments.
The authors would like to thank Pierrick Défossez, French Geological Survey (BRGM,) for his helpful review.
Jacquemin, P. (2006) The La Combe du Saut site rehabilitation (Salsigne), the Difpolmine project — application and evaluation of large
scale phytostabilisation, in Proceedings Difpolmine Conference, 12–14 December 2006, Montpellier, France.
Lescuyer, J.L., Bouchot, V., Cassard, D., Feybesse, J.L., Marcoux, E., Moine, B., Piantone, P., Tegyey, M. and Tollon, F. (1993) Le
gisemernt aurifère de Salsigne (Aude, France): Une concentration syntectonique tardivarisque dans les sédiments détritiques
et carbonatés de la Montagne Noire, Chronique de la Recherche Minière, Vol. 512, pp. 3–73.
Plumlee, G.S. (1999) The environmental geology of mineral deposits, in The Environmental Geochemistry of Mineral Deposits Part A:
Processes, Techniques and Health Issues, Reviews in Economic Geology, Vol. 6B, G.S. Plumlee and M.J. Logsdon (eds), Society
of Economic Geologists, Littleton, USA, pp. 71–116.
Plumlee, G.S. and Nash, J.T. (1995) Geo-environmental models of mineral deposits — fundamentals and applications, in Preliminary
Compilation of Descriptive Geo-environmental Mineral Deposit Models, E.A. du Bray (ed), Open File Report 95-831, U.S.
Geological Survey, Washington, DC, USA, pp. 1–9.
Plumlee, G.S., Smith, K.S., Montour, M.R., Ficklin, W.H. and Mosier, E.L. (1999) Geologic controls on the composition of natural waters
and mine waters draining diverse mineral deposits, in The Environmental Geochemistry of Mineral Deposits Part B: Case
Studies and Research Topics, Reviews in Economic Geology, Vol. 6B, G.S. Plumlee and L.H. Filipek (eds), Society of Economic
Geologists, Littleton, USA, pp. 373–409.
Tollon, F. (1970) Minéralisations développées au contact “schist X-Paléozoïque” dans la région de Salsigne (Aude), Bulletin de la
Société d’Histoire Naturelle de Toulouse, Vol. 105(3–4), pp. 464–470.
Trueb, L.F. (1996) The Salsigne gold mine: A world-class ore body in the south west of France, Gold Bulletin, Vol. 29, pp. 137–140.
Wanty, R.B., Berger, B.R., Plumlee, G.S. and King, T.V.V. (2002) Geo-environmental models, in Deposit and Geo-environmental Models
for Resource Exploitation and Environmental Security, Nato Science Partnership Subseries: 2, Kluwer Academic Publishers,
Dordrecht, Netherlands, Volume 80, pp. 3–42.

R.M. Nagare WorleyParsons Canada Services Ltd., Canada

Y.J. Park Aquanty Inc., Canada
J. Pal WorleyParsons Canada Services Ltd., Canada
Reclamation of oil sands tailings ponds will be a challenging long-term process. Management of saline pore-
water in the tailings sand is thought to be critical for the successful reclamation on both a landform (single
tailings pond) and site-wide scale. The broad objective of returning the disturbed landscape to a pre-
disturbance equivalency requires understanding and evaluation of environmental performance of conceptual
reclamation designs. Various research on challenges of reclamation of tailings ponds as well as research into
the hydrology of the natural boreal landscape has identified the following desirable hydrologic characteristics
of reclaimed tailings ponds: (1) a water balance that maximises outflow to limit “evapoconcentration” of salts
and sustain end-of-pit lakes; (2) a shallow subsurface flow system to limit transport of salts from within
tailings ponds; and (3) depth to groundwater table in the uplands that limits potential for upward migration
of salinity.
Understanding and evaluating the environmental performance of conceptual reclamation designs with
respect to the above desirable characteristics is challenging given the high potential of groundwater-surface
water interaction expected to occur within a reclaimed landscape. If based on a strong conceptual model,
integrated surface water–groundwater modelling (integrated modelling) can provide powerful tools to
evaluate the potential environmental performance of a reclaimed landscape. Transient simulations, driven
by long-term climatic forcing, could allow for detailed understanding of water balance under wet, average
and dry climatic cycles. Coupling the surface and sub-surface domains could allow for a more realistic
representation of water movement through the landscape than found in uncoupled models. The coupled
modelling approach is thought to be of critical importance, especially under boreal settings, where the
coupling is complex and occurs at many levels (e.g. shallow rooting zones and turning on and off of
evapotranspiration in wetland areas). Challenges remain with scaling of models, especially for coupled flow
and solute transport modelling. This paper discusses the practical issues and application of integrated models
to the complex problem of understanding and evaluating the potential environmental performance of
reclaimed tailings ponds.
In this paper, we discuss challenges and opportunities for application of integrated surface water and
groundwater flow and solute transport modelling (integrated modelling) codes to oil sands mine
reclamation. An integrated modelling tool was developed using HydroGeoSphere (HGS) for a hypothetical oil
sands mine reclamation design. The exercise was conducted to demonstrate the strength of integrated
modelling in evaluating environmental performance of oil sands mine reclamation designs. Such assessments
require integrated models to address:
 the strong coupling of surface water and groundwater in the Boreal Plains;
 the water deficit (annual potential evapotranspiration [PET] exceeds annual precipitation) climate
regime in the Boreal Plains;

Integrated surface water and groundwater modelling for oil sands reclamation R.M. Nagare, Y.J. Park and J. Pal
 crucial roles of unsaturated zone and evapotranspiration play in establishment of treed uplands
and organic wetlands in the reclaimed landscape.
Apart from environmental assessments of reclamation designs, integrated modelling helps in identifying the
extent of areas expected to behave as uplands, wetlands and transitional (lowland) areas irrespective of
topographical delineation. Such delineation is helpful in planning vegetation strategy during active
reclamation and helps save effort and money in long term. Further, assessment of water table configuration
under different climate scenarios can assist in identifying areas with geotechnical concerns. The applications
are demonstrated through the hypothetical case study.
Alberta’s oil sands are the third-largest proven crude oil reserve in the world (AER, 2015a). Oil sand is a
naturally occurring mixture of sand, clay or other minerals, water and bitumen (AER, 2015a). The oil sands
are vital to Canada’s economy. Bitumen production in 2012 stood at 1.9 million barrels per day and is
expected to increase to 3.8 million barrels per day by 2022 (AER, 2015a). While the oil sands underlie a
142,200 km2 area in north-eastern Alberta, the oil sands surface mining area is limited to a 4,800 km2 region
near Fort McMurray (Figure 1). The rest is economically recoverable by in situ methods only, wherein
bitumen is separated from the sand in the subsurface and pumped to the surface. Of the 4,800 km2, 767 km2
is currently disturbed by surface mining operations (Figure 1). Alberta’s land resource base is protected under
the Environmental Protection and Enhancement Act (EPEA). Under the provisions of the act, industry is
legally obligated to reclaim all disturbed land to an equivalent capability and return it to the Government of
Alberta.
Surface mining operations result in disturbances such as (but not limited to) forest clearing, peat (muskeg)
and soil displacement, modifications to wildlife habitat, soil and water contamination, changes to surface
and subsurface water quality and disturbances to geological (and hydrogeological) sequences. In general, a
reclaimed surface mining site includes the following landforms:
 In-pit structures: In the reclaimed landscape, the mine pits will be reclaimed to forested land or
function as end-of-pit lakes.
 Out-of-pit structures: These include the tailings ponds (settling basins), overburden dumps,
landfills and coke cells.

Modelling
The conceptualisation of reclaimed mine pits as a landscape consisting of treed uplands and organic wetlands
is based on the requirement to return the disturbed land to equivalent capability. Figure 2 shows a typical
upland, lowland and wetland conceptualisation under reclaimed conditions. The Fort McMurray area is a
part of the boreal forest setting. The natural landscape consists of mineral uplands and organic wetlands
(fens, bogs and organic lakes). Water is stored and slowly transmitted through the connected wetland
network, draining into streams and major rivers. Devito et al. (2012) presented a conceptualisation of water
movement in the Boreal Plains. In their conceptualisation, Devito et al. (2012) describe the Boreal Plains to
have evolved (within the hydrological context) under water deficit (sub-humid) climatic conditions.
The 1944–2011 annual precipitation and estimated PET data at Environment Canada’s Fort McMurray
weather station suggest that annual PET exceeds precipitation (Figure 3) in most years (sub-humid or water
deficit conditions). In the Boreal Plains, the interaction between the sub-humid climate and the geology
characterised by a large water-storage capacity leads to seasonal (annual) and inter-annual (decadal) soil,
landform and landscape-scale wetting and drying cycles. Actual evapotranspiration (AET) accounts for the
largest export of water from boreal watersheds, as shown by Devito et al. (2005).
This paper primarily focuses on mine pits reclaimed to a landscape consisting of forested upland and organic
wetlands. However, the discussion is also applicable to reclaimed tailings ponds and coke cells. Under the
current reclamation design conceptualisation, the different reclaimed landforms (reclaimed in-pits and out
of pit structures) will drain into end-of-pit lakes through a network of channels. The water reporting at the
outlets of each landform will be monitored for water quality and may be treated before it flows into an end-
of-pit lake. Owing to the low hydraulic conductivity of the McMurray Formation separating the different
landforms, groundwater flow from one landform to another is expected to account for a much smaller
component as compared to surface water discharge. However, surface water quality within each reclaimed
landform will be governed by groundwater discharging and mixing with overland flow.

The design of hummocks, lowlands and wetlands within each reclaimed mine pit may be constrained by some
or all of the following criteria:
 geotechnical stability requirements dictate the slopes and sizes of hummocks;
 dam safety requirements dictate the height of the hummocks;
 hydrotechnical requirements need to be considered to understand the stage of flooding under
extreme precipitation events. Hydrotechnical requirements combined with geotechnical and dam
safety requirements govern the placement, height and sizing of hummocks;
 material balance (e.g. availability of tailings sand, soil cover), combined with geotechnical, dam
safety and hydrotechnical criteria, governs the height and size of the hummocks;
 environmental criteria prescribe the requirements for successful growth of treed upland
vegetation and wetland vegetation.
It is clear from the design criteria requirements that a reclamation design will first have to satisfy the
geotechnical, dam safety, hydrotechnical and material balance criteria to create a stable reclaimed
landscape. In an iterative process, this design then needs to be evaluated for environmental performance
based on the environmental criteria developed through research over the years (e.g. Devito et al., 2012;
Dobchuk et al., 2012). Research on the challenges of tailings pond reclamation as well as research into the
hydrology of the natural boreal landscape has identified the following desirable hydrologic characteristics of
reclaimed tailings ponds:
1. A favourable water balance that maximises outflow to limit “evapoconcentration” of salts and
sustain end-of-pit lakes: Salts can be deposited in the uplands, lowlands and wetlands by upward-
moving water in the unsaturated zone, primarily during dry periods, and groundwater discharge.
In the absence of effective flushing, salts will concentrate because of continuous evaporation
(evapoconcentration). This is detrimental to vegetation growth and end-of-pit water quality
during wet years, when highly concentrated water flows to the end-of-pit lakes.
2. A shallow subsurface flow system to limit transport of salts from within tailings ponds: Deep flow
paths would result in deep flushing, bringing up more salts. This is not desirable from a water
quality perspective. Management of water quality is a major requirement for success of the
reclamation process, and limiting deep flushing through design is beneficial in the long term.

Modelling
3. Depth to groundwater table in the uplands that limits potential for upward migration of salinity
during dry climate cycles: Studies specify a separation of water table to 2 mbgs to avoid upward
migration of salts into the root zone during the dry climate cycles (Dobchuk et al., 2012). Salt
water moving into the root zone could be detrimental to treed vegetation in the uplands. During
wet cycles, the flow directions are mostly downward and the separation criterion is not as strict.
Although a complete and exhaustive review of all integrated modelling codes is out of scope of this paper, it
is important to mention model intercomparison studies that have reviewed most leading codes. Choice of
modelling code depends on the objectives of modelling. For oil sands reclamation design assessment and
reclamation support, the integrated modelling code should accurately represent surface water and
groundwater flow and solute transport processes in both saturated and unsaturated zones. Appropriate
representation of unsaturated zone is key because under water deficit climate conditions, surface and
subsurface coupling takes place through root water uptake and evapotranspiration. The model should also
calculate actual evapotranspiration accurately, given the largest portion of water export in the Boreal Plains
is through evapotranspiration, and represent the freezing and thawing process, given the frozen ground
conditions in winter.
HGS, ParFlow, MIKE SHE and GSFLOW are examples of leading integrated modelling codes. There are many
other modelling codes. All codes are peer reviewed — for example, Brunner and Simmons (2012) reviewed
HGS — and most are capable of some application in the oil sands region. Maxwell et al. (2014) reported a
model intercomparison study wherein they reviewed seven models. Most models reviewed reside in the
research domain. Another intercomparison study, now somewhat out-dated, was commissioned by the
Ontario Ministry of Natural Resources and conducted by AquaResources Inc. (AquaResources Inc., 2011). The
drawback of the study was the context in which winter processes were evaluated. The inclusion of snowmelt
processes as a criterion of evaluation does not help in evaluating the winter hydrological processes because
of the non-inclusion of pore water freezing. None of the integrated modelling codes can model winter ground
freezing processes at this time. One more crucial aspect of any coupling in numerical modelling is the non-
linearity of the processes. Such models are numerically expensive and take longer to solve. However, that is
a small price to pay for the insight integrated modelling offers into the complex process of a reclaimed mine
landscape.
A hypothetical mine was assumed to be reclaimed in the configuration shown in Figure 3. A three-
dimensional integrated surface water and groundwater flow and transport model was constructed using
HGS. The reclaimed mine was assumed to have two different watersheds (watersheds 1 and 2) based on the
reclaimed topography. The two watersheds were assumed to drain into two different end-of-pit lakes
through surface water channels (outlets 1 and 2). The mine pit was assumed to be incised into the McMurray
Formation, and therefore no-flow groundwater boundary conditions were assumed at the base and along all
sides of the pit. Thus, water could enter the reclaimed landscape as precipitation and leave through
evapotranspiration and surface outflow through outlets 1 and 2. Water was allowed to overflow out of the
model if the water head at the outlet exceeded the outlet elevation. Three subsurface materials were
assumed: cover soil, tailings sand and fine tailings. The conceptual layering is shown in Figure 4. The cover
soil was assumed to be 0.5 m thick all over the model domain. The tailings sand was assumed to be 20 m
thick, except in areas under the hummocks. Additional tailings sand was assumed to be placed under the
hummocks, resulting in a maximum increase of 5 m in tailings sand thickness under hummock crests. The
tailings sand was assumed to be underlain by a constant 40 m thickness of fine tailings. Triangular prisms
were used to construct the irregular surface (overland) domain grid. The surface domain was discretised
using 37,987 nodes and 75,209 triangular prisms of different sizes. The grid was refined in areas of steeply
sloping ground. The typical hummock widths were assumed to be between 150 m and 400 m, while the
lengths were assumed to be between 500 m and 1,250 m. The typical element side length within watershed
1 was 20 m or less, whereas in the rest of the model domain, it was 35 m or less. The subsurface domain was
spatially discretised in the same manner as the overland domain. Thus, each layer in the subsurface domain
consisted of 37,987 nodes and 75,209 triangular prisms of different sizes. The surface and subsurface
domains were separated by a coupling length of 0.001 m.
Table 1 summarises the model parameters used for simulations. The material hydraulic properties are based
on values reported in literature (Tuller and Or, 2004; Meier and Barbour, 2002; Shurniak and Barbour, 2002;
McKenna, 2002). Transport simulations assumed dispersivity values of 1 m (αL), 0.1 m (αT) and 0.01m (αv).
Horizontal
Anisotropy Residual
saturated Specific
ratio (Kh/Kv) Porosity water
Hydrostratigraphic Unit hydraulic storage
(m3/m3) content
conductivity (-) (1/m)
(m3/m3)
(m/s)
Cover soil 1E-04 10 0.05 0.66 0.16
Tailings sand 2E-06 16 1E-05 0.41 0.06
Fine tailings 1E-08 10 1E-05 0.38 0.15
Table 2 gives the vegetation-dependent model parameters. The model simulations were conducted for the
period spanning 1944 to 2011. The leaf area indices were based on Barr et al. (2004) and Bonneville et al.
(2007). The same leaf area index (LAI) sequencing was repeated annually (Figure 5). The properties given in
Table 2 and Figure 5 were applied to upland, lowland and wetland extents as shown in Figure 2. The
precipitation and PET time series used to drive the model are shown in Figure 2. Precipitation and PET time
series for 1944–2011 were applied with a daily input time step.
The initial conditions for flow simulation were derived from spin-up simulations using a constant
precipitation value of 433 mm (average annual precipitation for Fort McMurray weather station) and PET
value of 556 mm (average annual PET). The transport model simulations were carried out using relative
concentrations (C/C0; C0 = 1) assigned to the entire model domain. Solute was assumed to conservative,
meaning that the solute moves at the same velocity as water in both surface and subsurface domains (no
retardation).

Modelling
Interception Surface water flow

Category References
storage/transpiration (n: roughness coefficient)
cint: 0–0.0026
Uplands
RDF: linear 0.08–0.11 (full vegetation) Cint: Vrugt et
(Forests) al. (2003)
RD: 0.8–1.2 m
Lowlands cint: 0–0.0026
RD: Van Rees
(Wetlands or RDF: linear 0.03–0.08 (full vegetation)
(1997)
grasslands/shrublands) RD: 0.3–0.45 m
n: Sturm
cint: 0–0.0026 (2010);
Wetlands Arcement and
RDF: linear 0.01–0.02 (full vegetation)
Schneider
RD: <0.25 m (1984)
Model results were analysed for outflow and water balance; upland, wetland and lowland extents; seepage
face formation; and flushing efficiency and evapoconcentration under different climatic conditions. The
transient nature of simulation spanning 68 years (1944–2011) allows for such evaluation, as the
meteorological time series consists of wet cycles (precipitation > 433 mm) and dry cycles (precipitation < 433
mm).
Figure 6 shows the intra-annual water balance for dry and wet cycles. The dry and wet cycles between 2005
and 2011 and 1971 and 1977 have average annual precipitation of 350 mm and 520 mm, respectively. The
model results suggest that the maximum export of water will take place through evapotranspiration. The
total AET during different cycles does not vary greatly, primarily owing to availability of soil water. The
outflow generation will therefore be governed by precipitation regime. This means that there will be outflow
if precipitation exceeds AET, and there will be no outflow if AET exceeds precipitation. This has implications
for the flushing of salts. Maximum AET demand is during the months of May to September. The maximum
change in soil storage will occur during prolonged dry periods. This is primarily due to precipitation deficit,

especially in the summer months. The model results suggest that there will be relatively smaller changes to
soil storage during wet climate cycles owing to replenishment by precipitation.
Figure 2 shows the extent of upland, lowland and wetland based on topographic delineation. The model
results were analysed for depth to water table and near surface soil saturation to understand the extent of
upland, lowland and wetland in the reclaimed landscape under transient climatic scenarios. The groundwater
table under hummocks in most of the model domain satisfied the 2 mbgs separation criteria in the dry
climate cycles. In the wet cycles, the water table separation was less than 1 m in some areas. The model
results indicate that the extents will vary from season to season as well as during dry and wet climatic cycles.
This insight is crucial, as it shows that assumptions merely based on topographic delineation may not be
appropriate. Such delineation also helps in vegetation plantation planning. This would not be possible merely
through topographic delineation. Further, the model allows delineation of areas of potential geotechnical
concern identified as areas where the water table intersects the hummock surface. Such delineation allows
for measures to modify the design or sand type and placement along hummock slopes to mitigate seepage
face formation.

Modelling
The solute transport modelling provides insights into the extent of areas in which evapoconcentration may
occur. Figure 8 shows the extent of areas expected to experience evapoconcentration of salts in wet and dry
climate cycles. The areas delineated in Figure 8 are regions within watershed 1 where salt concentration
exceeds 3.0. It can be seen that the extent of such areas is much larger in dry climate cycles than in wet
cycles. Such delineation allows the right vegetation species to be chosen for areas where evapoconcentration
is predicted to occur. Further, design modification could be implemented to promote more effective flushing
and reduction in extent of evapoconcentrated areas in dry climatic cycles.
This paper discusses the application of integrated models to evaluate mine reclamation designs in the oil
sands region. A conceptual design was evaluated by constructing a three-dimensional integrated surface
water and groundwater flow and transport model using HGS. The modelling results provide important

insights. The model results indicate that for water deficit climate conditions, evapotranspiration is the
dominant process of water export. This insight would not be possible through a decoupled model. Decoupled
models are not able to calculate the recharge. Most decoupled models also require evapotranspiration to be
assigned as a boundary condition, leading to over- or underestimation of AET. Inappropriate calculation of
AET would lead to significant changes to water balance and misleading results with respect to delineation of
landform types or areas of geotechnical concern. Overall, the exercise demonstrates the usefulness of
integrated models in evaluating performance of a reclamation design. The same is applicable for closure and
reclamation plans wherein decoupled models are generally used to evaluate the water balance of reclaimed
landscape and water quality of end-of-pit lakes. This approach may be flawed, and gaps could be closed by
using an integrated, fully coupled modelling approach. From the model capability point of view, integrated
models need to simulate frozen ground processes to extend their applicability for oil sands reclamation
purpose. Currently, none of the integrated modelling codes incorporate pore water freezing processes.
Alberta Energy Regulator (AER) (2015a) Oil sands, viewed 25 February 2015, http://www.energy.alberta.ca/ourbusiness/
oilsands.asp.
Alberta Energy Regulator (AER) (2015b) Oil sands, viewed 03 March 2015, http://www.energy.alberta.ca/OilSands/791.asp.
AquaResources Inc. (2011) Integrated surface and groundwater model review and technical guide, Ontario Ministry of Natural
Resources, 116 pp.
Arcement, G.J., Jr. and Schneider, V.R. (1984) Guide for selecting Manning’s roughness coefficients for natural channels and flood
plains, Report No. FHWA-TS-84-204, Federal Highway Administration.
Barr, A.G., Black, T.A., Hogg, E.H., Kljun, N., Morgenstern, K. and Nesic, Z. (2004) Inter annual variability in leaf area index of a boreal
aspen-hazelnut forest in relation to net ecosystem production, Agricultural and Forest Meteorology, Vol. 126, pp. 237–255.
Bonneville, M.C., Strachan, I.C., Humphreys, E.R. and Roulet, N.T. (2007) Net ecosystem CO2 exchange in a temperate cattail marsh
in relation to biophysical properties, Agricultural and Forest Meteorology, Vol. 148, pp. 69–81.
Brunner, P. and Simmons, C.T. (2012) HydroGeoSphere: A fully integrated, physically based hydrological model, Ground Water, Vol.
50(2), pp. 170–176.
Devito, K.J., Creed, I.J. and Fraser, C.J.D. (2005) Controls on runoff from a partially harvested aspen-forested headwater catchment,
Boreal Plain, Canada, Hydrological Processes, Vol. 19, pp. 3–25.
Devito, K., Mendoza, C. and Qualizza, C. (2012) Conceptualizing water movement in the Boreal Plains, Implications for watershed
reconstruction, synthesis report prepared for the Canadian Oil Sands Network for Research and Development, Environmental
and Reclamation Research Group, 164 pp.
Dobchuk, B.S., Shurniak, R.E., Barbour, S., O’Kane, M.A. and Song, Q. (2012) Long-term monitoring and modelling of a reclaimed
watershed cover on oil sands tailings, International Journal of Mining, Reclamation and Environment, Vol. 1, pp. 1–22.
Maxwell, R.M., Putti, M., Meyerhoff, S., Delfs, J-O., Ferguson, I.M., Ivanov, V., Kim, J., Kolditz, O., Kollet, S.J., Kumar, M., Lopez, S.,
Niu, J., Paniconi, C., Park, Y-J., Phanikumar, M.S., Shen, C., Sudicky, E.A. and Sulis, M. (2014) Surface-subsurface model
intercomparison: A first set of benchmark results to diagnose integrated hydrology and feedbacks, Water Resources
Research, 50, 1531-1549. McKenna, G.T. (2002) Sustainable mine reclamation and landscape engineering, PhD Dissertation,
University of Alberta, Edmonton, Canada, 269 pp.
Meier, D.E. and Barbour, S.L. (2002) Monitoring of cover and watershed performance for soil covers placed over saline-sodic shale
overburden from oil sands mining, American Society of Mining and Reclamation, Lexington, USA.
Shurniak, R.E. and Barbour, S.L. (2002) Modeling of water movement within reclamation covers on oil sands mining overburden piles,
American Society of Mining and Reclamation, Lexington, USA.
Sturm, T.W. (2010) Open channel hydraulics, 2nd Edition, McGraw-Hill Series in Water Resources and Environmental Engineering,
New York, USA, 493 pp.
Tuller, M. and Or, D. (2004) Retention of water in soil and the soil water characteristic curve, in Encyclopedia of soils in the
environment, Vol. 4, Hillel, D. (Ed.), Elsevier Ltd., Oxford, UK, pp. 278–289.
Van Rees, K.C.J. (1997) Rooting patterns of Boreal tree species, Prince Albert Model Forest Association, Inc., Prince Albert, Canada.
Vrugt, J.A., Gupta, H.V., Bouten, W. and Sorooshian, S. (2003) A shuffled complex evolution metropolis algorithm for optimization
and uncertainty assessment of hydrologic model parameters, Water Resources Research, Vol. 39(8), pp. 1201–1214.

C.E. Lucas ARCADIS Canada Inc., Canada

B.E. Halbert ARCADIS Canada Inc., Canada
M.S. Webster Cameco Corporation, Canada
Although mining activities ceased in the early 1980s, dissolved radium-226, selenium and uranium levels
remain above background in the area surrounding the former Beaverlodge uranium mine and mill facilities in
northern Saskatchewan, Canada. In recent years, work has been undertaken to develop a final site clean-up
strategy that would allow for ultimate close-out of the area. Contaminant sources include residual mine
features such as small open pit mines, underground mine workings, waste rock piles, the former mill site and
tailings deposits. In order to aid the decision-making process, contaminant dispersion modelling has been
carried out in the post-decommissioning period for the immediate and downstream environment employing
a tool called the Beaverlodge quantitative site model (QSM), developed by ARCADIS Canada Inc. (formerly
SENES Consultants). Model calibration was carried out using 13 years of water quality and occasional
sediment data employing a Markov chain Monte Carlo computational procedure. In addition to modelling
contaminant dispersion, the QSM has a built-in pathways assessment module to calculate the resulting risks
to human and ecological receptors that may frequent the site.
The QSM was developed to assess a wide range of remedial activities and predict future environmental
conditions (surface water, sediment quality and potential risks) both with and without site remediation.
Potential remedial activities incorporated into the QSM include cover of waste rock piles, sediment
remediation (cover or dredge), stream diversion, elimination (infill) of waterbodies, sedimentation rate
manipulation, tailings area remediation and water treatment.
Since development of the QSM, the model has been used in numerous ways, including as a communication
tool during discussions with regulatory bodies (provincial and federal) and during stakeholder consultation,
when it has played an important role in the consensus-building process. In addition, the QSM provided insight
into the potential short- and long-term benefits of various remediation strategies that would otherwise been
difficult to ascertain. QSM results combined with data on capital and operating costs highlighted remedial
activities that were generally felt to be a good use of resources and identified others stakeholders felt were
unjustified. The final remediation plan for the site is based on this feedback.
The former Beaverlodge mining and milling properties are located north of Lake Athabasca in the northwest
corner of the province of Saskatchewan, Canada, approximately eight kilometres east of the northern
settlement of Uranium City (Figure 1). The Beaverlodge/Uranium City area is remote and accessed year-
round by aircraft and via winter road for a couple of weeks each winter. In June 2012, the population of
permanent residents in Uranium City was estimated to be 89, according to the Saskatchewan Ministry of
Health (2012).

Use of contaminant dispersion modelling to guide close-out of the former Beaverlodge Lake mine site C.E. Lucas, B.E. Halbert and M.S. Webster
Uranium-bearing minerals were first discovered in the Beaverlodge area of northern Saskatchewan in 1934
(MacLaren Plansearch Inc., 1983). Eldorado Mining and Refining Ltd. (Eldorado), a Crown corporation owned
by the Government of Canada, commenced detailed exploration in 1944, leading to the start-up of a mine
and mill in 1952. The primary focus of mining activity was north and east of Beaverlodge Lake, where three
mine shafts led to the development of a significant underground operation. Production from this mine and
numerous “satellite mines” continued until 1982. Mined ore was processed at a mill established on site in a
location near to the main underground workings. The majority of the generated tailings were deposited in a
nearby watershed.
By modern standards for Saskatchewan uranium deposits, the uranium content of the ore was relatively low.
The generally clean nature of the orebody in terms of secondary metal contaminants, as well as its carbonate
nature, made the waste rock relatively benign. During the initial period of operation, comprehensive
environmental protection regulations did not exist. It was not until the mid-1970s, more than 20 years after
operations began, that a federal Atomic Energy Control Board (AECB) licence was issued and effluent
treatment processes were initiated in response to discussions with provincial and federal regulatory
authorities (Webster and Hockley, 2013).
Operations ceased in 1982, at which time Eldorado Nuclear Ltd. initiated site decommissioning activities. The
site was decommissioned following a regulatory approved plan from 1983 to 1985. The decommissioning
met the regulatory requirements of the day, and post-decommissioning monitoring was initiated.
Decommissioning activities related to radiation protection included covering areas of exposed tailings in the
tailings management area with waste rock, sand and/or gravel, as well as removal of tailings and minewater

Modelling
treatment sludge from other areas for disposal underground. The decision was made to leave tailings in
forested areas in place to avoid damaging the vegetation.
The majority of the site remains in a transitional monitoring phase, which was initially expected to last for
about 10 years following completion of site remediation work in 1985. Despite meeting most of the predicted
recovery targets soon after decommissioning was completed, the transition phase is now in its 30th year.
Failure to bring an end to the transitional monitoring phase can be attributed to many factors, including
changing regulatory requirements, shifting attitudes of those involved and lack of clearly defined endpoints
for the ultimate close-out of the site.
In the meantime, levels of radium-226, selenium and uranium are elevated above background levels at a
number of locations across the Beaverlodge site and in the downstream environment. Current contaminant
sources include contaminated sediments, waste rock piles, tailings deposits/spills and seep from
underground mine workings. All of the underground mine workings as well as the majority of the waste rock
are located within the Ace Creek watershed, where the majority of the mining activities occurred. The bulk
of the tailings and contaminated sediments are located in the Fulton Creek watershed, which contained the
Beaverlodge tailings management area. The discharges from both watershed systems flow into Beaverlodge
Lake, which has a substantial source load contained in the sediments as a result of inputs to the lake over the
past 70 years. The locations of these watershed areas and key site features are indicated in Figure 2.
The current focus has been on achieving final close-out for the site, a task that involves determining which
remedial activities are deemed to be worthwhile and obtaining buy-in from the wide range of stakeholders

involved; the Beaverlodge QSM was developed to support this phase of work. The prime objective in
developing the QSM was to provide a tool to assist all interested parties in understanding the relationships
between contaminant loads from the licensed properties under current conditions, evaluating the potential
benefit of remedial actions related to recovery of the Beaverlodge area lakes and facilitating the process of
moving the properties to the Province of Saskatchewan’s institutional control (IC) program once it is
demonstrated that risks are managed and that conditions are stable.
The Beaverlodge QSM is a tool developed to support risk-based decision making regarding the benefit of
additional remediation of historical contaminant sources at the Beaverlodge study area properties. A
watershed dispersion model (LAKEVIEW) coupled with a pathways module to assess potential risks to human
and ecological health (INTAKE) serve as the framework for the QSM. ARCADIS Canada Inc. developed both
the LAKEVIEW and INTAKE models; LAKEVIEW predicts future water and sediment concentrations throughout
watershed systems, while INTAKE uses those predictions to assess contaminant movement along exposure
pathways and potential risks to human health and ecological receptors. The QSM is focused on areas within
the Beaverlodge study area and the immediate downstream area (Beaverlodge Lake), although future
conditions are predicted downstream to Lake Athabasca as well. In addition to predicting base case
conditions, the QSM was designed to assess the effectiveness of more than 65 different potential remedial
activities, either alone or in combination. The model was set up to run deterministically, decreasing the run
time to around 20 minutes. This means that results based on new scenarios or remedial options can be
generated quickly during meetings to immediately address stakeholder questions. While the Beaverlodge
QSM tool has been developed using a proprietary model to address the specific issues at the Beaverlodge
site, the purpose of this paper is to detail the overall process of how the tool was used to guide close-out of
the site, which can be undertaken with any suitable contaminant dispersion model adapted to allow
assessment of remedial activities.
Based on past sediment and surface water observations, selenium, uranium and radionuclides were
identified as parameters of particular interest in the Beaverlodge study area and were included in the
Beaverlodge QSM assessment. Total dissolved solids (TDS) was also included in the model as a general
indicator of water quality.
To address both current and future conditions, watershed dispersion modelling was carried out to predict
levels of radium-226, selenium, uranium and TDS in water and sediments. For this project, these constituents
were modelled using the LAKEVIEW code, which was developed specifically to assess mine waste effects on
the environment. The code may be run in either a deterministic or a probabilistic manner and has been
applied on numerous occasions in Canada, including in northern Saskatchewan, to assess the impact of
uranium and other mining operations.
LAKEVIEW is essentially a three-dimensional water quality transport and constituent speciation model for
lakes and rivers in a watershed. Important processes incorporated into the LAKEVIEW model include
horizontal (lateral) and vertical transport of dissolved species, chemical and biochemical reactions, settling
of particulate matter and sediment exchange. Transport between segments (water column–sediment, for
example) is computed by solving ordinary differential equations, while constituent speciation within a
segment is computed by thermodynamic stability criteria for equilibrium processes and kinetic
considerations (Stumm and Morgan, 1996; Scharer et al., 1994). Possible model output for lakes includes
surface water (multiple horizontal layers in water bodies that exhibit stratification) and sediment porewater
concentrations of dissolved constituents, constituents adsorbed to suspended matter and solid phase
concentrations in the sediment. Sediment samples collected across the site were analysed using the modified
Tessier’s extraction test (Mossop and Davidson, 2003), which involves a series of sequential extraction steps
to estimate the fraction of constituents in the sediment adsorbed onto calcite, quartz and organics (ion-
exchangeable), complexed with iron hydroxide (reducible), embedded in sulphide matrix (oxidisable) and

Modelling
present as solid solution, pure mineral and recalcitrant (un-extractable) fractions. Information collected
through the sequential extraction of the sediment samples was used within the model to specify the
availability of constituents in the sediments.
The Beaverlodge QSM was developed to encompass both main Beaverlodge site watersheds (Ace Creek and
Fulton Creek) and the downstream environment from Beaverlodge Lake to Lake Athabasca. In total, the
model predicts aquatic (sediment and surface water) conditions in more than 18 linked waterbody segments.
More than 20 studies were completed over the period between 2009 and 2011 to fill information gaps
required to develop the Beaverlodge QSM. Model calibration was carried out using 13 years (1998 to 2010)
of water quality and occasional sediment data employing a Markov chain Monte Carlo computational
procedure. The measured data were used to inform the selection of initial surface water and sediment
concentrations as well as the portion of sediment contaminants assumed to be exchangeable with the water
column. Additional parameter values estimated through calibration included external contaminant loads to
the system, the mass transfer coefficient between surface water and sediment porewater (Kl), the adsorption
coefficient in the sediment for a given constituent (KD_SED) and the adsorption coefficient of settling solids in
the water column for a given constituent (KD_WAT).
In calibrating the model parameters, the parameter values applicable to each lake were sampled one at a
time following the Markov chain Monte Carlo procedure using predicted and observed surface water
concentrations as criteria. The Markov chain Monte Carlo procedure was applied as follows. First, the
parameter being investigated was assigned a symmetric, triangular prior (proposal) distribution with
minimum, mean and maximum parameter values. The triangular parameter spaces were sampled by random
draw. The distribution of accepted parameter values was used to estimate the mean parameter value for the
study lake.
The external source loads were calibrated assuming an exponentially decreasing trend with both time-
dependant (weathering) and constant (continuous source and background) components.
Although the calibrated sediment-to-water mass transfer showed a slight dependence on the type of metal
being transported, consistent with experimental experience reported in the literature (Carbonaro et al.,
2005), the difference was not sufficient to warrant the use of an individual Kl value for each constituent, and
Kl values were averaged on a “per lake” basis. Similarly, the parameters KD_WAT and KD_SED were averaged on
a “per constituent” basis across all water bodies. The decision to average contaminant transport coefficients
was supported by further investigation, which involved applying the Akaike information criterion adapted for
small sample size (Hurvich and Tsai, 1989) to evaluate the trade-off between averaging parameters and the
change in residual error between measurements and observations (Hamer et al., 2012).
The sensitivity of the calibrated watershed dispersion model was examined with respect to changes in: (i)
model input parameter values, and (ii) source loads. Different approaches were used to evaluate the
sensitivity of the model response to changes in these factors.
For parametric sensitivity analysis, the sensitivity of model predictions to perturbations of parameters over
a 300-year period was investigated for two model parameters: settling velocity of suspended solids in the
water column and partition coefficient of a contaminant in the sediment (KD_SED). As shown by previous
analysis (SENES, 2003), there is considerable uncertainty in the values assigned to these parameters in the
long term. For maximum impact, the surface water concentration in the outflow of Beaverlodge Lake was
used as the criterion of evaluation. To simplify the interpretation of the model results, the values of all other
model parameters were held constant; however, the external source loads were allowed to vary with the
model year in accordance with the trends obtained through model calibration.

Uranium was selected as the constituent of interest for the sensitivity evaluation. Both investigated
parameters varied over what are believed to be reasonable ranges; the selected settling velocity values
capture typical values reported in the literature (Reckhow, 1994; Eadie et al., 1991), while the range of
sediment partition coefficients examined is based on results of the calibration procedure for this parameter.
The response in uranium concentration at the outlet of Beaverlodge Lake to changes in settling velocity is
shown in Figure 3(a), while the response to sediment partition coefficient is shown in Figure 3(b). The
response curves shown in Figure 3 were generated by varying the investigated parameter while holding all
other model inputs constant.
The overall effect of an increase in the settling velocity from 24 to 60 m/a (a 2.5 times increase) is a reduction
in the long-term uranium concentration at the outlet of Beaverlodge Lake from approximately 53 µg/L to 46
µg/L after 300 years (i.e. a decrease of only 12% in the long-term concentration). The reason that a greater
decrease is not realised may be mainly that removal from the water column on settling solids is largely offset
by diffusion from the sediment to the water column of dissolved uranium in the sediment porewater.
Unlike the effect of the settling velocity variation, the impact of KD_SED variability on surface water quality is
greatest within the first 50 years of the simulation and then diminishes with simulation time. This is because
of re-partitioning between the exchangeable solid phase and the porewater concentration as the result of
variation from the baseline KD_SED value. Since the flux to and from the sediment is proportional to the
difference between porewater and surface water concentrations, KD_SED values led the sediment to become
either a source or a sink for uranium in the surface water. The net effect in the long term on the uranium
level at the outlet of Beaverlodge Lake is seen to be minimal, largely owing to the buffering capacity (ability
to act as a source or a sink under different equilibrium conditions) of the sediments within Beaverlodge Lake.
A number of different sensitivity measures (algebraic, integral, gradient) have been employed (Isukapalli and
Georgopoulos, 2001) to quantify system sensitivity to changes in parameter or process variations. The
sensitivity value quantifies the relative change in a calculated variable evoked by a relative change in a system
parameter (Varma and Palsson, 1994). In the load sensitivity analysis carried out in the Beaverlodge QSM,
the objective was to rank sources affecting water quality at the outlets of the Ace Creek, Fulton Creek and
Beaverlodge Lake watersheds. Since the absolute value of the sensitivity response varies substantially with
respect to both the constituent of interest and time, the normalised gradient sensitivity, also known as the
logarithmic sensitivity, was used in the current evaluation. The assessment focused on uranium and radium-
226, as levels of selenium are not high enough above the method detection limits in the Ace Creek Watershed
to allow for meaningful evaluation.
To evaluate the normalised gradient sensitivity, the studied parameter was assigned a normal distribution
based on the Markov chain Monte Carlo calibration procedure; typical standard deviations of these

Modelling
distributions were approximately 20% of the nominal value. This distribution was then randomly sampled up
to 100 times to generate a set of new parameter values. The average normalised sensitivity over the
evaluation period was calculated for each new parameter value. Table 1 presents the sensitivities of the
examined constituents at the outlet of Beaverlodge Lake to changes in selected external and internal
(sediment) loads to the water column of Beaverlodge Lake; these values are an average of the calculated
averaged normalised sensitivities for all assessed values of the examined parameter. In order to evaluate the
sensitivity to external loads, the loads themselves were parameterised, while sensitivity to the upward flux
of constituents of concern from the sediments was assessed by parameterising the Kl values.
For the purposes of this sensitivity analysis, external loads were held constant over the evaluation period,
and constant (lake-dependent) mass transfer coefficients (Kl) were employed on all sources and lakes with
the exception of the source or lake segment under investigation. Each load (external or sediment) in each
lake was evaluated individually. The source load sensitivity is equivalent to the ratio of per cent change in
output concentrations (i.e. examined constituent level in the water column at a control location) to the per
cent change in the load under study. For example, a sensitivity value of 0.1 in Table 1 corresponds to a 1%
change in water column concentration in response to a 10% change in the studied load. For this work,
sensitivity gradient values above 0.10 are regarded as indicators of potentially significant contaminant
sources.
The results of the sensitivity analyses demonstrate that aquatic systems are complex networks, within which
any analysis is complicated by a multitude of interactions. As such, care must be taken in estimating the effect
of remedial work on downstream environments; the downstream response will not be 1:1, as effects of
remediation measures will be offset by downstream interactions. Overall, the analysis showed that water
quality at the outlet of Beaverlodge Lake is relatively unaffected by reduction in upstream sources, with a
maximum observed parametric sensitivity gradient of 0.46, which corresponds to a 4.6% change in water
quality in response to a 10% change in the uranium load from the sediments.
Parametric sensitivity gradient

Examined load
Uranium Radium-226
Total load from the Ace Creek watershed 0.29 0.27
Total load from the Fulton Creek watershed 0.11 0.42
Other external loads to Beaverlodge Lake 0.01 0.04
Sediment load from Beaverlodge Lake 0.46 0.37
The approach taken in the human health and ecological risk assessment involved estimation of exposures or
doses for each of the selected receptors and comparison of the exposure estimates to toxicity reference
values (TRVs). TRVs are used in risk assessments to judge whether the predicted (estimated) exposures (or
doses or intakes) may potentially have an adverse effect on ecological species or human health. The
comparison of predicted exposures/intakes to TRVs was undertaken by calculating the screening index (SI)
values for ecological species and hazard quotients (HQs) for human receptors; each is a ratio of the
exposure/intake estimates to the TRVs. SI and HQ values provide an integrated description of the potential
hazard, the exposure (or dose) response relationship and the exposure evaluation (U.S. EPA, 1992; AIHC,
1992). For radionuclides, the total dose rate received by a receptor was divided by the selected dose rate
guideline.
Ecological receptors selected as valued ecosystem components (VECs) included aquatic (aquatic plants,
phytoplankton, zooplankton, benthic invertebrates, forage fish and predatory fish) and terrestrial (muskrat
and waterfowl) species. Species with a solely terrestrial diet were not evaluated, as a previous risk

assessment (SENES, 2003) found that there were no potential adverse effects predicted for caribou, grouse
or hare. The selected ecological receptors were evaluated at a total of 16 exposure locations/waterbodies.
These are indicated in Figure 4.
Human receptors included both adults and children at eight exposure locations. The human exposure analysis
focused on two pathways only: the ingestion of drinking water and the ingestion of fish. The QSM analysis
focused on these pathways, as a country foods study undertaken in 2012 found that there were no risks to
people who reside in the area from consuming other local foods (SENES and CanNorth, 2013).
In setting up the human health and ecological risk evaluation, conservative assumptions were made
regarding receptor characteristics; for example, waterfowl are assumed to spend all their time on the site at
which they are evaluated, and humans are assumed to obtain all water and fish from their location while
they are present, despite regional water/fish consumption advisories.
As an environmental management tool, the QSM was developed with a user-friendly interface to allow easy
investigation during meetings with stakeholders, and to allow the client to understand the effects of various
remedial measures that could be undertaken on licensed properties as the sites are prepared for transfer to
the Province of Saskatchewan’s IC program. This remedial measure assessment feature of the model predicts
contaminant load reduction as well as future water quality throughout the Beaverlodge study area resulting
from implementation of selected remedial measures. The Beaverlodge QSM was designed to allow the user
to assess the effects of more than 65 different potential remedial activities, either as individual measures or
in any selected permutations. For each selected activity, the implementation year and activity effectiveness
can be selected. For sediment cover activities, cover material and material characteristics are specified.

Modelling
Remedial activities incorporated within the QSM for evaluation include:

 cover of waste rock piles;
 sediment remediation (cover or dredge);
 stream diversion;
 plug-flowing boreholes;
 elimination (infill) of waterbodies;
 sedimentation rate manipulation;
 tailings area remediation;
 water treatment.
The QSM output consists of predicted surface water quality results for uranium, selenium and radium-226
(with and without remediation) as well as the corresponding SI values and HQs for the evaluated receptors.
A benchmark of one was adopted for SI values and HQs. For general context, SI and HQ values below one are
inferred to imply negligible risks, and values above one infer there is an increased likelihood of potential risk
that warrants further investigation, such as the use of a weight-of-evidence approach to determine whether
an effect is occurring.
Overall, the QSM found that future radium-226, uranium and selenium levels are predicted to remain above
the provincial water quality objectives in some areas of the Beaverlodge site and the downstream
environment into the future. This finding results in predicted SI values above one for a number of aquatic
and terrestrial receptors. Predicted selenium and uranium levels also resulted in HQ values above one for
human receptors in the immediate downstream area (Beaverlodge Lake and one basin of Martin Lake);
however, there is a fish and water consumption advisory in place for Beaverlodge and Martin lakes (Public
Health Unit and Saskatchewan Ministry of Environment, 2009). Based on the fish consumption advisory, it is
not expected that locals will consume large quantities of water or fish from these waterbodies
The example results shown in Figure 5 evaluated the benefit of plugging boreholes at one of the satellite
mine locations in the Ace Creek watershed (the Dubyna area) in 2015 to stop contaminated mine water flow
into Dubyna Lake. Only uranium is shown in this case, as radium-226 and selenium levels are not elevated
above applicable guidelines in Dubyna Lake. As can be seen from Figure 5, this activity is predicted to
decrease uranium levels in Dubyna Lake, which will then ultimately expedite recovery of the waterbody.
Predictions show that this improvement in local surface water quality is not translated into a noticeable
change in downstream surface water quality either immediately downstream in Ace Lake or further
downstream, where the watershed discharges into Beaverlodge Lake (Ace Bay). Although concentrations
exceeded the SSWQO in Dubyna Lake, predicted SI values for the receptors considered for this area were
below the benchmark (SI of one) prior to remediation and, owing to the small nature of the waterbody, no
human receptors were evaluated on Dubyna Lake.

The example shown in Figure 6 assumed a large number of remedial activities would be implemented
concurrently starting in 2015. The selected remediation activities include diversion of two fresh water
sources around contaminated areas, plugging flowing boreholes in two areas, cover of the main waste rock
pile, elimination of seeps and applying sediment cover in the Fulton Creek watershed and in the bays of
Beaverlodge Lake. The graphs showing HQ for considered human receptors provide a breakdown of total risk
into contributions from ingestion of fish, ingestion of water and baseline intakes (soil, air and all other dietary
components). Although the selected remedial measures address every large upstream source, there is
predicted to be minimal influence on surface water quality in Beaverlodge Lake; this is largely owing to the
substantial inventory contained within the water column and sediments of Beaverlodge Lake. Radium-226 is
not shown, as it is currently below surface water quality objectives in Beaverlodge Lake. Uranium is not
predicted to present a risk to ecological receptors; however, risks above the benchmark are predicted for
humans consuming water from Beaverlodge Lake for three months of the year. Potential risks are predicted
to merganser and fish as well as children (through ingestion of fish) owing to elevated selenium levels in fish
flesh.
Overall, the Beaverlodge QSM results showed that, as presented in Figure 5, some remedial measures are
predicted to have a positive effect on the immediate environment; however, no remediation strategies are
predicted to result in substantive improvement in the impacted areas of the downstream environment (i.e.
in Beaverlodge Lake and downstream).

Modelling

The Beaverlodge QSM was developed to serve as a communication and stakeholder engagement tool to aid
the development of a path forward for releasing licensed Beaverlodge properties to the Province of
Saskatchewan’s IC Program. The remediation plan for the site was refined through two remedial options
workshops as well as other options evaluation activities.
The first workshop was held in 2009 with the objective of bringing stakeholders into the process of assessing
potential options for remediation of the Beaverlodge site. A total of 41 people participated in the workshop,
including community members, First Nations representatives, government representatives, federal and
provincial regulatory agencies and industry representatives. The workshop participants identified all
methods potentially applicable to individual elements of the sites, created a short list of the most applicable
methods, identified evaluation factors that should be used to assess individual activities, identified
uncertainties that prevent a clear selection of the most appropriate method and made a list of investigations
that would be required to reduce the identified uncertainties (Webster and Hockley, 2013).
Based on outcomes from the 2009 workshop, numerous environmental investigations were undertaken, and
the Beaverlodge QSM was developed to allow for a quantitative assessment of the investigated options. The
QSM was designed to allow evaluation of all remediation activities discussed during the 2009 meeting as well
as a number of additional strategies. In addition to the development of the QSM, a high-level cost estimate
was developed for each of the discussed activities identified during the 2009 workshop (SENES and SRK,
2012).
A second workshop was held over two days in 2012 to obtain informed, clear and documented feedback
about the predicted benefits and estimated costs of a range of remedial options from a cross-section of
stakeholders (Webster and Hockley, 2013). The 2012 workshop had 46 participants whose affiliations
mirrored those from the 2009 workshop. Participants included individuals representing Uranium City, the
Northern Saskatchewan Environmental Quality Committee, the Northern Mine Monitoring Secretariat and
various federal and provincial regulatory agencies, including the CNSC, Environment Canada, Natural
Resources Canada, Fisheries and Oceans Canada and the Saskatchewan Ministry of Environment.
Representatives from the Mamawetan Churchill River Regional Health Authority, the Saskatchewan Research
Council, Canada North Environmental Services, Canada Eldor Inc. and Cameco Corporation were also in
attendance. The heart of the workshop process was a series of steps that allowed the participants, working
as stakeholder groups, to weigh the benefits and costs of potential remediation measures. This was done by
providing the groups with information (costs, predicted surface water quality, estimated risks to assessed
receptors) about nine pre-defined remedial strategies chosen based on the 2009 workshop discussions. A
number of additional remedial measures were proposed by participants during day one of the 2012
workshop; predicted costs and expected change to environmental conditions based on these suggested
scenarios were determined that evening so that the additional scenarios could be assessed as a group during
day two of the workshop. Individual remedial activities ranged in cost from an estimated C$ 75,000 to more
than C$ 55 million, while remedial strategies (combination of activities) discussed at the workshop ranged
from no cost to more than C$ 100 million. In addition to the financial implications, the other costs and
benefits (e.g. potential for local employment, technical feasibility/uncertainty and long-term maintenance)
of each strategy were discussed and considered in each group’s assessment of the potential remedial option.
Strategies considered ranged from “do nothing” to combinations involving a wide range of activities, such as
water treatment, plugging flowing boreholes, cover of waste rock, sediment remediation and stream
diversion.
The conclusion of the path forward development process is that while there are no reasonably practical
measures that can be implemented to meaningfully reduce the overall recovery time of Beaverlodge Lake
and other downstream water bodies, some practical things that can be done to (1) assure the properties
have been decommissioned in accordance with good engineering practice and (2) incrementally improve
water quality in local water bodies. Remedial activities proposed for implementation included the creation
of a channel through a waste rock pile to allow the creek flowing through the waste rock pile to return to its

Modelling
normal flow path, thereby eliminating contact with the waste rock, and to plug all identified boreholes on
Beaverlodge properties to reduce groundwater outflow from former mine workings. The screening level costs
for these activities were estimated to be approximately C$ 2 million in total; the bulk of these costs are short-
term costs for performing the required remediation work with a small allotment for monitoring and
maintenance moving forward.
While monitoring water quality will continue at the well-established monitoring stations in the Beaverlodge
study area, a regional monitoring program will also be developed. The program would be initiated in the near
term, following formal acceptance by the involved regulatory bodies, and will continue after all licensed
Beaverlodge properties have been transferred to the IC program. The long-term regional monitoring program
will provide a tool for assessing the long-term recovery predictions made by the Beaverlodge QSM (SENES,
2012) for Beaverlodge Lake and the downstream environment.
A set of site-specific performance indicators was derived using the Beaverlodge QSM in order to assess the
expected performance of implementing the proposed remedial measures. The performance indicators will
be used in the short term (five to seven years) as a site-specific target for assessing the effectiveness of the
remedial activities implemented. If performance indicators are met, indicating the remedial action was
successful, application will be made to transfer the applicable properties to the IC program.
The effects of uncertainty about key parameters in the Beaverlodge QSM on water-quality predictions were
examined to establish bounds on the predictions, which were then used as the basis for determining
reasonable site-specific water-quality objectives.
Supported by the QSM, the path forward plan provides clear direction concerning further remedial activities
on the Beaverlodge properties to facilitate their transfer to the Province of Saskatchewan. A number of the
peripheral licensed sites will require little or no further remediation, and it is expected that they will be
released to the IC program within five years. It is anticipated those areas requiring additional remediation
can be released to the province following implementation of the proposed remedial actions followed by a
period of monitoring to confirm that short-term performance indicators have been met.
Uranium, selenium and radium-226 levels remain above background in the area surrounding the former
Beaverlodge mine and mill facilities in northern Saskatchewan. Decommissioning of the site was completed
according to an approved plan in 1985; however, the majority of the Beaverlodge properties remain in a
transitional period to this day. Recent work has focused on developing a strategy to allow for ultimate close-
out of the area and release of the properties to the Province of Saskatchewan. In order to aid the decision-
making process, contaminant dispersion modelling has been carried out in the post-decommissioning period
employing a tool called the Beaverlodge QSM.
The Beaverlodge QSM was developed to assess a wide range of remedial activities and to predict future
environmental conditions (surface water, sediment quality and potential risks) both with and without site
remediation. Potential remedial activities incorporated into the QSM include cover of waste rock piles,
sediment remediation (cover or dredge), plugging of flowing boreholes, stream diversion, elimination (infill)
of waterbodies, sedimentation rate manipulation, tailings area remediation and water treatment.
Informed by a stakeholder consultation workshop process and other engagement activities, a path forward
for management and remediation of the Beaverlodge properties was developed. This process used the QSM
and cost estimates to communicate potential benefits and costs of a wide range of remedial activities in order
to determine which activities the various stakeholder groups feel are justified.
The Beaverlodge QSM will remain useful in the future for assessing the effectiveness of implementing the
selected remedial activities as well as for ensuring the Beaverlodge and downstream environments continue
to recover as predicted.

American Industrial Health Council (AIHC) (1992) Improving risk characterization, Proceedings of a Symposium, 26–27 September
1991, AIHC, Washington, DC, USA.
Carbonaro, R.F., Mahony, J.D., Walter, A.D., Halper, E.B and di Toro, D.M. (2005) Experimental and modeling investigation of metal
release from metal-spiked sediments, Environmental Toxicology and Chemistry, Vol. 24, pp. 307–319.
Eadie, B., Bell, G. and Hawley, N. (1991) Sediment trap study in the Green Bay mass balance program: Mass and organic carbon fluxes,
resuspension, and particle settling velocities, National Oceanic and Atmospheric Administration (NOAA) technical
memorandum ERL GLERL-75. Available online at http://www.glerl.noaa.gov/ftp/publications/tech_reports/glerl-075/tm-
075.pdf.
Hamer, C.E., Halbert, B.E., Webster, M. and Scharer, J.M. (2012) Assessment of model adequacy and parameter identifiability for
predicting contaminant transport in the Beaverlodge Lake area, Canada, in Proceedings Eleventh International Conference
on Water Pollution Modelling, Monitoring and Management (Water Pollution XI) 10-12 July 2012, WITPress, Vol. 164,
pp. 159–170. Lyndhurst, England, July 2012.
Hurvich, C.M. and Tsai, C.L. (1989) Regression and time series model selection in small samples, Biometrica, Vol. 76, pp. 297–307.
Isukapalli, S.S. and Georgopoulos, P.G. (2001) Computational methods for sensitivity and uncertainty analysis for environmental and
biological models, U.S. Environmental Protection Agency, Washington, DC, USA. EPA/600/R01/068 (NTIS PB2004-102518).
Mossop, K.F. and Davidson, C.M. (2003) Comparison of original and modified BCR sequential extraction procedures for the
fractionation of copper, iron, lead, manganese and zinc in soils and sediments, Analytica Chimica Acta, Vol. 478, pp. 111–118.
Public Health Unit and Saskatchewan Ministry of Environment (2009) Fish advisory notice: Users of Beaverlodge and Martin Lakes,
updated 20 May 2009.
Reckhow, K.H. (1994) Water quality simulations modeling and uncertainty analysis for risk assessment and decision making, Ecological
Modeling, Vol. 72, pp. 1–20.
Saskatchewan Ministry of Health (2012) Covered population 2012, 80175 Uranium City, viewed 14 January 2015, http://population.
health.gov.sk.ca/80175.htm.
Scharer, J.M., Nicholson, R.V., Halbert, B.E. and Snodgrass, W.J. (1994) A computer program to assess acid generation in pyritic
tailings, ACS Symposium Series, Vol. 550, pp. 132–152.
Stumm, W. and Morgan, J.J. (1996). Aquatic chemistry: Chemical equilibria and rates in natural waters, 3rd edition, Wiley and Sons
Inc., Hoboken, NJ, USA.
United States Environmental Protection Agency (U.S. EPA) (1992) Framework for ecological risk analysis, EPA/630/R-92-001,
Washington, DC, USA.
Varma, A. and Palsson, B.O. (1994) Metabolic flux balancing: Basic concepts, scientific and practical use, Bio/Technology, Vol. 12,
pp. 994–998.
Webster, M. and Hockley, D. (2013) Stakeholder engagement and additional remediation of the decommissioned Beaverlodge
uranium mine site, in Proceedings 37th British Columbia Mine Reclamation Symposium, 16-19 September 2013, British
Columbia Technical and Research Committee on Reclamation, Vancouver, Canada.
Cameco Corporation (Cameco) (2012) Beaverlodge mine site path forward report, prepared with support from SENES Consultants,
December.
MacLaren Plansearch Inc. (1983) Operating history and environmental conditions — decommissioning of the Beaverlodge mine/mill
operations and reclamation of the site, created for Eldorado Resources Ltd., Toronto, Canada, Report #2 (unpublished), pp. 1-
1–3-22.
SENES Consultants (SENES) (2003) Beaverlodge mine site environmental effects reassessment, prepared for Cameco Corporation,
February.
SENES Consultants (SENES) (2012) Beaverlodge quantitative site model, prepared for Cameco Corporation, May.
SENES Consultants and Canada North Environmental Services (SENES and CanNorth) (2013) Uranium City country foods study year 2,
January.
SENES Consultants and SRK Consulting (SENES and SRK) (2012) Costing study — potential remedial options, former Beaverlodge Mine,
prepared for Cameco Corporation, March.

P.L. Defferrard MMG Century, Australia

T.K. Rohde EMGA, Australia
M.T. Lord MMG Century , Australia
In 2010 a store and release cover was constructed on the south waste rock dump (SWRD) at Century Mine.
The function of the cover is to reduce percolation into the underlying potentially acid-forming (PAF) waste
rock. The performance of the cover is currently being assessed using volumetric water content and matric
suction sensors. Monitoring has been ongoing since October 2012. This paper describes the design premise
for the store and release cover, monitoring instrumentation installation methodology and early measured
performance of the cover trials.
The south waste rock dump (SWRD) at Century Mine was commissioned in 1998 and received both
potentially acid-forming and non-acid-forming waste rock. The SWRD has produced near-neutral high salinity
seepage throughout its operational life. In 2010 the SWRD was decommissioned and a store and release
cover was constructed to limit future infiltration. Limiting infiltration will reduce long-term saline seepage
from the SWRD.
Century Mine is located 250 km north-north-west of Mount Isa, in Queensland, Australia. The topography of
the site comprises of undulating hills and ridgelines. The site vegetation is low and sparse, as is typical of the
semi-arid environment of northern Australia.
Rainfall data is available from the Bureau of Meteorology (BoM) weather station (Century Mine: station
029167). The average annual rainfall is 544 mm, with a dry season from April to September and a wet season
from October to March. January and February account for about 50% of total rainfall. In the last year Century
Mine has experienced lower than average annual rainfall of about 300 mm.
The SWRD was constructed from about July 1998, and was raised in three benches of 16 m, 16 m and 32 m
to a maximum height of 64 m. The average height of the SWRD is 55 m, and comprises about 50.5 Mm3 (65%)
of Type I waste rock (acid-neutralising dolomite, mainly in the base and sides), 0.9 Mm3 (1%) of Type II waste
rock (non-acid-forming [NAF] siltstone and sandstone), and 25.9 Mm3 (34%) of Type III (PAF shale, mainly in
the centre of each lift) (Scott and Defferrard, 2013).
The majority of the acid-neutralising dolomite, which was excavated at shallow depth in the open pit, was
located in the bottom bench of the SWRD, and the deeper PAF shale was placed above it. A 20 m wide NAF
dolomite perimeter was dumped around the sides of the PAF shale benches, effectively encapsulating the
PAF shale with neutralising dolomite.

Instrumentation and early monitoring results of the South Waste Rock Dump at Century mine P.L. Defferrard, T.K. Rohde and M.T. Lord
Near-neutral, saline seepage occurs from the toe of the SWRD, with a pH of approximately 6.5 and total
dissolved solids ranging from 7,000 mg/L during the wet season to 8,100 mg/L during the dry season. The
near neutral drainage is likely a result of the neutralising effect of the dolomite pad at the base of the SWRD.
A closure options study identified a store and release cover to prevent infiltration as the preferred closure
strategy for the SWRD.
A comprehensive review of factors that must be considered when considering a final cover for a waste rock
dump are presented in GARD (2009). In the long term the cover system must interact with climate, hydrology,
human activity, vegetation, animals, and settlement of the underlying waste (GARD, 2009). A chart of
appropriate cover types as a function of climate is reproduced in Figure 1 (from the GARD Guide, 2009).
Figure 1 shows that store and release covers will perform best in wet or dry climates where potential
evapotranspiration is at least two times greater than rainfall, such as applies at Century Mine (median annual
rainfall of about 500 mm, an average annual rainfall of 700 mm, and an average annual potential evaporation
of 2,800 mm, which is highlighted in Figure 1)
Cover systems for PAF waste rock dumps in Australia have traditionally been mounded to shed rainfall run-
off and hence minimise infiltration into the oxidising dump. A disadvantage of a rainfall-shedding cover is
that it may not have enough plant available water all year round to support sustainable vegetation growth,
and will likely desiccate during dry periods. This often results in excessive erosion of the cover during the
subsequent wet season. The store and release cover system developed for the SWRD recognised the need
to avoid rainfall run-off and desiccation, instead relying on storage within the cover of wet season rainfall

Covers
and its release during the long subsequent dry season through evapotranspiration. Australian experience
with store and release covers has typically restricted infiltration to less than 10% of annual rainfall. At Kidston
Mine (Williams et al., 2003) store and release covers restricted infiltration to less than 1%. At Mt Whaleback
the store and release cover restricted infiltration to between 5-13% (O’Kane and Waters, 2003). At Cadia
Mine (Rohde et al., 2011) the store and release cover restricted infiltration to less than 3%.
In 2010 a store and release cover was constructed on the SWRD (Williams, 2008). The purpose of the cover
is to reduce infiltration into PAF waste rock by maintaining a compacted reduced permeability layer (RPL) at
near-saturated conditions. This is achieved by overlaying the RPL with a 1.5 to 2 m thickness of soil/rock
mulch (Figure 2). The soil/rock mulch has a hummocky final surface that captures rainfall, allowing it in part
to infiltrate, with excess water removed by evaporation and transpiration. The store and release cover has
included crushed dolomite aggregate in the rocky soil mulch layer. The purpose of the crushed dolomite
aggregate is to produce acid-neutralising capacity (carbonate) as infiltrating rainfall passes through the pores
of the soil/rock mulch matrix. Infiltration into the underlying PAF waste rock may result in neutralisation of
acidity by interactions between the acid-neutralising pore water and potentially acid-producing waste rock.
The RPL is novel because it is predominantly the traffic-compacted surface of the SWRD (Figure 2). Some
sections of the RPL have been constructed from compacted clay when it was available during the construction
of a nearby sediment dam.
Store and release cover
Trafficked waste rock
Loose-dumped waste rock
During dumping operations, the main haul roads at Century Mine received about 500 loaded haul truck
movements per day. This, together with watering to suppress dust, is highly effective in breaking down the
surface waste rock and forming a well-compacted surface of low permeability. The SWRD was retrofitted
with a traffic compacted surface using the following method. The surface was first prepared by removing
large boulders by ’dozer followed by grading. The surface was graded towards the open-cut pit so that once
the store and release cover was installed and operating any lateral water movement would flow towards the
open-cut pit and not the receiving environment. Depressions on the graded surface of the SWRD were
backfilled to minimise subsequent ponding of any infiltration through the cover. Where these depressions

were deeper than 500 mm the fill was compacted in 500 mm layers using loaded haul trucks to minimise any
subsequent differential settlement that would reinstate the depressions. The graded and backfilled surface
was compacted with a minimum of 100 passes using a loaded Komatsu 830E haul truck, together with
watering for dust suppression. A rocky soil/rock mulch layer was then placed 1.5 m to 2 m thick by paddock
dumping, and finally aerial seeding was undertaken using locally sourced native plant seed to establish the
“release” component of the store and release cover.
The SWRD store and release cover is monitored at five locations (Figure 3).
Sensor tree installation was completed in three parts: wiring and transport of trees to site, backfilling of the
RPL and backfilling of the soil/rock mulch. The installation process is illustrated in Figure 4.
Sensor trees were fitted with sensors (paired matric suction and volumetric water content sensors) using
scaffolding prior to being transported to the SWRD using a Franna (Figures 4a, Figure 4b and Figure 4c).
The backfilling of the RPL (Figure 4d) reproduced the approximate density (by compactor plate) and
horizontal bedding pattern that was present at each location prior to excavation. Backfilling to reproduce the
correct horizontal beds and depths was assisted by stockpiling of materials during the excavation of the pits
and backfilling in the correct order of removal.
Backfilling of the RPL (Figure 4d) at each location applied different backfill techniques (number of lifts and
compaction effort) based on visual observation of the RPL profile.
Backfilling of the soil/rock mulch was completed by excavator (Figure 4e). Once the desired backfill height
was achieved the sensors were buried by hand-shovel, ensuring that a thin layer of fines was placed around
the sensor. Fines within the cover will migrate over time. Embedding the sensors in fines is a trade-off
between initial sensor to soil/rock mulch contact for good sensor performance and measurements that are
representative of the soil/rock mulch. The nature of the soil water characteristic curve (sensor

Covers
measurements) changes over time as fines migrate. This has been observed in the monitoring dataset, but is
not discussed in this paper. Backfilling continued until all the excavated fill was placed back into the
excavation (Figure 4f).
(a) (b)
(c) (d)
(e) (f)

The sensors were buried so that there is:

 six matric potential and six volumetric water content sensors in the soil/rock mulch layer of the
cover;
 one matric potential and one volumetric water content sensor in the RPL; and
 one matric potential and one volumetric water content sensor in the traffic compacted layer
below the RPL in waste rock.
The SWRD store and release cover had five test pits dug to analyse the cover profile (Rohde, 2012). The test
pit locations became the monitoring locations.
Rohde (2012) found during test-pitting and instrument installation that the enhanced store and release cover
layer showed variation in cover thickness and material types. Figure 5 presents representative cross-sections
for each of five test pit locations. The locations of the test pits on the SWRD are shown in Figure 3. In general:
 the soil/rock mulch layer is 2 to 2.25 m thick;
 the RPL ranges from 0.1 m thick to 0.25 m thick;
 the material used to construct the RPL is highly variable, ranging from dolomite at location one,
two and four; PAF at location three and clay at location five; and
 the traffic-compacted layer beneath the RPL is highly variable, ranging from dolomite at locations
two and four to PAF at all other locations.
Figure 6 shows the performance of the store and release cover as volumetric water content (VWC) over the
first 24 months of monitoring for location 2 (shown in Figure 3). Figure 6 shows where the degree of
saturation (S) is equal to one (fully saturated), and the corresponding gravimetric water content (w) for each
layer of the store and release cover. Figure 6 also shows rainfall versus time and includes highlights to indicate
six rainfall periods that occurred during the first wet season of monitoring.
Table 1 summarises the response of the store and release cover to the six rainfall periods. The periods
between rainfalls are drying periods. The general response of the store and release cover during these
periods is a drying (reducing VWC) trend. The drying trend may be a result of evaporation, transpiration or
infiltration of pore water into the waste rock beneath the store and release cover.

Covers
(a) (b) (c)
Dolomite
soil/rock mulch
Dolomite
soil/rock mulch Dolomite
RPL (Dolomite) soil/rock mulch
Trafficked
Dolomite
RPL (Dolomite)
Trafficked PAF RPL (PAF)
Trafficked PAF
(d) (e)
Dolomite
soil/rock mulch
Dolomite
soil/rock mulch
RPL (Dolomite)
RPL (Clay)
Trafficked
Dolomite
Trafficked PAF

45 
40 Rainfall
 
35
RAINFALL (mm)
30
25
20
15  
10 
5
0
0.6
VOLUMETRIC WATER CONTENT
0.5
DATE S~1, w~10%

0.4
S~1, w~17%
0.3
S~1, w~19%
0.2
0.1
DATE
near surface of storage layer (0.33 m bgl) middle of storage layer (1.23 m bgl)
base of storage layer (2.23 m bgl) RPL (2.63 m bgl)
below RPL (3.03 m bgl)
Period Description
1 Total rainfall for period ~ 13.6 mm.
Initially rainfall is removed from the cover as evaporation and transpiration. The S and VWC
remain unchanged.
At the end of the period there is 11.8 mm of rainfall. The rainfall wets the near-surface layer
of soil/rock mulch (0.33 m bgl) up to a S ~0.66 corresponding to a VWC of ~0.21.
The middle of the soil/rock mulch (1.23 m bgl), the base of the soil/rock mulch (2.23 m bgl),
and the RPL (2.63 m bgl) are neither wetting nor drying during this period.
2 Total rainfall for period ~ 17 mm.
This period results in re-wetting of the near-surface layer (0.33 m bgl) of the soil/rock mulch
to a S ~0.67 corresponding to a VWC ~0.22.
The middle of the soil/rock mulch begins to wet-up.

Covers
Period Description
The base of the soil/rock mulch (2.23 m bgl) is neither wetting nor drying.
The RPL (2.63 m bgl) shows some redistribution of pore water, resulting in minor wetting of
this layer. Because the base of the soil/rock mulch (2.23 m bgl) has not wet-up during this
period it is unlikely that the pore water redistribution is a result of rainfall infiltration.
3 Total rainfall for period ~ 104 mm (period of significant rainfall).
There are two main rainfall events during this period.
31.8 mm of rainfall fell on 22 December 2012 and continued to wet the soil/rock mulch to a
depth of 1.23 m bgl (middle of soil/rock mulch).
25.2 mm of rainfall fell on 29 December 2012. At this point the near-surface (0.33 m bgl) of
the soil/rock mulch was at a S ~0.66 corresponding to a VWC ~0.23. The middle of the
soil/rock mulch (1.23 m bgl) was at a S ~0.62 corresponding to a VWC of ~0.23.
Immediately after the initially (31.8 mm) rainfall event the VWC and S near-surface in the
soil/rock mulch and the middle of the soil/rock mulch (0.33 m bgl and 1.23 m bgl) remain
unchanged.
Immediately after the second rainfall event (25.2 mm) the base of the soil/rock mulch (2.23 m
bgl), the RPL (2.63 m bgl) and waste rock below the RPL (2.96 m bgl) all become near-
saturated (~1) corresponding to a maximum VWC ~0.41.
4 Total rainfall for period ~ 44.8 mm.
Rainfall during this period is stored in the near-surface layer of the soil/rock mulch (0.33 m
bgl) and removed by evaporation and transpiration.
The middle of the soil/rock mulch (1.23 m bgl), the base of the soil/rock mulch (2.23 m bgl)
and the RPL (2.63 m bgl) all continue to dry, probably as a result of percolation into the
underlying waste rock.
The underlying waste rock (3.03 m bgl) remains saturated.
5 Total rainfall for period ~ 115 mm.
This period is characterised by single-rain events intermittently spaced by short dry periods.
Rainfall is captured and removed from the near-surface of the soil mulch (0.33 m bgl) almost
immediately, with daily data showing a drying trend.
The middle of the soil/rock mulch (1.23 m bgl), the base of the soil/rock mulch (2.23 m bgl)
and the RPL (2.63 m bgl) all continue to dry, probably because of percolation into the
underlying waste rock.
On 8 March 2013 a single rain event of 41.6 mm occurred. This creates piston flow with all
layers of the store and release cover showing an immediate response to the rain event
resulting in a S ~0.68 to ~0.97 corresponding to a VWC range of ~0.22 to ~0.4
6 Total rainfall for period ~ 6 mm (minor rainfall).
Rainfall is captured and removed from near-surface (0.33 m bgl) of the soil/rock mulch almost
immediately, with daily data showing a drying trend.
All layers of the enhanced store and release cover continue to dry-out due to evaporation
and/or transpiration (near-surface) or percolation into the underlying waste rock.
Figure 7 shows the average volumetric water content profile with depth plotted for each month of
monitoring of the store and release cover. Figure 7 indicates that the effective depth of influence of
evapotranspiration in the store and release cover is approximately 1.25 m bgl. The volumetric water content
below this depth becomes wetter as it approaches the RPL.

It should be noted that the store and release cover is currently sparsely vegetated. It is expected that the
influence of evapotranspiration will extend the full thickness of the storage layer of the store and release
cover once vegetation becomes established.
VOLUMETRIC WATER CONTENT

0 0.1 0.2 0.3 0.4 0.5 0.6
0
Nov 12
Dec 12
0.5 Jan 13
Feb 13
Mar 13
1 Infiltration Apr 13
storage layer May 13
DEPTH (m bgl)
1.5 Jun 13
Increasing VWC
with depth Jul 13
Aug 13
2
Sept 13
Oct 13
2.5 RPL
3
Waste rock
3.5
Figure 8 shows estimated infiltration or seepage (as a percentage of cumulative rainfall) for location 2 (shown
in Figure 3). It should be noted that the result shown is for a sparsely vegetated store and release cover. Net
infiltration through the store and release cover system has trended towards 8% of incident rainfall over the
12 months of monitoring. The calculated seepage has been calculated using the volumetric water content
sensor located in the waste rock below the RPL and assumes all infiltration that reaches the RPL can no longer
be removed from the cover by evapotranspiration. This assumption is supported by Figure 7, and is described
in the previous section.

Covers
45 
40 Rainfall
 
35
RAINFALL (mm)
30
25
20
15  
10 
5
0
25
SEEPAGE (% cumulative rainfall)
20
DATE
15
10
DATE
The appropriate choice and sustainability of vegetative cover on a store and release cover is essential to its
effectiveness in limiting rainfall infiltration. The store and release cover systems reported within this study
were seeded, but vegetation has been slow to establish, in part due to drought conditions experienced at
Century Mine over the past 18 to 24 months. Species selection and seeding density are also being refined to
achieve greater density and diversity of species. The long-term performance of the store and release cover
on the SWRD will improve as vegetation becomes established, increasing the evapotranspiration rate and
depth of influence within the storage layer of the store and release cover.
Century Mine has demonstrated through instrumentation that store and release cover systems have the
potential to address the possible future release of acid rock drainage to the receiving environment. It is
anticipated that the performance of the store and release cover will improve as vegetation becomes
established. After 12 months the store and release cover net percolation had risen, but remained below 10%
of cumulative rainfall. The store and release cover has wet-up and dried-out in response to rainfall and
subsequent dry periods.
The soil/rock mulch is a suitable substitute for natural materials.

It is clear that the early performance of store and release cover is very much dependent on how quickly
vegetation can establish. The vegetation establishment is slow and it will take several years for the likely
store and release cover long-term behaviour to become apparent. There is a need, therefore, to maintain
monitoring for the order of 10 years to assess the cover system at equilibrium under a range of climatic
conditions.
Gard Guide (2009) Global acid rock drainage guide, viewed www.gardguide.com.
O’Kane, M. and Waters, P. (2003) Dry cover trials at Mt Whaleback – a summary of overburden area cover system performance, in
Proceedings Sixth International Conference on Acid Rock Drainage, 14-17 July 2003, pp. 419–426, Australian Centre for Mining
Environmental Research, Brisbane.
Rohde, T., Williams, D.J. and Burton, J. (2011) Store and release cover performance at Cadia Hill Gold Nine, Australia, in Proceedings
Sixth International Conference on Mine Closure, 18-21 September 2011, A. Fourie and M. Tibbett (eds), pp. 333-343, Perth:
Australian Centre for Geomechanics.
Rohde, T. (2012) Century Mine – as built cover instrumentation, SKM consulting report for MMG Century Mine.
Scott, P. and Defferrard, P. (2013) Quantifying reactive waste rock distribution in a constructed rock storage facility for effective
closure, in Proceedings 2013 Workshop on Australian Mine Rehabilitation, Adelaide, South Australia, 7-8 August 2013, l. Kerr
(ed), pp. 9-13, Brisbane: JKTech Pty Ltd.
Williams, D.J. (2008) As-built report – south dump stage 1 cover, Uniquest consulting report for MMG Century Mine.
Williams, D., Currey, N., Ritchie, P. and Wilson, G. (2003) Kidston waste rock dump design and store and release cover performance
seven years on, in Proceedings Sixth International Conference on Acid Rock Drainage, 14-17 July 2003, Australian Centre for
Mining Environmental Research, Brisbane.

L.A. Barber O’Kane Consultants Inc., Canada

B.K. Ayres O’Kane Consultants Inc., Canada
B.J. Schmid Safety, AREVA Resources Canada Inc., Canada
Cluff Lake uranium mine, owned and operated by AREVA Resources Canada Inc., is located in northern
Saskatchewan’s Athabasca Basin. Cluff Lake mine operated from 1980 to 2002, and decommissioning work
began in 2004 following an environmental assessment. Decommissioning of Cluff Lake mine included
construction of reclamation cover systems over a tailings storage facility and waste rock pile. The objective of
this paper is to review the design, construction, and performance of the cover systems based on field
monitoring data. The primary design objectives of the cover systems are to reduce percolation of meteoric
water into the waste, attenuate radiation emanating from stored waste to acceptable levels, and provide a
growth medium for development of a sustainable vegetation cover. The waste rock cover system is an
enhanced water store-and-release design, while the tailings cover system is a water store-and-release design.
Both cover systems, which were completed in 2006 and seeded with agronomic and native plant species,
incorporate positive drainage to promote runoff during wet precipitation conditions. Instrumentation was
installed during construction of the cover systems to facilitate evaluation of their hydrologic performance over
time under site-specific climate conditions. Field data have been collected and analysed since 2006, for a total
of eight full years of monitoring. The cover systems are performing as expected and are on a trajectory to
design net percolation rates and sustainable vegetation cover. Soil-plant-atmosphere numerical modelling is
planned to facilitate further understanding of the current and long-term performance of the cover systems at
Cluff Lake.
Cluff Lake uranium mine, owned and operated by AREVA Resources Canada Inc. (AREVA), is located in
northern Saskatchewan’s Athabasca basin, approximately 75 km south of Lake Athabasca and 15 km east of
the provincial border with Alberta. The mine operated from 1980 to 2002, and decommissioning work began
in 2004 following an Environmental Assessment. The majority of decommissioning work was complete by
the end of 2006. The project is now in the post-decommissioning and follow-up monitoring stage. AREVA has
demolished the last buildings on site including the camp, airstrip facilities, and warehouse. AREVA is
continuing its environmental monitoring program through four site visits per year.
Decommissioning of Cluff Lake mine included placement of a cover system over a waste rock pile (WRP)
known as Claude WRP, and a tailings management area (TMA). The cover systems were completed in 2006,
and monitoring of performance has been on-going since then. This paper reviews the design and construction
of the cover systems for reclamation of the Claude WRP and TMA, and focuses on hydrologic performance
of the cover systems based on eight full years of field monitoring data.
The Cluff Lake mine site is situated in a region with a mean annual precipitation and potential evaporation of
approximately 450 mm and 600 mm, respectively. Approximately 30% of annual precipitation occurs as snow.
Numerous lakes, swamps and rivers dominate the relatively flat topography of the region.

Performance evaluation of reclamation soil cover systems at Cluff Lake mine in northern Saskatchewan L.A. Barber, B.K. Ayres and B.J. Schmid
The Claude WRP was constructed between 1982 and 1989 and contains waste rock from the Claude pit. The
WRP is approximately 30 m high and covers an area of 26.4 hectares to the south of the Claude pit (Figure 1).
It contains approximately 7.23 million tonnes of waste, with an estimated volume of approximately 4.1 Mm3
based on a dry density of 1,750 kg/m3. The Claude WRP was developed by end-dumping and contains well-
developed traffic surfaces between lifts of dumped material. No attempt was made to segregate waste
placed in the Claude WRP by chemical composition. The Claude WRP has shown high levels of uranium
(200 mg/L) and nickel (43 mg/L) in piezometers around the toe of the pile (COGEMA, 2001). The high levels
of uranium and nickel indicate that acid mine drainage is occurring and will continue to occur until the source
is depleted. The occurrence of acid mine drainage was confirmed and quantified through a detailed waste
rock characterisation program completed in 1999. AREVA determined that the Claude WRP would be
decommissioned in-place, meaning that an engineered cover system would be required for closure.
The TMA contains tailings leftover from milling operations throughout the life of the mine (COGEMA, 2001).
The TMA contains approximately 2.67 Mm3 of tailings, and was comprised of four major components: a solids
containment area, water decantation area, water treatment facilities and settling ponds, and diversion
ditches. The containment and decantation areas were located in a topographic low, and tailings solids and
liquids were retained behind a main dam. Thickened tailings were pumped to solids ponds where
consolidation and liquid decantation occurred. Liquids then collected in the decant area where further solids
were settled out. Liquids from the decant area, as well as tailings thickeners and pumped water from the pits
were then fed through treatment facilities. Settling ponds allowed further settling of precipitates, after which
treated water was fed to another treatment facility. Treated water was then pumped to a final settling pond
prior to discharge into the environment. Monitoring wells adjacent to the main dam have shown that mean
concentrations of U, Cl, Ra-226, and SO4 have stabilised (CNSC, 2003). AREVA determined that the TMA would
be decommissioned in-place, meaning that an engineered cover system would be required for closure.
Cover system field trials, known as test plots, were constructed and instrumented in 2001 on the Claude WRP
to examine construction feasibility and hydrologic behaviour of the preferred cover system design
alternative. One test plot (TP#1) was constructed on a relatively horizontal surface, while a second (TP#2)
was constructed on a 4H:1V sloped surface. Both test covers had the same profile design, consisting of a
nominally 10 cm thick reduced permeability layer (RPL) overlain by a nominally 100 cm thick layer of local
silty-sand till. The RPL comprised weathered waste rock material compacted in situ. Field compaction trials

Covers
were completed in advance to determine the preferred techniques for RPL construction. Field data collected
over a 5-year period included in situ volumetric water content (automated and manual measurements),
matric suction (negative pore-water pressure), and temperature of the cover system and waste materials.
These data were interpreted and used to develop estimates of net percolation through the cover systems
into the underlying waste. The collected field data were used to aid in calibration of a soil-plant-atmosphere
(SPA) numerical model as part of the cover system design process for full-scale WRP decommissioning.
Based on the success of the cover system field trials, an enhanced store-and-release cover system was
selected as the preferred design for closure of the Claude WRP. The final design included a 20 cm thick layer
of compacted waste rock overlain by 100 cm of non-compacted silty-sand till with a grass and legume
vegetation cover. The primary design objectives of the cover system were to:
1. Reduce percolation of meteoric waters to attenuate peak concentrations of contaminants of
concern in natural watercourses to levels that can be assimilated without adverse effects to the
aquatic ecosystem.
2. Attenuate radiation emanating from stored waste to acceptable levels.
3. Provide a growth medium for development of a sustainable vegetation cover.
Decommissioning of the Claude WRP was completed between 2005 and 2006 and involved the following
primary work activities:
 Re-contouring the side-slopes to a maximum slope angle of 4H:1V.
 Compacting the WRP surface to meet density specifications over a minimum depth of 0.2 m.
 Placing nominally 1 m of local silty-sand till material over the compacted waste rock surface.
 Constructing surface water drainage channels to handle the 24-hour, 100-year design storm
event.
 Applying re-vegetation seed and fertilizer mixture with a drill seeder.
Compaction of the waste rock surface was accomplished using a Caterpillar CS583 roller (AREVA, 2007).
Generally, two passes were required to meet the required minimum dry density of 95% of Standard Proctor
Maximum Dry Density. Due to the unseasonably wet weather encountered during the waste rock compaction
effort, water conditioning was not required to achieve the specified density. In general, the majority of the
re-graded waste rock surface contained sufficient finer-textured materials to produce a relatively smooth
surface (Figure 2). Areas of the WRP that were visually determined to have too much void space were re-
graded to blend in additional fines with either waste rock or till and re-compacted. The estimated field
saturated hydraulic conductivity of the compacted waste rock layer was 10-5 to 10-6 cm/s.

Instrumentation was installed in August 2006 to enable monitoring the hydrologic performance of the Claude
WRP cover system over time under site-specific climate conditions. Field data being collected on the cover
system include precipitation, net radiation, runoff, and volumetric water content, matric suction, and
temperature of the cover system and upper waste rock materials (Figure 3). Monitoring stations were located
at various slope positions and aspects due to potential differences in cover system performance at these
different locations. Also, the monitoring system was automated to the extent possible to avoid missing
collection of field response data during key times of the year, such as during spring snowmelt and storm
events.
A cover system test plot was constructed and instrumented in the fall of 1999 on the TMA to examine
construction feasibility and hydrologic behaviour of the preferred cover system design alternative. One test
plot consisting of a 1.5 m thick layer of sandy-till was constructed over a relatively horizontal surface in a
coarse tailings area. Several instruments were installed at the time of construction to collect data at the soil-
atmosphere interface and within the unsaturated and saturated zones at the test cover system site. The
collected field data were used to aid in calibration of a soil-plant-atmosphere (SPA) numerical model as part
of the cover system design process for full-scale TMA decommissioning.
A store-and-release cover system was selected as the preferred design for closure of the TMA, based on the
success of the cover system field trial. The final design of the TMA soil cover system consisted of a nominally
100 cm thick layer of non-compacted, silty-sand till with a grass and legume vegetation cover (Figure 4). The
primary design objectives of the cover system are the same as the design objectives for the Claude WRP
cover system.

Covers
Decommissioning of the TMA was completed in 2006 and involved the following primary work activities:
 Re-contouring of the tailings surface to provide positive drainage and promote runoff.
 Placing nominally 1 m of local silty-sand till material over the tailings surface.
 Infilling the liquids pond with local till material.
 Constructing surface water drainage channels to handle the 24-hour, 100-year design storm
event.
 Buttressing the main dam structure to ensure long-term stability.
 Applying revegetation seed and fertilizer mixture with a drill seeder.

Gamma radiation measurements were completed following WRP and TMA cover system construction to
confirm gamma radiation emanating from the stored waste materials were below maximum dose criteria.
Instrumentation was installed in August 2006 to enable monitoring the hydrologic performance of the TMA
cover system over time under site-specific climate conditions (Figure 5). Field data being collected on the
cover system include precipitation, air temperature, relative humidity, wind speed and direction, net
radiation, as well as volumetric water content, matric suction, and temperature of the cover system and
upper tailings. In addition, groundwater level data at two wells (CN1000G and CS1100G) are collected.
Performance of cover systems is reflected primarily in estimated net percolation rates and in qualitative
assessment of vegetation growth. Water balances are a tool to determine net percolation rates that use field
measured, calculated, or residual data as inputs to solve the water balance equation (1) on a daily basis
during frost-free periods. In this way, the water dynamics of the cover system can be characterised.
PPT = R + AET + NP + ΔS + LD (1)
Where:
PPT = precipitation (rainfall plus snow water equivalent (SWE)),
R = runoff and sublimation,
AET = actual evapotranspiration,
NP = net percolation,
ΔS = change in water storage, and
LD = lateral drainage.
Daily water balances for monitoring locations on the cover systems were estimated during frost-free periods
(Tables 1 through 4). The period of April 1st to October 31st was used as the frost-free period for the purposes
of this analysis.
Year PPT (mm) Water balance fluxes (mm and % of PPT)

AET ΔS R NP
2007 450 231 (51%) 34 (8%) 6 (1%) 179 (40%)
2008 272 297 (109%) -96 (-35%) 6 (2%) 66 (24%)
2009 387 290 (75%) 31 (8%) 5 (1%) 61 (16%)
2010 358 303 (85%) 12 (3%) 2 (1%) 40 (11%)
2011 271 182 (67%) 9 (3%) 2 (1%) 58 (21%)
2012 430 317 (74%) 33 (8%) 5 (1%) 105 (24%)
2013 408 255 (63%) -4 (-1%) 40 (10% 67 (16%)

Covers

AET ΔS R LD NP
2007 419 239 (57%) 17 (4%) 58 (14%) 0 (0%) 104 (25%)
2008 261 308 (118%) -85 (-33%) 50 (19%) -57 (-22%) 45 (17%)
2009 396 314 (79%) 15 (4%) 41 (10% 0 (0%) 26 (7%)
2010 371 320 (86%) 4 (1%) 22 (6%) -21 (-6%) 46 (12%)
2011 295 231 (78%) 3 (1%) 19 (6%) 25 (9%) 44 (15%)
2012* 422 310 (73%) 51 (12%) 24 (6%) 0 (0%) 92 (22%)
2013 482* 247 (51%) 12 (2%) 113 (23%) 0 (0%) 68 (14%)
*Upslope data only.

AET ΔS R LD NP
2007 449 317 (70%) 67 (15%) 25 (6%) 0 (0%) 39 (9%)
2008 226 241 (107%) -87 (-39%) 31 (14%) 0 (0%) 41 (18%)
2009 409 365 (89%) 51 (12%) 46 (11%) -112 (-27%) 60 (15%)
2010 355 334 (94%) 1 (1%) 31 (9%) -54 (-15%) 42 (12%)
2011 356 116 (48%) 61 (25%) 50 (14%) 0 (0%) 8 (3%)
2012 464 311 (67%) 17 (4%) 53 (11%) 0 (0%) 90 (19%)
2014* 220 154 (70%) 10 (5%) 48 (22%) 0 (0%) 21 (10%)
*Data missing

AET ΔS R LD NP
2007 449 266 (59%) 100 (22%) 25 (6%) 0 (0%) 58 (13%)
2008 226 265 (117%) -76 (-34%) 20 (9%) 0 (0%) 17 (8%)
2009 408 299 (73%) 18 (5%) 45 (11%) 0 (0%) 46 (11%)
2010 355 269 (76) 19 (5%) 31 (9%) 0 (0%) 34 (10%)
2011 351 101 (42%) 68 (29%) 47 (13%) 0 (0%) 22 (9%)
2012 464 294 (63%) 13 (3%) 53 (11%) 0 (0%) 105 (23%)
2013 434 273 (63%) 0 (0%) 64 (15%) 0 (0%) 44 (10%)

Thermal conductivity (TC) sensors are used to monitor temperatures in the Claude WRP cover system and
upper waste rock profiles, and in the TMA cover system and upper tailings profiles. Freeze-thaw cycles in the
cover profile are important when interpreting runoff data for the water balance, and understanding when
net percolation can be expected throughout the year. The timing and rate of freezing and thawing of a cover
system profile in the fall and spring depends on several factors including snow cover accumulation, ambient
temperatures, and soil water content. The cover system profiles at the Claude WRP and TMA generally begin
to freeze around the end of October or early November, while thaw typically occurs in April or May. Delayed
thaw of a cover system surface impedes infiltration into the profile, increasing runoff volumes and increasing
performance of the cover system via reduced net percolation. Net percolation through the Claude WRP and
TMA cover systems generally occurs in May following thaw of the cover profile, in the summer following
intense rainfall events, and in the fall following large storm events and lower rates of evapotranspiration.
The timing of the precipitation can be of consequence to the amount of net percolation. If storm events occur
in September and October when evapotranspiration is no longer able to remove stored water from the cover
system, greater than expected net percolation can occur. In addition, if the entire soil profile does not
completely freeze or freezing is delayed, the water at the base of the cover system can continue to percolate
into the underlying waste, further causing conditions for higher net percolation.
Performance of cover systems will evolve over time in response to site-specific physical, chemical, and
biological processes (INAP, 2003). Substantial growth of the various grass and legume species has occurred
between August 2007 and August 2014 on the Claude WRP cover system (Figure 6) and the TMA cover system
(Figure 7), with only minor observed erosion. A mature vegetation cover contributes to lower net percolation
volumes through increased interception and transpiration rates, and will also serve to inhibit erosion.
(a) August 2007 – Plateau Station (b) July 2008 – Plateau Station (c) September 2009 – Plateau Station
(c) June 2012 – Plateau Station
(d) August 2014 – Plateau Station

Covers
(a) August 2007 – CN1000L (b) July 2008 – CN1000L (c) September 2009 – TMA
Meteorological Station
(c) June 2012 – CN1000L (e) August 2014 – CN1000L
Net percolation has averaged 22% and 16% for the top and sloping areas of the WRP respectively, and 12%
for both the CS1100L and CN1000L stations at the TMA since construction of the cover systems in 2006
(Figure 8). In general, the cover systems appear to be performing as designed and net percolation rates will
likely continue to improve with evolution of the vegetation cover. Variability in net percolation as a
percentage of precipitation occurred during the eight years of monitoring, which is to be expected in
response to normal cycles in the local climate. General trends in performance of cover systems cannot be
inferred from a single year of monitoring data; it is only when examining net percolation over the long term
and in the overall context of normal climate variability that trends in performance can be determined. Natural
climatic variability is to be expected and was accounted for during design of the Claude WRP and TMA cover
systems.
The Claude WRP and TMA cover systems are stable landforms supporting the growth of productive native
plant species and attenuating radiation emanating from stored waste material to acceptable levels. Net
percolation rates generally reflect expected performance of the cover systems, which will likely continue to
improve with vegetation cover evolution. The performance of these cover systems underlines the importance
of maintaining a long-term perspective when evaluating cover system performance in terms of reducing the
net percolation of meteoric waters. Field monitoring data are currently being used to calibrate a soil-plant-
atmosphere model of the cover systems, which will be used to develop a higher level of confidence in the
estimated long-term performance of the cover systems.

40
Net Percolation (% of incident precipitation)
30
WRP Plateau
TMA CS1100L
20 WRP Slope
TMA CN1000L
10
0
2007 2008 2009 2010 2011 2012 2013 2014
AREVA (2007) Cluff Lake Project, as-built – Claude mining area decommissioning.
CNSC (Canadian Nuclear Safety Commission) (2003) Comprehensive study report for Cluff Lake Decommissioning Project.
COGEMA (2001) Cluff Lake Project comprehensive study report.
INAP (International Network for Acid Prevention) (2003) Evaluation of the long-term performance of dry cover systems, final report.
Prepared by O’Kane Consultants Inc., Report No. 741/1-03 prepared for AREVA, September.
OKC (O’Kane Consultants Inc.) (2006) Claude Waste Rock Pile and TMA Cover Performance Monitoring Reference Manual. Report
No. 739/2-04 prepared for AREVA, February.
OKC (2015a) Cluff Lake Project Claude Waste Rock Pile Cover System 2013 to 2014 Performance Monitoring Report. Report No.
741/10-01 prepared for AREVA, February.
OKC (2015b) Cluff Lake Project Tailings Management Area Cover System 2013 to 2014 Performance Monitoring Report. Report No.
741/10-02 prepared for AREVA, February.

G.C. Allen O’Kane Consultants, Canada

L.K. Tallon O’Kane Consultants, Canada
D. Flemming North American Tungsten Corporation Ltd., Canada
S. Shaw pHase Geochemistry, Canada
B.K. Ayres O’Kane Consultants, Canada
The North American Tungsten Corporation Ltd.’s (NATCL) Cantung tungsten mine (Cantung) is located in the
Nahanni area of the Northwest Territories, Canada (62°N). Intermittent mining has occurred at Cantung since
1962, resulting in five current tailings and exfiltration ponds (TPs), three planned dry stack tailings storage
facilities (TSFs) and a thin deposit of historic tailings located on the floodplain of the Flat River. As part of
ongoing reclamation activities at Cantung, cover systems are being evaluated for the planned TSFs to reduce
the acid-generating potential of these proposed landforms. The proposed dry stack TSF cover system design
will be informed by data gathered at the current TP1 and TP2 tailings cover system. To understand the
influence of water and gas flow dynamics on performance of cover systems at TP1 and TP2, a numerical model
with inputs informed by the tailings containment area monitoring plan (TCAMP) program was used to
estimate water flux and diffusive oxygen flux to the tailings material. The numerical model was constructed
to allow for a changing oxidation rate over time for each discrete layer of tailings material within the model,
representing the consumption of pyritic materials with time. The final cover system design for the planned
dry stack TSFs will depend on acceptable infiltration and oxygen ingress rates that reduce loadings to
acceptable levels in the receiving environment over the long term.
The North American Tungsten Corporation Ltd.’s (NATCL) Cantung tungsten mine (Cantung) is located in the
Nahanni area of the Northwest Territories, Canada (62°N). Intermittent mining has occurred at Cantung since
1962. This has resulted in historic deposition of tailings on the floodplain of the Flat River as well as managed
tailings disposal in five current tailings and exfiltration ponds (TPs) and three planned dry stack tailings
storage facilities (TSFs). The Cantung tailings are considered potentially acid generating, with pyrrhotite being
the primary sulphide mineral of concern, though with a significant lag phase (decades) owing to the presence
of calcite in the tailings. Currently, these tailings remain buffered to near-neutral pH values. An initial
conceptual model for a cover system to reduce acid-generating potential was developed based on current
understanding of performance of the cover systems at existing TPs (TP1/2) as part of water license
amendment plans required for the construction of the new dry stack TSFs. As ongoing reclamation activities
at Cantung have progressed, so has the conceptual model for cover systems at the new dry stack TSFs. Several
studies, including physical monitoring and characterisation of existing TPs and cover system numerical
modelling, have informed changes to the initial conceptual model.
The tailings containment area monitoring plan (TCAMP) program was initiated in 2011 to provide
geochemical, hydrological, and geotechnical information for several mining units to inform closure planning.
Specifically, the TCAMP program initiated monitoring and characterisation of the cover systems and tailings
at the reclaimed TP1/2. The location of TP1/2 and the general location of the historic Flat River floodplain

Evaluation of cover systems for closure of TSFs at NATCL’s Cantung mine G.C. Allen, L.K. Tallon, D. Flemming, S. Shaw and B.K. Ayres
tailings are shown in Figure 1. Geotechnical and geochemical sampling of the cover system and tailings
material was completed, and ongoing monitoring at TP1/2 aims to investigate the specific meteorological,
unsaturated zone and saturated zone conditions in the vicinity of TP1/2.
Flat River
Floodplain Tailings
TP 1 TP 2
An initial conceptual model for a cover system to reduce acid-generating potential at future TSFs was
developed based on the understanding of cover system performance. Data collected as part of the TCAMP
program were reviewed in order to refine the conceptual model for cover systems. Data reviewed from the
TP1/2 program included ambient meteorological data and in situ soil and tailings data that included matric
suction, volumetric water content and pore gas oxygen concentrations
Geochemical analyses completed on both the Flat River floodplain tailings and TP1/2 tailings were also
reviewed. The geochemical analyses included acid base accounting, metals analysis, leachate extraction
analyses, kinetic data (humidity cell and field barrel tests), and particle-size distributions (PSDs).
Following the review of available background studies, the initial conceptual model of performance was
updated and further investigated using numerical modelling. The purpose of numerical modelling was to
understand the influence of water and gas flow dynamics on performance of cover systems at TP1/2.
Numerical modelling using VADOSE/W (GEO-SLOPE International Ltd., 2014) with inputs from the TCAMP
program was completed to estimate water and diffusive oxygen flux to the tailings material.
VADOSE/W is a two-dimensional (2-D) finite element model that predicts pressure head (suction) and
temperature profiles in the soil profile in response to climatic forcing, such as evaporation, and lower
boundary conditions, such as a water table. VADOSE/W is a fully coupled (through the vapour pressure term)
heat and mass transfer model capable of predicting water vapour movement and gas diffusion.

Covers
VADOSE/W requires material property inputs, climate inputs, vegetation inputs and a lower boundary
condition. Material property inputs include water retention curves, hydraulic conductivity functions, thermal
conductivity function and volumetric-specific heat functions. Material property inputs were developed using
material PSDs and results from other geotechnical test work. A 100-year climate database was developed for
the site based on historic climate data collected at the Environment Canada Tungsten Station (Government
of Canada, 2014a), the Environment Canada Rabbit Kettle Station (Government of Canada, 2014b) and the
TCAMP weather station, and from high-resolution gridded temperature and precipitation datasets (CRU,
2000, 2014). A lower boundary condition developed from local groundwater conditions at TP1/2 indicated
that the water table was generally below the base of the tailings.
VADOSE/W assumes that oxygen diffusion is the dominant oxygen transport mechanism across the cover
system and does not model advective transport. Aubertin et al. (2000) describe the use of an effective
porosity term to represent the porosity available for diffusion based on the air-filled porosity plus a portion
of the water-filled porosity. This relationship transforms the water-filled porosity into an equivalent air-filled
porosity by portioning it with Henry’s Law coefficient as follows:
θeff = θa + HθW (1)
Where:
θeff = effective porosity (m3/m3),
θa = volumetric air content (m3/m3),
θw = volumetric water content (m3/m3), and
H = Henry’s Law coefficient (approximated as 0.028 for O2 in air and water at 25°C).
The rate of diffusion is described through the use of Fick’s Law, as follows:
δC
q = −θeff D × δx (2)
Where:
q = mass flux of oxygen (kg/m2/s),
θeff = effective porosity (m3/m3),
D* = bulk diffusion coefficient (m2/s),
C = oxygen concentration in the gas phase (kg/m3), and
x = distance (m).
The diffusion coefficient through dry soil is nearly four orders of magnitude higher than it would be for
saturated soil, which is similar to the coefficient of diffusion through water. There is a substantial decrease
in the oxygen diffusion coefficient at higher degrees of saturation.
VADOSE/W calculates the oxygen flux through porous media based on diffusion through the gas phase, but
does not account for gas transfer due to advection in the air phase. The model makes the assumption that
the air phase in the soil is always at atmospheric pressure; thus, gas movement through the soil is by diffusion
only in the air phase. Gas can be dissolved into the water phase but is not transported in the water phase.
Oxygen mass can be lost to decay processes (GEO-SLOPE International Ltd., 2014).
An oxidation rate was set for each discrete layer of tailings material using the decay process function of
VADOSE/W. The oxygen concentration gradient used to calculate the oxygen mass flux was then based on
the concentration at the user-defined node and atmospheric oxygen concentration (i.e. 280 g/m3).
VADOSE/W automatically calculates the effective porosity based on the total porosity of the material and the
water content in an element at a specific time step. VADOSE/W then calculates the diffusion coefficient (D*)
based on the method proposed by Collin and Rasmuson (1988), which is strongly dependent on the degree
of water saturation in the soil.

Oxygen consumption was modelled using an oxidation coefficient (referred to as the decay coefficient
constant in VADOSE/W), which removes oxygen within each element at each time step. The decay rate is
interpreted by the solver as a “decay half-life” parameter (t1/2), which is computed as follows:
1⁄
t 2 = ln 2⁄λ (3)
Where:
λ = decay coefficient constant (1/time).
A first-order reaction rate was assumed to be sufficient for predicting pyrrhotite oxidation in the Cantung
tailings materials. As such, the following relationship can be used to describe the rate of change of
concentration in the pore space with time:
δC
δt
= −λCm (4)
Where:
C = oxygen concentration in the pore space,
M = reaction order assumed equal to 1, and
t = time.
The numerical model was constructed in discrete 1 m blocks, which allowed for a changing oxidation rate
over time in each layer of tailings material to represent the consumption of pyrrhotite with time. Based on
geochemical characterisation of the tailings, an initial total sulphur content of 9%, equivalent to 24.6% FeS
assuming no other sulphur species were present in the tailings, was assumed for the entire tailings profile.
An initial decay coefficient constant and the reactivity of pyrrhotite with oxygen were derived from sulphate
production rate in tailings humidity cell testing. The following oxidation reactions were considered in
developing an oxygen decay rate:
FeS + 2O2 → Fe2+ + SO42- (5)
FeS + 2.25O2 + 2.5H2O → Fe(OH)3 + SO42- + 2H +
(6)
An empirical formula for pyrrhotite of FeS was assumed, as there is likely a mixture of both hexagonal and
monoclinic pyrrhotite mineral forms at Cantung. For the purpose of this analysis, it was assumed that only
Eq. 5 was involved in sulphate production as a conservative measure to maximise the depth of oxygen
diffusion into the model.
Oxygen flux through each discrete model layer was calculated in VADOSE/W for a five-year time step. The
amount of pyrrhotite consumed was then calculated using Eq. 6 based on the calculated five-year cumulative
oxygen flux. As all oxygen consumption was assumed to be driven by pyrrhotite oxidation, as pyrrhotite was
oxidised, demand for oxygen (decay coefficient λ) decreased with time.
To calculate a decreasing decay coefficient, a constant reactivity of pyrrhotite with oxygen was calculated
from the reactivity rate of the tailings material (from humidity cell testing) and the proportion of sulphide
materials within the tailings using Eq. 7 as proposed by Collin (1987) and Mbonimpa et al. (2002). This method
has previously been applied to materials containing pyrite:
6
K r = K′ D (1 − n)Cp (7)
H
Where:
Kr = the oxygen reactivity rate coefficient (s-1),
K’ = reactivity of pyrrhotite with oxygen (m3[O2]/m2[FeS]/s),
DH = the equivalent particle-size diameter (m),
n = porosity (m3/m3), and

Covers
CP = pyrrhotite content per mass of dry tailings (kg [pyrrhotite]/kg [tailings])

Equivalent particle size can be estimated from the PSD using the following equation (Aubertin et al., 1998,
Mbonimpa et al., 2002):
DH = [1 + 1.17log(CU)]D10 (8)
Where:
CU = coefficient of uniformity, and
D10 = diameter at which 10% of all sample mass is smaller
and CU can be described by:
CU = D60/D10 (9)
Where:
D60 = diameter at which 60% of all sample mass is smaller.
The decay coefficient (λ) is related to the oxygen reactivity rate coefficient (Kr) by the following relationship:
Λ = Kr/θeff (10)
Effective porosity is calculated within VADOSE/W as described by Eq. 1. Therefore, the oxygen reactivity rate
coefficient (Kr) was entered into VADOSE/W as a material parameter to determine the “decay half-life” of
oxygen. For the purpose of this study, the half-life decay parameter was updated for each discrete model
layer for each time step as a function of the decreasing pyrrhotite content using Eq. 7. This allowed for the
calculation of the depth of oxidation within the tailings material over time.
The development of a conceptual model is an iterative process. It requires continued changes as new
information becomes available. The objective of this study was to validate a particular iteration of the
conceptual model using the various tools at a cover system designer’s disposal. Therefore, the results of the
study are both a set of model inputs developed through physical studies and their interpretation, as well as
outputs of a numerical model, which provided validation or previous conceptual models. The development
of inputs to the numerical model and the model outputs are discussed below.
The main objective of cover systems for the tailings facilities at Cantung is to reduce the potential for acid
generation of the tailings landforms. Thus far, the cover system at TP1/2 has successfully managed acid
generation. Prior to this study, the conceptual model of performance at TP1/2 was that the cover system in
place was limiting meteoric water and oxygen ingress to the tailings, thereby reducing sulphide oxidation and
the resulting potential acid load from the tailings. As this approach appeared successful in meeting cover
system objectives, preliminary design of the future dry stack TSFs proposed that cover systems for these
facilities should be designed to promote surface water runoff, reduce erosion and prevent the infiltration of
water and air onto the facility (NATCL, 2014).
While the cover system at TP1/2 has been effective in reducing sulphide oxidation and acid generation, the
conceptual model for performance needed to be refined to optimise a cover system design for future dry
stack TSFs. The probable performance of a given cover system is a function of its climatic setting, the
materials available for construction, the geotechnical and geochemical material properties of the cover
system and waste material, the hydrological context of the site, and the quality control measures in place
during construction of the cover system. The effects of quality control measures and of altering the cover
system materials already in place at TP1/2 were not considered in this investigation.

The climatic setting of a mine site is the most important parameter to consider when attempting to
determine how well a soil cover system will limit the infiltration of water. The ratio of rainfall to potential
evapotranspiration (PE) is of particular importance. Based on the 100-year climate database developed,
Cantung fell within the Dfc classification for the Köppen climate classification system and was broadly
described as continental subarctic or boreal (taiga) climate. The D in the Köppen climate classification
describes the temperature regime where the hottest month is greater than 10°C and the temperature of the
coldest month averages less than or equal to 0°C. The f denotes the moisture regime being mesic (wet),
without a pronounced dry period, and cool summers. Total precipitation at the site averages 620 mm per
year, and is split approximately evenly between rainfall and snowfall. Average annual potential evaporation
has been estimated as 320 mm.
A tri-linear plot of climate classification is often a good starting point for determining what type of soil cover
system is likely to be successful for a given site (INAP, 2009) (Figure 2). The conceptual model of performance
of the TP1/2 cover system is consistent with an infiltration control/water shedding cover system, but based
on the climate classification discussed above, Cantung falls within the zone where water cover systems and
low-permeability oxygen barriers or organic cover systems would be recommended. The need for further
review of the geotechnical and geochemical material properties and hydrological context was identified as a
result of the disagreement between the conceptual model of performance and the climate-predicted cover
system performance.
Geotechnical and geochemical investigations of the TP1/2 cover system material and tailings were completed
at Cantung as part of the TCAMP program and general closure planning. The geotechnical properties of both

Covers
the tailings and cover system material were reviewed and compared to the conceptual model of
performance. Tailings mill samples and grab samples from TP3 and the Flat River floodplain tailings were
reviewed, as were cover system material grab samples from TCAMP drilling at TP1/2 (Figure 3).
Grab samples of cover system material at TP1/2 determined an average of 3% clay, 18% silt, 42% sand and
37% gravel. It was also determined that two distinct PSDs were present within the cover system materials,
which likely reflected the two major borrow sources of fluvial outwash and reworked talus material.
Atterberg testing was also completed on the cover system material samples. The average Unified Soil
Classification System (USCS) classification for the cover system material was determined to be silty sand (SM)
(ASTM, 2011). Tailings taken from the mill and TP1/2 were similar in texture, while samples from TP3 and the
Flat River floodplain tailings were slightly coarser. Cantung tailings typically comprised approximately 56%
silt, 39% sand and 5% clay-sized particles (Tetra Tech EBA, 2014) and have been described as inorganic silts
(ML) based on the USCS soil classification system (ASTM, 2011).
The coarse texture of the cover system materials suggested that the cover system was likely highly permeable
with low water-holding capacity. This was visually confirmed by the lack of vegetation found at TP1/2, as
coarse soils drain quickly, resulting in insufficient water available for plant use. Healthy vegetation is often
both a marker and a requirement of a successfully performing store-and-release cover system. The textural
contrast between the cover system and tailings materials also suggested that limiting meteoric water from
infiltrating the tailings material may not be the mechanism limiting acid generation, as proposed by the initial
conceptual model of performance.
An alternative conceptual model of performance was developed because of the incongruity between field
data collected and the conceptual model of performance. The refined conceptual model of performance
presented in OKC (2014) stated that the cover system at TP1/2 is not thought to be limiting meteoric water
from infiltrating to the tailings material. It is proposed that the high net percolation rates are currently
contributing to maintaining high levels of saturation within the tailings material, which are in turn
contributing to negligible sulphide oxidation being observed at TP1/2.

1.E-02
1.E-03
Cantung Tailings
Hydraulic Conductivity (cm/s)
1.E-04
Coarser Cover Material
1.E-05
Finer Cover Material
1.E-06
1.E-07
1.E-08
1.E-09
1.E-10
1.E-11
1.E-12
1.E-13
1.E-14
0 1 10 100 1,000 10,000 100,000 1,000,000
Suction (kPa)
A numerical modelling program was conducted to confirm the updated conceptual model. Two one-
dimensional (1-D) column models of TP1/2 were modelled using VADOSE/W. The first 1-D model comprised
a 1 m cover system of the coarser cover system material encountered in sampling overlying 9 m of tailings
material, while the second model comprised a 3 m cover system of the finer cover system material overlying
7 m of tailings material. Lateral drainage through the cover system was not considered within the model, as
the cover system is not graded and thus there is no driver for lateral flow through the cover system. The 100-
year climate database was applied as a boundary condition at the surface of the 1-D models, and vegetation
was not considered, as TP1/2 was not vegetated. A lower boundary condition was set as a unit hydraulic
gradient, as ground water levels were below the base of tailings and water movement was thus dominated
by gravity near the base of TP1/2. Material property functions used in VADOSE/W for each material are
shown in Figures 4 and 5.
0.40
Volumetric Water Content (m3/m3)
Cantung Tailings
0.35
Coarser Cover Material
0.30 Finer Cover Material
0.25
0.20
0.15
0.10
0.05
0.00
0.1 1 10 100 1000 10000 100000 1000000
Suction (kPa)

Covers
The tailings produced at the Cantung mine have remained relatively unchanged in terms of primary
mineralogy throughout the life the mine and are not expected to change as a result of future mining. The
tailings have demonstrated a potential for generating acid, with pyrrhotite comprising the principal acid
generating mineral. Pyrrhotite is present in the Cantung tailings at an average concentration of 5 to 10% as
total sulphur. The primary mineral with neutralising potential is calcite. The ratio of neutralising potential to
acid potential has historically been just below 1; that is, given the right conditions, the tailings at the site will
exhibit a tendency to produce acid unless measures are taken to prevent the process. Humidity cell testing
on fresh mill tailings confirmed the acid-generating potential. Following approximately 180 wetting and
drying cycles, the mill tailings cell became acidic. Sulphate production and pH from the humidity cell test
work of the mill tailings are shown in Figure 6. The vertical line in the right pane shows the onset of acidic
conditions.
3000 9
2500 8
7
2000
SO4 (mg/L)
pH
1500
5
1000
4
500 3
0 2
0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350
Wetting and Drying Cycles Wetting and Drying Cycles
Average sulphate production between cycles 71 and 100 of 611mg/kg/week prior to the onset of acidic
leachate was selected to represent the initial conditions for modelling the TP1/2 Cantung tailings. The
sulphate production rate resulted in an initial oxygen reactivity rate coefficient (Kr) of 3.9 × 10-6 s-1 and a
resulting reactivity of pyrrhotite with oxygen of 2.4 × 10-10 m3 (O2)/m2(pyrrhotite)/s. For comparison, the
reactivity of pyrite with oxygen at another site has been stated as 5 × 10-10 m3 (O2)/m2 (pyrite)/s (Martin et
al., 2006).
Annual oxygen flux to the tailings through the cover system was modelled over a 100-year period in five-year
time steps. Figure 7 presents the probability that annual oxygen flux at the cover system/tailings interface
will exceed a given value. For example, the right pane shows that in any given year there is a 60% chance that
diffusive oxygen flux through a 3 m finer cover will exceed 150 g/m2. This probability of exceedance approach
was used rather than presenting an annual average oxygen flux to tailings in order to quantify the uncertainty
in modelling results, as oxygen flux is not a static process. This approach is often used in flood analysis and
has implications for how conservatively a future cover system is designed. The shaded area in the left pane
is enlarged in the right pane of Figure 7. For reference, a water cover of 0.5 to 1 m is generally expected to
limit oxygen ingress to approximately 64 g/m2 (2 moles/m2/yr). A line denoting 64 g/m2 is shown in the right
pane of Figure 7. Below the cover system/tailings interface, at a tailings depth of 1 m, modelling showed
oxygen flux was limited to much less than 64 g/m2. The results at a 1 m tailings depth are not shown, as the
probability that oxygen flux would exceed 64 g/m2 was essentially 0%. Beyond a 1 m tailings depth, modelling
showed oxygen flux was limited to much less than 64 g/m2, which was consistent with the conceptual model
of performance proposed.

2500 500
1m Coarser Cover System
450
3m Finer Cover System
Annual Oxygen Flux (g/m2)

3m Finer Cover System
Annual Oxygen Flux (g/m2)
2000 400
350
1500 300
250
1000 200
150
500 100
50
0 0
0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100%
Probability of Exceedance (%) Probability of Exceedance (%)
The results of this study have several implications for future TSFs at Cantung. Because of the relatively wet
climatic context of the site and the water retention properties of the tailings, diffusion of oxygen and
oxidation of pyrrhotite-rich tailings will be limited if saturation of the tailings can be maintained. This counters
the initial conceptual model, which aimed to shed meteoric water and limit infiltration to the tailings.
Modelling results that agree with observations of very minimal sulphide oxidation occurring below the
existing cover system will be further confirmed through in situ pore gas monitoring instrumentation in place
at TP1/2, and could be even further confirmed through continued test pitting over time through the TP1/2
cover system and tailings material to observe pyrrhotite content with depth.
Although the acid-generating potential of the tailings at Cantung is currently still buffered to near neutral pH
by the presence of calcite in the tailings, the tailings are expected to generate acid once calcite has been
depleted and in the presence of sufficiently high oxygen concentrations. Therefore, the primary means of
controlling the potential for acid generation is to limit the amount of either water or oxygen available to drive
the sulphide oxidation reaction. Given that the reaction can proceed using only the water present in the air,
limiting water is not a practical approach given the climatic context. The alternative is to limit the amount of
oxygen available for the reaction, and the primary means for limiting oxygen is to provide a barrier to oxygen
diffusion by maintaining a high degree of saturation in the tailings. Therefore, the conceptual model of a
successful closure system is one in which a cover system allows for the highest geotechnically feasible degree
of saturation in the tailings. Limiting net percolation to the lowest level possible may not only be technically
challenging given the climate and materials available at the site, but also incur the negative outcome of
increasing sulphide oxidation and the potential for acid production over time.
ASTM D2487-11 (2011) Standard practice for classification of soils for engineering purposes (Unified Soil Classification System). ASTM
International, West Conshohocken, USA.
Aubertin, M., Aachib, M. and Authier, K. (2000) Evaluation of diffusive gas flux through covers with a GCL, Geotextiles and
Geomembranes, Vol. 18, pp. 215–233.
Aubertin, M., Ricard, J.F. and Chapuis, R.P. (1998) A predictive model for water retention curve: Application to tailings from hard-rock
mines, Canadian Geotechnical Journal, Vol. 25, pp. 55–69.
Climatic Research Unit (CRU), University of East Anglia (2000) Global land precipitation, viewed 2 October 2014,
http://www.cru.uea.ac.uk/data/precip, last updated 2 June 2000.
Climatic Research Unit (CRU), University of East Anglia (2014) Global land temperature, viewed 2 October 2014,
http://www.cru.uea.ac.uk/data/temperature, last updated March 2014.

Covers
Collin, M. (1987) Mathematical modelling of water and oxygen transport in layere soil covers for deposits of pyritic mine tailings.
Licenciate Treatise, Department of Chemical Engineering, Royal Institute of Technology, Stockholm, Sweden.
Collin, M. and Rasmuson, A. (1988) Gas diffusivity models for unsaturated porous media, Soil Science America Journal, Vol. 52, pp.
311–333.
GEO-SLOPE International Ltd. (2014) VADOSE zone modelling with VADOSE/W — An engineering methodology, May 2014 edition,
GEO-SLOPE International Ltd., Calgary, Canada.
Government of Canada (2014a) Daily data reports for Tungsten, NWT Climate ID 2203922 1966–1990, viewed 18 September 2014,
http://climate.weather.gc.ca/index_e.html#access, last updated 8 August 2014.
Government of Canada (2014b) Daily data reports for Rabbit Kettle, NWT Climate ID 2203342 1995–2014, viewed 18 September
2014, http://climate.weather.gc.ca/index_e.html#access, last updated 8 August 2014.
International Networks for Acid Prevention (INAP) (2009) Global acid rock drainage guide (GARD Guide), viewed 5 January 5 2015,
http://www.gardguide.com/, last updated 18 December 18 2014.
Martin, V., Aubertin, M., Bussière, B., Mbonimpa, M., Dagenais, A-M. and Gosselin, M. (2006) Measurement of oxygen consumption
and diffusion in exposed and covered reactive mine tailings, in Proceedings 7th International Conference on Acid Rock
Drainage, 26–30 March 2006, St. Louis, USA: pp. 1156−1169.
Mbonimpa, M., Aubertin, M., Chapuis, R.P. and Bussière, B. (2002) Practical pedotransfer functions for estimating the saturated
hydraulic conductivity, Geotechnical and Geological Engineering, Vol. 20(3), pp. 235–259.
North American Tungsten Corp. Ltd (NATCL) (2014) Water license MV2002L2-0019: Permit amendment application dry stack tailings
facilities, Cantung Mine, NT, date of submission 28 March 2014.
O’Kane Consultants Inc. (2014) Cantung mine – tailings containment area cover system design study report. Report No. 937/2-01.
Prepared for North American Tungsten Corporation Ltd December 2014.
Tetra Tech EBA (2014) Preliminary design report TSF6: Dry stack tailings storage facility, Cantung mine, Tungsten, NT, date of
submission March 2014.


D. Sanscartier Saskatchewan Research Council, Canada

J. Lambert McElhanney Consulting Services Ltd., Canada
C. Reid Saskatchewan Research Council, Canada
D.E. Allen Saskatchewan Research Council, Canada
Project CLEANS (Cleanup of Abandoned Northern Sites) involves the decommissioning and rehabilitation of
remote abandoned uranium mines on public land in northern Saskatchewan, Canada, with the objective of
transferring them to the provincial Institutional Control Program for long-term maintenance and monitoring.
The majority of the 37 sites in Project CLEANS have vertical openings to the underground that require proper
closure to ensure long-term public safety. In Saskatchewan, the legislated requirement is to close openings to
the underground with a reinforced concrete bulkhead designed by a professional engineer. This is challenging
in remote locations with no road access or where access is limited to an ice road for a six-week period during
the winter. Since 2005, stainless steel covers have been used as a closure method for Project CLEANS sites.
Recently, an evaluation of closure methods was carried out to identify the most appropriate closure method
for these mine sites. High-performance concrete and several types of steel covers (weathering, galvanised
and stainless) and aluminium were compared with respect to the following design criteria: sufficient service
life (200 years) and durability, ease of transport, prevent unauthorised access, endure climate change, forest
fires, and additional loading (e.g., animals, snowmobiles), minimise field fitting, and cost effectiveness.
Structurally reinforced stainless steel plate was identified as the preferred material. The approach to
designing a closure is to first perform a site investigation to determine the closure size and points of
attachment to the surrounding bedrock. A 3-D model of the terrain is then developed in order to optimise the
shape of the closure. This provides a complete and secure closure with a limited amount of material use, while
maintaining a clear load path to the anchor points. The covers are fabricated at a qualified facility in
Saskatoon, Saskatchewan, and then trucked to the sites on ice roads and site access trails. The covers are
installed by a local contractor and the design allows some flexibility for field fitting. Quality assurance and
control includes inspections during fabrication and after installation. Between 2005 and 2012, ten mine
openings were closed with stainless steel covers. Since 2013, two shafts and five raises have been successfully
closed using the more stringent design and quality assurance process described above. Although the stainless
steel covers require an exemption from the Saskatchewan Mines Regulations prior to implementation, they
are now accepted by regulators as a safe and durable mine closure method for remote sites.
The Saskatchewan Research Council (SRC) manages the rehabilitation and closure of 37 legacy uranium mine
sites on public land in Northern Saskatchewan, Canada, on behalf of Saskatchewan Ministry of Economy. This
project is also known as Project CLEANS (Cleanup of Abandoned Northern Sites). The objective is to transfer
the sites to the provincial institutional control program for long-term maintenance and monitoring.
There are 37 mine sites in the vicinity of Uranium City, Saskatchewan, and two sites near La Ronge,
Saskatchewan (Figure 1). Activities related to uranium exploration, extraction, and/or milling took place at
the mine sites. The majority of the 37 sites in Project CLEANS have vertical openings to the underground that
require proper closure to ensure long-term public safety.

Chapter name D. Sanscartier, J. Lambert, C. Reid and D.E. Allen
When the mines were closed or abandoned, closure methods for the vertical openings were rudimentary
and inadequate. Openings were loosely covered with lumber, steel grates and debris, partially backfilled with
waste rock or covered with soil, or simply left open. Some remediation of the site was done in the late 1980s
by Saskatchewan Ministry of Environment (e.g., building demolition, fencing, installation of concrete covers).
None of the closure methods applied had been designed with long-term safety in mind, and most of the
covers were failing or had failed when investigation of the legacy mine sites was initiated in the early 2000s.
In many cases, accumulation of leaves, vegetation growth, soil subsidence, and rock slides over decades
made it difficult or impossible to find the openings.
Fortunately, to the author’s knowledge, inadequate closures did not result in public safety incidents. The
remoteness of the sites limits site accessibility, which reduces the public safety risks associated with the
openings. That said, Uranium City residents reported visiting several of the legacy sites for different purposes,
such as hunting, trapping, wood harvesting, berry picking, harvesting of fire wood, or simply travelling
through the site in motorised vehicles. In addition to the local population, several other stakeholders may be
affected by the adequate closure of the openings: First Nations in the Athabasca region, Saskatchewan
Ministry of Economy, Saskatchewan Ministry of Environment, geologists regularly visiting the sites, mining
exploration companies, etc.
In Saskatchewan, the legislated requirement and accepted method for closure of mine openings to the
underground is a reinforced concrete bulkhead designed by a professional engineer (article 407(2) of the
Saskatchewan Mines Regulation, 2003). The application of other closure methods requires an exemption
from the Saskatchewan Mines Regulation.
The legislated method is challenging to apply in remote locations such as the Uranium City area, where road
access is limited to an ice road for a six-week period during the winter. Once in Uranium City, several of the
legacy sites are accessible with the local network of gravel roads and trails. However, several sites are only
accessible by air (float plane, helicopter) or water (barge, boat). This makes it difficult or very costly to bring
the equipment and the high quality material and aggregates needed for the construction of a concrete
bulkhead which meets the required longevity and durability criteria.
In 2005, an exemption was obtained to use stainless steel covers as a closure method for a shaft, a raise and
another vertical opening at the Lake Cinch/Cenex Uranium Mine Sites, one of the Project CLEANS sites.
Uranium City Contracting Ltd. (UCC Ltd.), a local contractor in Uranium City and Kova
Engineering (Saskatchewan) Ltd. participated, with the Government of Saskatchewan Department of

Covers
Northern Affairs, in developing the idea, designing the closure, and obtaining approval of the first-generation
stainless steel covers.
Between 2005 and 2012, ten mine openings were closed with stainless steel covers by UCC Ltd. and Kova
Engineering (Saskatchewan) Ltd. Over seven years, the design of the first-generation covers has evolved from
covers anchored to the former mine openings concrete collars (Figure 2) to covers anchored to bedrock with
other improved features (Figure 3). The covers effectively close the openings, eliminate public safety hazards,
and meet regulatory requirements. However, in some cases, these designs used large quantities of stainless
steel and required extensive field fitting. Note that for conciseness we will refer to stainless steel covers
installed between 2005 and 2012 as “first-generation covers.”
With the intent to identify other suitable and potentially less costly material, SRC retained McElhanney to
carry out an evaluation of different materials for closure of mine openings and identify the most appropriate
material. McElhanney has been involved with the closures design for Project CLEANS since 2012 and has been
involved in closing seven openings using a new, refined stainless steel cover design. For conciseness we refer
to stainless steel covers designed by McElhanney as “second-generation covers.” The second-generation
cover design builds on and provides refinements to the first-generation cover design.

The objectives of this paper are three-fold:

1. Present an evaluation of different materials for permanent closure methods for vertical openings
to the underground at remote mines.
2. Present the design process and features of the second-generation covers applied at Project
CLEANS sites.
3. Provide cost estimates of stainless steel covers installed at Project CLEANS sites.
The first step in the design process was to evaluate different closure methods to identify the most
appropriate cover based on the specific criteria for the remote Uranium City area mine sites. High-
performance concrete; aluminium; weathering, galvanised, and stainless steel covers were compared with
respect to the following design criteria: sufficient service life (200 years) and durability, ease of transport,
prevent unauthorised access, endure climate change, forest fires, and additional loading (e.g., animals,
snowmobiles), minimise field fitting, and cost effectiveness. Details on each design consideration and on
materials are presented below. Backfilling of the openings and underground workings and polyurethane
foam plugs are other methods used to close mine openings. However, they are not examined in this paper.
The following design considerations were identified for the design of the second-generation covers in order
to meet Project CLEANS needs:
 Sufficient service life and durability: the integrity of the covers must not be reduced as a result of
annual freeze-thaw cycles, corrosion, acid rock drainage and other hazards, over a design life of
200 years.
 Ease of transport: many mines are in remote locations. Specifically for the mines in the Uranium
City area, bulk materials are shipped to the area over a winter ice road, or in the open water
months via barge. From Uranium City, the access roads to the mines in many cases have long
since deteriorated. Access to some of the mines included in Project CLEANS is by water or air only.
So, any solution must be constructible without the use of large cranes or other conveniences
available in less remote areas. Due to road and trail conditions, loads are typically restricted to a
maximum width of four metres.
 Prevent unauthorised access: in order to maintain public safety, the installed covers must resist
unauthorised access. Historically abandoned man-made structures have attracted curious people
who may venture into the dangerous and unstable underground workings. Furthermore,
depending on the type of minerals mined at the site and current industry demands, prospecting
and exploration may occur in the future.
 Endure climate change: covers must be designed for the possibility of more extreme weather,
increased precipitation including snow loads, and more dramatic temperature changes.
 Forest Fires: wildfires are inevitable in the boreal forest. As such, the covers must be able to
withstand the intense heat of a fire.
 Additional loading: vegetation may cause organic build-up. Large animals may pass over the
covers. There may be snowmobiles and other vehicles driving over the covers. It is foreseeable
that exploration and mining may return to the area as well. All of these loading scenarios must be
taken into consideration when designing the covers.

Covers
 Minimise field fitting: thorough design will reduce the requirement for field fitting during
installation. However, the design has to be flexible to accommodate unavoidable alterations when
unexpected site conditions are encountered. In cases of very irregular rock surround or
installation challenges, field fitting will be required and must be incorporated into the design.
 Economically effective: as the mine sites in Project CLEANS are the responsibility of the provincial
government, there is a need to safeguard taxpayers’ dollars. Life cycle considerations imposed a
need to minimise future replacement or repair costs.
Several materials were investigated in order to determine which material to use for the closures. Table 1
outlines variables evaluated during the decision-making process.
Based on SRC’s design requirements, the remote location, and the aforementioned material qualities,
stainless steel was determined to be the material of choice. This material offers the best balance of structural
strength, durability, ease of fabrication, field modifications, availability, and familiarity with fabricators and
installation contractors. It has also been previously approved to be used for mine opening closures by the
provincial regulators. Furthermore, shop fabrication can also easily be monitored for quality control. The
initial cost may be offset by the reduction in on-going maintenance costs, although no life cycle analyses have
been performed. Two stainless steel grades were considered to be appropriate for the project; AISI 304 and
AISI 316. It was determined that AISI grade 304 stainless steel offers the appropriate level of corrosion
resistance in the northern Saskatchewan environment which has been considered to exhibit low atmospheric
corrosion (non-marine, no industrial pollution). The 304 grade offers cost saving over the 316 grade ($0.50-
1.50/kg difference in the last five years [Garand, 2015]). The 316 grade is typically used for more corrosive
environments (marine, processing plants, etc.). Both grades offer suitable welding and fabrication properties.
Pickling has been specified post-fabrication (both shop fabrication and field fabrication) in order to foster the
re-creation of the protective passive layer.
An acceptable second choice for the closures would be galvanised steel or a combination of a galvanised steel
frame and stainless steel covers (including galvanic corrosion isolators). Significant cost savings may be
realised by using galvanised steel over stainless steel and, while published values suggest only a 100 year life
span, a higher life span may be possible in the climate of northern Saskatchewan.
Had the locations been closer to a ready-mix concrete plant, high performance concrete with hot-dipped-
galvanised or stainless steel rebar would have provided acceptable closures at lower cost. These materials
would easily conform and anchor to the undulations of the bedrock profile.
Weathering steel was considered. However, it does not perform well when continuously wet (as would be
found during the snow-melt season) as the surface patina may not have the ability to form.
Aluminium structures, while offering satisfactory corrosion resistance in a non-marine environment, are
inherently fragile and require a higher degree of maintenance.

Criteria Material options

Galvanised Weathering
Stainless steel Aluminium Concrete
steel steel
Raw material High Mid High

Mid Low
cost ($4.00-$8.00/kg) ($2.50/kg) ($5.50-$7.00/kg)
Plate usually in All structural Not available in
Common shapes
stock in fabrication steel shapes up large quantities
are typically
Availability shops. Hot rolled to 6.5 m long Pre-order at remote sites
available in
sections 1-2 weeks may be (Must be hand-
fabrication shops
shipping time galvanised mixed)
Low Mid Mid
Very low Low if mix
Corrosion / (approximately (approximately (approximately
(Approximately designed for site
degradation 0.8-2.0 1.3-3.1 0.025-0.050
0.0025 µm/year) conditions
µm/year) µm/year) µm/year)
Low. Subject to
Medium. Zinc High. See
fatigue in weld
coating may be corrosion/
Durability High High areas. Easily
damaged due to degradation
damaged due to
abrasion notes
low hardness
See steel note. Easy. Cutting,
Easy. Cutting, Hard. While field Hard. It is
Field drilling, and
drilling, and drilling and cutting difficult to
Field modifications welding
welding operations are easy, quality accommodate
installation require touch operations may
may be performed field welding is geometrical field
up to zinc be performed in
in the field difficult to achieve changes
coating the field
Not difficult. No Subject to fatigue
appreciable loss in Not difficult, but Not difficult. No and loss of strength
Easy. See notes
strength due to requires post- appreciable loss in adjacent to weld
Fabrication on field
welding. Requires fabrication in strength due areas. Easily
installation
post-fabrication galvanising to welding damaged due to
pickling low hardness
May be May be
May be fabricated May be fabricated May be pre-cast
fabricated off- fabricated off-
Shipping off-site and shipped off-site and shipped off-site and
site and shipped site and shipped
to site to site shipped to site
to site to site
Shop fabrication Shop fabrication Quality control of
Quality Shop fabrication Shop fabrication
may be may be materials on site
control may be inspected may be inspected
inspected inspected is difficult
Material
Exemption to Exemption to
approved by Exemption to Exemption to
Mines Mines
Provincial Mines Regulation Mines Regulation Yes
Regulation Regulation
Mines needed needed
needed needed
Regulation

Covers
The approach to designing the second-generation closures begins with establishing and documenting realistic
structural design criteria (in addition to the aforementioned design criteria). A set of design loads and load
combination factors were prepared based on established design criteria borrowed from the Canadian Bridge
Design Manual and the National Building Code. Local input was also included to establish foreseeable future
loading possibilities. Such loading scenarios included:
 truck, ATV, and snowmobile traffic;
 wildlife ;
 snow accumulation; and
 accumulation of organic debris (overburden).
Depending on the orientation of the cover and the surrounding topography, the governing design criteria
was either the wheel load from a service truck, or a snow drift.
The goal of the initial site visit is to obtain the data required in order to remotely design closures for each
unique location. This site investigation is performed to determine the overall size of the opening, the location
of sound bedrock, the surrounding site grading (for drainage), and the risks presented by up-slope rock fall.
The site visit provides the opportunity to discuss the closure design with the local contractor, who has
experience installing several other closures, which allows the engineers to determine the contractor’s
abilities for field installation and construction.
Prior to evaluating an opening, the original closure and any accumulated overburden are removed with heavy
machinery. By doing so, the extent of the opening and locations of sound bedrock for anchorage are revealed.
A reflectorless total station is utilised to survey a dense grid of points (approximately 20 shots/m2) on the
surrounding bedrock, and control points are established in order to align the custom design lid during
installation the following field season (Figure 4). Using the survey data, a 3-D model of the terrain is
developed and the shape of the closure is optimised to fit the contours of this model (Figure 5). This step
allows the design team to optimise each cap by reducing the volume of material needed while providing a
secure closure and maintaining a clear load path to the anchor points.

The covers are designed to be raised from the ground by a marginal amount. This decision is based on
providing a well-defined load path to the ground, allowing for easier future inspections, accommodating the
undulating bedrock surface without requiring excess bedrock removal, and making the structures less likely
to be driven on.
Readily available structural shapes are chosen in an effort to eliminate the fabrication costs associated with
custom-fabricating structural shapes. Furthermore, the lid is designed to have a minimum of complex
compound angle joints, while maintaining a low profile and providing drainage. The Hollow Structural Section
(HSS) beams and channels are stitch welded to the cover plate which results in a composite action, further
reducing material costs. A 3/16″ minimum element thickness was maintained for the main structural
elements (beams, joists, and posts) for durability and for weld-ability, regardless of the structural
requirements. However, the side panels, which are subject to less load, are chosen to utilise 1/8″ plate to
ease field cutting and installation and to save on material costs. The corrosion rates are negligible and
minimum support spacing has been specified in the drawings to limit deflections and resist
tampering/vandalism. The upper assembly of the closure (top plate, main beams, joists, and perimeter
channel) are pre-fabricated off-site in a shop to which will have the necessary equipment to efficiently
fabricate the closures, while also providing a higher degree of quality control (details below). The overall
structural arrangement provides redundancy (at reduced load rating) to mitigate any potential failure of
structural connections.
The closure concept utilises a series of main HSS beams up to 4 m long which generally span in the short
direction of the openings. Inverted L-section joists span between these beams and are designed to act
compositely with the 6.4 mm plate lid. Where the edge of the closure is perpendicular to the main beams, a
channel has been installed. This channel provides robust edge stiffness as well as a location for the side walls
to weld to. Where required, the downslope perimeter channel is skewed to accommodate the installation
angle of the lid, and to maintain the vertical alignment of the tee posts.

Covers
Rolled tee sections are specified as both vertical post supports and horizontal struts. The tees provide two
flanges for the side plates to mount behind, and provide stiffness in three directions. The tees are welded to
the mounting plates on the structure (both vertical orientations and horizontal orientations) to provide an
increased stiffness over bolted connections. The structure also utilises oversized baseplates and mounting
plates which provide the contractor with some flexibility when laying out the posts and field-cutting the
required post lengths and compound angles.
Side skirting, which consists of pre-cut stainless steel plate, is a requirement to prevent people from crawling
under the caps and into the openings. In addition to preventing access to the hole for the sides, the side walls
are welded to the tee posts to provide localised lateral restraint to the posts, as well as overall lateral load
resistance to the structure. A seal weld is provided between the side walls and the perimeter members of
the lid to prevent water from draining along the inside surface of the side walls. The skirting is field-fitted
according to guidance developed to balance a maximum permissible gap opening between the bedrock and
the closure with the costs associated with field cutting and welding.
An L-section clip is bolted to the bedrock and field welded to the side wall at a specified spacing to provide
an adequate level of stiffness while reducing field installation times and fabrication costs.
The skirting is comprised of 4.8 mm (3/16 inch) sheet if installed horizontally and subject to an accumulation
of overburden, or 3.2 mm (1/8 inch) if installed vertically. In locations where lateral earth pressure may exist,
the side plates are strengthened to resists bending.
Anchor bolts are not subject to significant lateral loading or uplift. Therefore, anchor bolt diameters and
embedment lengths are chosen to provide an acceptable level of structural integrity while requiring a lower
installation effort. The bolt embedment length is also chosen to extend through the bedrock surface layer,
which may be fractured. The drawings specify that the contractor should remove any loose surface rock prior
to installation of the anchor bolts.
Epoxy grouting of the anchor bolts provides lateral load resistance, prevents general movement of the
structure, and resists vandalism. While the epoxy resists exposures to typical adverse ground conditions, the
resistance to the heat generated during a forest fire is uncertain, and as such the closures are not designed
to rely upon withdrawal resistance of the anchor bolts. Therefore, should the epoxy grout fail the structure
should remain in position.
Fabrication is performed at an off-site shop that is certified by the Canadian Welding Bureau. This ensures
that the appropriate quality control measures are available and eliminates shipment of extra material to the
project site. Fabrication inspections are performed by the project engineer and third-party testing is
performed by a registered non-destructive testing firm. This testing includes visual assessment of the welds
as well as Liquid Dye Penetration testing on a selection of structural welds. Welder qualifications and mill
certificates are also reviewed as part of the quality control program. Robust quality control in Saskatoon is
essential as once the covers arrive on site the cost of performing these activities escalates.
The closure designs are limited to widths of 4.0 m to satisfy transportation width restrictions. In cases where
openings exceed 4.0 m, the covers are fabricated in two sections and field welded prior to installation. The
closures weigh between 730 kg (1,600 lb) to 2,800 kg (6,200 lb).
Once the closures arrive on site, they are positioned over the opening with an excavator using pre-installed
and balanced lifting lugs (Figure 6). Following positioning of the cover, the anchor plate holes are drilled in
the bedrock, and the stainless steel anchor rods are installed. In some cases small quantities of friable rock

are removed to achieve better anchorage or the design slope of the closure. The supporting columns are
then cut to length and welded to the cap and anchor plate. Once the epoxy cement has cured, the anchors
are tightened and rendered tamperproof. Following installation and compliance inspection, the stainless
steel cover is pickled to improve its corrosion resistance.
Installation instructions are provided as an approved method for the contractor to refer to while installing
the closures. Engineering support is also available should the need to deviate from the proposed installation
method arise.
Since 2013, two shafts (e.g., Figure 7) and five raises have been successfully closed using the stringent design
and quality assurance process described above.
A cost analysis was performed to compare the first- and second-generation cover systems. It is challenging
to compare the installed cost of the two cover systems as each cover is unique and all covers are of different

Covers
sizes and shapes. For this reason, normalised installed costs per square meters of cover were calculated for
individual cover design. The following equation was used:
∑ 𝐶𝑜𝑠𝑡𝑖𝑗
𝑁𝐼𝐶 =
∑ 𝐴𝑗
Where:
NIC = normalised installed cost (2014 $CAD/m2)
∑Costij = Sum of paid costs for individual cost category i for covers of generation j (2014 $CAD)
∑Aj = Sum of area covered by covers of generation j (m2)
Cost data readily available to the authors (2009 to present) were compiled for five first-generation covers
(five shafts) and seven second-generation covers (two shafts and five raises). Raises are typically smaller than
shafts. All of the first-generation covers were pooled together in this analysis, although this design has
evolved between 2009 and 2012 (as discussed in the Introduction), while the second-generation covers are
all of a very similar design. The cost analysis is nevertheless a valid and informative comparison. Costs were
converted to 2014 Canadian dollars using the Canadian non-residential construction price index for industrial
structures (Statistics Canada, 2014) to limit the effect the change in costs between 2009 and 2014 may have
on the analysis. The cost estimate results are presented as normalised installed costs broken down by cost
categories in Figure 8. The following cost categories are included in the cost estimate analysis, and costs are
aggregated as follows (as shown in Figure 8): (1) “Site preparation” for design; (2) “Design and inspection”
includes survey of the site for design, engineering design, and final inspection; (3) “Fabrication” includes the
cost of material; (4) “Transport to site” of the cover; and (5) “Installation” of the cover at the site.
$4,000.00
NIC (2014 $CAD/m²)
$3,500.00
$3,000.00
Installation
$2,500.00
Transport to site
$2,000.00
Fabrication
$1,500.00 Design and inspection
Site preparation
$1,000.00
$500.00
$-
First Second
Our analysis shows a 7% decrease in the overall normalised cost between the first- and second-generation
cover designs. A decrease in fabrication costs, was counterbalanced with increases in overall engineering
costs (design, inspection, etc.) and installation costs. The decrease in fabrication costs were due to reduced
size of covers and amount of material used, and lower cost material used (304 stainless steel in second-
generation covers vs. 316 stainless steel in first-generation covers). As discussed above, the 304 steel

provides comparable corrosion resistance to 314 stainless steel for the current and foreseeable environment
in northern Saskatchewan at lower cost than 316 stainless steel.
Decrease in fabrication costs were related to the design approach that aimed to reduce the amount of
material, and simplifying installation, while providing a safe closure meeting project needs. Low design cost
in the first-generation covers may be due to the simple, yet effective, design approach where the contractor
and engineer collaborated closely in designing the closures. Increasing experience and refinements in design
are expected to reduce fabrication cost for future closures, for example by reducing the number of compound
angles and reducing the thickness of baseplates.
Although the difference in normalised installed cost (per m2 of cover) discussed above is not substantial, it is
interesting to note that seven second-generation cover a total of 124 m2, while five first-generation cover a
total of 160 m2. This exemplifies how the second-generation design process optimises the size of covers and
results in material (and cost) saving. Furthermore, the average cost of typical covers for shafts went from
$83,500 (n=4) for first-generation covers to $77,000 (n=2) for second-generation covers, demonstrating a
reduction in cost with the new design. It should also be mentioned that the new approach also provides a
higher level of quality control documentation during the design and installation process. Table 2 summarises
the benefits and drawbacks of costs-saving initiatives which were taken while designing the second-
generation closures:
Initiative Benefits Drawbacks

Composite design Reduction in material costs Marginal increase in required welds
Reduction in side wall height Reduced material costs None
Increase in side wall gap tolerance Reduction in field fitting costs Aesthetics
Reduction in compound angles Reduced fabrication costs None
Topographic survey Reduction in lid footprint Survey costs
Second pilot car not required during
Design 2-piece units Requires assembly on site
shipping
Use hot rolled stainless steel
Reduced fabrication costs Requires pre-ordering
structural shapes
Less expensive than 316 grade while
Specified AISI304 grades stainless
maintaining adequate corrosion None
steel
resistance.
Requires intermediate supports to
Reduced side wall thickness Reduced material costs
limit deflections
Reduced anchor bolt embedment Reduced material costs and
None
length installation time
Although the stainless steel covers require an exemption from the Saskatchewan Mines Regulations prior to
implementation, they are now accepted by regulators as a safe, durable, and long-term mine closure method
for remote sites in northern Saskatchewan, and may be applicable to other jurisdictions. This closure option
is a suitable alternative to legislated concrete bulkheads for closure of openings to the underground to
provide protection to the public and the natural environment from the hazard of legacy mines.
The basic strength properties, relatively simple handling and manufacturing practices, and the inherent
corrosion resistance makes stainless steel suitable for long-term closures in a wide range of climatic
conditions, potentially worldwide.

Covers
Future efforts will continue to reduce material use, fabrication and installation costs, as well as provide
closure solutions for increasingly remote sites. There are also on-going design efforts for a component-based
closure method, using pre-manufactured parts which will be field adjustable to suit any shaped mine opening
in competent or fractured bedrock.
Future challenges will be presented as more remote sites are remediated. In this case, helicopter or boat
access may be the only viable way of reaching these sites. No heavy equipment will be available and the
closures will have to be installed by hand using the aforementioned component system. However, there is a
possibility of installing a preassembled lightweight version of the closures by helicopter.
Cost savings will be realised by reducing the complexity of the closures. The possibility of using galvanised
steel, which is much less expensive than stainless steel, will also be evaluated.
With a design criterion of 200 years, geotechnical and environmental changes (e.g., failure of the rock mass
of the openings or crown pillars) that are difficult to predict or cannot be forecasted will, at some point, limit
the life of the closures before corrosion does.
We thank Saskatchewan Ministry of Economy for the opportunity to participate in the remediation of legacy
uranium mines in northern Saskatchewan. We acknowledge Uranium City Contracting Ltd. (UCC Ltd.) and
Kova Engineering (Saskatchewan) Ltd. for their contributions in developing the idea and designing the first-
generation stainless steel covers, which have contributed to improving public safety at the legacy mine sites.
Statistics Canada. (2014) Table 327-0043, price index of non-residential construction, by class of structure, viewed http://www5.
statcan.gc.ca/cansim/a26?lang=eng&retrLang=eng&id=3270043&paSer=&pattern=&stByVal=1&p1=1&p2=37&tabMode=d
ataTable&csid
Garand K. (2015) Email communication between Krista Garand, Ryerson, and Jonathan Lambert, McElhanney, on the price of 304
and 316 stainless steel.


G. Meiers O’Kane Consultants Inc., Canada

M. O’Kane O’Kane Consultants Inc., Canada
D. Mayich David Mayich Consulting, Canada
P. Weber O’Kane Consultants (NZ) Ltd., New Zealand
C. Bradley O’Kane Consultants Inc., Canada
J. Shea Public Works and Government Services Canada, Canada
Closure and reclamation of waste rock piles using engineered cover systems and water treatment of seepage
is a common technique for management of acid rock drainage/metal leaching (ARD/ML) to mitigate adverse
impacts to the receiving environment. Linking cover system performance (i.e. net percolation and/or oxygen
ingress) to impacts to the receiving environment provides a rational basis for cover system design criteria.
Two general models are used in the mining industry to predict long-term seepage and closure costs. One
model assumes that loading of contaminants to the environment will remain unchanged under reduced flux
rates (net percolation) where contaminant concentrations increase proportionally as a function of decreasing
flow; the other assumes that a reduction in the flux rate will result in decreased contaminant loads for
constant contaminant concentrations in flow. Both models will have a transition point at which reduced
loading to the receiving environment will occur as the flux decreases.
Enterprise Cape Breton Corporation implemented a program for the closure of historic coal mines located
near Sydney, Nova Scotia, Canada, with Public Works and Government Services Canada providing project
management. The Victoria Junction waste rock pile was reclaimed with an engineered cover system
comprising a 60 mil HDPE geomembrane, a granular drainage layer and an overlying growth medium. It was
estimated following closure that active treatment of seepage waters impacted by ARD/ML would be required
for ~20 years before passive treatment systems could be established; however, this has occurred within seven
years with significant cost savings realised. Reclamation of the site reduced the risk of potential loading under
various failure scenarios. An acidity mass balance was used to provide an understanding of past (uncovered
waste rock pile and active treatment), current (covered waste rock pile and passive treatment) and long-term
(100 years) loading under progressive changes to water collection and treatment activities. The acidity mass
balance will serve to inform management decisions in ongoing closure planning. The case study presented
here demonstrates that it is possible to eliminate the need for active treatment of seepage for legacy sites.
Challenges exist in developing closure costs for mine waste storage facilities, primarily owing to uncertainty
in predicting long-term pore-water quality and the subsequent acidity load associated with a range of cover
system performance (i.e. oxygen and water ingress) scenarios. Two conceptual models are used in the mining
industry to estimate acidity generation for managing acid rock drainage/metal leaching (ARD/ML). In model
1, ARD/ML load increases linearly with flow. Concentration in model 1 is assumed to be constant, so the
increase in load is produced by the increase in flow. Conversely, the ARD/ML load in model 2 is assumed to
be constant with changes in flow. This is produced by high concentrations at low flows and low
concentrations at high flows due to the effects of dilution. Both models are depicted graphically in Figure 1.

Closure of legacy waste rock piles: can we achieve passive treatment to manage residual seepage in the short term? G. Meiers et al.
High
Model 2
Acidity Load
Low Water Ingress High
While neither model could exist in its perfect form in practice, given the range of site-specific controls such
as flow, residence time and other physiochemical factors, it is likely that a combination of the two models
will exist. Assuming physiochemical factors remain constant, it is anticipated that the concentration change
in model 2 will be more prolific during periods of high flow, reducing residence time and allowing for dilution.
Conversely, the constant concentration of model 1 will be more prolific under low rates of water ingress,
allowing adequate residence time for pore-water to equilibrate with the in situ conditions. O’Kane (2014)
proposed an alternate model for estimating acidity load generation in waste that takes into consideration
changes in physiochemical factors in response to changes in oxygen and water ingress. Given the uncertainty
in predicting long-term pore-water quality under various rates of water and oxygen ingress, challenges exist
in understanding the benefits of implementing a cover system compared to collection and treatment in
perpetuity.
While there are challenges in predicting pore-water quality, observed loadings to the downstream receptor
mark the culmination of progressive changes to mining and reclamation, and water collection and treatment.
It is essential that an understanding of observed loadings to downstream receptors be demonstrated prior
to developing predictions of long-term performance. Through the development of a conceptual model, this
paper has made use of an acidity mass balance to provide an understanding of past (uncovered and active
treatment), current (covered and passive treatment) and long-term loading under progressive changes to
water collection and treatment activities. Constant concentration (i.e. model 1) was observed for past and
current conditions. Constant concentration was used in the acidity mass balance to predict long-term (100
years) loading given the observed internal temperature and low water and oxygen ingress to the reclaimed
waste rock pile (WRP).
The conceptual model and mass balance provides an understanding for the site and, should it be required,
serve to inform on numerical modelling completed within subsequent components of the project. The
reclaimed Victoria Junction WRP is a unique case study because of its pilot scale, which allows a transition
from active to passive treatment to occur in a relatively short time period following cover system placement
compared to what would be expected for larger waste facilities typical of the mining industry.

Covers
Cape Breton Development Corporation (CBDC) was established as a Crown corporation in 1967 in order to
reorganise and rehabilitate the coal industry on Cape Breton Island, Nova Scotia, Canada. In 2009, CBDC was
dissolved, and its assets and liabilities were transferred to Enterprise Cape Breton Corporation (ECBC), a
federal Crown corporation. Under the transfer arrangement, ECBC acquired stewardship obligations
stemming from CBDC’s past operations, including land holdings and environmental remediation.
Properties covered under the environmental remediation program stem from mining operations that began
in 1685 and include more than 50 underground mines, which produced over 500 million tonnes of coal. The
history of coal mining in the Sydney coal fields included 720 individual parcels of land on which there were
95 coal related operations covering more than 1,000 km2. Some of the properties required remediation of
WRPs produced from the mining operations.
The reclaimed Victoria Junction WRP is located on the site of a historic coal preparation plant approximately
3 km east of Sydney and has a footprint of approximately 26 ha and a height of 40 m. Public Works and
Government Services Canada, under ECBC, provided project management for the remediation program. In
2014, ECBC was dissolved into Public Works and Government Services Canada, which took ownership of the
site.
The coal preparation plant was commissioned in 1976 and produced 750 short tons per hour (stph) of
metallurgical and thermal grade coal, which was upgraded to 1,000 stph in 1981/82. The coal preparation
plant operated until the Phalen Colliery closed in 2000, and then a blending facility operated on the site until
2001. The concern after decommissioning was ARD/ML into Northwest Brook, which flows from Grand Lake,
located 700 m south of the WRP, to Lingan Bay/the Atlantic Ocean, approximately 4 km north of the WRP.
Northwest Brook flows around the east side of the WRP, where it branches into a wetland area and reforms
into a channel before monitoring point 2016 (MP2016). Figure 2 is a plan view that shows the reclaimed
Victoria Junction site during the period of active treatment.
Post 2006 WRP Closure
MP2016 Northwest Brook
Leachate
Collection System
Pump and Treat Wells Wet Well
Smith’s Brook
Polishing Lower
Pond Surge Pond
Victoria Junction WRP
Upper
Surge Pond Passive
Treatment
2013
Active Treatment
The facility washed up to 4 million tonnes of raw coal per year, of which 15–20% was placed into the WRP
and 3% into fine tailings ponds. Coal tailings ponds (CTP) 1 and 2 were located south of the main WRP. During

operations, tailings within these ponds were relocated on an alternating basis and placed within the
northwest footprint of the WRP. These tailings ponds were replaced over a period of three years beginning
in 1980 by CTPs 3, 4 and 5, located within the central and eastern portion of the WRP, which were eventually
covered with waste rock.
In 1987 a 1 m thick bentonite wall was constructed along the north and east toe of the WRP to reduce
ARD/ML moving through shallow overburden into the wetland. A toe-drain collected upwelling seeps that
were treated through the active treatment plant. The toe-drain was upgraded to the leachate collection
system in 2006, during construction of the cover system.
A basic lime slurry addition water treatment plant was located adjacent to the WRP for the collection and
treatment of ARD/ML-affected runoff. This was replaced in 1994 with another lime slurry water treatment
plant that included various surge, settling and polishing ponds before discharge into Smiths Brook east of the
pile. Regulations do not apply for discharge from a coal preparation site on Cape Breton Island; however,
voluntary compliance with metal mining liquid effluent regulations was accepted to meet suspended solids,
pH and various dissolved metals guidelines. In 2003, six pump-and-treat wells were installed into bedrock to
the north of the WRP to intercept deeper impacted groundwater. A wet well was installed north of the lower
surge pond to intercept ARD/ML-impacted water percolating through the base/lining of the lower surge
pond. The lower surge pond received water from the pump-and-treat wells, the leachate collection system
and the wet well through a series of pumps.
An engineered cover system was implemented as part of the closure plan for the WRP in 2006. The cover
system was designed based on several closure objectives (AMEC, 2005a):
 providing a measurable positive effect on the environment compared to the impact pre-closure;
 minimising the impact to wetlands connected through Northwest Brook; and
 site aesthetics in accordance with industry standards and public expectations.
The cover system comprises a till growth medium layer placed over a granular drainage layer (GRDL) and a
60 mil HDPE geomembrane, as shown in Figure 3. The cover system was designed to limit the ingress of
meteoric water and oxygen, thus limiting the transport of stored acidity and generation of potential acidity.

Covers
In 2013, the active water treatment plant was decommissioned and a passive water treatment system
constructed in the lower surge pond to the east of the WRP, as illustrated in Figure 2. In reality, the passive
treatment system is not “passive,” as it is recharged with lime and polymer, and ARD/ML-impacted water is
actively pumped from the leachate collection system into the treatment system. However, for the purposes
of this paper, the system is considered passive given that the lime slurry plant was decommissioned, limited
lime/polymer is added to the natural system and annual operational costs decreased from approximately C$
200,000 to C$ 1,800.
A conceptual model was developed for the site to provide an understanding of past and current conditions
to provide information about long-term impacts from the WRP. The conceptual model included physical, flow
and geochemical components.
The Victoria Junction WRP contains an estimated 10 million tonnes of potentially acid-generating waste rock.
Water dynamics at the site include meteoric contributions as well as a groundwater component. The
groundwater table in the area is relatively shallow (1–5 mbgs). Tailings ponds within the WRP created strong
vertical hydraulic gradients, forcing the ARD/ML plume deep into the underlying bedrock. Migration of the
plume is northward with groundwater flow and then upward to the surface of Northwest Brook because of
higher water pressures in the underlying bedrock compared to the overlying till. As a result, loading from the
pile to the receiving environment can be quantified north of the pile at MP2016.
The saturated hydraulic conductivity (ksat) of the waste rock varies considerably, from 10-6 to 10-1 cm/s (JWEL,
2002), owing to the construction techniques used and heterogeneity in the material. While there is a
considerable range in the ksat of the waste rock, the primarily fine-textured material in the pile is in the range
of 10-6 to 10-4 cm/s. The tailings buried in the lower portion of the WRP are homogeneous with a ksat in the
range of 10-6 to 10-5 cm/s (JWEL, 2002), and act as a “lower” permeable layer. Through the completion of a
site water balance, AMEC (2005b) characterised water ingress through the bare waste rock at approximately
250–450 mm/year for the side slopes plateau, respectively. These rates of ingress are two orders of
magnitude larger than what would reasonably be expected for the engineered cover system.
Water levels in monitoring wells prior to placement of the cover system indicate the presence of perched
water within the WRP, as illustrated in Figure 4. The high pre-closure rate of water ingress and low hydraulic
conductivity at the base of the WRP are the main contributions to perched water within the WRP. An
observed decrease in water levels occurred in the pile following installation of the cover system and
associated reduction in water ingress. Line 1 of Figure 4 was generated from water levels observed prior to
placement of the cover system and shows water perched to an elevation of approximately 33 m,
approximately 10 m above the base of the WRP. Line 2 represents the current elevation of water levels, a
decrease of approximately 5 m since 2006. The predicted water table elevation at cessation of drain-down
was determined from water elevations up- and downstream of the WRP and wells within the footprint of the
pile screened in underlying till. The water elevations for wells completed in the till have not changed over
time and are reflective of Line 3, indicating that the final water table will likely mound within the base of the
pile at approximately 3 m and diminish to near zero at the toe drain/leachate collection system.

60
50 WASTE ROCK
40
ELEVATION (m)
TOE DRAIN 1
30 2
TAILINGS 3 TAILINGS TOE DRAIN
PEAT
20 TILL
SANDSTONE
10
There are two components to drain-down for the Victoria Junction WRP: saturated-unsaturated drain-down,
which will contribute greater loading over a shorter period; and unsaturated drain-down, which will occur
over a longer time frame and generate a lower loading. Figure 5 shows water levels (i.e. saturated-
unsaturated drain-down) in two of the monitoring wells that were completed in the tailings layer one year
prior to placement of the cover system. The decrease in water levels illustrates a more rapid drain-down
shortly after placement of the cover system, followed by a gradual decrease. The trend in the rate of drain-
down was extrapolated to estimate the cessation of saturated-unsaturated drain-down, as shown in Figure
5. It is estimated that the drain-down of water perched within the pile will continue for another 10–20 years,
at which time the unsaturated drain-down of water will result in a much lower rate of basal seepage. In terms
of unsaturated drain-down, under hydrostatic conditions, the maximum pressure head that could be
achieved in the uppermost portion of the WRP is 300 kPa. It is assumed that much of the unsaturated drain-
down in the upper reaches of the pile has already occurred given the waste rock water retention curve and
in situ matric suction. The suction sensors provide matric suction in the range of approximately 100 kPa.
32 1200
Lime
VJCP-05MW-03
30 1000
VJCP-05MW-09
28 800
Lime (Tonnes)
Elevation (m)
Estimated Time to
26 Saturated-Unsaturated 600
Drain-Down
24 400
Water Table
22 200
20 0
2002
2004
2006
2008
2010
2012
2014
2016
2018
2020
2022
2024
2026
2028
2030
2032
2034
2036
2038

Covers
Annual lime consumption for the active treatment plant from 2002 to 2013 and the passive treatment system
from 2013 is also presented in Figure 5. During the active treatment phase, the lime slurry feed was set to
achieve a pH of 8.5. There are two important trends worth highlighting in Figure 5, with the first being that
lime consumption is highly variable prior to placement of the cover system. The variability in lime
consumption is attributed to runoff and basal seepage being highly susceptible to atmospheric forcing (i.e.
rainfall intensity and volume). Subsequent installation of a cover system attenuated peak discharge and
reduced the burden on the water collection and treatment facility. In addition, installation of the pump-and-
treat wells in 2003 added to the acidity load for treatment.
The second and equally important trend in lime consumption is the constant decrease following placement
of the cover system. The rate of drain-down is currently on the order of 75 mm/year and currently contributes
the bulk of loading from the pile. Once drain-down has ceased, loading to the receiving environment will be
significantly reduced and mainly consist of net percolation and the residual effects of groundwater
mounding.
Net percolation, or leakage through the geomembrane, was estimated for a range of defects and heads of
water that could form above the layer. Water will flow through the geomembrane defect, then laterally some
distance between the geomembrane and underlying material and then into the underlying material. The
distance water flows between the geomembrane and underlying material is dependent on the contact
between the two layers. Analytical solutions are available for “good” and “poor” contact conditions (Giroud
et al., 1992). Leakage rates estimated for the Victoria Junction WRP are based on a poor contact given that
bedding sand was placed between the waste rock and geomembrane providing a conduit to flow. Net
percolation was estimated to be in the range of 2 to 5 mm/year.
While the aforementioned low net percolation and trend in lime consumption support lower loadings,
uncertainty exists in the relative benefit realised from the cover system given the culmination of factors
affecting loading to the receiving environment. For example, as identified in Figure 5, installation of the
pump-and-treat wells in 2003 would have increased loading to the active water treatment facility, while
decommissioning of the pump-and-treat and wet well in 2010 would have had the opposite effect. In order
to better understand long-term impacts to the receiving environment, acidity mass balances were developed
to quantify changes in loading associated with the installation of the cover system, water mounding, drain-
down and progressive changes to the water collection and treatment facility; these are discussed in the
following section.
An estimate of stored and potential acidity in the Victoria Junction WRP was developed based on a
preliminary geochemical assessment. It is assumed that all sulphate is present as melanterite (FeSO4) and
that two moles of H+ will be produced per mole of sulphur. This is a conservative assumption and assumes a
maximum stored acidity load. As noted above, there are two distinct materials in the WRP: waste rock and
tailings. The average total sulphur contents of the waste rock and tailings are 2.4 and 1.3 wt%, respectively,
and the acid potential can be calculated by:
𝐴𝑃 (𝑘𝑔 𝐶𝑎𝐶𝑂3/𝑡𝑜𝑛𝑛𝑒) = 𝑤𝑡% 𝑇𝑜𝑡𝑎𝑙 𝑆𝑢𝑙𝑓𝑢𝑟 × 31.25 (1)
Where:
AP = the acid potential in CaCO3
31.25 = a conversion factor based on stoichiometry of the neutralisation reaction of FeS2 by
CaCO3.
A conservative approach was assumed with the entire WRP having a sulphur content of 2.4 wt%. This will be
refined following refinement of the WRP biography. The potential and stored acidity for the WRP were
calculated using the acid-generating potential determined by JWEL (1995) and are summarised in Table 1.
The bulk of the acidity in the Victoria Junction WRP is present as potential acidity, which will be generated as

a function of oxygen ingress and then transported to the receiving environment as a function of net
percolation.
Acid-generating
Waste rock Sulphur potential Total acidity Potential acidity Stored acidity
(t) (wt%) (kg CaCO3/t) (kg CaCO3) (kg CaCO3) (kg CaCO3)
10 million 2.4 72.1 750,000 720,907 29,093
Potential and stored acidity were used to support the acidity mass balance and long-term predictions of
loading to the receiving environment.
Loading to the receiving environment has evolved over time and is characterised by three distinct periods:
pre-cover system with active treatment, post-cover system with passive treatment and long-term post-cover
system with passive treatment. An acidity mass balance was developed for each phase, with the pre-cover
mass balance shown in Figure 6. In all phases it is assumed that oxygen ingress is sufficient to generate
potential acidity at a rate greater than the mobilisation of stored acidity by net percolation. Under these
conditions, the generation of potential acidity will contribute to the total stored acidity and eventually be
mobilised with net percolation. The limiting factor on loading to the receiving environment is therefore net
percolation. This is a conservative assumption, as oxygen concentrations decrease with depth and diffusion
across the HDPE membrane is low. It is likely that there is a depth within the WRP at which oxygen
concentrations are insufficient to generate acidity at the mobilisation rate, given that oxygen transport is
limited to diffusion.
The load to the receiving environment pre-cover system consists of groundwater mounding, net percolation
and runoff from the site. There is also an alkalinity contribution from the polishing pond. The load for each
component was estimated based on average calculated acidity concentration and flow. Groundwater
mounding and net percolation used an average acidity concentration of 3000 mg/l and are conservatively
estimated to remain at this concentration until the total acidity in the WRP is depleted. A groundwater flow
rate of 1 × 10-5 cm/s was determined based on JWEL’s (2002) hydrogeological investigation. Mounding was
estimated to extend 3 m into the base of the waste rock (Figure 4). A net percolation rate of 350 mm/year
was used based on estimates from AMEC’s (2005b) closure plan feasibility design, and the cross-section was
considered to be the entire 26 ha footprint of the WRP.
The active treatment system collects flow from site runoff and the wet well, toe drain and pump-and-treat
wells. Active treatment system loads were estimated from JWEL’s (2002) hydrogeologic investigation and
resulted in an average input flow and acidity concentration of 1,500 l/min and 1,000 mg/l, respectively, and
a 5 mg/l output concentration. Input to the active treatment system was proportioned between runoff from
the site and pump/treat, where the pump/treat load consists of the toe drain, wet well and pump-and-treat
wells (AECOM, 2009). The pump/treat load was calculated from average monthly flow and concentration.
Runoff from the site was back-calculated from the pump/treat and active treatment system loads.
Natural groundwater alkalinity is conservatively assumed to be able to reduce acidity by 2 t/year based on
groundwater flow rates and background water quality. Considering all of the acidity contributions and “sinks”
for basal seepage, there is a calculated load of 161 tonnes per year. The observed load at MP2016 based on
historic concentrations (JWEL, 2002) and estimated flows (exp, 2014) is 150 t/year. The overall error of 1%
would suggest that the flow model used to close the acidity mass balance is representative of site conditions.

Covers
Oxygen Flux Net Percolation
656 t/yr
Oxygen 21%, Decreasing With Depth Runoff From Site
132 t/yr
30-40 m High Net Percolation ~350 mm/yr
PAF Waste Rock 788 t/yr
Active Treatment System
19 t/yr
Mounding Pump/Treat 3 t/yr
273 t/yr 160 t/yr 158 t/yr
Net Percolation MP2016
-2 t/yr Calculated = 161 t/yr
Groundwater Observed = 150 t/yr
Error = 1%
Alkalinity
The total acidity generated is 948 t/year, of which runoff is the largest contributor at 656 t/year. The active
treatment system neutralised 785 t/year, while 2 t/year were neutralised by natural alkalinity in
groundwater. In terms of basal seepage, approximately 45% is intercepted by the active treatment system,
with the remaining reporting to groundwater flow, which is key in developing an understanding of loading to
the receiving environment. Alkalinity
a) Oxygen Flux Net Percolation
Till Growth Medium Cover

0.4 m
Layer
0.4 m Granular Drainage layer Runoff From Site
1.52 mm HDPE Geomembrane
Oxygen < 1%, Decreasing With Depth
Low Net Percolation ~4 mm/yr
30-40 m Leachate Collection System
PAF Waste Rock
10 t/yr 0 t/yr
19 t/yr Passive Treament System

Mounding 10 t/yr 0.5 t/yr
59 t/yr
Drain-down 70 t/yr 68 t/yr
2 t/yr MP2016
Net Percolation -2 t/yr Calculated = 69 t/yr
Groundwater Observed = 66 t/yr
Error = 4%
Alkalinity
b) Oxygen Flux Net Percolation
Till Growth Medium Cover

0.4 m
Layer
0.4 m Granular Drainage layer Runoff From Site
1.52 mm HDPE Geomembrane
Oxygen < 1%, Decreasing With Depth
Low Net Percolation ~4 mm/yr
30-40 m PAF Waste Rock Leachate Collection System
5 t/yr 0 t/yr
Passive Treament System
19 t/yr
Mounding 5 t/yr 0.5 t/yr
0 t/yr
Drain-down 16 t/yr 14 t/yr
2 t/yr MP2016
Net Percolation -2 t/yr Calculated = 15 t/yr
Groundwater
Alkalinity

The mass balance substantially changes after installation of the low flux cover system and remediation of the
site, in terms of both the load produced from surface runoff and basal seepage (Figure 7a). Clean runoff from
the site resulted in a reduction of approximately 656 tonnes per year. The total acidity generated from the
site is reduced from 948 t/year to 80 t/year owing to the transition to passive treatment. However, the wet
well and pump-and-treat wells were decommissioned, and the leachate collection system contributes solely
to the passive treatment system. Acidity collected in the leachate collection system would be attributed to
basal seepage and groundwater mounding.
The reduction in net percolation from approximately 350 mm/year to 4 mm/year following placement of the
cover system introduces a drain-down component to the mass balance. As identified in the flow model, drain-
down is currently approximately 75 mm/year. The contribution from groundwater mounding remains
unchanged, given the total stored acidity in this region of the WRP has not been lost. Approximately 13% of
basal seepage is intercepted and treated before being discharged. The total calculated acidity load at MP2016
is 69 t/year, compared to the current observed load of 66 t/year.
While total acidity generated from the site was reduced by approximately 90% (948 to 80 t/year),
comparatively the acidity load at MP2016 decreased by approximately 56%. The pre-cover system (Figure 6)
and current (Figure 7a) mass balances provide context for the observed water quality at MP2016 in that a
proportional decrease in loading was not observed after changes in water collection and treatment that
included acidity loading from the decommissioned wet well and pump-and-treat-wells. Although a more
significant decrease in loadings at MP2016 was not observed, closure plan objectives are still being met.
Using the mass balances and conceptual model for pre-cover system and current conditions, a mass balance
was developed to predict loadings to the receiving environment 100 years post-cover system construction
(Figure 7b). It is estimated that saturated-unsaturated drain-down will be completed in another 10 to 20
years and the bulk of water from unsaturated drain-down will have occurred in 100 years; therefore, the
acidity load will be negligible. Loading due to mounding and groundwater flow will remain, given the total
acidity in this region is expected to be lost over 3,000 years. The total acidity in the remainder of the WRP
with net percolation occurring at a rate of 4 mm/year is estimated to be lost over 240,000 years.
Contributions to the leachate collection system, and, therefore, the passive treatment system, are
anticipated to decrease as drain-down completes. As a result, the load captured by the leachate collection
system was reduced to 5 t/year. The main contribution is anticipated to be from groundwater mounding and
seasonal fluctuations in groundwater elevation. With further development of the conceptual flow model,
loading to the leachate collection system will be adjusted accordingly. The mass balance closed for the 100-
year period would suggest a decrease in acidity to 15 t/year, approximately one order of magnitude from
current conditions.
This provides context for what benefit the passive treatment system currently provides. Based on the mass
balance, it captures approximately 13% of basal seepage, reducing the acidity generated from 80 t/year to
70 t/year. Loadings to the receiving environment without the reduction from the passive treatment system
would still be below pre-cover system loadings and site closure objectives would still be met. The load
captured by the passive treatment system is anticipated to decrease as unsaturated-saturated drain-down
completes. A strategy for decommissioning the passive treatment system may be to maintain its operation
for 10–20 years until unsaturated-saturated drain-down is complete and further improvements to the
wetland are observed.
Although these mass balances were completed during the post-closure phase, it is evident such an analysis
would be beneficial during the initial phases of operation. A mass balance completed within the closure
planning phase could be used to provide information about the cost/benefit of implementing various cover
system designs. For example, in this instance, a moisture store-and-release cover system would have reduced
the demand on the water collection and treatment system through the associated clean runoff. This
highlights the importance of developing a conceptual flow model and acidity mass balance (sources and
sinks) to inform management decisions in mine closure planning.

Covers
Although the acidity mass balances closed with small error, greater confidence in the results can be achieved
by using multiple lines of evidence to support the conceptual model. Basal seepage in the conceptual mass
balances is represented in Figure 8 as load versus flow. In general, the trend closely follows model 1, where
load is proportional to flow. As flow from the WRP is reduced, there is a corresponding decrease in load,
which demonstrates the benefit of installing the cover system from the perspective of loading.
350
Acidity Mass Balance

300
Linear (Acidity Mass Balance)
Pre Cover Sytem
250
200
Load (t/yr)
150
Post Cover Sytem

100
50
100 yrs Post Cover Sytem

0
0 50 100 150 200 250 300 350 400 450 500
Flow (mm/yr)
As discussed earlier, loading from the pile can be quantified at MP2016, and a plot of load versus flow can
be generated. It is anticipated that MP2016 will follow a similar trend to the conceptual model for calculating
long-term acidity (i.e. model 1) given that MP2016 is a downstream focal point for WRP loading. Figure 9
presents load versus flow observed at MP2016 over the 2010–2014 monitoring period. This period is
primarily influenced by basal seepage, as the WRP cover system was installed in 2006 and therefore is
comparable to load versus flow from the conceptual model.

700
MP2016
600
500
Load (kg/d)
400
300
200
100
0
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000
Flow (L/s)
Load versus flow at MP2016 also follows the general trend of model 1; an increase in load is observed with
an increase in flow. There is some spread in the data away from the linear trend line, which is primarily
attributed to changes in surface water inputs prior to MP2016. The flow path for MP2016 is approximately 1
km down-gradient of the WRP, and this distribution is likely introduced as a result of various processes
occurring in Northwest Brook, including background water levels, recharge, ephemeral flow, drought, spring
melt, and biological and other wetland processes.
A considerable amount of time was required to develop the conceptual model for the Victoria Junction WRP.
The conceptual model should be considered the primary step in demonstrating a strong understanding of
the system, as it provides the basis for moving into more advanced numerical modelling. A well-defined
objective for numerical modelling is made clear as part of this process. Without this step, it would be difficult
to validate results from the numerical modelling, likely resulting in greater time requirements.
There are significant initial capital investment and ongoing operation and maintenance costs associated with
operating a water treatment plant. The Victoria Junction water treatment plant began operation in the
1970s, was relocated in 1994 and transitioned to a passive treatment system in 2013. Operating costs for
water treatment at the site from 2002 to present are presented in Table 2.

Covers
Date Water treated Lime cost Polymer Electrical and Cost/1,000 Total cost
(US gal) cost labour costs US gal
2002 216,570,000 C$ 84,420 C$ 16,000 C$ 125,000 1.04 C$ 225,420
2003 231,563,000 C$ 46,920 C$ 16,000 C$ 130,000 0.83 C$ 192,920
2004 163,556,000 C$ 91,580 C$ 8,000 C$ 110,000 1.28 C$ 209,580
2005 201,535,000 C$ 118,650 C$ 7,000 C$ 120,000 1.22 C$ 245,650
2006 229,806,000 C$ 56,700 C$ 7,000 C$ 130,000 0.84 C$ 193,700
2007 120,327,000 C$ 40,320 C$ 4,000 C$ 100,000 1.20 C$ 144,320
2008 155,522,000 C$ 21,600 C$ 2,000 C$ 110,000 0.86 C$ 133,600
2009 123,142,000 C$ 16,200 C$ 1,500 C$ 100,000 0.96 C$ 117,700
2010 67,310,000 C$ 8,100 C$ 700 C$ 80,000 1.32 C$ 88,800
2011 31,137,000 C$ 2,700 C$ 200 C$ 40,000 1.38 C$ 42,900
1 1 1
2012 - C$ 2,700 C$ 200 C$ 40,000 - C$ 42,9001
2013 ~7,000,0002 ~C$ 6002 - ~C$ 1,200 0.27 C$ 1,800
2
2014 ~7,000,000 ~C$ 600 - ~C$ 1,200 0.27 C$ 1,800
1
Estimated based on previous years
2
Estimated based on average flow (AECOM, 2009)
Placement of the cover system in 2006 significantly reduced the ingress of meteoric water to the WRP and
collection of runoff, and subsequently a decrease in treatment costs was realised. Although the total cost of
active treatment began to decrease, there was an increase in cost per unit during 2010 and 2011. The active
water treatment system becomes less efficient at lower volumes, as electric and labour costs remain
relatively high. The transition to the passive treatment system reduces the total cost of treatment and the
per unit cost dramatically. The savings realised ultimately correspond to the significant decrease in electric
and labour costs and decrease in volume of water treated. A decrease in load was observed at MP2016
despite the decrease in water treatment, suggesting that the reduction in load is due to the reduction in flow
(i.e. model 1).
A long-term cost projection was completed for a 100-year period to include two scenarios: (1) the reclaimed
site with cover system and passive water treatment, and (2) uncovered WRP with active water collection and
treatment in perpetuity. Scenario 1 predicts cost based on the actual site reclamation, including placement
of the cover system and the progression from active to passive water treatment. Predicted costs are based
on actual costs incurred during site progression and consider an inflation rate of 2%. Expected costs in
scenario 2, had the cover system not been installed, were developed based on average water treatment costs
over a three-year period prior to placement of the cover system and consider a 2% inflation rate.
As shown in Figure 10, costs in both scenarios increase at the same rate until placement of the cover system
in 2006, when there is a C$ 13 million increase in the cover system scenario. The no cover system scenario
continues to increase at the original rate and exceeds C$ 70 million in 100 years. The Increase in the cost of
the cover system scenario from 2006 to 2013 reflects actual active treatment costs. Following construction
of the passive treatment system, the rate is significantly reduced. Although the cost of the cover system in
2006 is substantial, it is recouped in water collection and treatment costs by 2047, 41 years after
construction. The cumulative 100-year costs for scenarios 1 and 2 are approximately C$ 15M and C$ 70M,
respectively.

$80 1400
Predicted Costs With Cover System
Predicted Costs Without Cover System
$70 1200
Predicted Load With Cover System
Cumulative Treament Cost (Millions)
Predicted Load Without Cover System

$60
1000
$50
Load (t/yr)
800
$40
600
$30
400
$20
$10 200
$0 0
2000 2020 2040 2060 2080 2100 2120
While this analysis focuses on water treatment costs alone, factors such as upgrades and maintenance need
to be addressed in both scenarios. Given the limited mechanical components of the passive treatment
system, maintenance is anticipated to be a small component of ongoing costs. Although the cover system is
not a direct component of the passive treatment system, its functioning has direct implications on load to
the receiving environment.
The active treatment system has a greater reliance on mechanical systems and is expected to require greater
maintenance and upgrades. This system is also greatly influenced by fluctuations in the economy, as it relies
more heavily on inputs of lime and polymer.
Risk of failure must also be considered for each scenario, including the probability of occurrence and
magnitude of the impact. While the probability of a failure may be difficult to quantify, the impact of a
component failure can be developed based on the aforementioned mass balances. In scenario 2, the main
component mitigating load to the receiving environment is the active treatment system. A failure of the
active treatment system could involve the pump/treat system, runoff from the site or the active treatment
system as a whole and result in an additional load of up to 65 tonnes of acidity per month. While a complete
system failure may not be likely, it could have a significant impact, and even a low probability of this risk may
not be acceptable. The active treatment system inherently carries a greater probability of failure than the
passive treatment system, as it has a greater reliance on mechanical systems. The uncovered system is also
susceptible to the effect of peak discharges in response to climatic conditions where storage capacity could
be exceeded.
A complete failure of the passive treatment system under the reclaimed WRP could result in an additional
load of approximately one tonne per month. The impact of this failure is insignificant in comparison to the
active treatment system and impacts to downstream receptors. Potential failure of the cover system would
potentially occur in isolated regions, allowing risk to be effectively managed. The associated load would be
highly attenuated and not result in significant impacts to downstream receptors.

Covers
Through the remediation of Victoria Junction, it has been demonstrated that passive treatment can be
achieved to manage residual seepage in the short term while meeting closure plan objectives. An acidity
mass balance was key in developing an initial understanding of past, current and predicted loading. It was
particularly important for quantifying changes in loading associated with the cumulative effects of net
percolation, water mounding, drain-down and progressive changes to the water collection and treatment
systems. While a proportional improvement in water quality was not observed at the downstream
monitoring location (MP2016) compared to the overall decrease in loading, the acidity mass balance supports
the 56% reduction in acidity loading and is a considerable improvement in water quality. The conceptual
model was used to demonstrate a strong understanding of the site and provides the basis for moving into
more advanced numerical modelling should it be deemed necessary.
While advancements are still required in terms of better predictions of long-term pore water quality, acidity
mass balances were closed for the pre-closure and current reclaimed condition using constant concentration
(i.e. model 1). The acidity mass balance predicted at 100 years suggests that total acidity will decrease by
approximately one order of magnitude from current conditions solely under passive treatment. The acidity
mass balances also highlight that a comprehensive conceptual model is beneficial in developing water
treatment strategies and meeting closure plan objectives. Acknowledging this provides an opportunity to
optimise ARD/ML management in a cost effective manner and inform management decisions. While this was
completed in the closure phase, it is noted that there is significant value in such an analysis during the closure
design phase to evaluate risks among various closure options.
A wealth of challenges exists in determining if and when a particular site can transition from active to passive
water treatment. As demonstrated, a comprehensive conceptual model is fundamental to an effective
closure plan and requires adequate information pertaining to the site, including both technical and high-level
management. The availability of such information will dictate the ability to develop a comprehensive
conceptual model and highlights the need for adequate site study over the long term.
AECOM (2009) Victoria Junction ARD treatment assessment, report to Public Works and Government Services Canada, project no.
106928-LET-VJ-003 4a, January.
AMEC (2005a) Victoria Junction coal preparation plant closure plan final report, report to Public Works and Government Services
Canada, project no.TV411035, March.
AMEC (2005b) Victoria Junction coal preparation plant closure plan feasibility design report, Report to Public Works and Government
Services Canada, project no.TV411034, March.
exp (2014) Victoria Junction site closure monitoring report, report to Enterprise Cape Breton Corporation, project no. SYD-00203109-
A0/2.1, February.
Giroud, J.P., Badu-Tweneboah, K. and Bonaparte, R. (1992) Rate of leakage through a composite liner due to geomembrane defects,
Geotextiles and Geomembranes, Vol. 11(1), pp. 1–28.
JWEL (1995) Hydrogeological and environmental impact study, Victoria Junction coal preparation plant, report to Cape Breton
Development Corporation, project No.9418, April.
JWEL (2002) Hydrogeological investigation, course waste pile, Victoria Junction, NS, report to Cape Breton Development Corporation,
project no.NSD16172, June.
O’Kane, M. (2014) Cover systems … Are we there yet?, 67th Canadian Geotechnical Conference, Regina, SK, Canada. September.


R.N. Humphries Celtic Energy Limited, UK

M. Tibbett University of Reading, UK
The validity and practical use of the “novel ecosystem” concept and model, as applied to the restoration and
rehabilitation of vegetation and ecosystems, is currently being hotly debated amongst environmental
academics and conservationists. The use of the concept and model has been advocated to select alternative
target vegetation and ecosystems to those of the pre-mining situations for rehabilitating closed mines in some
circumstances.
The purpose of this paper is to examine the concept and its practical use in mine closure rehabilitation
planning and evaluation. The underlying basis for the concept and its threshold based model are examined
and tested in terms of the arguments for and against it, as proffered by its advocates and critics, using
historical and recent evidence from the UK and Australia.
The novel ecosystem concept and its model have merit as a theoretical framework for debate but might not
be particularly useful in the mine planning and regulatory context, especially if applied in an uncritical and
uninformed manner. For example, there may be risks of excusing poor mine closure outcomes as “novel” when
they are in fact wholly inadequate. We give examples of where novel ecosystems seem to work and where
the concept seems to fail, and suggest reasons. There is a case for matching more appropriate vegetation and
ecosystem targets with some post-mining conditions in the Australian context, and probably elsewhere.
Potential applications of the “novel ecosystem” concept to mine rehabilitation, in the form currently
construed, may be a potential risk to sustainable ecosystem outcomes if used as an excuse for restoration
failure. Given the lack of clarity and agreement on what constitutes a novel ecosystem, particularly in a post-
mining context, we conclude its adoption as a tool for mine rehabilitation is premature and encourage an
open debate between regulators, miners, and scientists to provide a safe framework for any future application
of the concept.
The “novel ecosystem” concept and the advocacy of its adoption has rapidly become widely known amongst
academics and practitioners in conservation ecology, owing to the publication of numerous articles in the
scientific press, as well as a recent book compiling the thinking behind and examples of the concept (Hobbs
et al., 2013). The concept has inspired and motivated others, particularly in Australia, to promote its
application to mine closure and rehabilitation planning. However, the basis of the concept and its
consequences have recently been heavily criticised (Murcia et al., 2014; Aronson et al., 2014). The purpose
of this paper is to examine the recently publicised concept of novel ecosystems in the restoration and
rehabilitation of vegetation and habitats, and to give a view of the merits of its application in mine closure
planning and rehabilitation.
The use of “novel” and similar terms in respect of ecosystems is not particularly new. Hobbs et al. (2006)
point out that Odum in 1962 had recognised and used the term “synthetic” to describe species assemblages
and their ecosystems that had not existed prior to the intervention or influence of humans. Bradshaw et al.
(1978) coined the term “novel” in 1976 as a label for deliberately concocted species assemblages (with

Does the concept of novel ecosystems have a place in mine closure and rehabilitation? R.N. Humphries and M. Tibbett
particular resilience or developmental traits), or unusual establishment techniques or soil amendments that
might be adopted to enable the restoration of functioning de novo ecosystems where the soil layer had been
entirely lost or severely altered or degraded due to mining activity. Chapin and Starfield (1997) used the term
“novel” to describe a native grassland-steppe vegetation that currently exists as a native North American
habitat, but which may expand northwards in its geographical distribution and replace tundra vegetation and
habitat over the next 150 to 250 years as global air temperatures rise.
Hobbs et al. (2006) and Mascaro et al. (2013) later adopted the terms “novel” and “novelty,” respectively, as
generic terms to describe species combinations not occurring in a currently recognisable biome that have
arisen because of human action. Subsequently the “novel” term was developed into a single unifying concept
relating to the reversibility and irreversibility of ecosystem dynamics and transformations (Hobbs et al.,
2009). Here, the underlying thinking was that historical vegetation and ecosystems had been and were being
transformed (due to human influences) to altered assemblages and ecosystems (referred to as hybrid types),
and in extreme instances to novel types where changes in the abiotic environment were so extreme that it
was unlikely that either the historical or hybrid types could be reasonably expected to be restored. Hobbs et
al. (2009) illustrate this by reference to historical Eucalyptus woodlands in southwestern Australia that
through livestock ranching had become modified (albeit still having representative main-tree components
(i.e. they were a hybrid type), and that could revert to the historical type through exclusion of grazing and
standard conservation measures of introducing lost species by seeding and/or planting. However, when
salinization occurred (through a rising water table caused by the deforestation), reversion was unlikely to be
achieved by the sort of restoration measures deployed to convert the hybrid to the historical type in the
context of our own life-spans. Under these circumstances, a very different and hence novel saline vegetation
and functioning ecosystem had developed.
Groffman et al. (2006) have debated the possibility of the occurrence and useful application of ecological
thresholds (tipping points) in the restoration and management of vegetation and ecosystems – a mechanism
relied upon for novelty by advocates of the current evolution of the novel ecosystem concept. Thresholds
are well established and used in the pollution assessment of sulphur, nitrogen, and acid-rain deposition and
their potential inducement of biotic changes in sensitive vegetation and ecosystems in the British Isles and
Europe (Department of the Environment and Rural Affairs, 2010; Centre for Ecology and Hydrology, 2014).
Proponents of “novel ecosystems” frequently refer to notional thresholds and changes in abiotic conditions
to differentiate in their conceptual model between hybrid and novel ecosystem states. The existence and
characterisation of the thresholds as a determining mechanism is a point of contention between those who
advocate the current concept and those challenging it (Murcia et al., 2014; Hobbs et al., 2014; Aronson et al.,
2014). This has led to a perception that it is being claimed that almost all ecosystems have been altered in
their composition and are therefore likely to be novel or can be treated as being novel for restoration planning
and conservation evaluation purposes. Accordingly, the notion that most terrestrial ecosystems are currently
in a novel state (i.e. requiring irreversible change in abiotic conditions) is seemingly implausible as novel
states (by definition) may only be induced in the most extreme conditions or events such as volcanic
eruptions, desertification, glaciation, landslips, or the built environment such as urbanisation, and mining
where the soil layer has been lost or severely degraded or changed. Hence, those ecosystems that have
undergone or are undergoing biotic change are most likely to be of a hybrid type capable of reversion to the
historical type.
In considering abiotic thresholds inducing novel states, and their irreversibility, the important dimensional
scale of time in the restoration of vegetation and ecosystems (Humphries, 2013b) is seemingly not factored
into the model, even though it is clearly involved and often acknowledged (e.g. Mascaro et al., 2013). It
appears that the proponents of novel ecosystems may be applying the abiotic threshold term to a “current”
and/or very restricted point(s) in time when actions and interventions might be implemented as a
programme of restoration within an individual human’s relatively short life-span. The Puerto Rico example
cited by Mascaro et al. (2013) is a good example of where the partial reversion to native forests following the
abandonment of agricultural land after only 40 years is described as being in a “novel” state owing to its
relatively incomplete species composition, whilst acknowledging structurally they “… greatly resemble the

Ecosystem Reconstruction
historical ecosystem.” Over a longer period of time it might reasonably be observed that, as colonisation by
the less mobile species takes place, the forest composition would become more similar to the historical.
Chapin and Starfield’s (1997) predicted changes in tundra vegetation have been cited as an example of a
novel ecosystem in support of the current concept and model. However, the change is clearly a gradual
process over a considerable period of time (some 150 to 250 years) and this accords with the Gleasonian
Framework said by Mascaro et al. (2013) to be a foundation of the novel concept. Similarly, the late E.M.
Nicholson (p. 277 in Bradshaw et al., 1978) pointed out that following the catastrophic coastal landslip at
Lyme Regis in 1840 it “… looked like a bare quarry for 60 years ...” and by the 1970s it was a mature typical
English ash woodland (sensu Rodwell, 1991a) and highly valued for its wildlife. Over a longer time
perspective, Dimbleby (1978) describes successive European Mesolithic and Bronze Age clearances of forest
in England and Scotland, and the subsequent forest regrowth and re-establishment of the original forest
composition (from the pollen stratigraphy) even though there was evidence of significant loss and
degradation of the soil layer between the periods of afforestation. These examples imply processes of
gradients and cascading constraints, seemingly over periods of time, thereby challenging the fundamental
notion of the existence of “thresholds” as embodied in the novel ecosystem concept.
Bradshaw et al. (1978) point out that the restoration of the original type of vegetation and ecosystem
functioning in mine closure schemes is unlikely (without intervention) in the short to medium term where
there has been a loss of the in situ soil resources. Where the soils have been carefully recovered and reused,
and appropriate landform and water resources and drainage conditions are reconstructed, similar vegetation
and ecosystems are possible. Murcia et al. (2014) in their criticism of the novel ecosystem concept cite the
rehabilitation of jarrah forest in Western Australia after bauxite mining as evidence that restoration of
severely disturbed ecosystems is possible and the outcome is closer to the historical state than being at all
novel. However, the rehabilitation example cited deploys careful landform reconstruction, soil recovery and
replacement, and the introduction of appropriate plant species. In these circumstances, whilst some early
differences in composition and ecosystem functioning might be detected, it would be expected according to
the Hobbs model that the rehabilitation in the absence of major abiotic change(s) would qualify as being a
hybrid, hybrid-historical or historical type according to the degree and importance of the difference, and
capability of reversion over time or with intervention.
In contrast, rehabilitated surface coal mines in the Bowen Basin of central Queensland, where soils are often
mixed with lower occurring lithologies or are only partially or even not at all recovered, the landform is much
altered and the resulting, often sodic or saline, conditions are extreme, resulting in very different composition
and stability of vegetation and ecosystem function to the original pre-mining types (Erskine and Fletcher,
2013). In these circumstances, as predicted by Hobbs et al. (2006, 2009), the initial outcome might be
considered to be of a novel type and of lower intrinsic value relative to the historical landscape and the
requirement for landform evolution and its stability (sensu Hopper, 2009), but not necessarily so in terms of
similar naturally occurring situations or in cases of historic abandoned mining sites where native vegetation
and their associated ecosystems are established.
Doley et al. (2012), in setting out to direct the attention of mine operators to more realistic rehabilitation of
severely altered landscapes and soils, particularly in arid Australian environments, adopted and developed a
version of the “novel-hybrid ecosystem concept.” This is subsequently re-rehearsed by Doley and Audet
(2013, 2014) with Doley and Audet (2014) advocating the adoption of the concept as a vehicle to drive a
change in thinking about mine rehabilitation planning and expectations, in a similar way as Hobbs et al. (2006)
were thinking for guiding restoration ecology in general. Doley and colleagues saw the novel element of the
concept as a means whereby alternative vegetation and ecosystems (that are more appropriate to the post-
mining conditions than the historic pre-mining types) could be adopted by the industry and regulators to
overcome the alleged poor performance of current schemes (Doley et al., 2012). However, in advocating the
adoption of novel vegetation and ecosystems it is not indicated or specified what the novelty might comprise
of in the essential traits of composition, structure, and function in order to guide both practitioners and

regulators in the application of the concept in practice – as would be required by mine closure completion
criteria in Australia (Glenn et al., 2014). It is in this context that Murcia et al. (2014) and Aronson et al. (2014)
contend that the novel ecosystem concept is of no practical use or relevance, and cannot be justified on the
basis of current evidence.
A test of the value of the novel ecosystem concept and its model is to apply them to actual rehabilitation
schemes and to judge the outcome. For example, the Nant Helen surface coal mine began excavation for coal
in the late 1980s and part was restored in the mid-1990s (Humphries et al., 1999; Humphries, 2014). The pre-
mining (historic) vegetation (Table 1) comprised recognisable British native habitat types (Joint Nature
Conservation Committee, 1990) and British plant community types (Rodwell, 1991b, 1992).
Pre-mining Post-mining
Remnant pre- NCC habitat NVC plant NCC habitat NVC plant
mining types* community types* community
vegetation types# types#
Semi-improved B1 U4b B1 U4
acid grassland1
Marshy grassland B5 M23 B5 M23
types2
Acid dwarf shrub D1 H12 D1 H12
heathland3
Wet modified E1.7 M19 and M25 E1.7 M19a and M25
(peat) bog4
*Joint Nature Conservation Committee (1990); #Rodwell, 1991b, 1992; 1
Benyon, 2013; 2, 4 Humphries, 2014; 3Smith, 2012
The excavation for coal has resulted in the almost complete destruction of the acid grassland habitat and
communities and the loss of the supporting soil fabric as in the context set out by Bradshaw et al. (1978).
Here, rehabilitation in the 1990s was achieved by providing a 0.5 to 1.0 m layer of crushed overburden rock
(a Carboniferous sandstone selected as a soil layer substitute) and placed over the Carboniferous mudstone
backfill, and an agricultural seed mixture (Humphries et al., 1999). The predicted outcome in applying the
novel ecosystem concept and model would be a novel type given the significant change in the soil physical
and chemical abiotic conditions (Jones, 2013).
Although the original excavation had resulted in extensive loss of the other three vegetation types, recovery
of some vegetation and/or soil material was possible. For the marshy grassland, soil with its seed bank was
recovered, stored, and reused in the restoration as a topsoil layer over the crushed sandstone soil substitute.
The peat fabric and the intact vegetated layer were recovered and replaced in prepared basins (Humphries,
2000). The predicted outcomes for the rehabilitation of these two vegetation and ecosystem types might be
a hybrid type rather than novel given the provision of the supporting abiotic soil component, with the
possibility of the hybrid types reverting to a historic type with or without intervention (Hobbs et al., 2006,
2009). In contrast, the acid dwarf shrub heathland and associated plant species was re-established on
crushed sandstone rather than the native soil, and in the context of the concept’s abiotic and irreversible
threshold, a novel outcome would be predicted.

The initial agriculturally sourced and sown prototype acid grassland seed mixture included the productive
forage grassland species of Lolium perenne and Poa compressa as a nurse crop for erosion control measures
(Humphries et al., 1999). Almost twenty years on and the sown acid grassland is of recognisable British
habitat and community types (Table 1) according to the published criteria (Joint Nature Conservation
Committee, 1990; Rodwell, 1992) and comparable to the pre-mining type (Benyon, 2013). In this example,
the outcome is not as predicted with the historic type becoming re-established rather than the novel type or
other novel type being maintained as a result of the drastic change in soil conditions and loss of plant species.
This has come about by the invasion and establishment of native plant species from the surrounding area
and propagules brought in by livestock in the grazing of the grassland.
For the re-establishment of the acid dwarf shrub heathland on the overburden material, this involved the
introduction and chequerboard replacement of 1.5 × 1.5m turves rescued from the heathland being lost at
the time, with reliance on the persistence of the introduced plants to facilitate the colonisation of the
remaining bare areas of overburden to recreate the pre-mining heathland habitat and vegetation
composition. In 2012, the heathland was surveyed in detail (Smith, 2012) with the outcome that the pre-
mining habitat and NVC community H12 type (Rodwell, 1991b) had re-established (Table 1). In this example,
the outcome is also not as predicted with the historic type becoming re-established rather than the novel
type or other novel type being maintained as a result of the drastic change in soil conditions. This has come
about by the invasion and establishment of native plant species from the introduced islands of vegetation.
The marshy grassland and modified bog habitat possibly aligns with the hybrid state as a starting point given
the recovery and use of the native soil fabric, and it is not surprising that they achieve a historic community
composition (Humphries, 2014), according to the “novel ecosystem” concept and model (Hobbs et al., 2006,
2009). Similarly, Pywell et al. (2003) have witnessed the same process and outcome for the rehabilitation
sowing of native grasslands on natural soils following arable farming in the UK, as have Humphries et al.
(2014) in the re-establishment of native floodplain grassland (Rodwell, 1992) on a degraded pasture.
Humphries (2013b) points out that there is necessarily a long timescale of decades and even centuries
depending on which vegetation and ecosystem is being considered, before judgement in the case of mine
site rehabilitation is possible given the rate of vegetation and structural development, and their life cycles.
He suggested that for grasslands this might be in the order of 15 years, for dwarf shrub communities 30 to
50 years and for woodland and forest types 150 to 350 years depending on main-tree composition. That for
woody vegetation was illustrated above by E.M. Nicholson’s observation (in Bradshaw et al., 1978) about the
Fraxinus excelsior NVC W8 woodland (Rodwell, 1991a) re-establishment taking in the order of 130 years
following the Lyme Regis landslip. The detailed studies on china clay waste (i.e. an absence of in situ native
soil cover) in the 1970s and 1980s demonstrated that there is a rapid successional change from an initial
novel (pioneer) vegetation and ecosystem comprising Lupinus arboreus or Ulex species – Sarothamnus
scoparius – Calluna vulgaris on the young mineral wastes to recognisable native/naturalised British
vegetation of Betula pendula – Quercus robur – (Rhododendron ponticum) NVC W16 woodland (Rodwell,
1991a) on the oldest wastes over a 100 year period (Humphries and Rowell, 1994). Whilst because of the
presence of the alien rhododendron, a hybrid label might be applied rather than a historical type. Putting the
presence of rhododendron aside, the succession to a recognisable native type suggests that novel types may
only be transitory over time on mine sites and drastically disturbed materials, as witnessed for naturally
occurring situations, with the time lag depending on their composition and structural traits, as pointed out
by Humphries (2013b).
Taking the above findings in the UK example of re-establishing the historic vegetation at the Nant Helen
Surface Mine Site, it would appear that the novelty and hence novel vegetation and ecosystem largely relates,
as set out by Bradshaw et al. (1978) and Humphries (1979), to the initial stage of site treatment and
revegetation methodology used to achieve a particular target land use and ecosystem.

In the context of Hobbs et al. (2006, 2009), the instigation of novel vegetation and ecosystems as long-term
outcomes in the UK is probably restricted in scope and detail. This may be related to the much younger and
more varied landscape of the British Isles (for example, Rudeforth et al., 1984) than those ancient landscapes
and soils in Australia (McKenzie et al., 2004; Hopper, 2009). Novelty and hence novel ecosystems may largely
be equitable in the UK with alien species and their vegetation assemblages. However, under current
European and UK legislation and policy provisions, the introduction of novelty will certainly exclude alien
plant species and their assemblages. The importation and establishment of foreign species as trialled in the
UK in the 1970s (Jobling and Stevens, 1980) would not now be allowed.
There is a long tradition in the UK of recognising man-made landscapes and soils, and their habitats (Joint
Nature Conservation Committee, 1990; Blackstock et al., 2010) and vegetation communities (Rodwell, 2000)
as legitimate entities in themselves and ones which often have high nature conservation value, besides the
role they have in effective mine and derelict land rehabilitation (Ratcliffe, 1974; Gemmell, 1977; Humphries
and Elkington, 1980; Johnson, 1978; Bradshaw and Chadwick, 1980). Mine-like habitat is readily recognised
and not usually regarded as a novelty, for example, as “artificial exposures and waste tips,” sub-type “spoil”
or “quarry” with their mature vegetation described as generic types such as heath, lichen/bryophyte heath,
dwarf shrub heath, scrub, scattered trees (Joint Nature Conservation Committee, 1990). The vegetation of
open habitats associated with the artificial exposures are also recognised (Rodwell, 2000), for example NVC
OV37 Festuca ovina – Minuartia verna – Cladonia spp sub-community is associated with heavy metal spoils.
Blackstock et al. (2010) recognise slate and other quarry and mining wastes, and their vegetation, as part of
the habitat and ecosystem fabric of Wales.
In the UK acceptable long-term novel mining related outcomes are likely to be very specific and relate to
situations without natural soils or where soil development is expected to be particularly slow, such as rock
dumps, toxic spoils, and tailings dams. For example, the stabilisation of toxic mine wastes as spoil heaps and
tailings dams with tolerant grass populations (Smith and Bradshaw, 1972; Johnson, 1980; Chadwick and
Bradshaw, 1980) has created grasslands of botanical importance alongside their natural counterparts
(Rodwell, 2000). Plant communities established for phyto-remediation (Dobson et al., 1997) would be
another specific situation. Recently, it has been recognised that coal mine overburden dumps provide
valuable habitat for the conservation of endangered lower plant “bio-crust” communities in South Wales
(Humphries, 2013a).
The landforms and substrates, and hence the habitats created by mining activities, may be foreign to the
Bowen Basin locality, but are unlikely to be unique in the wider Australian geography where it is more likely
vegetation and habitat (ecosystem) analogues are to be found. In Australia, broad descriptions as to types
are published as the Major Vegetation Groups (MVG) inventory (Department of the Environment and Water
Resources, 2007) and more detailed descriptions on a State by State basis (e.g. Regional Ecosystems
Descriptions, Queensland Government, 2014). On a national MVG basis, natural types of extreme habitat
(similar to those occurring at some mine sites on closure) are characterised as “Other Cover Types,” sub-type
“Naturally bare – sand, rock, clay-pan, mudflat.” At the Queensland State level, comparable natural extreme
vegetation and ecosystems appear to go unrecorded.
Given this, in relation to post-mining landscapes, perhaps the novel ecosystem based paradigm referred to
by Doley et al. (2012) is not so much about the detailing of the vegetation composition and ecosystem
function per se, but alternative habitat and vegetation types being simply “out-of-place” in the Australian
context. The “out of place” being the association with other locations such as arid landscapes elsewhere in
the Australia or the State, but not locally recognised or documented. It is in this context that there is, for
example, scope for exploring novelty (in the current time frame) in the form of adopting as mine closure and
rehabilitation outcomes the arid/semi-arid assemblages of plant assemblages which have been successful on
mine tailings and rock dumps elsewhere (Tongway and Hindley, 2003; Mulligan et al., 2006). This might be
extended to include the recently recognised important dryland biocrusts (Read et al., 2014).

The “novel ecosystem” concept as advocated by Hobbs et al. (2013) is now hotly debated as to why it is
needed and what use it is (Murcia et al., 2014; Hobbs et al., 2014; Aronson et al., 2014). As is only too clear
from the above discussion, and as Murcia et al. (2014) and Aronson et al. (2014) point out, there is much
confusion in the understanding and application of the “novel ecosystems concept” and its threshold based
model. Initially and traditionally, novelty in vegetation and ecosystems was equitable with alien species and
vegetation, such as described by Ewel (2013). This is a concept that is easily and universally understood.
Proving the existence of abiotic barrier(s), as a general mechanism impeding the reversion of novel to historic
types, is likely to be central in support of the case for the novel ecosystem concept and the model proposed.
The concept will need to factor in time as a key dimension, which may vary from decades to centuries or
longer in the case of sodic, saline, metal and other contaminated substrates, and the evolution and eventual
stabilisation of appropriate landforms. The Lyme Regis landslip and the china clay sand waste examples, cited
earlier, suggest the absence of irreversible thresholds impeding transformation from novel to historic types
over mere decades, making the notion of obligatory abiotic thresholds apparently redundant.
In the test example of re-establishment of the historic acid grassland and acid dwarf shrub heath types, it
might be counter argued that the sandstone substitute is locally a precursor of acidic soils, and all that is
needed is time for the establishment of plant nutrient accumulation and cycling for these vegetation and
ecosystems to become sufficiently established. If this interpretation were to be correct, then the loss of the
native soil and substitution of the parent-like geological material would constitute less of an abiotic change
than envisaged and more of a biotic change. This suggests the initial rehabilitation state to be of a hybrid
rather than a novel type. Less surprisingly, had the restoration deployed Carboniferous mudstone or
limestone material as the soil substitute that occurs within the geological profile and locally, then the
outcome would most certainly have been novel within the context of the Nant Helen Mine as the resulting
mesotrophic and calcareous grasslands, and their shrub counterparts, would have been out of place and not
of the historic types, although they occur in the wider landscape around the site. In this context, the resulting
vegetation and ecosystem types associated with the neutral and calcareous soil materials would not have
reverted or would have easily been transformed to the historic acidic types without reverting to the measure
of soil removal as described by Ewel (2013). This scenario is synonymous with novel salinization example
cited by Hobbs et al. (2009) and challenges the interpretation by Murcia et al. (2014) of the status of the
jarrah forest mine rehabilitation. Perhaps this insight reveals the nature of the so called abiotic barrier
referred to by Hobbs and colleagues as necessary for novelty. Clearly, in the mining and soil type contexts, it
is not the loss of the soil layer and its substitution with a parent-like material, but the absence of such or its
replacement with a material of a very different physical/chemical lithology or origin, which results in novelty.
However, from the UK evidence, the resulting vegetation and ecosystems would still be of a recognisable
native type and not irreversibly novel as a non-native type as required by the Hobbs model, but comprise
native and historical types that are simply out of place.
Seemingly less debated by the advocates for novel states is the outcome of adopting a laissez-faire approach
to restoration (Humphries, 1979) or the similar “open-endedness” strategy of leaving revegetation to natural
colonisation and successional processes as suggested by Hughes et al. (2011, 2012). This approach to
revegetation assumes the tacit acceptance of whatever vegetation and ecosystem outcome occurs in the
short and longer terms. In the context of mining, the laissez-faire and open-endedness approaches would
embody either leaving the closed mine (following any required initial treatments) to natural colonisation,
and vegetation and ecosystem development/successional processes (for example, as described by Rentel and
Rentel, 2009). It might be predicted, by default, that the outcome would result in novel vegetation and
ecosystems of an unusual but native composition (unless it is dominated by alien/introduced species as in
Ewel, 2013). However, as Gemmell (1977), Humphries and Rowell (1994), Jones and Jones (2008) and others
point out, the novelty of pioneer vegetation establishing on mine and industrial wastes is ultimately likely to
be excluded by later invading species and is likely to become a familiar native type.

The outcome of the above is that in the UK context of a long history of naturally revegetated abandoned
mine workings, the development of concomitant native vegetation and ecosystems is itself not a novelty.
Hence, the use of the “novel ecosystems” concept and its threshold model are likely to be of limited relevance
and application. In undertaking the rehabilitation of mine sites in the UK, the general use of the term novel
in the context of Bradshaw et al. (1978) is applicable and of practical use without the need to refer to that of
Hobbs et al. (2013). Elsewhere, like Australia, where there is a shorter history of naturally vegetated mine
workings and/or less interest in the recording of the resulting vegetation and their ecosystems, novel may be
a better description of the state of human knowledge rather than novelty per se.
Given the historical context, the rehabilitation of specific vegetation and ecosystems types associated with
the legacy of mining activities is recognised in the UK and may be acceptable (with appropriate evidence) to
the planning and regulatory authorities, and can be technically incorporated into mining consents. The same
is apparently the case in Australia. Here, government guidance to the mining industry (Department of
Industry, Tourism and Resources, 2006) readily recognises that mining operations can result in a range of
features, such as waste rock dumps, tailings storage facilities, water filled voids etc., and that rehabilitation
and land use targets should take into account the specific characteristics of the climate, landform, substrate
etc. The guidance advocates the zoning of the features as separate entities (“domains”) and the setting of
specific rehabilitation goals and targets during the mine closure planning stage. It is accepted that these may
be different from the pre-mining land use and vegetation. This includes the use of pioneer vegetation and
ecosystems and reliance on successional processes for the targeted outcome. The provision for non-pre-
mining land use and vegetation is translated into the provisions of current state government guidance (e.g.
Department of Mines and Petroleum and Environment Protection Authority, 2011; Department of the
Environment and Heritage Protection, 2014). It is within this context of aligning rehabilitation to the physical
and chemical requirements of the target vegetation and ecosystem types that Doley and colleagues argue
their case for adopting (novel) alternatives for mine closure and rehabilitation for a more successful outcome
(where vegetation and ecosystems are better matched with site, landform and substrate conditions). Whilst
the principles are well understood and provided for in the Australian planning and permitting process, what
is seemingly missing is the identification of the viable alternative vegetation and ecosystems that provide the
requisite stable and self-sustaining conditions consistent with the nominated post-mining land use
(Department of Industry, Tourism and Resources, 2006). To this end, there appears to be a gap in the
knowledge that needs a systematic research programme to define the target appropriate habitat, vegetation
composition, and ecosystem types for the domain types associated with mining operations in Australian, and
possibly elsewhere too. For example, bio-crusts might be an alternative and novel kind of “vegetating” mine
tailings, instead of the conventional practice of seeding or planting local forest types. In order to address the
concern that the adoption of the novelty is one of faith and open to criticism of “lowering the bar” for
rehabilitation standards (Perring et al., 2013; Perring et al., 2014), there will be a need for long-term
monitoring programmes of pilot and eventually established schemes in order to create the kind and level of
understanding that is well established for former mine sites and wastes in the UK.
Accepting that mine workings can be “out of place” in the landscape, as described by Erskine and Fletcher
(2013), the establishment of non-locally occurring vegetation and ecosystems characteristic of the pre-
mining landform and soils would be a local novelty and might be novel in this context. So, in response to the
question posed by this paper: “is there a place for the ‘novel ecosystem’ concept in mine closure and
rehabilitation?” The answer is “yes” in principle as there is provision in the current best practice and
regulatory guidance, but in practice it is a “no” in the context it is currently being marketed, given the
confusion around the concept and model and apparent lack of supporting evidence (Aronson et al., 2014). It
might be misleading and unsafe in mine planning and regulation terms for the present time for it to be
adopted without further critical scrutiny.
Could it be useful as a debating tool in mine rehabilitation planning? The answer is “yes” as Doley and Audet
(2014) suggest there is benefit for a more active airing as to what is to being proposed for mine rehabilitation

rather than to adopt the pre-mining historical vegetation by default. Do we currently know enough about
the type and composition for, and the sustainability of, rehabilitation alternatives for mine sites? The answer
is probably “yes” in the UK, but possibly “no” in the Australian context and maybe elsewhere too, where
alternatives to the pre-mining situation might be viewed as a novelty.
Given the above, is there a way forward? “Yes” there is. If novel ecosystems are to be considered in mine site
reclamation, some fundamental rules need to be established. Firstly, Mascaro et al. (2013) set out clear
direction that ecosystems are not ones that are part of: “… a historical range or variability … intensively
managed systems … managed with the purpose of reproducing historical ecosystem …”. Secondly,
reconstructed novel ecosystems should be expected to have ecosystem services greater than or equal to the
historic system and its own intrinsic demonstrable ecological value unless, thirdly, alternative candidate
systems are adopted from ecosystems with analogous landforms and soils, and possibly also from abandoned
historic mine sites. In these cases, in areas of high conservation value, local stakeholders and regulators may
not accept these novel systems, but see them as alien to the region.
As a final word on the “novel ecosystem” concept and its application to mine rehabilitation, in the form
currently construed, it may be a potential risk to sustainable ecosystem outcomes if used as an excuse for
restoration failure. Given the lack of clarity and agreement on what constitutes a novel ecosystem, we
conclude its adoption as a tool for mine rehabilitation should either be avoided or used with caution for the
present time. However, the usage of the term to describe techniques and methodologies in the context of
Bradshaw et al. (1978) is simple to understand and has practical application.
Queensland University’s Centre for Mined Land Rehabilitation kindly provided study facilities for RNH as a
Visiting Academic. The views expressed in this paper are the authors’ and do not necessarily reflect those of
CMLR nor Celtic Energy Limited.
Aronson, J., Murcia, C., Kattan, G.H., Moreno-Mateos, D., Dixon, K. and Simberloff, D. (2014) The road to confusion is paved with
novel ecosystem labels: a reply to Hobbs et al., 2014. Trends in Ecology and Evolution, Vol. 29(12), pp. 646–647.
Benyon, P.B. (2013) Nant Helen Remainder: survey of grassland within restored area. Celtic Energy, Caerphilly.
Blackstock, T.H., Howe, E.A., Stevens, J.P., Burrows, C.R. and Jones, P.S. (2010) Habitats of Wales. University of Wales Press, Cardiff.
Bradshaw. A.D. and Chadwick, M.J. (1980) Studies in ecology, Vol. 6: the restoration of land: the ecology and reclamation of derelict
and degraded land. Blackwell Scientific Publications, Oxford.
Bradshaw, A.D., Humphries, R.N., Johnson, M.S. and Roberts, R.D. (1978) The restoration of vegetation on derelict land produced by
industrial activity, in The breakdown and restoration of ecosystems, M.W. Holdgate and M.J. Woodman (eds), Plenum Press,
New York, pp. 249–278.
Centre for Ecology and Hydrology (2014) APIS, viewed 16 January 2015, http://www.apis.ac.uk/critical-load-function-tool.
Chapin. F.S. and Starfield, A.M. (1997) Time lags and novel ecosystems in response to transient climatic change in the arctic Alaska.
Climate Change, Vol. 35, pp. 449–461.
Department of the Environment and Heritage Protection (2014) Guideline: Rehabilitation requirements for mining resource activities.
Queensland Government, Brisbane, viewed 16 January 2015, http://www.ehp.qld.gov.au/era/mining/mining-rehabilitation-
em1122.pdf.
Department of the Environment and Rural Affairs (2010) Critical loads, viewed 16 January 2015, http://cldm.defra.gov.uk/Critical_
Load.htm.
Department of the Environment and Water Resources (2007) Australia’s native vegetation: a summary of Australia’s major vegetation
groups. Australian Government, Canberra.
Department of Industry, Tourism and Resources (2006) Mine rehabilitation: leading practice, sustainable development for the mining
industry. Commonwealth of Australia, Camberra.
Department of Mines and Petroleum and Environment Protection Authority (2011) Guidelines for preparing mine closure plans:
June 2011. Government of Western Australia, Perth, viewed 16 January 2015, http://www.dmp.wa.gov.au/documents/
Mine_Closure(2).pdf.
Dimbleby, G.W. (1978) Prehistoric man’s impact on environments in North West Europe, in The breakdown and restoration of
ecosystems, M.W. Holdgate and M.J. Woodman (eds), Plenum Press, New York, pp. 129–144.
Dobson, A.P., Bradshaw, A.D. and Baker, A.J.M. (1997) Hopes for the future: restoration ecology and conservation biology. Science,
Vol. 277, pp. 515–522.

Doley, D. and Audet, P. (2013) Adopting novel ecosystems as suitable rehabilitation alternatives for former mine sites. Ecological
Processes, Vol. 2: p. 2.
Doley, D. and Audet, P. (2014) Changing restoration priorities in the 21st century – opportunities for novel ecosystems in mine closure,
in Proceedings of Australian Institute of Mining and Metallurgy Life-of-Mine 2014: Delivering sustainable legacies through
integrated life-of-mine planning, Brisbane, pp. 649–662.
Doley, D., Audet, P. and Mulligan, D.R. (2012) Examining the Australian context for post-mined land rehabilitation: reconciling a
paradigm for the development of natural and novel ecosystems amongst post-disturbance landscapes. Agriculture,
Ecosystems and Environment, Vol. 163, pp. 85–93.
Erskine, P.D. and Fletcher, A.T. (2013) Novel ecosystems created by coal mines in central Queensland’s Bowen Basin. Ecological
Processes, Vol. 2, pp. 33.
Ewel, J.J. (2013) Case study: hole-in-the donut, Everglades, in Novel ecosystems: intervening in the new ecological world order, R.J.
Hobbs, E.S. Higgs and C.M. Hall (eds), Wiley-Blackwell, Chichester, pp. 11–15.
Gemmell, R.P. (1977) Studies in biology 80: colonisation of industrial wasteland. Edward Arnold, London.
Glenn, V., Doley, D., Unger, C., McCaffrey, N., McKenna, P., Gillespie, M. and Williams, E. (2014) Mined land rehabilitation – is there
a gap between regulatory guidance and successful relinquishment? Mine Rehabilitation: Bulletin, Vol. June, pp. 48–54.
Groffman, P.M., Baron, J.S., Blett, T., Gold, A.J., Goodman, I., Gunderson, L.H. … Wiens, J. (2006) Ecological thresholds: the key to
successful environmental management or an important concept with no application? Ecosystem, Vol. 9, pp. 1–13.
Hobbs, R.J., Higgs, E. and Harris, J.A. (2009) Novel ecosystems: implications for conservation and restoration, Trends in Ecology and
Evolution, Vol. 24, pp. 599–605.
Hobbs, R.J., Higgs, E.S. and Hall, C.M. (eds) (2013) Novel ecosystems: intervening in the new ecological world order. Wiley-Blackwell,
Chichester.
Hobbs, R.J., Higgs, E.S. and Harris, J.A. (2014) Novel ecosystems: concept or inconvenient reality? A response to Murcia at al. Trends
in Ecology and Evolution, Vol. 29(12), pp. 645–646.
Hobbs, R.J., Arico, S., Aronson, J., Baron, J.S., Bridgewater, P., Cramer, V.A. … Zobe, l. M. (2006) Novel ecosystems: theoretical and
management aspects of the new world order. Global Ecology and Biogeography, Vol. 15, pp. 1–7.
Hopper, S.D. (2009) OCBIL theory: towards an integrated understanding of evolution, ecology and conservation of biodiversity on
old, climatically buffered, infertile landscapes. Plant Soil, Vol. 322, 49–86.
Hughes, F.M.R., Adams, W.M. and Stroh, P.A. (2012) When is open-endedness desirable in restoration projects? Restoration Ecology,
Vol. 20(3), pp. 291–295.
Hughes, F.M.R., Stroh, P.A., Adams, W.M., Kirby, K.J., Mountford, J.O. and Warrington, S. (2011) Monitoring and evaluating large-
scale, “open-ended” habitat creation projects: a journey rather than a destination. Journal for Nature Conservation, Vol. 19,
pp. 245–253.
Humphries, C.E.L., Humphries, R.N. and Wesemann, H. (1999) The role of a fertiliser trial in reconciling agricultural expectations and
landscape ecology requirements on an opencast coal site in South Wales, United Kingdom, in Proceedings of American Society
for Surface Mining and Reclamation 16th Annual National Meeting, S.A. Bengson and D.M. Bland (eds), Scottsdale, Arizona,
pp. 262–272.
Humphries, R.N. (1979) Some alternative approaches to the establishment of vegetation on mined land and on chemical waste
materials, in Ecology and coal resource development, M.K. Wali (ed), Pergamon Press, New York, pp. 461–475.
Humphries, R.N. (2000) Biodiversity: curse or opportunity? Mining Quarrying and Recycling Vol. 29(10), pp. 26–29.
Humphries, R.N. (2013a) Case Study: The contribution of active surface mines in the conservation of lichen communities in the South
Wales Coalfield, United Kingdom. Journal American Society of Mining and Reclamation, Vol. 2(1), pp. 80–98.
Humphries, R.N. (2013b) Understanding and delivery of the components of structure, diversity, and function in the restoration of
ecosystems on mined land: working towards a practical methodology. Journal American Society of Mining and Reclamation,
Vol. 2 (2), pp. 1–31.
Humphries, R.N. (2014) Why reinvent the wheel when there are established methodologies to aid the design and assessment of
restored natural ecosystems on mined land? in Proceedings of Australian Institute of Mining and Metallurgy Life-of-Mine
2014: Delivering sustainable legacies through integrated Life-of-Mine planning, Brisbane, pp. 675–692.
Humphries, R.N. and Elkington, T.T. (eds) (1980) Special issue: reclaiming limestone and flourspar workings for wildlife. Reclamation
Review Vol. 3 (4), pp. 193–240.
Humphries, R.N. and Rowell, T.A. (1994) The establishment and maintenance of vegetation on colliery Spoils. British Coal Corporation,
Eastwood.
Humphries, R.N., Lunn, J. and Benyon, P.R. (2014) Establishment of Meadow Foxtail – Great Burnet meadows (MG4) on former
pasture at South Grange Farm, in the East Riding of Yorkshire. The Naturalist, Vol. 139, pp. 197–209.
Jobling, J. and Stevens, F.R.W. (1980) Occasional Paper 7: Establishment of trees on regraded colliery spoil heaps. Forestry
Commission, Farnham.
Johnson, M.S. (1978) Land reclamation and the botanical significance of some former mining and manufacturing sites in Britain.
Environmental Conservation, Vol. 5, pp. 223–228.
Johnson, M.S. (1980) Revegetation and development of wildlife interest on disused fluorspar tailings dams. special issue: reclaiming
limestone and fluorspar workings for wildlife. Reclamation Review Vol. 3(4), pp. 209–216.
Joint Nature Conservation Committee (1990) Handbook for Phase I habitat survey: a technique for environmental audit. JNCC,
Peterborough.
Jones, D.L. and Jones, G. (2008) CCW Science Report 824: A strategic conservation assessment of heathland and associated habitats
on the coal spoils of South Wales. Countryside Council for Wales, Bangor.

Jones, T.A. (2013) Plant materials for novel ecosystems, in Novel ecosystems: intervening in the new ecological world order, R.J.
Hobbs, E.S. Higgs and C.M. Hall (eds), Wiley-Blackwell, Chichester, pp. 212–227.
Mascaro, J., Harris, J.A., Lack, L., Thompson, A., Perring, M.P., Richardson, D.A. and Ellis, E.C. (2013) Origins of the novel ecosystems
concept, in Novel ecosystems: intervening in the new ecological world order, Hobbs, R.J., Higgs, E.S. and Hall,C. M. (eds),
Wiley-Blackwell, Chichester, pp. 45–57.
McKenzie, N., Jacquier, D., Isbell, R. and Brown, K. (2004) Australian soils and landscapes. CSIRO Publishing, Collingwood.
Mulligan, D.R., Gillespie, M.J., Gravina, A.J. and Currey, N.A. (2006) An assessment of the direct revegetation strategy on the tailings
storage facility at Kidston gold mine, North Queensland, in Proceedings First International Seminar on Mine Closure, A.B.
Fourie and M. Tibbett (eds), Australian Centre for Geomechanics, Perth, pp. 371–381.
Murcia, C., Aronson, J., Kattan, G.H., Moreno-Mateos, D., Dixon, K. and Simberloff, D. (2014) A critique of the “novel ecosystem”
concept. Trends in Ecology and Evolution, Vol. 29(10), pp. 548–553.
Perring, M.P., Standish, R.J. and Hobbs, R.J. (2013) Incorporating novelty and novel ecosystems into restoration planning and practice
in the 21st century. Ecological Processes, Vol. 2: pp. 18.
Perring, M.P., Audet, P. and Lamb, D. (2014) Novel ecosystems in ecological restoration and rehabilitation: innovative planning or
lowering of the bar? Ecological Processes, Vol. 3, pp. 8.
Pywell, R.F., Bullock, J.M., Roy, D.B., Warman, L., Walker, K.J. and Rothery, P. (2003) Plant traits as predictors of performance in
ecological restoration. Journal of Applied Ecology, Vol. 40, pp. 65–77.
Queensland Government (2014) Regional ecosystems descriptions, viewed 15th January 2015, https://environment.ehp.qld.gov.au/
regional-ecosystems/list/.
Ratcliffe, D.A. (1974) Ecological effects of mineral exploitation in the United Kingdom and their significance to nature conservation.
Proceedings Royal Society London, Vol. A339, pp. 355–372.
Read, C.F., Duncan, D.H., Vesk, P.A. and Elith, J. (2014) Biocrust morphogroups provide an effective and rapid assessment tool for
drylands. Journal of Applied Ecology, Vol. 51, pp. 1740–1749.
Rentel, U. and Rentel, M. (2009.) Determining the rehabilitation success of the old tailings storage facility of Navachab Gold Mine,
Karibib, Namibia, in Proceedings of the Fourth International Conference on Mine Closure, A.B. Fourie and M. Tibbett (eds),
Australian Centre for Geomechanics, Nedlands, pp. 109–121.
Rodwell, J.S. (ed) (1991a) British plant communities, volume 1: woodlands and scrub. Cambridge University Press, Cambridge.
Rodwell, J.S. (ed) (1991b) British plant communities, volume 2: mires and heaths. Cambridge University Press, Cambridge.
Rodwell, J.S. (ed) (1992) British plant communities, volume 3: grasslands and montane communities. Cambridge University Press,
Cambridge.
Rodwell, J.S. (ed) (2000) British plant communities, volume 5: maritime communities and vegetation of open habitats. Cambridge
University Press, Cambridge.
Rodwell, J.S. (ed) (2006) National vegetation classification: user’s handbook. Joint Nature Conservation Committee, Peterborough.
Rudeforth, C.C., Hartnup, R., Lea, J.W., Thompson, T.R.E. and Wright, P.S. (1984) Bulletin No. 11: Soils and their use in Wales. Soil
survey of England and Wales, Harpenden.
Smith, P.L. (2012) Nant Helen: bryophyte and lichen survey. Celtic Energy, Caerphilly.
Smith, R.A.H. and Bradshaw, A.D. (1972) Stabilisation of toxic mine wastes by the use of tolerant plant populations. Transactions of
the Institution of Mining and Metallurgy, Vol. 81A, pp. 230–237.
Tongway, D. and Hindley, N. (2003) Final report – indicators of ecosystem rehabilitation success: stage 2 – verification of EFA
indicators. Australian Centre for Mining Environmental Research, Darra.


M.E. Wen Environmental Resources Management (ERM), Canada

C.A. Pelletier Environmental Resources Management (ERM), Canada
K. Norlund Environmental Resources Management (ERM), Canada
G.W.R. Wolff Environmental Resources Management (ERM), Canada
D. Berthelot BHP Billiton, Canada
The flooding of open pits is a key feature of closure plans for many metal mines. Once filled, pit lakes typically
discharge to the environment. Pit lake water requiring treatment prior to discharge can represent a significant
cost of long-term closure care and maintenance. We present phytoremediation technology that has been
successfully used in many industries as a relatively low-cost technique to attenuate contaminants in water.
Phytoremediation describes the use of plants to mitigate environmental contamination. In the context of mine
pit lakes, phytoremediation is the biologically mediated removal of contaminants using photosynthesizing
organisms to improve water quality. The relatively low-cost option of phytoremediation can make it an
attractive alternative to conventional lime neutralisation for metal contamination resulting from metal
leaching and acid rock drainage. Phytoremediation has also been investigated for treatment of other
contaminants in mine water.
Metal mine pit lakes present an opportunity to use phytoremediation to treat waters affected by metal
leaching and acid rock drainage. Within a pit lake, the greatest opportunity for phytoremediation of water is
in the limnetic zone, which usually occupies a much greater area and volume than the littoral zone. However,
other important variables may influence the potential success of phytoremediation.
Phytoremediation has been used successfully for more than a decade to treat metal-contaminated pit lake
waters at the closed Island Copper Mine. We discuss how the phytoremediation program at the Island Copper
Mine has been designed to treat pit lake waters to meet permit requirements and significantly reduce metal
loads in the lake. Through this case study we also explore the relationships between the physical and
biogeochemical characteristics of the water column and show how an understanding of this relationship is
critical to planning, designing and identifying the potential success and risks of phytoremediation treatment
of mine pit lakes.
Phytoremediation describes the use of plants to mitigate environmental contamination. In the context of
mine pit lakes, phytoremediation is the biologically mediated removal of contaminants using
photosynthesizing organisms to improve water quality. The relatively low-cost option of phytoremediation
can make it an attractive alternative to conventional lime neutralisation for metal contamination resulting
from metal leaching and acid rock drainage (ML/ARD; Pelletier et al., 2009). Phytoremediation has also been
investigated for treatment of mine water for other contaminants, including using phosphorous amendments
to treat nitrate loads (Rescan, 2010) and uranium and other metal loads (Dessouki et al., 2005).
Within a pit lake, the greatest opportunity for phytoremediation of water is in the limnetic zone, which
usually occupies a much greater area and volume than the littoral zone. However, other important variables
may influence the potential success of phytoremediation. Evaluation of the physical structure of the water
column — in addition to evaluating the source and nature of contaminants — is critical to identifying the
potential success and risks of pit lake phytoremediation options. Phytoremediation has been used

Phytoremediation to improve pit lake water quality M.E. Wen, C.A. Pelletier, K. Norlund, G.W.R. Wolff and D. Berthelot
successfully for more than a decade to treat metal-contaminated pit lake waters at the closed Island Copper
Mine. In this paper, we describe the phytoremediation technique used to attenuate metal concentrations in
the Island Copper pit lake.
Island Copper (Figure 1) was an open pit porphyry copper mine that operated from October 1971 to
December 1995. At start-up, the mine had an ore reserve of 254 million (metric) tonnes, averaging 0.52%
copper and 0.017% molybdenum. After receipt of agency permits, the decision to proceed with the
development was made and construction of the mill, ancillary support facilities and deep-sea port began.
Over the 24-year life of the mine, mineral reserves expanded to 408 million tonnes and ore production
increased to 55,000 tonnes per day.
More than one billion tonnes of rock were removed from the open pit. The mine produced metal
concentrates containing more than 1.3 million tonnes of copper, 31,000 tonnes of molybdenum, 31.7 tonnes
of gold, 336 tonnes of silver and 27 tonnes of rhenium. By the time it closed in 1995, Island Copper was
Canada’s third-largest mine and the lowest man-made depression on Earth.
1995 1996
2002
A unique feature of the operational closure plan was the 1996 decommissioning of the open pit by flooding
it with seawater from adjacent Rupert Inlet and capping it with fresh water to create a 355-metre-deep lake
covering 174 hectares. As part of the mitigation plan, this lake provided a receiving environment for acid rock
drainage from the mineralised rock deposited around the rim of the pit during operations. The rapid
development of meromixis was expected to result in the long-term development of anoxic conditions deep
in the lake. These conditions would cause metals to precipitate as insoluble metal-sulphides, providing a
means of treatment for injected ML/ARD. Table 1 lists the physical characteristics of the Island Copper pit
lake and the water inputs to the lake.

The site receives high rainfall (an average of 1.9 m/yr from 2009 to 2013), little snow (less than 15% of
precipitation) and has relatively cool summers and mild winters, resulting in a long growing season ideal for
vegetation growth. Construction of the upland waste rock dumps began in the early 1970s and land
reclamation in the 1990s. Land reclamation included a nominal half-metre soil reclamation cap, contouring
and planting of the waste rock dumps with woody vegetation. This cover was designed to reduce erosion and
was not intended to limit or reduce infiltration of meteoric water into the stockpiles. The grading and
vegetation cover were designed to maintain an aesthetic quality consistent with the surrounding landscape.
Today, approximately 88 million tonnes of waste rock stored over 44 hectares of land have waters reporting
to the pit lake for remediation. The approximate annual quantity and quality ML/ARD reporting to the surface
and middle layers of the pit lake from 2009 to 2013 are shown in Table 2.
Basic parameter Approximate value

Horizontal dimensions of pit lake Approximately 1 by 2 km, oval
Major rock type in walls Andesitic pyroclastic, quartz feldspar porphyries, Bonanza
volcanic, minor sulphides
Surface area of pit lake 1.74 million m2
Perimeter of pit lake 5,600 m
Maximum depth of pit lake 355 m
Total volume of pit lake (five-year average) 243 million m3
Annual direct rainfall to pit lake 1.9 m/yr, equivalent to 3.4 million m3/yr
(five-year average)
Mine water directed to pit lake 4.3 million m3/yr
(five-year average)
Run-in water from pit walls 1.1 million m3/yr
(five-year average)
Lake evaporation (from regional estimates) 1.0 million m3/yr
Note: Five-year averages use 2009 to 2013 data.
Variable To surface To middle layer

North catchment Other Southeast catchment
Total annual flow (five-year 2.0 0.8 1.5
average; million m3)
pH 6.64–8.08 3.69–8.32 4.30–7.83
Conductivity (µs/cm) 726–1,220 187–2,840 1,210–2,302
Sulphate (mg/l) 293–638 44.0–1,960 653–1,621
Dissolved zinc (mg/l) 0.646–4.22 0.191–23.2 0.564–11.1
Dissolved copper (mg/l) 0.0107–0.201 0.00823–3.78 0.00323–2.52
Dissolved cadmium (mg/l) 0.00396–0.0234 0.00085–0.120 0.0027–0.0541
Note: Estimates are from 2009 to 2013. Ranges are from quarterly sampling.

Because the pit lake was intended as a water treatment facility, near-shore habitat features were
intentionally left out its design. The surface of the lake receives direct precipitation (3.4 million m 3/yr), run-
in from the pit walls (1.1 million m3/yr), drainage from the catchment north of the lake (2.0 million m3/yr),
and mine drainage inputs from other sources (0.8 m3/yr). The lake normally has open water year-round owing
to the mild climate, and evaporation loss at the surface is estimated at 1.0 million m 3/yr based on regional
data from Environment Canada (2015). There is no evidence of significant groundwater input to the pit lake.
Water from the southeast catchment (1.5 million m3/yr) is injected into the lake’s middle layer through the
south injector system (SIS). The SIS was installed after flooding to convey drainage from waste dump areas
to a depth of about 220 m. The system has a capacity in excess of 800 l/s. The SIS (Figure 2) consists of a 36-
inch high-density polyethylene (HDPE) pipe approximately 380 m long extending from an elevation of
approximately 3 m below sea level (−3 m amsl) to approximately −220 m amsl. The end of the SIS pipe is
equipped with a multiport diffuser, through which the SIS drainage discharges at the bottom of the middle
layer of the pit lake. In addition, a 40-metre section of 63-inch HDPE pipe, installed from −3 m up to +5 m
amsl, penetrates the surface of the pit lake and provides a deaeration chamber for the injected flow. Another
approximately 75-metre section of 36-inch HDPE pipe connects the deaerator to the SIS settling pond to the
south, which in turn is fed by ditches and piping from the northeast areas of the property.
Settling Pond
36” HDPE Pipe
Deaeration Chamber
Pit Lake
The pit lake design takes advantage of the strong density contrast between seawater and freshwater to
create a meromictic lake. The lake is now a three-layer system, with each layer having distinct physical and
chemical characteristics (Table 3). The strong salinity contrast between the top and middle layers forms the
density gradient between them (the pycnocline). Although there is a temperature difference between these
two layers, it is the salinity contrast that provides the water column stability between the two layers. The

elevation of the pycnocline at the Island Copper pit lake is of importance in determining which layer can
discharge to Rupert Inlet. The diffuse discharge to the inlet is also important in eliminating salt loads from
the top layer, thereby preventing salt build-up and water column instability.
In February 2005, the middle layer lifting (MLL) system was commissioned to control the pycnocline
elevation. The Island Copper MLL system is an example of the maintenance of the physical water column
structure of a pit lake through designed manipulation. Before the commissioning of the MLL system, the
pycnocline was rising at a rate of approximately 0.5 m/year, mostly during the rainy winter period. The
pycnocline’s rise was caused by upward displacement from the injection of mine drainage at depth. The
concept of the MLL system is simple. Water from the upper portion of the middle layer is “pumped” to the
surface at a rate slightly greater than the rate at which mine water is injected at depth through the SIS
(Figure 2). The action of the MLL system counteracts the upward displacement of the pycnocline, and in fact
results in a lowering of the pycnocline over time. The system is described in Pelletier et al. (2009).
Figure 3 shows four density profiles of the pit lake measured in 2013, expressed as t units. The salinity
contour plot of Figure 4 shows the typical fluctuations in water level, the clearly defined pycnocline, and the
drop in pycnocline elevation since the MLL was commissioned.
Variable Top layer Middle layer Bottom layer

Thickness (five-year 6.5 215 130
average; m)
Volume (five-year average; 11.4 217 14.8
million m3)
Sampling depths (m) 0, 2 10, 30, 100, 200 300, ~338
Five-year ranges (2009–2013)
pH 7.17–9.76 6.52–7.81 6.46–7.81
Salinity (PSU) 5.5–7.4 19.5–24.7 26.1–28.8
Temperature (C) 4.51–25.1 10.7–13.7 13.1–13.2
Dissolved oxygen (mg/l) 8.48–15.8 <0.04–1.91 <0.04
Hydrogen sulphide (mg/l) <0.1–1.61 <0.1–2.7 0.7–8.8
Dissolved zinc (mg/l) <0.0008–0.247 <0.0008–0.0414 <0.0008–0.0096
Dissolved copper (mg/l) 0.0012–0.0124 <0.00005–0.00079 <0.00005–0.00118
Dissolved cadmium (mg/l) 0.000024–0.00476 <0.00002–0.000336 <0.00002–0.000050
Note: Estimates are from 2009 to 2013. Ranges are from quarterly sampling.

March 13, 2013 June 11, 2013 September 3, 2013 December 17, 2013
Density ( t) Density ( t) Density ( t) Density ( t)
0 5 10 15 20 25 0 5 10 15 20 25 0 5 10 15 20 25 0 5 10 15 20 25
0
50
100
Depth (m)
150
200
250
300
350
Salinity
Gradient (ppt)
24
22
20
2
18
16
0
Elevation (masl)
14
12
10
0
2002 2004 2006 2008 2010 2012
Year

As discussed earlier, the Island Copper pit lake receives ML/ARD at the surface and at depth, which
contributes to the lake’s total metal load. As part of the phytoremediation program, weekly nutrient
amendments are made to the lake’s surface. Commercial-grade fertiliser is purchased in batches of
approximately 24,000 l of liquid fertiliser comprising 8,130 kg of ammonium polyphosphate (10-34-0) and
22,870 kg of urea ammonium nitrate (28-0-0). The mixture is delivered to the site by tanker truck and stored
in a steel tank. Each delivery of fertiliser lasts approximately three to four months under the current
fertilisation schedule. The liquid fertiliser is applied to the surface of the lake from the stern of a purpose-
built motorboat with an integrated 434 l fertiliser tank. The fertiliser is mixed into the surface waters through
the propeller wash of the boat. This procedure is repeated over a number of transects, for a total of
approximately 2,150 l of fertiliser per week. In total, approximately 2% of the lake’s surface area is fertilised
by boat; however, sampling at multiple locations immediately after fertilisation shows that nutrient
concentrations are uniform within the mixed layer. Based on the principle of a well-mixed surface layer,
testing is underway to automate injection of fertiliser to the surface of the pit lake without the use of a boat.
In either case, the weekly application rate is equivalent to 382 mg N/m2 and 63 mg P/m2. Lake fertilisation is
suspended during July and August, when metal loads to the lake are low because of the dry conditions. A
monitoring program evaluates the evolving ML/ARD and the effects of the fertilisation program on water
quality, primary production, and the overall biogeochemical performance of the pit lake.
Phosphorous is typically the limiting nutrient for phytoplankton populations in natural lakes. Nitrogen is
abundant in the atmosphere as nitrogen gas, and some phytoplankton (cyanobacteria) can use this form of
nitrogen and are thus never limited by nitrogen in surface waters. Consequently, these organisms can often
out-compete other phytoplankton species when nitrate and other forms of nitrogen are at low
concentrations. The forms of nitrogen most utilised by phytoplankton are nitrate and ammonia. Nutrient
amendments to the pit lake provide nitrogen and phosphorus in a ratio of 6:1 by weight. Although this is
slightly less than the Redfield ratio (Redfield, 1958), it has proven effective for the Island Copper pit lake
system.
Nutrient amendments from 2009 to 2013 produced an average nitrate concentration in the top layer of
0.0188 mg/l (as N) and ranged from below the instrument detection limit to 0.122 mg/l. Surface
concentrations were slightly greater in winter than in summer because of the suspension of lake fertilisation
in July and August. The average ammonia concentration in surface waters ranged from below the detection
limit (0.0050 mg/l) to 0.107 mg/l (as N) from 2009 to 2013. Urea concentrations were as high as 0.339 mg/l
(as N) in surface waters.
From 2009 to 2013, total phosphate in surface waters ranged from 0.0020 mg/l to a peak of 0.142 mg/l in
May 2009. Dissolved phosphate concentrations at the surface are generally close to or below the best
available detection limit (<0.002 mg/l) for the first half of the year, indicating near-complete utilisation of
this nutrient by phytoplankton cells. As with nitrate, dissolved phosphate concentrations in surface waters
peak in the wet period. From 2009 to 2013, the peak concentration of dissolved phosphate was 0.0180 mg/l
in November 2010. Ortho-phosphate was typically below the analytical detection limit (<0.001 mg/l) because
of its preferential consumption by phytoplankton. Diatoms are not ideal candidates for the pit lake treatment
system, as they often have a low surface-area-to-volume ratio and a high sinking rate (low residence time in
the top layer). Although silicic acid is a required macronutrient for diatoms, this nutrient is not used as a
component of the fertiliser to avoid the dominance of diatoms in the phytoplankton.
Figure 5 shows Secchi depths and chlorophyll a concentrations measured in the top and middle layers of the
lake from 2009 to 2013. Chlorophyll a concentrations at the surface ranged from below detection (August
and September of most years) to 142 g/l (April 2013), and averaged 16.1 g/l. The low chlorophyll
concentrations measured in August and September coincide with the period when fertiliser was not applied
to the pit lake. In the middle layer (100 m depth), chlorophyll a measurements ranged from below detection

to 12.6 g/l, and averaged 5.2 g/l. The presence of any measurable biomass in the middle layer is caused
by phytoplankton cells sinking out of the productive top layer and eventually settling to the lake bed.
The stimulation of phytoplankton growth by fertilisation increases the pH of surface waters by reducing the
concentration of dissolved carbon dioxide, which is consumed through photosynthesis. The pH in the top
layer of the pit lake trends approximately with chlorophyll in the top layer. The observed pattern follows that
of increased primary productivity. In the top layer, the pH from 2009 to 2013 ranged from 7.17 to 9.76 and
averaged 8.73.
The phytoplankton biomass found in the top and middle (10 m) layers of the pit lake is primarily composed
of the chlorophytes Coccomyxa granulata and Oocystis cf. parva in the winter, spring and summer months
and by the chlorophyte Ankistrodesmus falcatus and the diatom Chaetoceros muelleri in the fall. The same
trends in phytoplankton abundance are observed in the middle layer, with the predominant species O. parva
dominating the assemblages from February through October.
C. granulata is a mostly solitary cell, 3 to 4 m in diameter, which sometimes forms spherical colonies (Caljon,
1983). It is a “spring-summer species” that tolerates a wide range of salinities. A. falcatus is a needle-like,
mostly solitary cell that is 2 to 6 m in diameter and 25 to 100 m long (Prescott, 1982). This species is most
commonly found in acidic water with high temperatures. O. parva can generally tolerate a wide variety of
habitats (Wehr and Sheath, 2003), including bogs. Its small cells, 4 to 7.5 m in diameter, can occur as a single
cell or as families surrounded by a mother cell (Prescott, 1982). Colonies can be up to 43.9 m in diameter
(Prescott, 1982). Based on these characteristics, the phytoplankton species that colonise the Island Copper
pit lakes are well suited to the ambient water quality conditions. Overall, the biological composition of the
pit lake appears to be dominated by chlorophytes.
The effect of the observed production of these small phytoplankton species is to provide adsorption surfaces
for the removal of dissolved metals in the top layer. The lake has met its discharge permit limits, which are
based on dissolved metal concentrations in the top layer. Figure 6 shows concentrations of the particulate
fraction of zinc from 2009 to 2013 in the top layer and in the upper portion of the middle layer, immediately
below the upper pycnocline (at a sampling depth of 10 m). Particulate zinc in the top layer primarily
represents the fraction of zinc adsorbed to phytoplankton cells and available for downward export to the
lower layers. In the top layer, particulate zinc concentrations are at their highest when phytoplankton

biomass is high and there is a high metal load to the surface of the lake. At Island Copper, this occurs from
early winter until the late spring of each year. In the summer months, before significant metal loading to the
surface occurs, particulate zinc is low because of the limited metal loads for the biological system to treat
despite the season’s generally low primary production. Similar responses to biological treatment in the top
layer are observed for particulate copper and cadmium (Fisher and Lawrence, 2006; Pelletier et al., 2009). In
contrast to the top layer, particulate zinc, copper and cadmium concentrations at a depth of 10 m do not
correspond closely to chlorophyll a measurements because of the relatively low metal concentrations
present in the middle layer. These low metal concentrations have been present since 2006, when
biogeochemical processes caused metal-sulphide precipitation. This is discussed in the following section.
Surface to 3 m Depth
150 1.4
1.2
Particulate Zinc (mg/L)

Chlorophyll a (ug/L)
1.0
100 Chlorophyll a
Particulate Zinc 0.8
0.6
50
0.4
0.2
0 0.0
2009 2010 2011 2012 2013
Year
The in situ phytoremediation is designed to treat lake surface waters for discharge to Rupert Inlet, and to
export organic carbon to depth to induce metal removal through biological sulphide generation. Pelletier et
al. (2009) discuss the success of metal-sulphide precipitation in the sulphidic zone of the lower layers of the
lake.
The biogeochemical zonation in the pit lake is delineated into three redox zones (oxic, transition and
sulphidic). These zones differ from the physical delineation separating the pit lake into layers (top, middle
and bottom) based on water density described in Table 3. The main driver behind the redox potential of the
water is related to the bacterially mediated oxidation of organic matter, which is coupled with the reduction
of a redox indicator species. Redox reactions occur in a predictable sequential order (i.e. the redox ladder)
based on the energy yields known as Gibbs Free Energy (Table 4).

Redox reaction Chemical equation ∆Gro (kJ mol-1)

Aerobic respiration CH2O(s) + O2(aq) → CO2(g) + H2O(l) −519
Denitrification CH2O(s) + 0.8NO3-(aq) + 0.8H+(aq) → CO2(g) + 0.4N2(g) + 1.4H2O(l) −510
Manganese reduction CH2O(s) + 2MnO2(s) + 4H+(aq) → CO2(g) + 2Mn2+(aq) + 3H2O(l) −503
Ferric reduction CH2O(s) + 2αFe2O3(s) + 8H+(aq) → CO2(g) + 4Fe2+(aq) + 5H2O(l) −297
Sulphate reduction CH2O(s) + 0.5SO42-(aq) + 0.5H+(aq) → CO2(g) + 0.5HS-(aq) + H2O(l) −124
Methanogenesis CH2O(s) → 0.5CO2(g) + 0.5CH4(g) −85
Notes:
1. ΔGro = ΣΔGfoproducts − ΣΔGforeactants.
2. ΔGfo CH2O= −129 kJ mol-1 (Charlson and Emerson, 2000).
3. ΔGfo O2 = 16.37 kJ mol-1 (Weast, 1983).
4. All other ΔGfo values from Drever (1997).
5. MnO2 is pyrolusite and αFe2O3 is hematite.
In oxic environments, oxygen is preferentially reduced to water as organic matter is oxidised (oxic zone).
Once oxygen is depleted, consumption of nitrate (i.e. denitrification) ensues. Denitrification is not discussed
in detail here because of the interferences caused by the nutrient amendment program and phytoplankton
metabolism. Manganese and iron are subsequently reduced from insoluble oxides and hydroxides to soluble
metals. We refer to the zone over which manganese and iron are reduced as the “transition zone,” located
between saturated oxic and sulphidic waters. The reduction of sulphate to hydrogen sulphide (sulphide)
occurs once all the available manganese and iron have been reduced.
The sulphidic zone is characterised by the presence of reduced species (dissolved manganese and iron and
sulphide), as well as by the absence of oxygen and manganese and iron (hydr)oxides. Insoluble metal-
sulphides can precipitate and accumulate in the bottom sediments in a sulphidic environment. In strongly
reducing environments, methanogenesis, the lowest rung on the redox ladder, will ensue. The high sulphate
(naturally occurring in the seawater initially added to the pit) concentration likely prevents the onset of
methanogenesis in the Island Copper pit lake. Table 5 lists the redox couples associated with each redox
reaction in Table 4.
Oxidised member Reduced member

O2 (oxygen) H2O (water)
Mn4+ (manganese IV) Mn2+ (manganese II)
Fe3+ (iron III) Fe2+ (iron II)
SO42- (sulphate) HS- (sulphide)
C0 (carbon) CH4 (methane)
The redox potential of the water is also modified by the upwards diffusion of reduced species into zones
located higher on the redox ladder. Reduced species diffusing into an increasingly oxic environment may
react with oxidised species. As a result, oxidised species can be reduced and reduced species can be oxidised.

As a result of nutrient-enriched primary production at the surface, the concentration of organic matter in the
Island Copper pit lake is highest in surface waters and gradually decreases with depth. Aerobic surface waters
(oxic zone) overlay anoxic waters dominated by reduced species. The transition zone is situated directly
beneath the oxic zone and is characterised by a suboxic to anoxic environment and a steady increase in
dissolved manganese and iron. The sulphidic zone is located beneath the transition zone. The majority of the
pit lake volume is classified as sulphidic. This biogeochemical model of the Island Copper pit lake is
summarised in Figure 7. The idealised model focuses on the dominant biogeochemical processes and
excludes the physical transport processes caused by the MLL system, downward erosion of the pycnocline
caused by wind-induced turbulent currents, turbulent mixing at the side walls caused by breaking internal
waves of the pycnocline, diffusion across the pycnocline, convective mixing and rainwater dilution.
Figure 8 shows the redox indicator species in the pit lake plotted against depth. Redox indicator profiles are
plotted for two distinct sampling events (June and December 2013) to highlight the contrast between the
geochemical characteristics observed during the summer and winter months. The overall conceptual model
of the pit lake does not change from summer to winter; however, the concentrations of the redox indicator
species do change significantly, with a notable increase in sulphide concentrations in the winter. Total organic
carbon (TOC) is plotted in Figure 8 alongside the redox indicator species. Near the surface, high loads of TOC
indicate a high adsorption potential for metals. During the fall and winter, the TOC in the surface declines;
however, TOC at depth increases. Bacteria oxidise the TOC in the water by reducing the redox species.
The key characteristics of the Island Copper pit lake biogeochemical model are:
 The upper portion of the pit lake (above 2 m) is oxic.
 The anoxic zone, below depths of 2–10 m, can be classified as sulphidic and constitutes the
majority of the pit lake volume.
 The main mechanism of metal removal in the surface oxic waters is likely adsorption onto organic
matter, although there is also potential for mineral precipitation (e.g. [hydr]oxides).
 Dissolved iron and manganese concentrations increase at depths below 200 m, likely because of
the reductive dissolution of oxides. The potential for iron and manganese to precipitate as
sulphides is lower than that of other metals in solution.
 The likely mechanism for dissolved metal removal from the anoxic zone is mineral precipitation:
○ Mineral precipitation in the anoxic zone is dictated by the redox state of the water, which is a
function of the organic matter delivered from the fertilisation program.
○ The formation of sulphates, hydroxides and possible carbonates is a mechanism for dissolved
metal removal.
○ As the lower zones progress towards an increasingly sulphidic state, the potential for metal
removal in these zones via metal-sulphide precipitation remains high.


June
Dissolv ed Oxygen and Total Organic Carbon (mg/L)
0 2 4 6 8 10 12 14 16
0
100
Depth (m)
200
300
0.1 1 10 100 1,000 10,000

Sulphide, Sulphate, Iron, and Manganese (mg/L)
Sulphide
Sulphate
Dissolved Iron
Dissolved Manganese
Dissolved Oxygen
December T otal Organic Carbon
Dissolv ed Oxygen and Total Organic Carbon (mg/L)

0 2 4 6 8 10 12 14 16
0
100
Depth (m)
200
300
0.1 1 10 100 1,000 10,000

Sulphide, Sulphate, Iron, and Manganese (mg/L)
ICM-0400-004

The Island Copper pit lake can be divided into physically distinct and stable layers as a result of meromixis.
The top layer is approximately 6 m thick, brackish and well mixed. The middle layer is saline and occupies
approximately 90% of the pit lake volume. The bottom layer’s salinity is the closest to seawater, indicative of
remnant floodwater from Rupert Inlet. The injection of ML/ARD in the middle layer causes an upward
displacement of the pycnocline, and this effect is actively managed using the innovative MLL system, which
requires no external power.
The pit lake can also be divided into three zones to distinguish between dominant geochemical processes.
The zone closest to the surface is oxic and characterised by detectable levels of oxygen. The oxic zone extends
from the surface to depths of between 2 and 10 m. Below the oxic zone is the transition zone, the position
of which varies seasonally. In June, the transition zone extends to below the 10 m interval, whereas in
December the zone is truncated in response to elevated sulphide concentrations. Beneath the transition
zone, the sulphidic zone extends to the bottom of the pit lake. Although the pit lake is characterised by three
distinct redox zones, a clear delineation between the zones is not easily discernable because of mixing
processes. Sulphide is detectable within the oxic zone and the transition zone. Therefore, metal attenuation
results can be further segregated into two zones: oxic surface waters and sulphidic waters.
Two attenuation mechanisms likely control metal concentrations in the oxic surface waters: metal adsorption
by phytoplankton cells and secondary mineral precipitation. The pit lake appears to undergo slight seasonal
cycles in biogeochemical redox zonation and, therefore, metal attenuation mechanisms within the pit lake.
The largest variations are observed in the oxic upper layers. During the summer, the metal concentrations
are lowest in the surface oxic zone. This corresponds to lower observed rainfall when compared to December
rainfall and, therefore, lower metal loadings, as well as lower observed sulphide concentrations.
During the winter months, the highest metal concentrations are observed in the oxic layer. This corresponds
to an increase in metal loadings from ML/ARD as a result of heavy rainfall. The primary mechanism for
removal of metals in waters in the surface oxic zone is likely metal adsorption onto organic matter.
The primary mechanism for the removal of metals in the sulphidic zone is likely the result of metal-sulphide
precipitation. The concentration of sulphides in the lower layers of the pit lake have increased since pit
flooding, and this has resulted in the development of a sulphidic zone where soluble metal concentrations
are low and concentrations of hydrogen sulphide are high. Sulphide mineral precipitation is the most
important geochemical factor for the attenuation of metals at depth in the pit lake. The process is anticipated
to continue to assist in metal sequestration to the lake bed.
Phytoremediation at Island Copper pit lake has been successful in treating metal loads from ML/ARD. The
designed system takes advantage of the compartmentalised physical characteristics and geochemical
zonation of the meromictic lake and integrates treatment of surface waters using nutrient amendments, with
the resulting phytoplankton biomass providing a carbon source for microbial oxidation occurring in the
majority of the lake’s water volume. The system has required an engineered manipulation of the physical
structure of the water column using an innovative system to control the pycnocline elevation. The
phytoremediation techniques at Island Copper have been successful because of the integrated approach to
linking the physical and biological systems operating in the pit lake.
Having been flooded with seawater and capped with freshwater, the Island Copper pit lake represents an
extreme case in the continuum of meromixis. Although few other lakes may have such a strong density
contrast between layers, the Island Copper case shows that an integrated physical and biological pit lake
design can provide a long-lasting and effective treatment system. The site-specific conditions at Island Copper
resulted in challenges for water management, but these same conditions allowed for innovative
management solutions integrating physical and biological systems into a semi-passive treatment design.
Proximity to the sea was a challenge to mining but facilitated flooding the pit to create a meromictic pit lake.

The challenge of managing ML/ARD in a high-precipitation environment was tempered by the exploitation of
isolated layers of the lake where contaminants could undergo in situ treatment. The downward flux of
primary production from the top layer fuelled and continues to fuel this in situ treatment in the isolated
lower layers, and links the physical and biological systems.
The converse is likely true, in that ignoring the inter-relation between the physical and biological
characteristics of a pit lake or future pit lake might lead to unintended or non-optimal results.
The authors thank our colleagues at BHP Billiton and ERM, who have contributed to the many years of
successful innovation at Island Copper.
Caljon, A. (1983) Brackish-water phytoplankton of the Flemish Lowlands, Dr. W. Junk Publishers, The Hague, Netherlands, 272 p.
Charlson, R.J. and Emerson, S. (2000) Acid-base and oxidation-reduction balances of the Earth, in Earth system science, R. Charlson
(ed), Academic Press, New York, USA, pp. 421–437.
Dessouki, T.C.E., Hudson, J.J., Neal, B.R. and Bogard, M.J. (2005) The effects of phosphorus additions on the sedimentation of
contaminants in a uranium mine pit-lake, Water Research, Vol. 39(13), pp. 3055–3061.
Drever, J.I. (1997) The geochemistry of natural waters: Surface and groundwater environments, 3rd edition, Prentice Hall, Upper
Saddle River, USA, 436 pp.
Environment Canada (2015) Climate Data Online. Canada’s National Climate Archive.
http://www.climate.weatheroffice.gc.ca/climateData/canada_e.html
Fisher, T.S.R. and Lawrence, G.A. (2006) Treatment of acid rock drainage in a meromictic mine pit lake. J. Environ. Eng. 132:515-526.
Pelletier, C.A., Wen, M.E. and Poling, G.W. (2009) Flooding pit lakes with surface water, in Mine pit lakes — characteristics, predictive
modeling, and sustainability, Volume 3: Management technologies for metal mining influenced water, D.N. Castendyk and
L.E. Eary (eds), Society for Mining, Metallurgy & Exploration, Inc. (SME), Littleton, USA, pp. 187–202.
Prescott, G.W. (1982) Algae of the western Great Lakes area, Otto Koeltz Science Publishers, Koenigstein, Germany, 977 pp.
Redfield, A. (1958) The biological control of chemical factors in the environment, American Scientist, Vol. 46, pp. 205–221.
Rescan (2010) EKATI Diamond Mine: 2008/2009 Long Lake containment facility nitrate in situ treatment test, prepared for BHP Billiton
Diamonds Inc. by Rescan Environmental Services Ltd., April 2010, viewed 6 June 2011, http://www.mvlwb.ca/WLWB/
Registry.aspx.
Weast, R.C. (1983) CRC handbook of chemistry and physics, 64th edition, CRC Press, Cleveland, USA.
Wehr, J.D. and Sheath, R.G. (eds) (2003) Freshwater algae of North America, ecology and classification, Academic Press, New York,
USA, 917 pp.


M.W. Palmer Jones & Wagener (Pty) Ltd, South Africa

A. van Zyl Terrasim, South Africa
C. Smith Anglo American Coal South Africa, South Africa
The ingress of water to opencast coal mine workings, both during operations and post closure, is an ongoing
problem in the coal mining industry. This is particularly true in the Mpumalanga coal fields of South Africa,
where mines typically operate with a significant water surplus. The use of soil stripped ahead of mining is the
current best practice for capping of mine spoils. The thickness and quality of soil used as capping material
varies according to the availability of soil, which is often extremely limited due to poor utilisation of soils or
only thin soils present in the pre-mining soil profile. Consequently, the rehabilitated area often has severely
limited land use potential.
This paper details infiltration modelling and water balance modelling carried out for a proposed opencast
mine in the Mpumalanga Province of South Africa. The mine expects that, by carefully managing topsoil and
subsoil utilisation during mining, the soil cover thickness on the rehabilitated spoils can be maximised.
Expected soil cover thicknesses range from a minimum of 500 mm to 3,500 mm or greater, with an average
of 900 mm topsoil overlying a subsoil layer. The objective is twofold. Firstly to minimise the water make in the
opencast workings, thereby minimising water management and treatment costs, and secondly to create a
rehabilitated environment post closure that is sustainable in the long term and can potentially be suitable for
a range of commercial land uses. An analysis was done on the typical soils found within the proposed mining
area, to be used for topsoil and subsoil. The soils will be stripped ahead of mining and used to create the
rehabilitated cover. Infiltration modelling was conducted to determine the rates of seepage through the
opencast backfill, for different soil cover thicknesses. These seepage rates were then applied to the mine water
balance model to estimate the effect on the overall water make and consequently the water treatment
requirements. From these analyses, it was deduced that as the topsoil thickness was increased, there was a
marked decrease in the infiltration rate, and consequently the mine water make and required treatment rate.
However, for subsoil thicknesses greater than 500 to 1,000 mm, no noticeable decrease in infiltration rate
was realised for increased topsoil thicknesses. Various scenarios of topsoil and subsoil thickness were
modelled to estimate the impact of layer thickness on the mine water make. The modelling indicates that a
reduction in post closure mine water make of between 15% and 30% can be practically achieved at the mine,
with a reduction of up to 35% being achievable with thicker cover layers. This reduction in water make will
have a direct impact on operational costs, in terms of volumes of water pumped from the rehabilitated
workings, as well as water treatment costs both during operations and post closure.
The ingress of water to opencast coal mine workings, both during operations and post closure, is an ongoing
problem in the coal mining industry. This is particularly true in the Mpumalanga coal fields of South Africa,
where mines typically operate with a significant water surplus. The use of soil stripped ahead of mining is the
current best practice for capping of mine spoils. The objective of rehabilitation should be to produce a final
land form that is free draining and is sloped to minimise the risk of erosion, while having a topsoil cover that
has sufficient thickness and is of sufficient quality to minimise infiltration into the backfilled pit and provide
a sustainable growth medium for vegetation cover.

Potential impact of improved rehabilitation methods on the water balance of an opencast coal mine M.W. Palmer, A. van Zyl and C. Smit
In practice this is seldom achieved. In the Mpumalanga coal fields of South Africa, the rehabilitation of
opencast mine spoils is typically carried out as follows (Waygood and Ferreira, 2009):
 Material is stripped ahead of mining. In some instances the full depth of soils (soft material) is
stripped, but in many cases, only the volume required for the intended land capability post mining
is stripped. Based on arable land usage, as defined by the Chamber of Mines of South
Africa/Coaltech (2007), 600 mm or more of soil cover would generally be considered adequate.
 The stripped material is then placed over the spoils after they have been levelled, typically as one
mixed layer of soils and subsoils.
 The area is then grassed with a mix of vegetation species.
The thickness and quality of soil used as capping material varies according to the availability of soil, which is
often extremely limited due to poor utilisation of soils or only thin soils present in the pre-mining soil profile.
Good quality topsoil is often mixed with poor quality subsoils, making the establishment of sustained
vegetation cover difficult. Typically, topsoil is placed directly onto levelled and graded spoils in a layer 300 to
600 mm thick. In order to minimise materials movement, the rehabilitated land form often comprises a
combination of steep slopes on the ramp scars, which can be prone to erosion, and very gentle slopes in
between the ramp scars and on the free draining flow paths (often as gentle as 1:1000). Differential
settlement of the backfilled spoils can result in significant areas of ponding on the rehabilitated pit.
Anglo American Coal South Africa (AACSA) is in the process of developing a new opencast coal mine near
Ogies in the Mpumalanga Province of South Africa. The New Largo Colliery will supply coal to a new power
station, which is being constructed adjacent to the mine. The opencast workings will cover some 5,500 ha
and the associated surplus water make will need to be treated to meet the Resource Water Quality Objectives
(RWQO) for the relevant Catchment Management Unit, as specified by the South African Department of
Water and Sanitation (DWS). This will require a reverse osmosis (RO) water treatment plant.
AACSA aims to rehabilitate the New Largo opencast mine in such a manner so as to achieve a sustainable
final land form that will minimise the post closure environmental impacts and liabilities, minimise the post
closure water make (and therefore water management costs) and will be an asset to the company (or
subsequent land owner) post closure, by designing for a land use that will have economic value, rather than
just a wilderness area.
It is fortuitous that the in-situ soils at New Largo are relatively deep, with a well-developed topsoil layer
overlying deep residual sandstone subsoils. With careful materials management and rehabilitation planning,
a rehabilitated soil profile of sufficient thickness to support agricultural activity and to minimise infiltration
of water to the workings, should easily be achieved. AACSA plans to provide a dual layer cover to the
backfilled mining spoils by first placing a layer of clayey subsoils over the waste rock backfill, followed by a
final capping layer of topsoil.
Infiltration modelling of seepage through various layer thickness combinations on the opencast backfill was
conducted in order to evaluate the benefits in terms of reduced water make vs increased rehabilitation costs
as input to the rehabilitation strategy for the mine. The infiltration rates obtained were then applied to the
mine water balance model. In this way the effect of variations in topsoil and subsoil thickness on the mine
water make and required water treatment rate was assessed.
The following methodology was employed for the investigation:

 The soils specialist report (Earth Science Solutions (ESS), 2011) was reviewed and key soil
types/units on the proposed mining footprint were identified.
 Soil sampling and test pitting was performed at a location where the dominant soil units were
encountered.

 Laboratory testing was performed on the soil samples, including Atterberg limits, density,
permeability and soil retentivity.
 Infiltration modelling was conducted based on the soil parameters obtained from the laboratory
tests, to estimate the seepage rate for various soil thicknesses.
 A detailed mine water balance model was compiled to model the mine water make on a monthly
time step over the entire life of mine and post closure.
 The results of the infiltration modelling were applied to the water balance and various topsoil and
subsoil thickness scenarios were run and compared.
The soils specialist report (ESS, 2011), prepared for the New Largo environmental applications, was reviewed
to identify the key soil types/units found within the mining area. This was used to identify four locations at
which test pits were excavated for soil sampling purposes.
The agricultural soil mapping (ESS, 2011) and test pitting undertaken for this study indicated that moderate
to deep sandy soils (comprising a hillwash sand) cover the majority of the proposed opencast area. These
soils are characterised largely as Hutton and Clovelly soils and will be used as the final capping/cover layer in
the rehabilitation process. They are the “arable” nutrient rich soils typically encountered in the upper 1.0 to
1.5 m of the profile.
The material underlying the agricultural topsoils (hillwash horizons) generally comprises reworked residual
sandstone or siltstone, clayey sands to sandy silts, derived from in situ weathering of the siltstones and
sandstones of the Karoo Sequence. The initial material placed immediately on top of the waste rock backfill
will be derived from this material, which are characterised by clayey sands to sandy clays that typically exhibit
low nutrient properties.
The above two horizons are the soft overburden soils that typically would be used as cover materials in
rehabilitation programmes.
In addition, some hydromorphic soils (with clays) are present in the low lying alluvial and gully wash areas.
The soil properties, as determined from the laboratory tests, are presented in Table 1.
The bulk density of 1,691 kg/m3 represents a density equivalent to 85% Proctor Density. The permeability at
a remoulded density of 1,500 kg/m3 is considered to represent conditions where the soil has been “dumped”,
i.e. loose bulk density. However, the physical characteristics, such as the grading and Atterberg Limits suggest
that this hillwash material will rework and compact fairly easily. Consequently, an additional permeability
test was undertaken on a sample compacted to 93% Proctor Density. This compactive effort is considered a
reasonable but probably conservative value that will be achieved during the rehabilitation, as some
reworking of the material is likely (i.e. during placing, dozing and levelling, and some compaction is expected
to occur). The permeability at 93% Proctor (4.4 × 10-6 cm/s) was therefore considered a feasible parameter
to use in the modelling, while the density of 1,500 kg/m3, representing a low compactive effort of 75% Proctor
Density, was considered unlikely.

Hillwash Reworked residual soils

Material property
(topsoil) (subsoil)
Liquid Limit SP 38 to 42
Plasticity Index SP 15 to 21
Gravel % 1 to 2 0
Sand % 70 to 80 36 to 46
Silt % 9 to 15 9 to 16
Clay % 10 to 12 45 to 50
Proctor Maximum Dry Density (kg/m3) 1,982 1,555
Bulk Density (kg/m3) 1,691 1,713
3
Dry Density (kg/m ) 1,624 1,367
Permeability (cm/s):
- at Dry Density 1.5 × 10-5 1.1 × 10-5
- at remoulded to approximately 1,500 kg/m3 2.3 × 10-5 4.9 × 10-7
- at 93% Proctor Maximum Dry Density 4.4 × 10-6 −
8.3 20.7
Field capacity (-33 kPa) (% m/m)
(7.8 to 8.9) (19.6 to 23.0)
4.5 15.2
Wilting point (-1,500 kPa) (% m/m)
(4.1 to 5.0) (13.6 to 17.2)
The densities of these soils can range from 1,350 kg/m3 to 1,713 kg/m3 and the permeability of the in situ
material is in the order of 1.1 × 10-5 cm/s. When remoulded to a density of approximately 1,500 kg/m3
(approximately 96% Proctor Density), the permeability reduced to 5 × 10-7 cm/s. The Unified Classification of
the reworked residual material is CL (C = clay; L = low plasticity) and this material generally exhibits a fair to
good relative workability (Sherard et al., 1967) and also low permeabilities. Consequently, a permeability of
4.9 × 10-7 cm/s was considered likely to be achieved.
The site climate data sets were based on data obtained from climatic stations in the areas surrounding the
New Largo site. The mean annual precipitation (MAP) for the site is 715 mm, while the mean annual Symons
Pan (S Pan) evaporation (MAE) amounts to 1,650 mm.
Modelling of the unsaturated flow through the rehabilitated cover was carried out in order to determine
effective infiltration rates to the opencast backfill. The key objectives of the cover through-flow modelling
were as follows:

 To model the expected cover through-flow rates based on site climatic conditions, hydraulic
properties of topsoil cover and backfilled materials, vegetation characteristics and cover thickness.
 To determine the optimum topsoil cover thickness in minimising moisture ingress into the backfilled
opencast pit spoil material, based on the cover through-flow rates.
The modelling approach considered one dimensional (1-D) flow profiles through an evapotranspirative (store
and release) cover and underlying backfill material. Cover through-flow rates were predicted from the
interaction between site climatic conditions, hydraulic properties of the cover and backfill materials,
vegetation characteristics and cover thickness for the 28 cover configuration options shown in Table 2.
Cover Layer thickness Number of

Layer Material
system (mm) scenarios
Single Hillwash
Surface 300 | 600 | 900 | 1,200 4
layer (topsoil)
Hillwash
Surface 300 | 600 | 900 | 1,200
(topsoil)
Double
24
layer Residual material
Sub surface 300 | 500 | 1,000 | 1,500 | 2,500 | 3,500
( subsoil)
Unsaturated flows were simulated with the 1-D finite element variable saturated-unsaturated flow model
code of SVFlux (Soil Vision Systems, 2012).
Climate data sets were prepared representing an average rainfall year and an 18 year time series of recorded
meteorological data. The climate data sets were based on a hydrological year running from October to
September, with October representing the onset of the summer rainy season.
Daily rainfall, temperature and relative humidity data from the meteorological station at Witbank and daily
S-pan evaporation data from the Bronkhorstspruit dam meteorological station was used to prepare the
18 year climate data set.
A data set for an average rainfall year was prepared, which was used in multiple iterations to establish climate
equilibrium conditions for the modelling profiles before the 18 year time series was used for estimating the
cover through-flow rates.
The properties of the materials used are summarised in Table 1. The sandy hillwash material from the upper
1.0 to 1.5 m of the profile is more suitable for use as the cover surface layer and as a plant growth medium,
having a higher soil fertility and organic matter content and lower risk of desiccation cracking than the
underlying reworked residual materials.
The clayey reworked residual materials with lower soil fertility are more suitable as a store and release layer
below the hillwash surface layer due to their higher water retention and moisture storage capacity.
Consequently, larger volumes of infiltrated rain could temporarily be stored during wet periods for
evapotranspiration during drier periods. The lower permeability of the reworked residual materials will also

result in a shallower wetting front during wet periods. These materials have a higher risk of development of
cracks with intense drying and wetting if used as the surface layer.
SVFlux requires the following material property functions:
 Water retention function: Describes a material’s ability to store and release water as the material
de-saturates and also the capillary potential of the material.
 Hydraulic conductivity function: Describes the material’s ability to conduct water under saturated
and unsaturated conditions.
The water retention function for the cover materials was defined by applying the Fredlund and Xing (1994)
closed-form solution method to the water retentivity test results. The SoilVision database was used to predict
the water retention function for the opencast spoils backfill material through the use of pedo-transfer
functions (Fredlund, 2011). The prediction was based on a typical opencast spoils particle size distribution,
derived from testing at active mines in the region, for the < 4.75 mm fraction of the spoils material and the
correction for gravel contents according to the method of Hendrickx et al. (1991). Experience has shown that
a gravel content of 70% can be assumed for the backfilled spoils material.
The hydraulic conductivity function was defined from the water retention function and the permeability of
the material by applying the Fredlund and Xing (1994) closed-form solution method. Permeabilities of
1.5 × 10-5, 4.4 × 10 -6 and 4.9 × 10 -7 cm/s were used to define the hydraulic conductivity functions for a 150 mm
surface layer with a bulk density of 1,700 kg/m3, and Proctor maximum densities for the hillwash and
reworked residual materials respectively. The density of the surface layer is based on the Guidelines for the
Rehabilitation of Mined Land (Chamber of Mines of South Africa/Coaltech, 2007), which recommends that
compacted surface layers should be ripped to achieve a maximum density of 1,700 kg/m3 for root
development for sandy materials such as the hillwash. The permeability at 93% Proctor was used for the
hillwash and reworked residual cover materials since some compaction is expected during the placing, dozing
and levelling of the cover materials during rehabilitation.
SVFlux simulates plant water uptake from the root zone and the transpiration thereof on a daily basis based
on the interaction between:
 Potential evaporation which relates to the atmospheric evaporative demand applied to the leaves.
 Vegetation characteristics such as:
○ Leaf area that relates to the area from where transpiration occurs.
○ Root zone profile that relates to the potential moisture uptake through the roots.
○ Plant water limiting function that relates to the soil suction when transpiration starts to
decrease till plant wilting.
 Matrix suction of the root zone material that relates to the material’s moisture availability.
The vegetation characteristics applied in SVFlux for re-vegetated annual grasses are based on information
collated from literature for grasslands and rehabilitated land since this information was not available (Dye et
al., 2008; Jarmain et al., 2004; Versfeld et al., 1998).
The seasonal leaf area index (LAI) was simulated based on the growing season that commences mid-
September till senescence at end April. Peak LAI was simulated from end December till end February
whereafter it declined to a LAI of 0 during the winter. The peak LAIs were varied from 1.5 for dry years to 2.7
for wet years to account for the expected variability in the condition of the vegetation during average rainfall,
dry and wet years.
A rooting depth of 300 mm was specified for the 300 mm hillwash surface layer. This was increased to
450 mm for the 1,200 mm hillwash layer. A triangular root distribution was specified with a denser root

distribution in the upper section of the root system. A linear plant water limiting function was defined
between the moisture limiting point of -100 kPa, at which transpiration starts to decrease, till wilting point
at -1,500 kPa.
The assumptions on which the unsaturated flow modelling was based are the following:
 Sample representativeness. The defined material property functions are spatially representative of
the respective hillwash and reworked residual cover materials.
 Flow profile representativeness. The modelled one dimensional flow profiles are spatially
representative of the flows that would occur at the rehabilitated backfilled opencast pit.
 Preferential flow processes. Flows were simulated as matrix flows through a material that does not
account for preferential flows such as flow through cracks and burrow holes.
 Material hydraulic properties. Water retention and permeability of the cover materials are
governed by the particle size distribution and not by clods or cracks.
 Loss of cover material. The effect of a decreasing cover thickness due to erosion or loss of cover
material into the larger pores of the underlying backfill material on cover through-flow rates is not
accounted for.
The cover through-flow rates represent the net infiltration / moisture ingress into the backfill material after
accounting for rain infiltration, evaporation, plant transpiration and changes in material moisture storage.
The cover through-flow rates relate to climate equilibrated conditions. The likely cover through-flow rates
are presented in Table 3 as recharge volumes in mm/year to the opencast workings and as a percentage of
MAP in Table 4. The results are shown graphically in Figure 1.
Thickness of Cover through-flow rate (mm/y) for

hillwash (topsoil) thickness of residual material (subsoil) (mm)
(mm) 0 300 500 1,000 1,500 2,500 3,500
300 148 83 67 63 61 59 57
600 101 72 59 55 54 52 50
900 84 66 61 60 57 57 57
1,200 68 61 60 57 57 55 53
The unsaturated flow model was calibrated for the single cover layer system against the results from a study
with recorded outflow rates at an experimental site near Newcastle, KwaZulu-Natal. The experimental site
was setup to represent rehabilitated areas of opencast coal mines. Outflow rates were continuously recorded
with tipping buckets over a period of 7 years for leachate collected at the base of constructed lysimeters
having a surface area of 100 m2 and a depth of 3 m (Vermaak et al., 2004).

Thickness of Cover through-flow rate (% of MAP) for

hillwash (topsoil) thickness of residual material (subsoil) (mm)
(mm) 0 300 500 1,000 1,500 2,500 3,500
300 20.6 11.6 9.3 8.9 8.5 8.3 8.0
600 14.2 10.1 8.2 7.7 7.6 7.4 7.0
900 11.7 9.2 8.5 8.3 8.0 7.9 7.9
1,200 9.5 8.6 8.3 8.0 7.9 7.6 7.5
20.0
Annual seepage rate (% of MAP)
17.5
15.0
12.5
10.0
7.5
5.0
0 500 1000 1500 2000 2500 3000 3500
Thickness of sub-soil (mm)
300 mm Topsoil 600 mm Topsoil 900 mm Topsoil 1200 mm Topsoil
The following inferences are made from the predicted cover through-flow rates:
 The rates for the single layer cover system are about 3 to 10 times higher than the expected natural
recharge of 2 to 3%, depending on cover thickness. This can be ascribed to the high sand fraction
and related higher permeability and lower water retention determined for the hillwash. The higher
permeability is associated with higher infiltration rates and a deeper wetting front during rain
events. The lower water retention is associated with lower water storage capacities resulting in
smaller amounts of infiltrated rain to be stored during wet periods. Low water retention is also
associated with low capillary potential resulting in smaller amounts of deep infiltrated rain moving
upwards to the surface for evapotranspiration.
 The through-flow rate for the single layer cover system decreases with increasing thickness due to
increasing moisture storage capacity of the hillwash layer.
 The optimum thickness for the single layer cover system with hillwash is at least 1,200 mm to
minimise cover through-flow to rates comparable with the double layer cover system.

 The rates for the double layer cover system are significantly lower when compared to the single
layer cover system. This can be ascribed to the lower permeability and higher water retention and
moisture storage capacity of the clayey reworked residual material. Consequently, a larger fraction
of infiltrated rain is stored in the cover during wet periods, which would move upwards for
evaporation and plant transpiration during dry periods.
 The through-flow rate for a double layer cover system decreases with decreasing thickness of the
hillwash layer. This can be ascribed to the larger interaction between the climatic evaporative
demand (potential evaporation) and the higher water retention and capillary potential of the
reworked residual materials, resulting in higher evapotranspirative losses.
 No significant decreases in cover through-flow rates were predicted for the reworked residual
material layer thicker than 1,000 mm.
 The optimum thicknesses of the hillwash and reworked residual material layers is about 600 to
900 mm of hillwash and 1,000 mm of reworked residual material, for a double layer cover system
that is re-vegetated with grass species.
 Cover through-flow rates estimated for a wet year are about 1.8 to 2.5 times higher than the likely
rates for an average rainfall year. This can be ascribed to the following, in addition to higher
precipitation volumes:
○ More rain days during the wet year compared to average rainfall years.
○ More consecutive rain days during the wet year.
○ Precipitation during the wet year is comparable to potential evaporation during the rainy
season.
The objectives of the mine water balance model are to estimate the mine water make over the life of mine,
accounting for the contributions from all components of the mine, including the opencast mine, historical
underground mining, dewatering operations and surface infrastructure, including mine residue deposits, as
well as the coal processing plant. The expected water surplus over the life of mine needs to be quantified to
facilitate the planning of water management measures and infrastructure, including dirty water storage
requirements and potentially water treatment rates.
The mine water balance was computed by means of a water balance model compiled in Microsoft Excel,
programmed in Visual Basic. Daily rainfall data from the selected rainfall station, together with monthly
evaporation data were entered into a hydrological model based on the Soil Conservation Service (SCS)
method to determine runoff on a daily basis using antecedent rainfall conditions. The method, as adapted to
South African conditions by Schmidt and Schulze (1987), is believed to be highly suitable to the site, having
been developed on catchments of around 8 km2.
From the daily rainfall and runoff, the monthly rainfall, runoff and evaporation were entered into the
spreadsheet-based model mentioned above for evaluation of inflows from the various mine components and
areas. The results of the cover through-flow modelling detailed above were input to the water balance model.
The groundwater inflows to the opencast mine were derived by JMA Consulting (2011).
The mine water balance was modelled on a monthly time step. This is considered adequate for water balance
modelling on large mines, where the long response time of the mine system to rainfall events makes analysis
on a daily time step meaningless.

Four combinations of topsoil and subsoil layers were selected, based on the results of the cover through-flow
modelling and these combinations were each modelled as a scenario in the water balance model. The
selected combinations are presented in Table 5.
Recharge (through-flow rate) Hillwash layer thickness Subsoil layer thickness

Scenario
(% of MAP) (mm) (mm)
Base case 14.2 600 0
1 11.7 900 0
2 8.3 900 1,000
3 7.5 1,200 3,500
The base case scenario represents “conventional” rehabilitation, while scenario 1 represents an
improvement on the “conventional” single layer rehabilitation design. Scenario 2 represents the most likely
average thicknesses of the two layers, while scenario 3 represents an extreme case in terms of layer thickness
for both layers.
Key input data to the water balance model included the following:
 Climatic information in the form of monthly rainfall and evaporation data for a 95 year period.
 Annualised surface areas of the various surface components of the mine, including historical
underground mining (for aerial based estimation of recharge), opencast workings (broken down
into active pit, ramps, spoil heaps, levelled spoils, rehabilitated spoils that are free draining and
rehabilitated spoils that are not free draining), dirty surface areas (coal processing plant, stockpiles,
mine residue facilities, run of mine tips, etc.).
 Water usage information (dust suppression, coal processing, etc.).
 Modelled surface runoff, cover through-flow (as per the scenarios outlined above) and
groundwater ingress parameters.
 Dirty water storage capacity (surface dams, in-spoils storage, etc.).
The following key assumptions were made:

 It was assumed that rehabilitation will take place concurrently, with a minimal lag on rehabilitation
(maximum 200 m width of spoil heaps and a further 200 m width of levelled spoils between the
advancing opencast pit and the rehabilitation area).
 All rehabilitation will be made free draining, both during operations and post closure.
The expected surplus mine water make over the life of mine for average rainfall, for each scenario modelled,
is shown graphically in Figure 2.

30000
28000
26000 Base case: Conventional rehab
600mm topsoil only
24000
Scenario 1: 900mm topsoil only
22000
20000
Water make (m3/day)
Scenario 2: 900mm topsoil

18000 1000mm sub-soil
16000 Scenario 3: 1200mm topsoil
14000 3500mm sub-soil
12000
10000
8000
Rehabilitation
Post Closure
6000
Phase
Phase
4000
Operational Phase
2000
0
2015
2018
2021
2024
2028
2031
2034
2037
2040
2043
2046
2049
2052
2055
2058
2061
2064
2067
2070
2073
Base Case Scenario 1 Scenario 2 Scenario 3
The estimated post closure water make, with the percentage reduction in water make, as compared with the
base case scenario, is presented in Table 6.
Topsoil Subsoil Water make Water make Percentage

Scenario
thickness (mm) thickness (mm) (Ml/day) (% of MAP) reduction
Base case 600 0 22.1 20.4 -
1 900 0 18.9 17.5 14
2 900 1,000 15.2 14.0 30
3 1,200 3,500 14.4 13.3 35
The modelling indicates that a reduction in post closure water make of around 30% can be reasonably
achieved at New Largo by implementing a double layer rehabilitation cover system, with the topsoil and
subsoil materials available on site.
The feasibility of implementing an improved cover system needs to be carefully evaluated against the
additional cost of the improved cover system, the long term cost savings on water treatment and the
associated brine management, as well as the long term economic benefit, in terms of agricultural activities,
that can be derived from the land by implementing an improved cover system.
The following conclusions were drawn from the study:

 The incremental impact of subsoil layer thickness on the cover through-flow rate becomes small for
layer thicknesses greater than 1,000 mm.
 The thickness of the topsoil layer has little defining effect on the through-flow rate when combined
with subsoil thicknesses greater than 500 to 1,000 mm.
 The optimum layer thickness for topsoil is therefore between 600 and 900 mm. To ensure adequate
rooting depth, it is suggested that a layer thickness greater than 750 mm be utilised.
 The optimum layer thickness for subsoil is 1,000 mm. Thicknesses greater than this will have little
additional effect on the infiltration rate.
 A reduction in post closure mine water make of around 30% could reasonably be achieved at New
Largo.
 A full cost-benefit analysis has still to be undertaken to assess the feasibility of the improved
rehabilitation cover system. This analysis must take account of the additional costs associated with
the improved cover system, the reduction in water treatment costs, the reduction in brine handling
costs, as well as the long term economic benefit that can be gained from the rehabilitated land.
Chamber of Mines of South Africa/Coaltech (2007) Guidelines for rehabilitation of mined land, Chamber of Mines of South Africa and
Coaltech Research Association.
Dye, P.J., Jarmain, C., le Maitre, D., Everson, C.S., Gush, M. and Clulow, A. (2008) Modelling vegetation water use for general
application in different categories of vegetation, Water Research Commission Report No. 1319/1/08, Water Research
Commission, Pretoria, South Africa.
Earth Science Solutions (2011) Anglo American Inyosi Coal (Pty) Ltd., New Largo Colliery, Specialist soils and land capability
assessment, EIA report SO 403-NLC-S01-00-Soil Specialist Assessment.
Fredlund, D.G. and Xing, A. (1994) Equations for the soil-water characteristic curve, Canadian Geotechnical Journal, Vol. 31, pp. 521–
532.
Fredlund, M. (2011) SoilVision: A knowledge-based soil database, SoilVision Systems, Ltd., Saskatoon, Canada.
Hendrickx, J.M.H., Khan, A.S., Bannink, M.H., Birch, D. and Kidd, C. (1991) Numerical analysis of groundwater recharge through stony
soils using limited data, Journal of Hydrology, Vol. 127(1-4), pp. 173–192.
JMA Consulting (2011) Groundwater specialist study report, Report No. New Largo/Groundwater/Ver-01/2011.
Schmidt, E.J. and Schulze, R.E. (1987) Flood volume and peak discharge from small catchments in South Africa, based on the SCS
technique, Agricultural Catchments Research Unit Report No. 24, Water Research Commission Report No. TT31/87, Water
Research Commission, Pretoria, South Africa.
Sherard, J.L., Woodward, R.J., Gizienski, S.F. and Clevenger, W.A. (1967) Earth and earth-rock dams: engineering problems of design
and construction, John Wiley & Sons, 3rd printing.
Soil Vision Systems (2012) SVFLUX Saturated/unsaturated finite element 1D/2D/3D seepage modelling software, SoilVision Systems
Ltd., Saskatoon, Canada.
Terrasim (2013) New Largo opencast backfill – cover through-flow modelling, Terrasim Report No. TSR1302/JW003.
Vermaak, J.J.G., Wates, J.A., Bezuidenhout, N. and Kgwale, D. (2004) The evaluation of soil covers used in the rehabilitation of coal
mines, Water Research Commission Report No. 1002/1/04, Water Research Commission, Pretoria, South Africa.
Waygood, C.G. and Ferreira, S. (2009) A review of the current strategy for capping mining spoils, Paper presented at Mine Water
2009, South Africa.

J. Straker Integral Ecology Group Ltd., Canada

T. Baker Integral Ecology Group Ltd., Canada
S.L. Barbour Civil and Geological Engineering, University of Saskatchewan, Canada
S. Carey School of Geography and Earth Sciences, McMaster University, Canada
D. Charest Teck Resources Limited, Canada
Understanding reclamation effects on surface water balances in mine-affected watersheds is critical to both
prediction of, and design for, water movement through the post-closure landscape, and for development of
appropriate reclamation and revegetation treatments for mine closure. Substantial effort has been invested
in increasing knowledge of the effects of mine-waste cover systems on key water-balance terms such as net
percolation, but tools to extend that knowledge to effects of cover systems on vegetation establishment and
the subsequent effects of these vegetation-substrate interactions on water-balance terms are lacking. The
concept of a “soil moisture regime” is used worldwide to understand edaphic conditions and plant
communities. However, in most applications, soil moisture regime is a relative or unquantified parameter
estimated from the presence of indicator plants or soil properties observed in natural ecosystems.
Applications of these approaches to post-mining landscapes are challenging because soils/surficial materials
are reconstructed and reference plant communities are often not fully re-established. Some quantitative
approaches to estimation of properties that influence soil moisture regime (e.g. available water storage
capacity) have been developed, but these are generally based on agricultural soil science, and have limited
utility to many post-mining materials.
The authors propose new methods for estimating soil moisture regime in post-closure landscapes, using
concepts from existing biogeoclimatic ecosystem classification systems and analyses of effects of particle-size
distribution on soil water retention. Key variables in the proposed estimation model include regional and local
climate, material particle-size distributions (including distributions typical of mine-waste materials), organic-
matter accumulation, and topography. This paper discusses methods development of this estimation
approach, and presents suggestions for broader application of the approach for quantifying surface-water-
balance components (e.g. net percolation) and for reclamation planning (e.g. revegetation species selection)
in closure landscapes.
One of the principal knowledge gaps in mine reclamation involves characterising soil water dynamics within
surficial materials used in reclamation cover systems. The concomitant water balance in these materials is a
dominant control on both ecosystem development and watershed performance, and better understanding
of these water balances is critical for improved reclamation planning and projection of long-term
characteristics of reclaimed ecosystems. It is also critical for improving our understanding of the hydrologic
behaviour of reconstructed mine-affected watersheds, including the role of net percolation on flushing of
constituents of interest (CIs) from the mine wastes within these watersheds. Although substantial effort has
been invested from an engineering perspective in investigating the effects of cover systems on rates of net
percolation, this effort has generally not been coupled with understanding soil water dynamics within the

Mine reclamation and surface water balances: ecohydrologic classification J. Straker, T. Baker, S.L. Barbour, M. O’Kane, S. Carey and D. Charest
same cover systems for the purposes of reclamation planning and execution. To date most ecological
approaches used to address the above knowledge gap, where it has been addressed at all, have been
borrowed from ecosystem classification systems, and are limited by the following factors:
 they are qualitative or semi-quantitative, and there is limited attempt to evaluate and
demonstrate their hydrologic validity;
 they rely substantially on the presence of existing natural vegetation communities to provide
information on edaphic conditions — these techniques are not applicable to mine-reclamation
settings where vegetation communities are absent or introduced; and/or
 they have a narrow focus on a single aspect of the surface water balance, such as estimating
water retention for revegetation planning. In these approaches, there is no attempt to provide a
more comprehensive understanding of surface water balances, or of how water retention and use
by vegetation may influence deeper percolation and the water balance of the underlying mine-
waste landform.
This paper presents a quantitative method to relate the properties of landforms and cover-system materials
to plant-available water and surface water balances, based on particle-size distributions (PSDs) and
biogeoclimatic ecosystem classification (BEC).
BEC is an ecological classification system developed primarily in the western Canadian province of British
Columbia (B.C.), in which biogeoclimatic units (“zones”) represent broad geographic areas of similar
macroclimate and are recognised as influencing the biological characteristics of the resulting ecosystems
(Meidinger and Pojar, 1991). In the BEC system, biogeoclimatic zones can be subdivided into subzones, which
can in turn be subdivided into variants, with each subdivision representing a reduction in climatic variability
and geographic area (Lloyd et al., 1990). Within each subzone or variant, there are sequences of distinct
ecosystems (“site series”), with associated vegetation communities reflecting differences in topography and
soil depth, texture, drainage, moisture regime and nutrient regime. In this system, soil water availability is
believed to have the greatest influence on ecosystem development. This availability is in part determined by
climate, but since climate is relatively uniform within a biogeoclimatic subzone or variant, variation in soil
water availability at this level of classification results from influences of soil and topography on surface water
balances (Lloyd et al., 1990). These influences are manifested in the resulting plant associations, that is, each
site series has an assemblage of plants adapted to its edaphic conditions — a fundamental principle of the
BEC system is that sites with similar physical properties have similar vegetation potential (Meidinger and
Pojar, 1991). A subset of plants on a site — “indicator plants” — are diagnostic of edaphic conditions owing
to their adaptation to narrow ranges of conditions such as soil water availability.
In the BEC system, soil water availability is estimated using a concept termed “soil moisture regime” (SMR),
which reflects “the average amount of soil water annually available for evapotranspiration by vascular plants
over an extended period of time (several years)” (Pojar et al., 1985). The BEC system incorporates nine SMR
classes ranging from driest (Class 0, or very xeric) to wettest (Class 8, or hydric); this spectrum is referred to
as a hygrotope (Pojar et al., 1985; Meidinger and Pojar, 1991). The most common classifications of hygrotope
— and those used in the BEC system — are classifications of potential hygrotope based on subjective
inferences from site and/or vegetation features (e.g. source of water, rate of water removal, slope position,
soil textural class), and represent a relative ranking of sites in terms of potential soil water availability.
Meidinger and Pojar (1991) provide a common example of this approach, although more complex and semi-
quantified examples exist in the BEC system (e.g. Lloyd et al., 1990). Actual hygrotopes integrate the above

information with climate and surface-water-balance inputs and losses such as precipitation and
evapotranspiration to provide estimates of absolute rather than relative water availability.
Quantified estimates of both potential and actual hygrotope or SMR are uncommon, and those that exist are
limited in their application to specific geographic regions and/or ecosystems (e.g. Waring and Major, 1964).
In B.C., Green et al. (1984, summarised in Pojar et al., 1985) used a water-balance approach to develop an
actual hygrotope, but it is based on intra-annual duration of water deficits and on presence of water tables,
and although the authors provide defining features for their classes, methods for classifying sites according
to this system are not provided.
Various land-capability classification systems in Canada — beginning with agricultural land-capability systems
— have used available water storage capacity (AWSC) as an index of potential soil-water availability. AWSC
is defined as the volume of water per unit area held within the active or rooting zone of the soil profile
between the volumetric water content (VWC) at field capacity (FC) and the permanent wilting point (PWP).
The FC is the VWC at which the rate of gravitational drainage becomes negligible relative to the current rate
of evaporation or evapotranspiration (Zettl, 2014). This water content is often taken to be the water content
at negative pore-water pressures of 10–33 kPa, depending on soil texture. The PWP is the VWC at which soil
water is no longer available for plant uptake. Although this water content varies by plant species, by
convention it is defined as the water content at a negative pore-water pressure of 1500 kPa.
AWSC is generally expressed as a depth of water (mm) over a specified soil depth, or as a depth of water per
unit depth of soil (mm water/cm soil). A common practice has been to assign AWSC values based on soil
texture: for example, the document Land capability classification for agriculture in British Columbia (B.C.
Ministry of Environment, 1983) provides AWSC values in mm water/cm of soil depth for soils of different
textural classes. However, these systems, being initially focused on agriculture, do not link AWSC to SMR or
to the occurrence of typical natural ecosystems and/or larger hydrologic performance.
In northeast Alberta, the Land capability classification system for forest ecosystems in the oil sands (or “LCCS”
— Cumulative Environmental Management Association, 2006; first published in 1996) attempted to use
earlier concepts (and values) of assigning AWSC to textural classes for application to mine-reclamation and
forest-ecosystem settings. The LCCS equates a potential hygrotope to numeric values calculated from
texture-class-based AWSC and some topography and surficial-material-depth modifiers such as slope
position and depth to impermeable layers. This approach represents an advancement in producing an
objective and quantified relative hygrotope, but still has limitations for broader application. First, texture-
based AWSC values in the LCCS apply uniformly across texture or material classes, and do not recognise or
account for variation in PSDs within these classes. For instance, the LCCS applies an AWSC value of 1.0 mm/cm
to oil sands tailings, regardless of actual PSD and whether these tailings are complete or are cyclone overflow
or underflow products. Second, although there has been substantial investigation and validation of the LCCS
AWSC values (e.g. Barbour et al., 2010), and thus of their use as a relative hygrotope, there has been limited
evaluation of the relationship between these values and actual soil water contents (i.e. the actual hygrotope),
and of the relationship between these values and ecosystem development and landscape/watershed
hydrologic performance.
The concept of SMR has been applied globally based on duration or magnitude of growing-season water
deficits but typically involves relatively broad classes that can be mapped at a continental scale (e.g. Soil
Survey Staff, 1999) rather than applied to differentiate between ecosystems and hydrologic behaviours at a
local or regional scale.
The classification system proposed here is substantially informed by the biogeoclimatic, hygrotope/SMR, and
land-capability classifications described above, but is intended to derive estimates of plant-available water,
surface-water-balance performance, and associated ecosystem characteristics from landscape, landform and

surficial-material properties using objective and quantified methods that can be consistently and easily
applied. Further, the proposed system is designed to be broadly applicable to a range of climatic,
physiographic and surficial-material conditions (i.e. globally), yet have sufficient resolution to differentiate
ecosystem characteristics and hydrologic performance on a local scale. Additional goals for the classification
system are that it
 be capable of derivation solely from information on material properties, topography and climate,
and not rely on observations of intact above-ground ecosystems for diagnosis;
 be capable of evaluation and validation or adjustment through analysis of related empirical
observations, including relationships with non-mine ecosystems classified through standard BEC
methods; and
 provide useful interpretations for a range of mine-planning and reclamation-management
considerations, including both cover placement/revegetation and understanding hydrologic
behaviour on the mine landform-landscape-watershed scale.
The proposed classification framework is based on three primary factors with decreasing geographic scales
of application (Table 1, adapted from Devito et al., 2005). For the first classification factor, AWSC is
determined from PSDs of materials in the upper one metre of surficial material. This determination can be
applied globally, as it is based on universal principles of soil physics. The next classification factor involves
modification of the profile AWSC estimate for a topography-based energy regime. These modifications are
specific to latitudinal ranges and thus must be developed specifically on the continental to sub-continental
scale. The final classification factor applies regional and local climate information to the potential hygrotope
resulting from application of the first two factors to generate an actual hygrotope and identify ecosystems
associated with this hygrotope. Thus, application of the first classification factor (AWSC) requires only
information contained in this paper; application of the second factor (topography/energy) may require
modification of information contained in this paper, depending on latitude of application; and application of
the third factor (climate) requires information on climate local to the application site and biogeoclimatic or
similar classification information.
AWSC is estimated from PSD of the sub-100-mm fraction, using the following classes: clay (<0.005 mm), silt
(0.005–0.05 mm), sand (0.05–2 mm), Unified System of Soil Classification (USSC) sand (2–4.75 mm), and
coarse fragments (>4.75 mm). To facilitate more cost-effective and reliable classification, low-technology
field equipment has been developed to allow rapid determination of the cobble-and-gravel separate (>4.75
mm) based on a large volume of material, with subsequent determination of the finer fractions based on
laboratory analyses of smaller subsamples.

Scale of
Factor Range of factor applicability Classification outputs
High silt and clay High sand, cobble and
PSD of
contents, low sand, gravel contents, low silt Profile AWSC in
surficial Global
gravel and cobble and clay contents: surface 1 m
materials
contents: higher AWSC lower AWSC
Tropical and sub-
High latitude: slope and
tropical: slope and Continental Adjusted AWSC;
Topography aspect significantly
aspect do not to sub- relative SMR or
and energy affect energy
significantly affect continental potential hygrotope
distribution
energy distribution
Dry, arid to sub-humid
(P < PET) Wet, humid (P > PET) Actual SMR and
hygrotope;
Regional and storage and ET runoff and NP Regional to
identification of
local climate dominant dominant local
associated
runoff and NP may be ecosystems
reduced
The AWSC of each material profile is calculated in two steps from PSD data. First, the AWSC of the fine
fraction (<2 mm) is calculated using the empirically derived soil-water-characteristic equations developed by
Saxton and Rawls (2006). (Bulk density for these equations is assumed to be 1.44 T/m 3, which is based on a
database of mine-waste material properties, as is the estimated rock particle density of 2.71 T/m3.) These
equations include adjustment for organic matter content, which is measured at all sites. In the second step,
additional AWSC is assigned to the USSC sand fraction (2–4.75 mm) in recognition of the potentially
substantial volume of water held within the pores of rocks, particularly those of sedimentary origin (Waring
and Major, 1964; Flint and Childs, 1984; Childs and Flint, 1990). Rock porosity is calculated using measured
rock bulk-density values, which are converted to AWSC estimates based on Flint and Child’s (1984) work that
shows 50% of the internal porosity of rocks is released at suctions between 10 and 1500 kPa. Coarse
fragments (>4.75 mm) are assumed to contribute negligible AWSC to the overall profile given the physical
inaccessibility of the majority of their internally stored water.
AWSC estimates from the Saxton-Rawls equations were compared to results generated from application of
Arya and Paris’s (1981) physicoempirical model to predict soil moisture characteristics from PSD data, which
are known to be useful for coarse-textured mine-waste materials. For this comparison, we analysed data for
a wide range of materials, representing a range of PSDs and gradations, and including the 12 study sites
discussed further in Section 4. The comparison of the two methods indicates R2 values of 0.66 to 0.90,
depending on assumed gradation, and indicates a tendency for the Saxton-Rawls methods to generate
slightly lower estimated AWSC than the Arya-Paris methods. An example of a comparison for materials with
an assumed uneven gradation favouring coarser particles is presented in Figure 1. The relatively good
agreement between the two models was used to support continuation of use of the Saxton-Rawls methods,
as these methods incorporate other useful factors such as organic-matter content, but we will continue to
evaluate different approaches to estimating AWSC, and to investigate zones of poorer agreement between
the Saxton-Rawls and Arya-Paris approaches.

Values derived from the above AWSC estimation methods are intended to represent a single material, and
to be aggregated across a standard material profile or control section (typically 100 cm, but lesser sections
could be used if stipulated). For instance, to estimate AWSC for a 50 cm soil cover placed on mining waste
rock or tailings, one AWSC value is calculated for the cover material, another is calculated for the mine-waste
material, and an aggregate AWSC is generated by summing the values. If multiple layers are present within
the soil cover (or mine waste), then an AWSC value is calculated for each layer corresponding to depth and
PSD data. For natural soils, calculation is based on horizon depths and characteristics. In the case of shallow
soils over non-rooting-zone materials, the AWSC for the control section is based only on the depth of the soil
material, and thus is reduced compared to a 1 m potential rooting zone.
Although estimated AWSC values in mm water/cm material depth are based on continuous equations, their
results can be summarised in a standard engineering ternary diagram as presented in Figure 1. (The values
in Figure 1 assume well-graded materials. Some surficial and mine-waste materials deviate from this
assumption; for instance, oil sands tailings from the McMurray formation in northeast Alberta, Canada, are
predominantly fine sands, and research has indicated an AWSC for these materials of 1 mm/cm (Barbour et
al., 2010), as opposed to the lower values shown in Figure 1 for these PSD positions.) Data are presented in
this format to facilitate communication between mine planners/engineers (who often use USSC PSD
classification systems) and reclamation specialists (who often use soil-science PSD or texture classification
systems).

In the British Columbia BEC system, the topographic effect on energy is recognised through “warm” and
“cool” site modifiers. These modifiers are applied to slope angles greater than 25% (14°), with warm aspects
being southerly or westerly (135°–285°), and cool aspects being northerly to easterly (285°–135°; Resources
Inventory Committee, 1998). The concepts behind this approach were adapted to the current classification
system by using modelled solar radiation differences across latitudes, slopes and aspects to produce
modifiers (additions or deductions) to the PSD-derived AWSC estimate. Short-wave radiation was calculated
for different slopes, aspects and latitudes as the sum of the direct- and diffuse-beam components. The
theoretical direct-beam component of solar radiation was determined after Garnier and Ohmura (1968,
1970). Diffuse clear-sky radiation was calculated after Iqbal (1983), assuming a standard atmosphere. These
modelled values were then converted to additions or deductions to AWSC (in mm) by treating values above
the median for each latitude band as deductions, and values below the median as additions. These modifiers
are intended not to imply actual reductions or additions to AWSC on different slopes and aspects but as
surrogate modifiers to AWSC based on increased or decreased evapotranspirative demand driven by varying
energy regimes.

An example of the described approach for 50° N latitude is presented in Table 2. Maximum and minimum
values for modifiers at this latitude have been set at 30 mm, which reflects a shift of one entire SMR class
(Section 3.5). Maxima and minima increase with increasing latitude and decrease to zero as latitude
decreases to approximately 23.5° N or S.
Energy
class Class definition AWSC modifier
Neutral Slopes <10°1 none, 0
Warm Slope >10°; aspects ~080-280°1 Calculated AWSC minus 1-30 mm2
Cool Slope >10°; aspects ~280°-080°1 Calculated AWSC plus 1-30 mm2
Note that specific classification depends on slope/aspect combination: some slopes greater than 10° have
neutral aspects, and aspects may vary between warm and cool classifications depending on slope. Modifier
values depend on specific slope and aspect, where the maximum value for warm slopes occurs at slope angles
of 25° and aspects of 170–190°, and the maximum value for cool slopes occurs at slope angles of 35° and
aspects of 350–010°.
Adjusted AWSC values (PSD-based AWSC plus any applicable energy modifiers from Table 2) are used to
determine SMR, as outlined in Table 3. This table uses the SMR classes of the BEC potential hygrotope, but
replaces the relative ranking of various criteria with quantified ranges of adjusted AWSC. AWSC ranges for
each SMR class are modified from the oil sands reclamation land capability classification system discussed
above. The AWSC method for SMR determination applies only to upland (very xeric to mesic) SMRs, as wetter
SMRs require input of seepage water or the presence of a water table within 100 cm of the soil surface, and
are not dependent on soil storage. Thus, determination of SMRs wetter than mesic in this system is based on
observations of shallow groundwater seepage and/or the presence of a water table within the top metre of
surficial materials. Note that these moisture regimes are intended to reflect dominant soil-water conditions
over a multi-year period, consistent with the BEC system hygrotope.
1
Specific classification depends on slope/aspect combination – some slopes >10° have neutral aspects, and aspects may vary between
warm and cool classifications depending on slope.
2
Modifier values depend on specific slope and aspect, where the maximum value for warm slopes occurs at slope angles of 25° and
aspects of 170-190°, and maximum value for cool slopes occurs at slope angles of 35° and aspects of 350-010°.

Water-table depth
Primary water (cm below ground Available water storage, surface 1 m
SMR source surface) (mm)
Precipitation and soil
Very Xeric (0) >100 <60
storage
Xeric (1) >100 60-89
storage
Subxeric (2) >100 90-119
storage
Submesic (3) >100 120-149
storage
Mesic (4) >100 >150
storage
Precipitation and
Subhygric (5) >100 >150, seepage contributes to supply
seepage
Hygric (6) Seepage 30-100 n/a
Seepage or
Subhydric (7) permanent water 0-30 n/a
table
Permanent water Water table permanently
Hydric (8) n/a
table at or above soil surface
The methods discussed above were developed and tested at reclamation-monitoring sites at seven mining
operations from 2012 to 2014: five steelmaking coal operations operated by Teck Resources Limited in south-
eastern B.C. and west-central Alberta; at the Teck Highland Valley Copper Partnership’s Highland Valley
Copper mine in south-central B.C.; and at Thompson Creek Metals’ Endako molybdenum mine in central B.C.
The five Teck steelmaking coal mines are particularly relevant to testing, because in 2011 Teck commenced
an integrated and multi‐disciplinary applied research and development program focused on managing water
quality in mining‐affected watersheds. In 2012–13, this program included installation of soil and
meteorological instrumentation and soils and vegetation assessments at 12 reclamation sites at these coal
mines to provide data on reclamation conditions co‐located and concurrent with information on
meteorological and soil‐moisture variables at each study site. This instrumented-site network and the data it
provides supports increased understanding of how surface water balances and SMRs are affecting
reclamation responses over time, and vice versa, as well as how reclamation approaches affect reconstructed
landform water balances and watershed hydrology.
PSD data from 65 mine-reclamation and non-mine reference sites were used to estimate AWSC values based
on the methods described above. These values provide quantification of the potential hygrotope, as they
indicate the capacity for soil water storage (and eventual release as evapotranspiration, interflow and/or net
percolation), not actual storage. Actual storage is a product of the interaction between the potential

hygrotope and local climate, which delivers precipitation for storage and energy for evaporation and
transpiration. To evaluate the relationship between potential (calculated) and actual hygrotope, VWC and
matric-potential (φm) data collected from the instrumented study sites were analysed to derive mean
growing-season available volumetric water contents (AWC) for each site. To do so, the VWC at PWP was
calculated for each VWC sensor from interpolated plots of VWC against φm for each sensor pairing. In cases
where VWC continued to decrease at matric potentials drier than PWP, the mean of the lowest 10 VWC
measurements was used as an alternative VWC-at-PWP value in recognition of potential error in the paired
sensor measurements. For sensors without sufficient data, VWC-at-PWP was assigned from neighbouring
sensors in the same material type. Each sensor’s VWC-at-PWP value was then subtracted from each of its
VWC measurements to calculate AWC (water content above PWP) for all sensors. To calculate mean AWC
from all sensors over a profile depth, sensor AWC values were mathematically weighted according to the
relative depth interval represented by each sensor within the upper metre of soil. The reported AWC values
are the means of all daily measurements made during the 2013 and 2014 growing seasons, defined as the
period from April 1 to October 31 based on modelled PET for the study region.
SMR was assigned for each of the 65 study sites using the PSD-based AWSC estimates with energy modifiers,
as described above. For the Teck steelmaking coal mine research sites where AWC data are available,
calculated AWSC values were evaluated using mean growing-season AWC (Figure 2). These data show
support for the proposed classification system, with growing-season AWC significantly (p < 0.001) positively
correlated with estimated AWSC (Figure 2); the derived linear relationship accounts for approximately 80%
of variation in observed growing-season water contents. At least a portion of the unexplained variation is
attributable to the various stages of vegetation development on the sites and the resulting differences in
transpirative demand for soil water. It is anticipated that as these sites mature, fitting errors will be reduced.
On average, very xeric sites have less than 45% of the plant-available water that subxeric sites have during
the growing season. Research sites at Endako and Highland Valley Copper lack continuous measurement of
soil water contents, and thus cannot be added to this database. However, reference sites in these studies
provide some ability to evaluate system fit, as predicted SMR based on methods proposed in this paper can
be related to potential hygrotopic classification using standard subjective keys and the presence of indicator
plants. All reference sites studied to date are zonal site series with mesic SMR: the mean AWSC for these
sites estimated with the proposed methods is 159 mm, which places them in the mesic SMR category
according to the criteria presented in Table 3.
It should be noted that there are currently no comparable methods for estimation of ecohydrologic
behaviours and SMR at the sites used for methods testing. Estimation of soil water retention for design of
revegetation programs is conducted using the expert knowledge of individual practitioners. The effects of
surficial materials on surface water balances and underlying waste-material water balances are either not
considered or modelled for specific conditions using standard engineering approaches, which generally do
not provide information for other aspects of reclamation planning.

This paper proposes a quantified and objective hygrotopic classification system broadly applicable to a range
of ecosystems, including mine-waste-based landforms and mine-affected watersheds. Although the
proposed classification system is initially based on potential hygrotope, the use of regional and local BECs
allows its translation into actual hygrotopes based on regional and local climatic conditions. This translation
from potential to actual hygrotope has been tested in two regions of western Canada on instrumented
reclamation study sites. The initial results show promising relationships between predicted SMR using the
proposed classification system and mean growing-season available water contents calculated from
continuous measurement by in situ sensors, with increasing observed available water contents as SMRs
predicted by the classification model progress from drier to wetter sites. In addition, the proposed
classification system shows concordance with traditional ecosystem classification of non-mine reference
sites where classification is based on indicator-plant presence and topographic/soil relationships.
The potential management applications of the classification system include
 Reclamation and revegetation planning: using methods discussed above, SMR can be estimated
for existing or planned landforms and covers, and locally appropriate candidate vegetation
species adapted to these hygrotopic positions can be selected for reclamation.
 Assessment of pre- and post-development land capability: using estimated hygrotopic position
and the BEC system or similar approaches allows comparison of anticipated post-closure
ecosystems to pre-development inventories. These comparisons can then be used to evaluate the
effects of mining on dependent values such as wildlife habitat, biodiversity or land productivity,

and can provide the basis for application for custodial transfer of reclaimed lands or assessment
of such applications.
 Quantification of the effects of surficial-materials management on landform surface water
balances: when combined with information on local climatic conditions, the proposed
classification system can be developed to provide relative estimates of surface-water-balance
terms such as evapotranspiration and net percolation. Reclamation practitioners can use this
information to understand the effects of cover placement for reclamation on movement of water
through the surface layers of the reclaimed landscape.
This proposed classification system and empirical approach to its evaluation are evolving, and will continue
to be updated and adapted as additional information is collected and as the classification system is refined.
Alberta Agriculture and Rural Development (2009) Agricultural atlas of Alberta: agricultural climate elements, Alberta Ministry of
Agriculture and Rural Development, Edmonton, Canada.
Arya, L.M. and Paris, J.F. (1981) A physicoempirical model to predict the soil moisture characteristic from particle-size distribution
and bulk density data, Soil Science Society of America Journal, Vol. 45(6), pp. 1023–1030.
B.C. Ministry of Energy, Mines and Petroleum Resources (2008) Health, safety and reclamation code for mines in British Columbia,
Province of British Columbia, Victoria, Canada.
B.C. Ministry of Environment (1983) Land capability classification system for agriculture in British Columbia, Province of British
Columbia, Victoria, Canada.
Barbour, L., Chanasyk, D., Hendry, J., Leskiw, L., Macyk, T., Mendoza, C. … Welham, C. (2010) Soil capping research in the Athabasca
Oil Sands Region, volume 1: technology synthesis.
Childs, S. and Flint, A.L. (1990) Physical properties of forest soils containing rock fragments, in Sustained productivity of forest soils:
Proceedings of the 7th North American Forest Soils Conference, S.P. Gessel, D.S. Lacate, G.F. Weetman, R.F. Powers (eds),
UBC Faculty of Forestry Publication, Vancouver, Canada, pp. 95–121.
Cumulative Environmental Management Association (2006) Land capability classification system for forest ecosystems in the oil
sands, 3rd edition, Alberta Environment, Edmonton, Canada.
Devito, K., Creed, I., Gan, T., Mendoza, C., Petrone, R., Sillins, U. and Smerdon, B. (2005) A framework for broad-scale classification
of hydrologic response units on the Boreal Plain: is topography the last thing to consider? Hydrological Processes, Vol. 19, pp.
1705–1714.
Flint, A.L. and Childs, S. (1984) Physical properties of rock fragments and their effect on available water in skeletal soils, in Erosion
and Productivity of Soils Containing Rock Fragments, Soil Science Society of America Special Publication 13, pp. 91–103.
Garnier, B.J. and Ohmura, A. (1968) A method of calculating the direct shortwave radiation income of slopes, Journal of Applied
Meteorology, Vol. 7(5), pp. 796–800.
Garnier, B.J. and Ohmura, A. (1970) The evaluation of surface variations in solar radiation income, Solar Energy, Vol. 13(1), pp. 21–
34.
Green, R., Courtin, P., Klinka, K., Slaco, R. and Ray, C. (1984) Site diagnosis, tree species selection, and slashburning guidelines for the
Vancouver Forest Region, Province of British Columbia, Victoria, Canada.
Iqbal, M. (1983) An introduction to solar radiation, New York, USA: Academic Press.
Lloyd, D., Angove, K. Hope, G. and Thompson, C. (1990) A guide to site identification and interpretation for the Kamloops Forest
Region, Province of British Columbia, Victoria, Canada.
Meidinger, D. and Pojar, J. (1991) Ecosystems of British Columbia, Province of British Columbia, Victoria, Canada.
Pojar, J., Klinka, K. and Meidinger, D. (1985) Biogeoclimatic ecosystem classification in British Columbia, Ministry of Forests, Victoria,
Canada.
Resources Inventory Committee (1998) Standard for terrestrial ecosystem mapping in British Columbia, Province of British Columbia,
Victoria, Canada.
Saxton, K.E. and Rawls, W.J. (2006) Soil water characteristic estimates by texture and organic matter for hydrologic solutions, Soil
Science Society of America Journal, Vol. 70(5), pp. 1569–1578.
Soil Survey Staff (1999) Soil taxonomy, 2nd edition, Agriculture Handbook Number 436, United States Department of Agriculture,
Washington, D.C., USA.
Waring, R. and Major, J. (1964) Some vegetation of California coastal redwood regions in relation to gradients of moisture, nutrients,
light, and temperature, Ecological Monographs, Vol. 34, pp. 167–215.
Zettl, J. (2014) Infiltration and drainage through coarse layered soil: a study of natural and reclaimed soil profiles in the Oil Sands
Region, Alberta, Canada, MSc Dissertation, University of Saskatchewan, Canada.

J. Straker Integral Ecology Group Ltd., Canada

T. Baker Integral Ecology Group Ltd., Canada
R. Shurniak O’Kane Consultants Inc., Canada
S.L. Barbour Civil and Geological Engineering, University of Saskatchewan, Canada
S. Carey School of Geography and Earth Sciences, McMaster University, Canada
A primary challenge in mine reclamation and ecological aspects of mine closure is the estimation of water
retention by surficial materials (mine wastes and/or other materials used in mine-waste cover systems). The
magnitude and timing of this retention affect both the types of ecosystems (plant communities) that can be
established on a given reclaimed site or landform and the partitioning of meteoric waters on the site or
landform into runoff, evapotranspiration and net percolation. The disciplines of ecology/soil science and
engineering have knowledge applicable to addressing this problem, but the application of that knowledge is
subject to limitations, and the work and approaches are generally de-coupled between the disciplines. In
ecology, the principle means of estimating soil water retention (often termed the “soil water regime” or “soil
moisture regime”) rely on the presence of intact landscapes and plant communities, and on the inferred
behaviour of natural surficial materials and geology. In mine reclamation and closure settings, this inferential
evidence is lacking, which makes direct application of established tools for estimating soil water regimes
difficult.
In a companion paper, the authors propose new methods of estimating soil water regimes in post-closure
landscapes, using concepts from existing biogeoclimatic ecosystem classification systems and analyses of
effects of particle-size distribution on soil water retention. Key variables in the proposed estimation model
include regional and local climate, material particle-size distributions (including distributions typical of mine-
waste materials) and organic-matter accumulation, and topography. This paper discusses the testing of these
proposed methods on a global basis, using data on surficial-material physical properties and water retention
from mines in different geographic and climatic settings.
The water balance in mine-waste cover systems is a dominant control of both ecosystem development and
watershed performance, and better understanding of these water balances is critical to improved
reclamation planning and projection of long-term characteristics of reclaimed ecosystems. It is also critical
to improving our knowledge of the hydrologic behaviour of reconstructed mine-affected watersheds,
including the role of net percolation in flushing of constituents of interest (CIs) from the mine wastes within
these watersheds. In a companion paper (Straker et al., 2015), we present a quantitative method to relate
the properties of landforms and cover-system materials to plant-available water and surface water balances,
based on particle-size distributions (PSDs) and biogeoclimatic ecosystem classification. This paper tests that
method using data from a variety of mine sites and materials in North America and Australia.

Ecosystem reconstruction: Methods of estimating soil water regimes J. Straker, T. Baker, M. O’Kane, R. Shurniak, S.L. Barbour and S. Carey
The proposed classification framework is based on three primary factors, with decreasing geographic scales
of application (Table 1, adapted from Devito et al., 2005). For the first classification factor, available water
storage capacity (AWSC) is determined from PSDs of materials in the upper one metre of surficial material.
This determination can be applied globally, as it is based on universal principles of soil physics. The next
classification factor involves modification of the profile AWSC estimate for a topography-based energy
regime. These modifications are specific to latitudinal ranges, and thus are developed specifically at the
continental to sub-continental scale. The final classification factor applies regional and local climate
information to the potential hygrotope resulting from application of the first two factors to generate an
actual hygrotope and identify ecosystems associated with this hygrotope. (The potential hygrotope is a
relative scale of potential soil water availability. The actual hygrotope uses information on water fluxes to
produce a scale of absolute [quantified] soil water availability.) This paper applies the first two factors to
estimate relative soil water availability for study sites, and compares these estimates to actual observed
available water contents.
Scale of
Factor Range of factors Classification outputs
applicability
High silt and clay High sand, cobble and
PSD of
contents; low sand, gravel contents; low silt Profile AWSC in
surficial Global
gravel and cobble and clay contents: surface 1 m
materials
contents: higher AWSC lower AWSC
Tropical and sub-
High latitude: slope and
tropical: slope and Continental Adjusted AWSC;
Topography aspect significantly
aspect do not to sub- relative SMR or
and energy affect energy
significantly affect continental potential hygrotope
distribution
energy distribution
Dry, arid to sub-humid
(P < PET); Wet, humid (P > PET); Actual SMR and
hygrotope;
Regional and storage and ET runoff and NP Regional to
identification of
local climate dominant; dominant local
associated
runoff and NP may be ecosystems
reduced
The classification system discussed above (with detailed methods provided in Straker et al., 2015) was
evaluated through application to multi-year databases from sites instrumented with soil and meteorological
sensors. These data are generated from 39 cover-system profiles at 31 sites on 12 mines located in a range
of climates in Canada, the United States and Australia, and encompass a combined 247 years of monitoring
records. The general locations of these sites are shown in Figure 1, with site characteristics presented in Table
2. Mines in the study included uranium, oil sands, coal, and base and precious metals operations.


Mean Mean
Material
No. of Years of Slope Aspect annual annual PET
Mine Site description of
profiles record (%) (° true) temp. precip. (mm)
upper metre
(°C) (mm)
1 1 2 7, 8 12 70 −0.7 450 430 Sandy till
2 1 9 0 - −0.7 450 430 Sandy till
3 2 8, 9 0 - −0.7 450 430 Sandy till
2 4 1 14 14 320 0.8 470 570 Peat-mineral mix
(18 cm) over fine-
textured till
5 1 16 14 320 0.8 470 570 Peat-mineral mix
(17 cm) over fine-
textured till (17–30
cm) over shale
6 1 8 14 320 0.8 470 570 Peat-mineral mix
(22 cm) over fine-
textured till (22–49
cm) over shale
7 1 14 0 - 0.8 470 570 Peat-mineral mix
(18 cm) over fine-
textured till (18-87
cm) over shale
3 8 2 7, 8 11 220 4 900 520 Topsoil (15 cm)
over pit-run
material
4 9 1 7 0 - 17.5 1,270 1,700 Soil cover
10 1 7 0 - 17.5 1,270 1,700 Soil cover (78 cm)
over tailings sand
11 1 7 0 - 17.5 1,270 1,700 Soil cover
5 12 1 3 34 182 3.3 1,060 440 Waste rock
13 1 3 31 12 3.3 1,060 560 Regolith cover (50
cm) over waste
rock
14 1 3 29 160 1.6 920 510 Salvaged soil cover
(50 cm) over waste
rock
6 15 1 3 33 248 4.5 965 630 Waste rock
7 16 1 3 48 53 2.9 1,060 1,150 Waste rock
17 1 3 0 - 1.8 1,000 670 Salvaged soil cover
(50 cm) over waste
rock

Mean Mean
Material
No. of Years of Slope Aspect annual annual PET
Mine Site description of
profiles record (%) (° true) temp. precip. (mm)
upper metre
(°C) (mm)
8 18 1 3 25 295 2.8 950 600 Salvaged soil cover
19 1 2 25 300 2.6 1,040 600 Waste rock
20 1 3 45 204 3.1 720 570 Waste rock
21 2 3 45 204 3.1 1,080 570 Waste rock
9 22 1 3 6 259 2.6 1,280 590 Waste rock
23 1 3 11 182 2.6 1,240 590 Waste rock
24 1 3 6 162 3.1 720 620 Waste rock
10 25 3 4, 4, 5 0 - 28 1,380 2,220 Fine-textured
saprolite cover
material
11 26 1 2 0 - 21.3 820 1,650 Cover material
27 1 2 25 217 21.3 820 1,650 Cover material
28 1 7 0 - 21.3 820 1,650 Cover material (50
cm) over
compacted clay
liner
29 1 7 25 217 21.3 820 1,650 Cover material (50
cm) over
compacted clay
liner
12 30 2 6 2 - 31.3 320 1,550 Run-of-mine
materials
31 2 17, 18 0 - 31.3 320 1,550 Run-of-mine
materials
The following sections describe application of the first two classification factors from Table 1 to the study-
site data
AWSC was estimated from the PSD of the sub-100-mm fraction, using the following classes: clay (<0.005 mm),
silt (0.005–0.05 mm), sand (0.05–2 mm), Unified System of Soil Classification (USSC) sand (2–4.75 mm), and
coarse fragments (>4.75 mm). PSD data for all study sites were obtained by sieve analysis to a particle
diameter of 0.075 mm, with hydrometer analysis conducted on the finer fractions for samples with an
appreciable (>~20% of total sample weight) silt and clay content. For coarse samples that were not analysed
by hydrometer, the per cent of material passing the finest sieve size was assigned to particle size classes by
the relative proportion of each class within the unanalysed fraction. The PSD of all samples used in this study
is summarised in Figure 2. Data are presented in this format to facilitate communication between mine
planners/engineers (who often use USSC PSD classification systems) and reclamation specialists (who often
use soil-science PSD or texture classification systems).

The AWSC of each material profile is calculated from PSD data using the empirically derived soil-water-
characteristic equations developed by Saxton and Rawls (2006). These equations include adjustment for
organic matter content and bulk density. Organic matter content was measured at 17 of 31 study sites, and
estimated based on material type and depth for the remainder of sites using an internal database of organic
matter contents in mine wastes and reclamation cover soils. Bulk density and particle density (specific
gravity) were measured at 15 of 31 study sites, and the remainder were assigned from available data
according to material types. Mean bulk density for all measured materials (excluding shale overburden and
compacted clay layers) was 1.44 T/m3, and mean particle density was 2.71 T/m3. Available data for rock bulk
density were insufficient to calculate internal water-storage capacity in coarse fractions, and therefore all
sites were assumed to have zero coarse fraction AWSC. An AWSC value was calculated for each PSD sample,
and these AWSC values from individual samples were averaged for each material type at each site. An
aggregate AWSC for the upper metre of each profile was then generated by summing AWSC values for each
material type present, arithmetically weighted by the depth of each material.
A secondary mechanism through which PSD may influence soil water drainage and thus storage in both
natural soil profiles and constructed mine-waste cover systems is the presence of layers with finer or coarser
PSDs than adjacent layers, and the resulting effects on water drainage (Zettl, 2014; Clothier et al., 1977).
These effects are not yet incorporated into the classification system presented in this paper, but are planned
for future iterations of the system.

Modelled solar-radiation differences across latitudes, slopes and aspects were used to produce modifiers
(additions or deductions) to the PSD-derived AWSC estimate. These modifiers are intended not to imply
actual reductions or additions to AWSC on different slopes and aspects, but as surrogate modifiers to AWSC
to reflect increased or decreased evapotranspirative demand driven by varying energy regimes. Short-wave
radiation was calculated for different slopes, aspects and latitudes as the sum of the direct- and diffuse-beam
components. The theoretical direct-beam component of solar radiation was determined after Garnier and
Ohmura (1968, 1970). Diffuse clear-sky radiation was calculated assuming a standard atmosphere after Iqbal
(1983). These modelled values were then converted to additions or deductions to AWSC (in mm water per m
material depth) through the following two general steps:
1. Values were indexed to a neutral energy regime defined by radiation received on flat ground;
values above this were defined as deductions proportional to their deviation from the neutral
values, and values below were defined as additions using the same methods. Thus, at northern
latitudes, north-facing slopes would receive less radiation than flat ground and have an AWSC
addition, while southern slopes would have an AWSC deduction.
2. Values were indexed to proportional additions or deductions calculated for approximately 50
degrees latitude, with the magnitude of adjustment increasing for higher latitudes and decreasing
for lower latitudes. All slope positions at latitudes less than 25 degrees were assigned a neutral
energy correction.
Energy corrections were compiled into separate tables for 5-degree latitude bands, consisting of 10-degree
aspect and 5-degree slope classes. Energy corrections for each site’s slope and aspect were determined by
interpolating between slope classes within the nearest aspect class, and the two energy corrections from the
nearest latitude tables were interpolated according to the site’s location.
Application of this adjustment approach to the 31 study sites resulted in 17 sites being assigned a neutral
energy correction, six sites being assigned deductions (ranging from 2 to 28 mm/m), and eight sites being
assigned additions (ranging from 1 to 9 mm/m).
AWSC modifiers additional to those for energy regime will be considered for future iterations of this
classification system in order to reflect the effects of soil’s chemical and physical properties, particularly the
osmotic effects of salinity and the effects on infiltration of surface crusting in fine-textured soils (e.g.
Bissonnais, 1996). As an example, saline materials may supply less plant-available water than estimated
simply from the PSD-AWSC relationships discussed above, owing to the decreased osmotic potential in the
soil. In this study, several sites had sensors within saline mine wastes and overlying salinity-affected cover
materials. Seasonal wetting and drying patterns in these materials were similar to those observed in non-
saline materials, indicating that water in the saline materials was still available for extraction by plants.
Adjusted AWSC values were used to determine the soil moisture regime (SMR) after Pojar et al. (1985; also
Meidinger and Pojar, 1990; Green et al., 1984). Table 3 uses the SMR classes of the British Columbia (B.C.)
Biogeoclimatic Ecosystem Classification potential hygrotope, but replaces the relative ranking of various
criteria with quantified ranges of energy-adjusted AWSC. AWSC ranges for each SMR class are modified from
values presented by the Cumulative Environmental Management Association (2006) for reclamation of oil
sands mines in northeast Alberta, Canada. The AWSC method for SMR determination applies only to upland
(very xeric to mesic) SMRs, as wetter SMRs require input of seepage water or the presence of a water table
within 100 cm of the soil surface, and are not dependent on soil storage. Thus, determination of SMRs wetter
than mesic in this system is based on observations of shallow groundwater seepage and/or the presence of
a water table within the top metre of surficial materials. Note that these moisture regimes are intended to
reflect dominant soil-water conditions over a multi-year period, consistent with the B.C. Biogeoclimatic
Ecosystem Classification hygrotope.

Water-table depth
Primary water Available water storage, surface
SMR (cm below ground
source metre (mm)
surface)
Very Xeric (0) >100 <60
storage
Xeric (1) >100 60–89
storage
Subxeric (2) >100 90–119
storage
Submesic (3) >100 120–149
storage
Mesic (4) >100 >150
storage
Precipitation and
Subhygric (5) >100 >150, seepage contributes to supply
seepage
Hygric (6) Seepage 30–100 n/a
Seepage or
Subhydric (7) permanent water 0–30 n/a
table
Permanent water Water table permanently
Hydric (8) n/a
table at or above soil surface
PSD and topography data collected from the 31 study sites were used to estimate energy-corrected AWSC
values based on the methods described above. Analysed volumetric-water-content (VWC) and matric-
potential (φm) data collected from the instrumented study sites were analysed to derive mean growing-
season available volumetric water contents (AWC) for each site. To do so, the VWC at permanent wilting
point (PWP) was calculated for each VWC sensor from plots of VWC against φm for each sensor pairing. In
cases where VWC continued to decrease at matric potentials drier than PWP, the mean of the lowest 10 VWC
measurements was used as an alternative VWC-at-PWP value in recognition of potential error in the paired
sensor measurements. For sensors without sufficiently dry conditions, VWC-at-PWP values were
extrapolated by referring to VWC-φm curves from neighbouring sensors in the same material type. Each
sensor’s VWC-at-PWP value was then subtracted from each of its VWC measurements to calculate its AWC
(water content above PWP). To calculate mean AWC from all sensors over a profile depth, sensor AWC values
were arithmetically weighted according to the relative depth interval represented by each sensor within the
upper metre of soil. Annual AWC values were calculated for region-specific growing seasons, which were
defined based on modelled PET.
Growing-season AWC values were compared to estimated energy-adjusted AWSC values, as shown in Figure
3. This comparison shows support for the proposed classification system, with growing-season AWC
significantly (p < 0.001) positively correlated with estimated AWSC, and the derived linear relationship
accounts for approximately 71% of observed variation in observed growing-season water contents. Fitting
errors are at least partially attributable to the various stages of vegetation development on the sites and the

resulting differences in transpirative demand for soil water. The two sites with the highest AWSC and AWC
values (Carolina) are composed of 78 cm and 100 cm of coarse-fragment-free silt-loam cover material
overlying tailings sand, and are located on flat surfaces with neutral energy regimes.
The 31 study sites were assigned to hygrotopic SMR classes based on the proposed classification methods
and the SMR AWSC ranges presented in Table 3. Means and variances for each SMR class based on mean
observed growing-season AWC at each site were calculated and are presented in Figure 4. These data again
support the proposed classification system, and show a progression from a mean observed growing-season
AWC of approximately 50 mm/m on very xeric sites to 175 mm/m on mesic sites. Further, with the exception
of the single and anomalously dry submesic site, these data indicate a mean separation of approximately 30
mm/m between SMR classes, consistent with the ranges in Table 3.

This paper presents the results of testing a proposed ecohydrologic classification system (described in more
detail in Straker et al., 2015) designed for estimation of soil water storage across a range of ecosystems,
including mine-waste-based landforms and mine-affected watersheds. Data on landform characteristics,
cover-material PSDs, and continuously measured VWCs at 31 instrumented sites in Canada, the United
States, and Australia were used to complete the proposed classification process and to evaluate results on
both a continuous (regression) and categorical basis. These evaluations show good agreement between
available soil water contents predicted by the classification system and observed growing-season water
contents, with approximately 71% of the variation in observed AWC explained by the classification model.
Future iterations of the model will seek to incorporate modifications to AWSC from textural discontinuities,
soil salinity and soil-surface crusting.
The ecohydrologic classification system can be used for a variety of applications in mine reclamation and
closure planning, including
 estimation of site moisture status and selection of appropriate species for revegetation programs;
 providing an explanatory variable for data analysis for reclamation research and monitoring
programs; and
 estimation of soil water storage and associated transpiration to support derivation of surface-
water-balance terms and flux estimates.
Bissonnais, Y.L. (1996) Aggregate stability and assessment of soil crustability and erodibility: I. Theory and methodology, European
Journal of Soil Science, Vol. 47(4), pp. 425–437.

Clothier, B.E., Scotter, D.R. and Kerr, J.P. (1977) Water retention in soil underlain by a coarse-textured layer: theory and a field
application, Soil Science, Vol. 123(6), pp. 392–399.
Cumulative Environmental Management Association (2006) Land capability classification system for forest ecosystems in the oil
sands, 3rd ed. Alberta Environment, Edmonton.
Devito, K., Creed, I., Gan, T., Mendoza, C., Petrone, R., Sillins, U. and Smerdon, B. (2005) A framework for broad-scale classification
of hydrologic response units on the Boreal Plain: is topography the last thing to consider? Hydrological Processes, Vol. 19, pp.
1705–1714.
Garnier, B.J. and Ohmura, A. (1968) A method of calculating the direct shortwave radiation income of slopes, Journal of Applied
Meteorology, Vol. 7(5), pp. 796–800.
Garnier, B.J. and Ohmura, A. (1970) The evaluation of surface variations in solar radiation income, Solar Energy, Vol. 13(1), pp. 21–
34.
Green, R., Courtin, P., Klinka, K., Slaco, R. and Ray, C. (1984) Site diagnosis, tree species selection, and slashburning guidelines for the
Vancouver Forest Region, Province of British Columbia, Victoria, Canada.
Iqbal, M. (1983) An introduction to solar radiation, New York, USA: Academic Press.
Meidinger, D. and Pojar, J. (1991) Ecosystems of British Columbia, Province of British Columbia, Victoria, Canada.
Pojar, J., Klinka, K. and Meidinger, D. (1985) Biogeoclimatic ecosystem classification in British Columbia, Ministry of Forests, Province
of British Columbia, Victoria, Canada.
Saxton, K.E. and Rawls, W. J. (2006) Soil water characteristic estimates by texture and organic matter for hydrologic solutions, Soil
Science Society of America Journal, Vol. 70(5), pp. 1569–1578.
Straker, J., Baker, T., Barbour, L., O’Kane, M., Carey, S. and Charest, D. (2015) Mine reclamation and surface water balances: an
ecohydrologic classification system for mine-affected watersheds, in Proceedings Mine Closure 2015. Australian Centre for
Geomechanics, Perth.
Zettl, J. (2014) Infiltration and drainage through coarse layered soil: a study of natural and reclaimed soil profiles in the Oil Sands
Region, Alberta, Canada, MSc Dissertation, University of Saskatchewan, Canada.


D.S. Whitlock Cranfield University, UK

J. Rickson Cranfield University, UK
R.N. Humphries Celtic Energy Limited, UK
R. Thompson Celtic Energy Limited, UK
M. Tibbett University of Reading, UK
Coal mine spoil tips have been recognised as a UK Biodiversity Action Plan habitat owing to the presence of a
host of unique flora and fauna assemblages and, in particular, “lichen heaths.” The natural revegetation of
abandoned spoil tips in South Wales, UK, has created some of the most dynamic habitats and historic
landscapes for lichen colonisation. The sensitivity of lichen habitat to vegetation successional processes and
anthropogenic factors is leading to their gradual extinction without intervention. Such is the threat that there
is a requirement for the reconstruction of coal spoil tip habitats following the eventual closure of a recently
consented surface mine.
While recent colonisation of coal spoil elsewhere has been documented, this has been opportunistic and has
not involved the deliberate reconstruction of the habitat following mining. Experience elsewhere indicates
that the physical properties of the landform and spoil material are important factors in colonisation. However,
little is known about the factors relating to the high rainfall environment in South Wales. In order to provide
guidance for the recreation of the lichen habitat, a detailed study was undertaken to determine the
relationships between the spatial distribution of vegetation class patches and the physical proprieties of coal
mine spoils in South Wales. The aim was to understand the key physical factors aiding the establishment and
development of lichen species on the mine spoils.
Lichen-dominant patches were observed in the field during a physical properties survey and sampled for
laboratory analysis alongside two other vegetation classes (bare ground and a moss-heather mix).
Relationships were identified between vegetation class and physical properties, including slope aspect, slope
gradient, slope form, erodibility (through laboratory simulated rainfall experiments) and the particle-size
distribution of the slope-forming materials.
The results demonstrate that several physical properties of the coal mine spoil have significant influence on
the spatial distribution of vegetation classes. It was found that the lichen-dominant patches were best
established on southwest-facing slopes, with slope gradients ranging from 22 to 32o, where higher
successional vegetation classes were less abundant.
While this study provides a better understanding of the spatial distribution of vegetation classes in the
replication of coal spoil lichen habitats, it also identified that a singular rule-based approach might not be
applicable across multiple coal mine spoil habitats without further research owing to the nature of mine spoil
sites. Even so, knowing the influence of key physical properties on vegetation distribution will aid design
decisions relating to the replication of lichen communities following mine closure.
In the UK, mine spoil tips have been recognised as biodiversity action plan (BAP) habitats hosting unique flora
and fauna assemblages (JNCC, 2010). Despite this, little is known about how or why rare community patches

The influence of coal mine spoil physical properties on the spatial distribution of lichen-rich communities D.S. Whitlock et al.
have established and thrive in what were initially barren and isolated landscapes. Lichen communities are
fragile and are increasingly under pressure from environmental and anthropogenic factors. The
establishment and continued presence of lichen communities in coal mine spoil habitats are under-
researched. In other landscapes, lichen species have been identified as having few opportunities in which
they can establish and thrive without being destroyed by disturbance or being outcompeted by vascular plant
species (Làzaro, 2004). Therefore, determining why coal mine spoils provide suitable habitats for lichen
species is fundamental to preventing the gradual loss of rare species.
Although lichen assemblages are poorly understood in coal mine spoil habitats, several studies have explored
the role of the landscape’s physical properties in determining habitat suitability for lichen. These studies span
several regions, mainly within arid and semi-arid environments, where soil, slope forming material and land
stability need to be understood.
The influence of slope aspect on the growth and spatial distribution of ground cover types has been studied
extensively (Pentecost, 1979; Kutiel and Lavee, 1999; Cantón et al., 2004; Lázaro et al., 2008; Johnson, 2011).
Armstrong (1991) identified that aspect preferences differ even amongst similar lichen species. Worsley
(1990) and St Clair et al. (1993) identified the potential significance of slope gradient on lichen distribution,
as it is a proxy for soil erosion and slope stability (Gray, 2013). Johnson (2011) found lichen growth was
negatively affected by steep slope gradients (where average slope gradient was above 28°). Nonetheless,
Lázaro (2004) found that lichen species establishment was more successful where the slope-forming
materials created a degree of disturbance .
Slope morphological features identified as influencing the spatial distribution include slope position (referring
to the position on a slope) and slope form (referring to shape of the slope profile) (Gray, 2013; Mao et al.,
2014). Cerdà (1998) identified that lichen species were most abundant on the uppermost parts of a slope.
The uppermost section of a slope (the summit or shoulder) is often associated with lower availability of water
and nutrients, thus reducing competition on the highest section of a slope face, leading to notable differences
in vegetation distribution (Cerdà, 1998; Maggi et al., 2005; Lázaro et al., 2000; Tang et al., 2010). Cantón et
al. (2004) found lichen did not inhabit the most stable concave slopes, preferring linear slopes for
establishment. As spoil tips exhibit a discontinuous variety of complex slope forms, this could account for the
patchy nature of vegetation across a spoil tip.
Lázaro et al. (2008) and Humphries (2013) suggest that a degree of erosion could be pivotal in maintaining
population assemblages of lichen and other lower plant species. The particle-size distribution (PSD) of spoil
material influences both soil water dynamics and susceptibility to erosion (Eldridge and Greene, 1994;
Sheoran et al., 2010); therefore, despite expected homogeneity of parent material, PSD changes across a
spoil type could cause the vegetation patches present in a mine spoil habitat, as Lalley et al. (2006) found
that PSD had a significant effect on lichen growth.
This study aimed to ascertain whether relationships exist between physical site conditions and the spatial
distribution of different vegetation species, particularly the lichen species found in coal mine spoil habitats.
Comparisons of physical properties across vegetation types and spoil tips were surveyed, sampled and
analysed to investigate whether causal relationships exist between physical site properties and lichen spatial
distribution.
The study was undertaken on a group of mine spoil tips to the north of Llwydcoed, Aberdare (+51° 44' 35.19"
N, −3° 25' 54.21" W), South Wales, UK (Figure 1). The climate in this region is temperate, and Aberdare is
subject to an annual mean of 1674.2 mm of rainfall (average 1981–2010), with the rain often falling in high-
intensity episodes (Chambers and Garwood, 2000; Met. Office UK, 2014). The spoil tip site was previously a
surface mine that operated post-1847. Although the spoil tips across the site are of unknown ages, it is
believed they are between 135 and 155 years old (Cotswold Archaeology, 2013).

The mine spoil forms a rolling landscape of predominately southwest-facing tips (Figure 1), but not all spoil
tips have all eight cardinal sides. Many spoil tips to the northeast of the site only have east to west/northwest
slope faces owing to the nature of the land beyond the coal spoil site, which comprises sheep-rearing fields.
Running down the south of the site is a small watercourse, which has incised the spoil tips directly next to it.
The spoils found at the lowest altitude across the site are dominated by bracken vegetation. No lichen-
dominant or bare ground vegetation patches are present across these spoils.
Humphries and Leverton (2012) identified that 73 of the 95 spoil tips at the study site are made up of laminar
grey mudstone and laminar hard grey mudstone. Five spoils were eliminated from the group because the
spoil tips exceed a safe slope gradient to survey and sample. In the lower section of the site, a further five
spoil tips were eliminated, as bracken had outcompeted all other vegetation. Finally, Humphries and
Leverton (2012) also found two spoil tips that did not host any lichen-dominant patches; these were
consequently removed from the subset under consideration for sampling. Owing to time constraints in both
the field (surveying and collecting samples) and the laboratory analysis, it was only practical to investigate
10% of the remaining 61 spoil tips. Of the six spoils selected via a random sampling technique, five were
made up of the laminar grey mudstone (spoil tips 33, 36, 56, 61 and 71) and one spoil was made up of the
laminar hard dark grey mudstone (spoil tip 5) (Figure 1).
The site survey investigated the physical properties of the selected spoils in relation to three of the six
vegetation classes identified by Dickinson et al. (2013) to create a comparative analysis across the three
vegetation classes (bare ground, lichen-dominant and moss-heather mix). For the physical properties survey,
each vegetation class was identified in three locations on each spoil tip and surveyed by using 20 by 20 cm
quadrats (following the methodology of Dickinson et al., 2013). This equated to 54 quadrats (as all three
vegetation classes were present on each of the six selected spoil tips, with three replications). Each quadrat
point was mapped using a Trimble GEOXH unit GPS with Terrasync, which records 10–30 cm accuracy range
when conditions are favourable.
Spoil material samples from each of the 54 sampling points were collected for laboratory analysis. Samples
for erodibility analysis were collected with minimal disturbance using containers with a base of 130 by 105
mm and a depth of 40 mm. An additional circa-50 g sample of the material (at approximately 0–10 cm depth)
below the vegetation cover was taken for particle-size determination.

All slope characteristics and vegetation classes of each quadrat were identified visually using the
photographic classification described in Dickinson et al. (2013). The quadrat orientation (aspect) was
determined using the eight cardinal directions with a compass. Slope gradient was determined in the field
with a clinometer for each sample. Slope position was determined by splitting each slope face into three

equal sections: the “top” samples found on the summit and typically in the upper shoulder of the slope, the
“middle” samples found within the shoulder and upper footslope, and the “bottom” samples at the footslope
and the toeslope.
To determine the spoil materials’ erodibility, a pressurised nozzle rainfall simulator located in the Cranfield
University Soil Management facility was used to produce rainfall events. High-intensity rainfall events were
chosen for the experiment. This was in order to obtain measurable quantities of eroded material for each
vegetation class and to recreate the high-intensity rainfall events that are becoming increasingly common
across the UK (Mullan, 2013; Foulds et al., 2014; Smith et al., 2014; Met. Office UK, 2014). The samples were
subject to a mean rainfall intensity of 42.2 mm hr-1 (standard deviation = 0.68 mm hr-1 and coefficient of
variance = 1.6 mm hr-1). The intensity was produced using a LECHLER 460.788.17.CE nozzle 2 m above the
sample. A water pressure of 0.7 bar was applied to each sample at a constant rate for 15-minute durations.
Initially, each sample was saturated in a water bath overnight to ensure all air was removed from the material
and bring all samples to a standard soil moisture condition before rainfall simulation was conducted.
The samples were contained in splash cups (internal diameter 74 mm, internal depth 50 mm) and a 19 cm
diameter sieve receiver was used to collect any eroded material. After the application of rainfall, eroded
material was placed onto pre-weighed Whatman no 2, 150 mm filter paper (weighed using a four decimal
place balance). The filter papers, after air-drying, were oven-dried for one hour and 20 minutes (at 105°) and
weighed.
The PSD analysis was conducted in accordance with the British Standard (BS) Operating Procedures 7755
Section 5.4:1998 Determination of particle-size distribution in mineral soil material — Method by sieving and
sedimentation. The material samples were initially air-dried for 24 hours before being sieved to separate
particles that were ≤2 mm in size. Samples of 10 g of particles ≤ 2mm in size were treated with hydrogen
peroxide to produce a slurry, which was dispersed using a buffered sodium hexametaphosphate solution
before the particle-size fractions were determined by pipette extraction.
Categorical data were analysed through the Fisher exact test (because the sample size did not comply with
the requirements of Chi-square). All other data were tested for normality using the Shapiro-Wilk test. If the
data were normally distributed, a factorial analysis of variance (ANOVA) was used for analysis. All statistics
were carried out using SPSS Statistics 22, applying a confidence level of 95%. All boxplot diagrams illustrate
the median and interquartile range in results exploring physical parameters across vegetation classes. The
whiskers show minimum and maximum values, excluding outliers.
The cardinal directions were analysed via the factorial ANOVA to determine the influence of slope aspect on
lichen spatial distribution. With a 95% level of confidence and a p-value of 0.031, the test indicated that slope
aspect does influence the spatial distribution of vegetation classes. Figure 2 shows that lichen-dominated
vegetation was found in abundance on southwest-facing slopes.

N
7
6
NW 5 NE
4
3
2 Bare
1
W 0 E Lichen
Moss-Heather Mix
SW SE
The range in slope gradient values for each vegetation class is shown in Figure 3. Lichen-dominant vegetation
had the most limited distribution based on slope gradient. The factorial ANOVA indicated that slope gradient
does influence the spatial distribution of lichen (p = 0.002).
40
30
Slope Gradient (degrees)
20
10
0
Bare Lichen Moss-Heather Mix
Vegetation Class
All three vegetation classes were present across the three slope positions (Figure 4). In particular, the moss-
heather mix alongside other higher successional classes was present in abundance across all slope positions.
Lichen and bare vegetation were less frequent across the top and bottom slope positions. However, there is
insufficient evidence to suggest that slope position affects lichen distribution (p = 0.392).

15 Vegetation Class
Bare
Lichen
Moss-Heather Mix
10
No. of Samples
0
Top Middle Bottom
Slope Position
The Fisher’s exact test identified a significant relationship between slope form and vegetation class (p = 0.01).
Figure 5 shows the distribution of samples across the three slope forms. All three vegetation classes favoured
linear slopes; however, moss-heather mix was limited to only linear and convex slopes.
Page 1

Before determining whether the erodibility of the slope-forming material found at the study site had a
significant influence on lichen spatial distribution, a normality test identified that the erodibility data
generated by the rainfall simulations was skewed; therefore, it was log10 transformed to make the dataset
normally distributed. This ensured a more robust parametric statistical test could be conducted. Figure 6
illustrates that the distribution of vegetation was affected by the erodibility values of the slope-forming
materials (p = 0.00).
.00
Log(Material Erodibility)
-.25
-.50
-.75
-1.00
Bare Lichen Moss-Heather Mix
Vegetation Class
To determine if the PSD of the material had a significant influence on lichen spatial distribution, the data
generated by the PSD analysis were first tested for normality. However, two samples were removed, as they
skewed the data (an anomalous lichen material sample from spoil tip 56 and an outlier from a bare material
sample on spoil tip 71). Figure 7 indicates the PSD samples were clustered, with the majority of samples
containing a high silt fraction percentage. However, the factorial ANOVA detected a significant difference
between PSD values for all three vegetation classes. Material samples with the lowest abundance of sand
fractions were associated with lichen. The lowest silt fractions were found for bare ground, and the lowest
clay fractions were associated with moss-heather samples.
Page 1

0
100
10
90
20
80
Clay Fraction 30 Silt Fraction
(<0.002mm) 70 (0.002-0.063mm)
40
60
50
50
60
40
70
30
80
20
90
10
100
0
100 90 80 70 60 50 40 30 20 10 0
Sand Fraction (0.063-2mm)
The results of this research concur with several studies indicating that slope aspect has a significant effect on
ground cover distribution. The northwest- to northeast-facing slopes were dominated by moss-heather mix
and vascular plant classes. Spoil tip 56 had no northwest to east slope faces, explaining the concentration of
moss-heather vegetation surveyed on the south-facing slopes. Although bare ground was observed on north
faces of some spoil tips at the study site, this is probably due to animal track erosion and sheep scars across
the landscape. Furthermore, incidence of sheep scars increased on the southerly slope faces, which might
explain the predominant slope orientation of the bare ground patches. It is unclear why the sheep scars are
in higher abundance on the south-facing slopes. However, this may be linked to selection in shelter, diet (as
sheep are highly selective in feeding preferences), the south-facing slopes’ exposure to the longest duration
of direct sunlight or the location of the animals’ winter feed (Armstrong, 2003). While no (bare soil) sheep
scars were surveyed, one lichen-dominant sample was taken within a sheep scar because of the low
occurrence of lichen-dominant vegetation elsewhere on the spoil tip (spoil 5). Overall, both bare ground and
lichen-dominant vegetation favoured similar slope aspects (Figure 2) because lichen species were the primary
colonisers of bare earth surfaces in this habitat.
Pentecost (1979), Armstrong (2002) and Johnson (2011) all confirm that species’ slope aspect preferences
are strongly associated with soil moisture content. With this in mind, moisture content can influence a slope

material’s susceptibility to erosion. Renner (1936) identified that north-facing slopes (in the northern
hemisphere) were more stable than south-facing slopes owing to differences in moisture and temperature.
This could explain the absence of lichen from north-facing slopes, as vascular species would outcompete
lichen species where disturbance was absent or minimal. Further research is needed to understand the
preferred requirements of individual lichen species and validate this relationship.
Furthermore, slope gradient is a proxy for soil erosion and slope stability processes. Therefore, it is
unsurprising that the results identify all bare ground areas to exceed 12°; four bare ground samples exceeded
a 30° slope gradient. Bare ground samples also had the highest degree of erodibility. Both bare ground and
the moss-heather mix were found across a wide range of gradients. The bare ground areas that had lower
slope gradients could have been influenced by previous disturbance events, including paths and sheep scars
that had been subject to erosion and had consequently collapsed. Furthermore, some bare ground sections
on sampled spoil tips could not be surveyed because of the steepness and instability of the material (spoils
61 and 71). Nonetheless, all but one of the samples taken on these slopes exceeded a 20° gradient, so this is
unlikely to have significantly impacted the results. The small slope gradient range in which lichen-dominant
patches were found (Figure 3) suggests they could have a slope gradient threshold in which they are less
subject to competition from other vegetation types. All but two lichen-dominant samples were within a 10°
gradient range (22–32°).
Slope morphology parameters (i.e. slope form and position on slope) were subjective. Slope position did not
identify a relationship with the spatial distribution of the vegetation classes, despite physical differences
observed when surveying; all vegetation classes were established across each of the slope positions. Perhaps
if a more extensive survey of each spoil tip had been conducted, identifying all vegetation class patches on
each spoil tip, the differences would have been detected.
Tang et al. (2010) split slopes into four sections, separating the summit from the rest of the slope. The four-
section was not applied in this study because the spoil tips were not uniform in slope length or shape across
each of the spoil tips. If the four-section method had been applied, the results would have been altered
significantly because there were notable differences and similarities across the six spoil tips that were unable
to be represented in the survey approach used. For example, all six spoil tips had flat spoil tip summits and
four of the summits were dominated by moss-heather, while the other two summits were dominated by
other vascular plant vegetation classes. Other studies identified the highest section on a slope to be most
abundant in lichen vegetation (Cerdà, 1998), but in the present study, lichen vegetation was sparse in the
upper sections of the slope (Figure 4). The lowest slope gradients at the study site were found on the summits
and toeslopes, where higher succession vegetation classes dominated, and perhaps this is why lichen patches
were not found in abundance on these sections. The two lichen-dominant patches positioned on the lower
slopes of the spoil tips also had the lowest slope gradients.
The slope forms on each spoil tip exhibited discontinuities along singular slopes, as expected (Cerdà, 1998;
Gray, 2013). All three slope forms were present across the landscape (Figure 5). Convex slopes were not
abundant within the study. As no prior study has explored the spoil site’s morphological characteristics, it is
unclear whether there were too few convex slopes across the site, as they are often uncommon in less
actively eroding landscapes, or whether the sampling technique limited the selection of convex slopes across
the study. Two moss-heather mix patches were sampled and surveyed on convex summits (associated with
higher rates of erosion, than are concave and linear slopes), despite vascular plants’ association with more
stable slope forms (Cantón et al., 2004). The most stable slope form, concave, did not host any moss-heather
mix in this study. This was perhaps owing to the abundance of other vascular plant species classes dominating
these stable slopes in the lower sections of the spoil tips. All vegetation classes favoured linear slopes.
Although slope form needs to be addressed further, the fact that vegetation class patches are influenced by
slope form demonstrates that another physical property influenced the spatial distribution of lichen-
dominant patches.

Bare ground material was more susceptible to rainfall-induced erosion than was material associated with
lichen and moss-heather vegetation classes, which concurs with previous research (Eldridge and Greene,
1994; Buxton et al., 2005; Lázaro et al., 2008; Malam Issa et al., 2011). While the material associated with
lichen and moss-heather vegetation classes shared similar susceptibility to rainfall-induced erosion, the
median erodibility for moss-heather mix was lower than that for material samples associated with lichen-
dominant patches (Figure 6). This result reiterates the stability associated with the spoil tip summits and
toeslopes where higher successional class species were most abundant, relegating lichen to the less stable
slope gradients (Lázaro et al., 2000). Furthermore, disturbances such as animal trampling can strongly
influence the materials’ susceptibility to erosion through selective grazing choices that concentrate
disturbance (Eldridge 1998).
Lázaro et al. (2008) outlined the importance of low-magnitude, low-intensity precipitation in enabling lichen
species to thrive in semi-arid environments. Although the study site is much more stable than the study area
explored by Lázaro et al. (2008), identifying the thresholds of disturbance lichen species can survive will
promote lichen distribution across coal mine spoil habitats. The impacts of a range of rainfall simulation
events (intensities and durations) were not explored in this study. However, the relatively high-intensity
rainfall event used in this study generated such small amounts of erosion across all vegetation classes (0.09–
0.5 g) that it is unlikely that a further study exploring a combination of rainfall intensities would add to the
findings of the present study. Instead, the relationship between erosion processes and lichen spatial
distribution should be investigated further by exploring the materials’ erodibility in terms of infiltration
capacity and resistance to detachment. This would enhance the understanding of the results found.
The PSD results identify significant differences across spoil tips and vegetation classes, even though most
samples were classified as silty clay loam (UK ADAS Textural Triangle). It was expected that bare ground
samples would have the largest silt fraction percentages, because silt particles are more susceptible to
erosion (Wischmeier and Mannering, 1969). Seventy-five per cent of the samples contained high silt content
(54–70%). Both lichen-dominant and moss-heather mix classes contained higher silt fractions than the bare
ground samples. The comparatively lower silt fraction found in bare ground samples is likely to be linked to
the increased mobility of the silt in this vegetation class, especially for bare ground, as there is no vegetation
cover protecting the material. None of the sampled material contained clay fractions below 10%, thus
improving the materials’ cohesive nature, which may have contributed to the three vegetation classes’ low
susceptibility to erosion. Material associated with lichen contained the smallest sand particle fractions. Lalley
and Viles (2008) suggest that sand particles help reduce the impact of disturbance and positively influence
the recovery rate of lichen recovery. Again, this raises the question of what degree of disturbance is required
to encourage lichen populations while preventing higher successional vegetation from establishing.
The findings of this study demonstrate that several physical properties of a coal mine spoil have significant
influence on the spatial distribution of vegetation classes at the study site near Llwydcoed, South Wales. The
results show that lichen species were present across east- to northwest-facing slopes; however, lichen was
only abundant on southwest-facing slopes within a refined range of slope gradients (22–32°). Aspect plays
an important role in the spatial distribution of lichen; however, this relationship could be dependent on the
lichen species across the site, as species preferences could vary on a site-by-site basis. The roles of slope
morphology, material susceptibility to erosion and PSD need to be explored further to increase
understanding of the relationships that could exist both at the study site and across other coal mine spoil
sites. Nonetheless, knowing the influence of physical parameters on vegetation spatial distributions at the
study site will aid future management decisions if the site is reworked for coal in the future. Prioritising slope
aspect, gradient, material erodibility and PSD would be fundamental to encourage the re-establishment of
lichen communities in the future. For example, loose tipping methods could be used to create as many
southwest-facing large bare linear slope surfaces with gradients between 22 and 32° as possible.

This study has clarified the importance of identifying the optimal conditions required to encourage lichen-
dominant vegetation to establish in coal mine spoil habitats in South Wales. Important physical properties
have been identified that could be explored across other coal mine spoil sites to develop an expansive
collection of optimal conditions for lichen species to thrive in rare and unique coal mine spoil habitats.
Further work is required if the application of a rule-based approach is to be established. Ultimately, research
should concentrate on understanding how micro-habitats change across multiple coal mine spoils as a
function of the physical characteristics of the site. This can be achieved by replicating and further developing
the methodology applied in this study.
This study was funded by Celtic Energy Limited.
Armstrong, H. (2003) Information and advisory note number 47: The grazing behaviour of large herbivores in the uplands. Scottish
Natural Heritage, UK, viewed 3 September 2014, http://www.snh.org.uk/publications/on-line/advisorynotes/47/47.htm.
Armstrong, R.A. (1991) Competitive interactions between four foliose lichens on north and south facing rock surfaces, Environmental
and Experimental Botany, Vol. 31(1), pp. 51–58.
Armstrong, R.A. (2002) The effect of rock surface aspect on growth, size structure and competition in the lichen Rhizocarpon
Geographicum, Environmental and Experimental Botany, Vol. 48, pp. 187–194.
Cantón, Y., Del Barrio, G., Solé-Benet, A. and Lázaro, R. (2004) Topographic controls on the spatial distribution of ground cover in the
Tabernas badlands of SE Spain, Catena, Vol. 55, pp. 341–365.
Cerdà, A. (1998) The influence of aspect and vegetation on seasonal changes in erosion under rainfall simulation on a clay soil in
Spain, Canadian Journal of Soil Science, Vol. 78(2), pp. 321–330.
Buxton, R.P., Stanley, J., Alspach, P., Morgan, C. and Ross, C. (2005) Mosses and other early colonizers: Pioneers for revegetation
mine waste rock dumps and rock walls. Paper was presented at the 2005 National Meeting of the American Society of Mining
and Reclamation, Breckenridge, Colorado, June, pp.170-180. Viewed 2 July 2014, http://www.asmr.us/Publications/
Conference%20Proceedings/2005/0170- Buxton-NZ%20.pdf
Chambers, B.J. and Garwood, T.W.D. (2000) Monitoring of water erosion on arable farms in England and Wales: 1990–1994, Soil Use
and Management, Vol. 16, pp. 93–99.
Cotswold Archaeology (2013) Environmental impact assessment chapter 15: Development of Bryn Defaid surface coal mine for the
extraction of coal and associated ancillary activities including restoration and subsequent aftercare of the full development
site, Celtic Energy, viewed 20 July 2014, https://planning.rctcbc.gov.uk/online-applications/application Details.do?
activeTab=details&keyVal=_RHOND_DCAPR_43419.
Dickinson, A., Tibbett, M., Pawlett, M. and Humphries, N. (2013) Changes in the microbial community during the ecological
development of mine spoil, MSc Thesis, Cranfield University, UK.
Eldridge, D.J. (1998) Trampling of microphytic crusts on calcareous soils, and its impact on erosion under rain-impacted flow, Catena,
Vol. 33, pp. 221–239.
Eldridge, D.J. and Greene, R.S.B. (1994) Microbiotic soil crusts: A review of their roles in soil and ecological processes in the rangelands
of Australia, Australian Journal of Soil Research, Vol. 32, pp. 389–415.
Foulds, S.A., Griffiths, H.M., Mackli, M.G. and Brewer, P.A. (2014) Geomorphological records of extreme floods and their relationship
to decadal-scale climate change, Geomorphology, Vol. 216, pp. 193–207.
Gray, D. (2013) Influence of slope morphology on the stability of earthen slopes, Geo-Congress 2013: Stability and performance of
slopes and embankments III, San Diego, California, USA, March (3-7), pp. 1895–1904.
Humphries, R.N., 2013. Case study: The contribution of lichen communities in the south Wales coalfield, United Kingdom, Journal of
the American Society of Mining and Reclamation, Vol. 2(1), pp. 80–98.
Humphries, R.N. and Leverton, R.E. (2012) An investigation of the nature of tipped surface mine material on the western slopes of
Aberdare Mountain above Llwydcoed, Rhondda Cynon Taff, Celtic Energy, Caerphilly, UK, 101 p.
Johnson, D. (2011) An investigation into lichen growth on terminal moraines on the glacial foreland of Nigardsbreen, South Norway,
Reinvention: A Journal of Undergraduate Research, British Conference of Undergraduate Research 2011 special issue, viewed
11 June 2014, http://www2.warwick.ac.uk/fac/cross_fac/iatl/reinvention/issues/bcur2011specialissue/johnson.
Joint Nature Conservation Committee (JNCC) (2010) UK biodiversity action plan — Priority habitat descriptions: Open mosaic habitats
on previously developed land, Defra, London, UK.
Kutiel, P. and Lavee, H. (1999) Effect of slope aspect on soil and vegetation properties along an aridity transect, Israel Journal of Plant
Sciences, Vol. 47(3), pp. 169–178.
Lalley, J.S. and Viles, H.A. (2008) Recovery of lichen-dominated soil crusts in a hyper-arid desert, Biodiversity and Conservation, Vol.
17, pp. 1–20
Lalley, J.S., Viles, H.A., Copeman, N. and Cowley, C. (2006) The influence of multi-scale environmental variables on the distribution of
terricolous lichens in a fog desert, Journal of Vegetation Science, Vol. 17, pp. 831–838.

Lázaro, R. (2004) Implications of precipitation on vegetation of water-limited lands, Recent Research Development in Environmental
Biology, Research Signpost, Kerala, India, pp. 553–591.
Lázaro, R., Alexander, R.W. and Puigdefábregas, J. (2000) Cover distribution patterns of lichens, annuals and shrubs in the Tabernas
Desert, Almería, Spain, in Vegetation mapping: From patch to planet, R.W. Alexander and A.C. Millington (eds), Wiley,
Chichester, UK, pp. 19–40.
Lázaro, R., Cantón, Y., Solé-Benet, A., Bevan, J., Alexander, R., Sancho, L.G. and Puigdefábregas, J. (2008) The influence of competition
between lichen colonization and erosion on the evolution of soil surfaces in the Tabernas badlands (SE Spain) and its
landscape effects, Geomorphology, Vol. 102, pp. 252–266.
Maggi, O., Persiani, A.M., Casado, M.A. and Pineda, F.A. (2005) Effects of elevation, slope position and livestock exclusion on
microfungi isolated from soils of Mediterranean grasslands, Mycologia, Vol. 97(5), pp. 984–995.
Malam Issa, O., Valentin, C., Rajot, J.L., Cerdan, O., Desprats, J-F. and Bouchet, T. (2011) Runoff generation fostered by physical and
biological crusts in semi-arid sandy soils, Geoderma, Vol. 167–168, pp. 22–29.
Mao, Z., Bourrier, F., Stokes, A. and Fourcaud, T. (2014) Three-dimensional modelling of slope stability in heterogeneous montane
forest ecosystems, Ecological Modeling, Vol. 273, pp. 11–22.
Met. Office UK (2014) Wales (Region for Aberdare), climate period: 1981–2010, viewed 26 May 2014,
http://www.metoffice.gov.uk/public/weather/climate/gcjw1qwkq.
Mullan, D. (2013) Soil erosion under the impacts of future climate change: Assessing the statistical significance of future changes and
the potential on-site and off-site problems, Catena, Vol. 109, pp. 234–246.
Pentecost, A. (1979) Aspect and slope preferences in a saxi-colous lichen community, Lichenologist, Vol. 11, pp. 81–83.
Renner, F.G. (1936) Conditions influencing erosion on the Boise River watershed, USDA Technical Bulletin No. 528.
Sheoran, V., Sheoran, A.S. and Poonia, P. (2010) Soil reclamation of abandoned mine land by revegetation: A review, International
Journal of Soil, Sediment and Water, Vol. 3(2), article 13, 22 p.
Smith, A., Bates, P., Freer, J. and Wetterhall, F. (2014) Investigating the application of climate models in flood projection across the
UK, Hydrological Processes, Vol. 28(5), pp. 2810–2823.
St Clair, L.L., Johansen, J.R. and Rushforth, S.R. (1993) Lichens of soil crust communities in the intermountain area of the western
United States, Great Basin Naturalist, Vol. 53(1), pp. 5–12.
Tang, X., Liu, S., Liu, J. and Zhou, G. (2010) Effects of vegetation restoration and slope positions on soil aggregation and soil carbon
accumulation on heavily eroded tropical land of Southern China, Journal of Soils and Sediments, Vol. 33, pp. 131–137.
Wischmeier, W.H. and Mannering, J.V. (1969) Relation of soil properties to its erodibility, Soil Science Society of America Proceedings,
Vol. 33, pp. 131–137.
Worsley, P. (1990) Lichenometry, in Geomorphological techniques, 2nd edition, A. Goudie, M. Anderson, T. Burt, J. Lewin, K. Richards,
B. Whalley and P. Worsley (eds), Routledge, New York, USA, pp. 422–428.


M.B Nadeau SymbioTech Research, Canada; Viridis Terra Innovations, Canada

D.P. Khasa Centre for Forest Research, Canada; Université Laval, Canada
Mining requires the removal of belowground rock-sand materials for extracting valuable minerals and fossil
fuel. Overburden and by-products, including waste rocks and fine tailings, are usually put back to original
locations or piled elsewhere in the surrounding mining area. These by-product materials create newly-formed
and vegetation-free ecosystems, in which soils are structurally disrupted, poor, without organic matter, and
often contaminated by heavy metals or organic compounds. These highly disturbed soils generally have low
beneficial microorganisms, which are fundamental to nutrient cycling and plant nutrition. The ultimate goal
of ecosystem reconstruction is to recover a healthy soil system capable of supporting plant communities.
Therefore, successfully integrating soil microorganisms in ecosystem reconstruction activities can be very
helpful in accelerating the recovery of healthy soil systems and sustainable plant communities on mining sites
after mine closure. In this paper, we will discuss the role and importance of saprophytic and symbiotic
microorganisms for trees, shrubs, and herbaceous plants in degraded lands, and for soil restructuring.
Saprophytic microorganisms play a very important role in ecosystem nutrient cycling. Soil organic matter,
which is mainly made of dead plant, animal, and microorganism tissues, contains significant nutrient pools
unavailable to plants, under insoluble forms. In the process of decomposition, dead tissues are degraded by
saprophytic fungi, bacteria, and actinomycetes and moved into soil horizons, thereby improving soil structure,
water infiltration, and retention before making their journey to plants through mycorrhizal associations.
Organic amendments and residual wastes are often used for capping tailings before revegetating
contaminated mining sites in order to enhance plant survival and establishment. Many of these amendment
materials such as wood chips and char do not usually harbor effective saprophytic microorganisms when
amended on tailings. Hence, some reclamation specialists use the technique of introducing consortia of
microbes before amending organic materials in order to favour rapid biomass decomposition. The
performance of different saprophytes in degrading and decomposing organic compounds will be presented
and their possible use in ecosystem reconstruction will be discussed.
Many studies have demonstrated the ability of root-associated symbiotic microorganisms (microsymbionts)
to protect plants against toxic environments and promote their health, growth, and nutrition in poor soils.
Mine tailings usually have low inoculum potential. Hence, inoculating seeds or seedlings with microbial
inoculants such as ectomycorrhizal (ECM) fungi, arbuscular mycorrhizal (AM) fungi, ericoid mycorrhizal fungi,
nitrogen-fixing bacteria (e.g. Rhizobium sensu lato, Frankia spp.), and plant growth promoting rhizobacteria
(PGPR), before outplanting, can tremendously increase plant health, growth, and establishment success on
mining sites. Case studies on using plant-microsymbiont associations for mine revegetation and ecosystem
reconstruction will be discussed with emphasis on in vivo and in situ experiments conducted by our research
team in Canada (Alberta oil sands and Quebec gold mine), Burkina Faso (gold mine), Madagascar (ilmenite
mine), and South Africa (iron mine).
Mining leads to the extraction of valuable fossil fuels (such as coal and bitumen) and minerals (such as iron,
copper, gold, diamond, and silver) from underground deposits. The extraction of these natural resources is
highly beneficial to the economies of many countries around the world. For example, the mining industry in
Canada contributed C$ 52 billion to Canada’s gross domestic product (GDP) in 2012 and employs, on average,

The triad plant-soil-microbes, an ecofriendly way to reconstruct healthy ecosystems on mining sites M.B Nadeau and D.P. Khasa
400,000 people (Mining Association of Canada, 2015). On the other hand, mining activities can cause negative
effects on the environment. Before mining, natural forests and woodlands are removed. Surface soil and
overburden are disrupted and holes are created at the surface and belowground. Adjacent natural
ecosystems are commonly destroyed in order to pile tons of by-products (waste rocks and fine tailings). At
the end, we are left with a vegetation-free, poor, rocky soil without organic matter and often contaminated
by sulphuric acids (acid mine drainage), hydrocarbons (e.g. naphthenic acids in oil sands) and/or heavy metals
(such as arsenic, mercury, chrome, nickel, iron, and lead) (Fung and Macyk, 2000; Renault et al., 2002). Then,
these contaminants, without proper controls in place, can be transported to surface and groundwater and
pollute their respective ecosystems. In order to mitigate these negative effects on the environment, healthy
ecosystems have to be rebuilt as quickly as possible following mine closure. However, plant communities
generally struggle to successfully colonise and establish themselves on these mining sites due to the extreme
nature of post-mining materials.
Newly-formed mining anthropogenic soils generally have low levels of the microbial activity essential for
ecosystem reconstruction, nutrient cycling, and plant nutrition. Many studies indicate that microbial status
of tailings and overburden is considerably reduced and nearly non-existent during storage and subsequent
disposal (Danielson et al., 1983; Malajczuk et al., 1994; Pfleger et al., 1994; among others). Bois et al. (2005)
found low mycorrhizal and endophytic status on oil sand tailings and overburden in Alberta. Nadeau and
Khasa (2015) also demonstrated low ectomycorrhizal status on pure fine tailings and waste rocks of a Quebec
gold mine. Furthermore, Vézina et al. (2012) showed an absence of ectomycorrhizal (ECM) fungi and a very
low presence of rhizobia and arbuscular mycorrhizal (AM) fungi on fine tailings and waste rocks of a Burkina
Faso gold mine. After the emergence of mostly sterile by-product piles, soil microorganisms, like plants, have
to recolonise the site by wind, water, and/or animal dispersal from surrounding healthy ecosystems
(Jumpponen et al., 2005). This process is known to be relatively slow in harsh environments such as mining
sites due to extreme conditions that only favour a few adapted microbial species and strains (Jumpponen et
al., 2002), and nature alone may take hundreds of years to retrieve plant-benefitting microbial activity over
all disturbed soil areas. The main objective of ecosystem reconstruction is to recover healthy soil systems
capable of harbouring sustainable plant communities.
In this paper, we aim to corroborate that ecosystem reconstruction on post-mining materials (overburden,
fine tailings, and waste rocks) can be accelerated by integrating soil microorganisms in our mine reclamation
and restoration activities. The role and importance of saprophytic and symbiotic microorganisms for early
soil and ecosystem development, soil restructuring, trees, shrubs, and herbaceous plants in degraded lands
will be discussed. Furthermore, the performance of different saprophytes in degrading and decomposing
organic compounds will be presented and their possible use in ecosystem reconstruction will be considered.
Finally, case studies on using plant-microsymbiont associations for mine revegetation and ecosystem
reconstruction will be examined with emphasis on in vivo and in situ experiments conducted by our research
team in Canada (Alberta oil sands and Quebec gold mine), Burkina Faso (gold mine), Madagascar (ilmenite
mine), and South Africa (iron mine).
Saprophytic microorganisms, the most important decomposers in nature, consist of bacteria, fungi, and
actinomycetes that acquire most of their nutrients by degrading and decomposing soil organic matter (dead
plant, animal, and microorganism tissues) (Bot and Benites, 2005). These microorganisms use dead tissues
as a carbon source and food. As they break down the tissues, excess nutrients are released into soil horizons
in forms available to plants (Bot and Benites, 2005). As they move into soil horizons, non-humic substances
of the organic matter (5 to 25%), which are nutritional carbohydrates easily degraded by saprophytes,
stimulate aggregate formation and greater soil structure thanks to their ability to bind to inorganic soil
particles (Van den Driessche, 1991). As a result, these substances play a fundamental role in restructuring
disturbed mineral soil. Humic substances, on the other hand, which are decay-resistant by-products of

decomposition, increase soil cation exchange capacity thanks to their negatively-charged binding sites (Van
den Driessche, 1991). This process of decomposition is called mineralization and is considered the core of
ecosystem nutrient cycling (Bot and Benites, 2005). Both degraded products enhance soil water retention
and infiltration capacity, buffer soil pH, reduce soil evaporation, and protect plants against cold, wind, heat,
and drought desiccation (Bot and Benites, 2005). Without these saprophytic microorganisms in partnership
with decomposer fauna (especially earthworms), ecosystem nutrient cycling and soil restructuring would be
substantially slower, indeed almost non-existent.
Few studies have looked at the performance of different saprophytic microorganisms in degrading organic
materials such as organic matter, residual waste, and hydrocarbons. Hattenschwiler et al. (2005) showed that
various species and diversity of decomposers can influence differently carbon and nutrient turnover rates in
litter. Other researchers such as Setala and McLean (2004) demonstrated that fungal community efficiency
in degrading raw coniferous humus and releasing it into soil can be enhanced by increasing the number of
taxa introduced in the system. White-rot fungi are more efficient in degrading wood from deciduous trees,
whilst brown-rot fungi prefers coniferous substrates (Tuor et al., 1995). Some white-rot fungi (e.g. strains of
Pleurotus sp. and Phanerochaete sp.) are more effective in degrading lignin and polycyclic aromatic
hydrocarbons than others (Hatakka, 2002; Andersson and Henrysson, 1996; Field et al., 1992). Several
symbiotic fungi (AM fungi, ECM fungi, and ericoid mycorrhizal fungi) are also capable of degrading and
capturing nutrients from organic matter, but their decomposition efficiency is generally low compared to
wood decomposing fungi (Bending and Read, 1997; Hodge et al., 2001; Rineau et al., 2012). In addition,
white-rot fungi are able to mineralise nutrients and degrade toxic xenobiotics from residual waste because
of the non-specific nature of their extracellular lignin-mineralising enzymes (Asgher et al., 2008). Working
with bacteria, Afzal et al. (2012) found that soil inoculation with Pantoea sp. and Pseudomonas sp.
considerably enhanced hydrocarbon degradation. The highest degradation rate was observed when both
strains were applied to soil (Afzal et al., 2012). In vitro selection and adaptation by prior exposure of microbial
communities to organic pollutants can improve microbial degradation rates (Leahy and Colwell, 1990).
Ghazali et al. (2004) showed that some consortia of bacterial strains, selected based on their ability to grow
in crude oil, were more efficient in degrading hydrocarbon pollutants than others. Consequently, under
similar environmental conditions, certain ecosystems, depending on their saprophytic communities (strain,
species, and diversity), may be more effective in degrading and decomposing organic materials than others.
Organic amendments and residual waste are often added to tailings before revegetating mining sites, which
is normally beneficial to plant survival and establishment. After being produced, some of these organic
materials are practically sterile (very low microbial activity – e.g. ash, char, wood chips, etc.) and, most likely,
do not harbor the most effective degrading and decomposing communities of saprophytic microorganisms
when amended on tailings. It may take years before highly efficient saprophytes colonise the amended site,
especially if we are using, as recommended, deciduous wood chips while being surrounded by pure
coniferous forests. In this case, we could potentially speed up the process of decomposition (thus,
accelerating ecosystem nutrient cycling and soil restructuring) through bioaugmentation by developing and
introducing specific consortia of highly efficient saprophytes (specific to the amended materials), which
would hasten soil restructuring and ecosystem reconstruction after mine closure; thereby benefitting mining
companies. Some of our specialist reclamation colleagues already use this technique of introducing consortia
of microbes before amending organic materials on tailings in order to favour rapid biomass decomposition
(Bordeleau and Juge, 2014). Their alkaline tailing soil amended with ash in the Quebec boreal forest has
already shown signs of podzolisation in the first 60 cm and the formation of an indurated layer preventing
the spread of contaminants less than ten years after reclamation (Bordeleau and Juge, 2014), which is
amazing. However, much research needs to be done in order to develop such a registered bioproduct that
can be commercially sold on the market. Our team hopes to have the opportunity to fulfil this task in the
coming years.

Mycorrhizal fungi form a symbiotic relationship with the roots of nearly all terrestrial plants (Fortin et al.,
2008). Arbuscular mycorrhizal fungi, which colonise around 70% of plants (bryophytes and vascular plants),
include approximately 200 species of the phylum Glomeromycota which form arbuscular and vesicular
tissues inside root cells (Fortin et al., 2008). They secrete a glycoprotein called glomalin (a substance that is
hard to decompose), which favours soil aggregate formation and contributes directly to the soil organic pool,
thereby improving soil structure (Smith and Read, 2008). On the other hand, ectomycorrhizal fungi, which
colonise around 5% of vascular plants (coniferous, deciduous, and evergreen tree species – most trees in the
boreal forest), are conservatively estimated to be around 7,750 species of the phyla Ascomycota and
Basidiomycota (Rinaldi et al., 2008). They produce an intercellular layer of hyphae in roots, called the hartig
net, and a mantle surrounding the exterior of fine roots (Horton and Bruns, 2001). These ECM fungi protect
plants against toxic environments (heavy metals, pollutants, etc.), thanks to the formation of mantle, all
materials have to go through the fungus first, which sequesters toxic elements in its tissues, before entering
fine roots (Khan, 2006). Then, ericoid mycorrhizal (ERM) fungi, which also colonise around 5% of vascular
plants (ericaceous shrubs), consist of few dozen species of mostly dark septate Ascomycota and form
intracellular mycelial squads in plant roots (Fortin et al., 2008). Today, these mycorrhizal fungi are widely
known for their ability to enhance plant mineral and water absorption in soil, improve soil aggregation, and
protect plants against soil pathogens and environmental stress (Smith and Read, 2008). Plants alone do not
have access to nutritional elements stored in rock particles (Berner et al., 2012). However, through the
production of organic acids, such as oxalic, malic and citric acids, mycorrhizal fungi with the aid of plant
growth promoting rhizobacteria solubilise nutritional elements (P, K, Ca, Mg, Fe, etc.) from rock particles
(biological weathering) and then transfer them to plant roots in exchange for photosynthates; thereby,
improving plant nutrition (Berner et al., 2012). Moreover, mycorrhizal fungi cover a greater soil surface area
than plant roots, thanks to their smaller mycelium size, which further promote plant mineral absorption and
boost plant water uptake; thus, enhancing plant resistance against drought (Allen et al., 2003). Mycorrhizal
fungi have been shown to play a very important role in improving plant survival, health, growth, and
establishment on degraded and contaminated sites, with some able to degrade petroleum hydrocarbon
(PHC) waste (Robertson et al. 2007).
Pure mine tailings, waste rocks, and spoils without organic matter do not contain any nitrogen, which can be
extremely problematic to plant development (Taner et al., 1986). Some symbiotic microorganisms associated
with plant roots have the ability to fix relatively large amounts of atmospheric nitrogen using their
nitrogenase enzyme; thus, improving plant nitrogen uptake, health, and growth (Franche et al., 2009). In the
process, atmospheric nitrogen is converted to ammonia, a form available to plants, in root nodules (Franche
et al., 2009). These nitrogen-fixing microorganisms consist of bacteria from the genus Rhizobium sensu lato
with six phylogenetically distinct genera of rhizobia (Allorhizobium, Azorhizobium, Bradyrhizobium,
Mesorhizobium, Rhizobium, and Sinorhizobium) and actinomycetes from the genus Frankia (Franche et al.,
2009). Respectively, they colonise and form nodules in roots of leguminous plants, shrubs, and trees (species
from the Fabaceae family with over 17,000 species) and in roots of actinorrhizal shrub and tree species from
8 families and 24 genera (Betulaceae: Alnus; Casuarinaceae: Gymnostoma, Casuarina, Allocasuarina,
Ceuthostoma; Coriariaceae: Coriaria; Datiscaceae: Datisca, Elaeagnaceae: Elaeagnus, Hippophae,
Shepherdia; Myricaceae: Myrica, Comptonia; Rhamnaceae: Colletia, Discaria, Kenthrothammus, Retanilla,
Talguenea, Trevoa, Ceanothus; Rosaceae: Dryas, Purshia, Cowania, Cercocarpus, Chamaebatia) (Franche et
al., 2009). They play a huge role in ecosystem nitrogen cycling, especially on nitrogen-free disturbed sites
(Franche et al., 2009). Some diazotrophic PGPR such as Azotobacter sp. and Azospirillum sp. also fix
atmospheric nitrogen freely in soil. However, they do not form nodules and are generally not as effective in

fixing nitrogen and contributing to plant nitrogen pool as nodular Rhizobium sensu lato and Frankia spp.
(Hayat et al., 2010).
Plant growth promoting rhizobacteria (PGPR) are bacteria that thrive in the rhizosphere of all terrestrial
plants (Chanway, 1997). These microorganisms promote plant health and growth by (1) synthesising plant-
benefitting hormones, (2) facilitating plant nutrient absorption by mycorrhizal fungi through organic acid
secretion, and (3) preventing pathogen attack through antibiotic production in plant rhizosphere (Chanway,
1997). PGPR include species from the genera Pseudomonas, Bacillus, Rhizobium, Azotobacter, Burkholderia,
Achromobacter, Agrobacterium, Micrococcus, Aerobacter, Flavobacterium, and Erwinia, among others
(Hayat et al., 2010). These microorganisms, like mycorrhizal fungi and N-fixing microsymbionts, may
contribute to rapid ecosystem reconstruction as a result of enhanced plant health, growth, nutrition, and
establishment in degraded lands.
In the last decade, our research team has done many studies on mine revegetation and reforestation in
mining areas using trees and their associated symbiotic microorganisms. We started our research in the
Alberta oils sands industry in Canada. In a greenhouse experiment, Bois et al. (2006) found that shoot dry
biomass of white spruce (Picea glauca) and jack pine (Pinus banksiana) seedlings was increased by 30% when
inoculated with Suillus tomentosus (an ECM fungus) on salt-affected sand tailings after four weeks of growth
under 200 mM of NaCl. On the other hand, Hebeloma crustulinforme and Laccaria bicolor had no effect on
shoot dry biomass (Bois et al., 2006). Seedling health (chlorophyll fluorescence) was improved by 80% when
jack pine seedlings were inoculated with H. crustuliniforme (Bois et al., 2006). Inoculation had no effect on
white spruce seedling health (Bois et al., 2006). In field trials, Quoreshi (2008) demonstrated, after three
growing seasons, that plot volume index of jack pine seedlings growing on overburden was enhanced by 50%
and 20% while plot volume index of white spruce seedlings was enhanced by 52% and 90% on two studied
sites when inoculated with ECM fungi (H. crustuliniforme, L. bicolor, S. tomentosus, or Amphinema byssoides).
Furthermore, white spruce and jack pine total dry biomass was significantly increased by 60 to 70% on both
sites when inoculated with the ECM fungi (Quoreshi, 2008). In another field study, Onwuchekwa et al. (2014)
showed, after two growing seasons, that height growth rate of jack pine seedlings was improved by 100%
when inoculated with H. crustuliniforme and between 40 and 95% when inoculated with other ECM fungal
treatments (L. bicolor, S. tomentosus, species mix) on piled top soil and overburden. Jack pine stem volume
was increased by 60% when inoculated with H. crustuliniforme and around 40% when inoculated with L.
bicolor, H. crustuliniforme + L. bicolor, H. crustuliniforme + S. tomentosus, or L. bicolor + S. tomentosus
(Onwuchekwa et al., 2014). On the other hand, height growth rate of white spruce was enhanced by 60%
when inoculated with H. crustuliniforme and between 20 and 50% when inoculated with the other ECM
fungal treatments (Onwuchekwa et al., 2014). Then, plot volume index of white spruce seedlings was
increased by more than 100% when inoculated with H. crustuliniforme + L. bicolor (Onwuchekwa et al., 2014).
Finally, survival rate of white spruce seedlings was improved by 75% when inoculated with H. crustuliniforme
+ S. tomentosus and around 65% when inoculated with S. tomentosus, H. crustuliniforme + S. tomentosus, or
H. crustuliniforme + L. bicolor + S. tomentosus (Onwuchekwa et al., 2014). As a result, the inoculation of
specific ECM fungi on white spruce and jack pine can greatly improve seedling health, growth, and
establishment on oil sand spoils. Today, one registered ectomycorrhizal product, TreeBoost HC EctoSlurry, is
available on the market. Furthermore, two other amazing biofertiliser products will be registered soon under
the fertiliser act of the Canadian Food Inspection Agency.
Other studies were conducted in the Alberta oil sands using green alder (Alnus viridis ssp. crispa) and speckled
alder (Alnus incana ssp. rugosa). In a greenhouse experiment, Bissonnette et al. (2014) found that dry
biomass of alders on sand tailings was 15 times greater when inoculated with Frankia (a nitrogen-fixing
actinomycete) compared to control without inoculation. In field trials, Quoreshi et al. (2008) demonstrated,
after two growing seasons, that alder stem volume was improved by 65% and 76%, alder height growth was

enhanced by 30% and 30%, and alder root collar diameter was increased by 20% and 50% when inoculated
with Frankia and planted on overburden on two studied sites. In another field study, Lefrançois et al. (2010)
showed, after two growing seasons, that Frankia-inoculated alders considerably improved soil quality – up
to 50% increase in petroleum hydrocarbon mineralization, decrease in soil pH (±1), 70% decrease in plant-
available sodium, and considerable increase in soil organic matter content (up to 6 times higher) on
overburden. Overall, Frankia had no effect on alder survival rate (Bissonnette et al., 2014; Lefrançois et al.,
2010; Quoreshi, 2008). Alders also form a symbiotic relationship with ECM fungi. In a growth chamber
experiment, alder seedling health, height, and dry biomass were two times greater when inoculated with a
specific Paxillus involutus strain on Canadian oil sands like sodic soils (manuscript in preparation). In the same
study, other ECM fungal species and P. involutus strains were not as effective in improving seedling health
and growth. Accordingly, both Frankia and ECM fungi have the ability to improve alder health and growth on
oil sand spoils.
Another study was carried out on an ilmenite mine in operation in Fort Dauphin, south-east of Madagascar.
The mining company has pledged to reclaim the mining sites after closure. Mimosa latispinosa is a promising
native pioneer nitrogen-fixing leguminous shrub of the region that could be used in ecosystem reconstruction
after mine closure. In a greenhouse trial, Sarasin et al. (2011) tested the effect of inoculating M. latispinosa
with Rhizophagus irregularis (RI – an AM fungus) and two Malagasy Bradyrhizobium strains (B1 and B2 –
nitrogen-fixing bacteria) on seedling growth. Sarasin et al. (2011) found that dual inoculation of M. latispinosa
roots with Bradyrhizobium sp. and R. irregularis significantly enhanced seedling growth by 100% (RI + B1) and
60% (RI + B2 and RI + B1 + B2) compared to control. Field trials are planned in the foreseeable future before
mine closure.
Our research team also did a reclamation research project on a gold mine located in the north-east of Burkina
Faso. In the greenhouse trial, Vézina et al. (2012) demonstrated that dry biomass of Acacia Senegal seedlings,
a native leguminous shrub, was increased by 56% on sand tailings when inoculated with local Glomus
aggregatum (an AM fungus) after 12 weeks of growth. On the contrary, Rhizobium sp. and Rhizophagus
irregularis inoculation had no effect on seedling growth (Vézina et al., 2012). In situ field results will be
published soon and they are quite promising.
In South Africa, our team did a greenhouse experiment using mine tailings from an iron ore mine of the
Northern Cape Province. Adeleke et al. (2012) inoculated Pinus patula seedlings with Pisolithus tinctorius, P.
involutus, L. bicolor, or S. tomentosus. After 24 weeks, the inoculation of S. tomentosus improved by
approximately 50% shoot and 200% root growth of P. patula seedlings on mine tailings while other ECM fungi
had no effect (Adeleke et al., 2012).
Finally, we are currently working on a phytorestoration research project on waste rocks and fine tailings of a
Quebec gold mine in Canada. In a greenhouse trial, Nadeau et al. (2014) inoculated white spruce seedlings
with different ECM fungi (Cadophora finlandia, Tricholoma scalpturatum, or H. crustuliniforme) and PGPR
(Pseudomonas putida, Rhizobium radiobacter, or Azotobacter chroococcum). Seedlings were grown for 32
weeks directly on waste rocks and fine tailings (Nadeau et al., 2014). The inoculation of seedlings with C.
finlandia and T. scalpturatum (two ECM fungi isolated from scarce healthy trees naturally regenerating on
the mining site) enhanced white spruce health (% of healthy foliage) by 130% and 65%, respectively (Nadeau
et al., 2014). H. crustuliniforme with provenance of natural forest had no effect on seedling health (Nadeau
et al., 2014). The inoculation of seedlings with A. chroococcum (a nitrogen-fixing PGPR) improved white
spruce health by 40% compared to control (Nadeau et al., 2014). Finally, the inoculation of seedlings with P.
putida (a PGPR isolated from scarce healthy trees naturally regenerating on the mining site) increased
seedling shoot growth by 25% compared to control (Nadeau et al., 2014). Many field trials are currently in
progress with different tree and shrub species of the boreal forest (white spruce, jack pine, green alder, and
speckled alder) and associated microsymbionts (Khasa et al., 2014). Results shall be available shortly.
As we have seen in these case studies, inoculating plants with specific symbiotic microorganisms can
considerably improve plant survival, health, growth, and establishment in mining areas; thereby, promoting
litter accumulation and rapid soil and ecosystem reconstruction. However, in vivo and in situ selection of the

most suitable plant-microsymbiont associations is crucial for the success of mine revegetation, reforestation,
and ecosystem reconstruction activities; because not all symbiotic microorganism species and strains are
efficient in enhancing plant fitness on these degraded lands. We are looking forward to following all current
and future field trials over time.
Phytobial restoration consists of using plants and their associated microorganisms for the restoration,
revegetation, and reforestation of degraded ecosystems. In the diagram below, we show the eight steps to
follow, as mine reclamation specialists, in order to integrate phytobial restoration in our ecosystem
reconstruction activities during mining operations or after mine closure (Figure 1). Firstly, we assess the
microbial status of mine spoils, waste rocks, and tailings. Do they contain active plant-benefitting microbial
communities? If they do, inoculating plants before outplanting and sowing may not be necessary; however,
proper biostimulation and management of indigenous beneficial microorganisms is recommended. Secondly,
if they do not, we have to find out whether they are any plants or seedlings naturally regenerating on the
mining site. Thirdly, if the answer is yes, we have to collect roots from the healthy plants or seedlings naturally
regenerating on the mining site and then, fourthly, isolate site-adapted microorganisms from the collected
roots under aseptic conditions in the lab. Scarce colonization of mining sites by plants happens relatively
quickly close to natural ecosystems. However, if the answer is no, we have to acquire microorganisms from
local microbial collections. Fifthly, we conduct a small laboratory experiment in which we select promising
microorganisms by growing several strains and species on mine tailings under aseptic conditions. Sixthly, we
select the best plant-microsymbiont associations, which trigger the greatest increase in plant fitness and
development, by growing a wide variety of combinations on mine tailings under greenhouse conditions.
Seventhly, we conduct some small-scale field testing on mine spoils and tailings in order to confirm our lab
and greenhouse results. Then, eighthly, we choose our best treatment for large-scale phytobial restoration
activities. Finally, microsymbiont inoculation on plants should be done by highly-qualified professionals,
because the success of microbial inoculation and subsequent root colonization is highly dependent on the
inoculum quality and method used (Afzal et al., 2012; Quoreshi, 2008).

Mining materials have very low levels of effective microbial activity compared to well-developed soils. These
materials are a very harsh environment for plant establishment and development. The triad plant-soil-
microbes is a dynamic interrelated system. All three components of this triad should be managed in order to
favour rapid soil and ecosystem development during mining operations and after mine closure. However,
today, most reclamation specialists still disregard soil microbes as an integrative and fundamental
component of dynamic functioning ecosystems. Here, we indicated that successfully integrating symbiotic
and saprophytic microorganisms in our ecosystem reconstruction and restoration activities can be very
helpful in accelerating the recovery of healthy soil systems and sustainable plant communities in mining
areas.
The authors would like to thank all colleagues who contributed to these studies. Furthermore, the authors
would like to show their appreciation to all industrial and academic partners: Suncor Energy, Imperial Oil,
Syncrude Canada, 2020 Seeds Labs, Rio Tinto, IAMGOLD, Kumba Iron Ore Resources, Century Mining,
Université de Sherbrooke (Team of Sébastien Roy), McGill University (Team of Charles Greer), the University
of Saskatchewan (Team of John V. Headley), the National Research Council of Canada, Centre SEVE, the
Centre for Forest Research, and NSERC of Canada.
Adeleke, R.A., Cloete, T.E., Bertrand, A. and Khasa, D.P. (2012) Iron ore weathering potentials of ectomycorrhizal plants, Mycorrhiza,
Vol. 22, pp. 535–544.
Afzal, M., Yousaf, S., Reichenaver, T.G. and Sessitsch, A. (2012) The inoculation method affects colonization and performance of
bacterial inoculant strains in the phytoremediation of soil contaminated with diesel oil, International Journal of
Phytoremediation, Vol. 14, pp. 35–47.
Allen, M.F., Swenson, M., Querejeta, J.I., Egerton-Warburton, L.M. and Treseder, K.K. (2003) Ecology of mycorrhizae: a conceptual
framework for complex interactions among plants and fungi, Annual Review of Phytopathology, Vol. 41, pp. 271–303.
Andersson, B.E. and Henrysson, T. (1996) Accumulation and degradation of dead-end metabolites during treatment of soil
contaminated with polycyclic aromatic hydrocarbons with five strains of white-rot fungi, Applied Microbiology and
Biotechnology, Vol. 46 (5-6), pp. 647–652.
Asgher, M. Bhatti, H.N., Ashraf, M. and Legge, R.L. (2008) Recent developments in biodegradation of industrial pollutants by white-
rot fungi and their enzyme system, Biodegradation, Vol. 19(6), pp. 771–783.
Bending, G.D. and Read, D.J. (1997) Lignin and soluble phenolic degradation by ectomycorrhizal and ericoid mycorrhizal fungi,
Mycological Research, Vol. 101(11), pp. 1348–1354.
Berner, C., Johansson, T. and Wallander, H. (2012) Long-term effect of apatite on ectomycorrhizal growth and community structure,
Mycorrhiza, DOI: 10.1007/s00572-012-0438-y.
Bissonnette, C., Fahlman, B., Peru, K.M., Khasa, D.P., Greer, C.W., Headley, J.V. and Roy, S. (2014) Symbiosis with Frankia sp. benefits
the establishment of Alnus crispa and Alnus rugosa in tailings sand from the Canadian oil sands industry, Ecological
Engineering, Vol. 68, pp. 167–175.
Bois, G., Piché, Y., Fung, M. and Khasa D.P. (2005) Mycorrhizal inoculum potentials of pure reclamation materials and revegetated
tailing sands from the Canadian oil sand industry, Mycorrhiza, Vol. 15, pp. 149–158.
Bios, G., Bigras, F.J., Bertrand, A., Piché, Y., Fung, M. and Khasa, D.P. (2006) Ectomycorrhizal fungi affect the physiological responses
of Picea glauca and Pinus banksiana seedlings exposed to an NaCl gradient, Tree physiology, Vol. 26(9), pp. 1185–1196.
Bordeleau, L.M. and Juge, C. (2014) Establishing permanent vegetation on the Opemiska tailing site in mid-north Quebec: the
successful soil-plant-microbes combination, Canadian Reclamation, Vol. 14(1), pp. 28–35.
Bot, A. and Benites, J. (2005) The importance of soil organic matter: key to drought-resistant soil and sustained food production, Food
and Agriculture Organization of the United Nations, Rome, Italy, 80 pp.
Chanway, C.P. (1997) Inoculation of tree roots with plant growth promoting soil bacteria: an emerging technology for reforestation,
Forest Science, Vol. 43, pp. 99–112.
Danielson, R.M., Visser, S. and Parkinson, D. (1983) Microbial activity and mycorrhizal potential of four overburden types used in
the reclamation of extracted oil sands, Canadian Journal of Soil Science, Vol. 63, pp. 363–375.
Field, J.A., Jong, E., Feijoo, G. and Bont, J. (1992) Biodegradation of polycyclic aromatic hydrocarbons by new isolates of white-rot
fungi, Applied Environmental Microbiology, Vol. 58(7), pp. 2219–2226.
Fortin, J.A., Plenchette, C. and Piché, Y. (2008) Les mycorhizes : la nouvelle révolution verte, 1st ed, Éditions MultiMondes, Québec,
Canada, 138 pp.

Franche, C., Lindstrom, K. and Elmerich, C. (2009) Nitrogen-fixing bacteria associated with leguminous and non-leguminous plants,
Plant and Soil, Vol. 321, pp. 35–59.
Fung, M.Y.P. and Macyk, T.M. (2000) Reclamation of oil sands mining areas, in Reclamation of drastically disturbed lands, R.I.
Barnhisel, R.G. Darmody, and W.L. Daniels (eds), Agronomy Monograph Vol. 41, American Society of Agronomy, Crop Science
Society of America, and Soil Science Society of America, Madison, WI, USA, pp. 755–774.
Ghazali, F.M., Rahman, R.N., Salleh, A.B. and Basri, M. (2004) Biodegradation of hydrocarbons in soil by microbial consortium,
International Biodeterioration and Biodegradation, Vol. 54, pp. 61–67.
Hatakka, A. (2002) Lignin-modifying enzymes from selected white-rot fungi: production and role in lignin degradation, FEMS
Microbiology Reviews, Vol. 13(2-3), pp. 125–135.
Hattenschwiler, S., Tiunov, A.V. and Scheu, S. (2005) Biodiversity and litter decomposition in terrestrial ecosystems, Annual Review
of Ecological Evolution Systems, Vol. 36, pp. 191–218.
Hayat, R., Ali, S., Amara, U., Khalid, R. and Ahmed, I. (2010) Soil beneficial bacteria and their role in plant growth promotion: a review,
Annals of Microbiology, Vol. 60, pp. 579–598.
Hodge, A., Campbell, C.D. and Fitter, A.H. (2001) An arbuscular mycorrhizal fungus accelerates decomposition and acquires nitrogen
directly from organic material, Nature, Vol. 413, pp. 297–299.
Horton, T.R. and Bruns, T.D. (2001) The molecular revolution in ectomycorrhizal ecology: peeking into the black-box, Molecular
Ecology, Vol. 10, pp. 1855–1871.
Jumpponen, A., Trappe, J.M. and Cazares, E. (2002) Occurence of ectomycorrhizal fungi on the forefront of retreating Lyman Glacier
(Washington, USA) in relation to time since deglaciation, Mycorrhiza, Vol. 12, pp. 43–49.
Jumpponen, A. and Egerton-Warburton, L.M. (2005) Mycorrhizal fungi in successional environments: a community assembly model
incorporating host plant, environmental, and biotic filters, in The fungal community: its organization and role in the
ecosystem, J. Dighton, J.F. White and P. Oudemans (eds.), CRC Press, Boca Raton, Florida , USA, pp. 139–168.
Khan, A.G. (2006) Mycorrhizoremediation – an enhanced form of phytoremediation, Journal of Zhejiang University SIENCE B, Vol. 7,
pp. 503–514.
Khasa, D.P., Nadeau, M.B., Roy, S. and Greer, C.W. (2014) Phytobial remediation of Sigma-Lamaque gold mine waste lands in Val d’Or,
Québec, Canadian Reclamation, Vol. 14(1), pp. 36–39.
Leahy, J.G. and Colwell, R.R. (1990) Microbial degradation of hydrocarbons in the environment, Microbiology and Molecular Biology
Review, Vol. 78(4), pp. 305–315.
Lefrançois, E., Quoreshi, A., Khasa, D.P., Fung, M., Whyte, L.G., Roy, S. and Greer, C.W. (2010) Field performance of alder-Frankia
symbionts for the reclamation of oil sands sites, Applied Soil Ecology, Vol. 46(2), pp. 183–191.
Malajczuk, N., Reddell, P. and Brundrett, M. (1994) Role of ectomycorrhizal fungi in mineland reclamation, in Mycorrhizae and plant
health, F.L. Pfleger and R.G. Linderman (eds), APS Press, St Paul, Minnesota, USA, pp. 83–100.
Mining Association of Canada (2015) Website, viewed February, 2015, www.mining.ca.
Nadeau, M.B., Azaiez, A., Roy, S., Greer, C.W. and Khasa, D.P. (2014) Development of a new green technology for the revegetation of
abandoned gold mine tailings using specific symbiotic microorganisms associated with Picea glauca, in Proceedings 9th
International Conference on Mine Closure, 1-3 October 2014, Johannesburg, South Africa (in press).
Nadeau, M.B. and Khasa D.P. (2015) Ectomycorrhizal fungi and bacterial PGPR improve white spruce seedling health, growth, and
nutrition on Precambrian rocky gold mine tailings, in Development of a new green technology for the revegetation of
abandoned mine tailings using specific symbionts associated with Picea glauca, Nadeau, M.B. (ed), Archimede, Université
Laval, Québec, Canada, pp. 125–166.
Onwuchekwa, N.E., Zwiazek, J.J., Quoreshi, A. and Khasa, D.P. (2014) Growth of mycorrhizal jack pine and white spruce seedlings
planted in oil sands reclaimed areas, Mycorrhiza, Vol. 24(6), pp. 431–441.
Pfleger, F.L., Stewart, E.L. and Noyd, R.K. (1994) Role of VAM fungi in mine land revegetation, in in Mycorrhizae and plant health, F.L.
Pfleger and R.G. Linderman (eds), APS Press, St Paul, Minnesota, USA, pp. 47–81.
Quoreshi, A.M. (2008) The use of mycorrhizal biotechnology in restoration of disturbed ecosystem, in Mycorrhizae: sustainable
agriculture and forestry, Z.A. Siddiqui, M.S. Akhtar and K. Futai (eds), Springer, ISBN: 978-1-4020-8769-1, pp. 303–320.
Renault, S., Sailerova, E. and Fedikow, M.A.F. (2002) Phytoremediation of mine tailings and bio-ore production: progress report on
seed germination, survival and metal uptake of seedlings planted at central Manitoba (AU) minesite, Trade and Mines,
Manitoba Geological Survey, Vol. 1, pp. 255–265.
Rinaldi, A.C., Comandini, O. and Kuyper, T.W. (2008) Ectomycorrhizal fungal diversity: separating the wheat from the chaff, Fungal
diversity, Vol. 33, pp. 1–45.
Rineau, F., Roth, D., Shah, F., Smits, M., Johansson, T., Canback, B., Olsen, P.B., Persson, P., Grell, M.N., Lindquist, E., Grigoriev, I.V.,
Lange, L. and Tunlid, A. (2012) The ectomycorrhizal fungus Paxillus involutus converts organic matter in plant litter using a
trimmed brown rot mechanism involving Fenon chemistry, Environmental Microbiology, Vol. 14(6), pp. 1477–1487.
Robertson, S.J., McGill, W.B., Massicotte, H.B., and Rutherford, P.M. (2007). Petroleum hydrocarbon contamination in boreal forest
soils: a mycorrhizal ecosystems perspective, Biological Reviews, Vol. 82(2), pp. 213–240.
Sarasin, G., Rakotoarimanga, N., Ramanankierana, H. and Khasa, D.P. (2011) Simple, dual, and triple inoculation of woody legume
(Mimosa latispinosa) with rhizobial bacteria and an arbuscular mycorrhizal fungus in glasshouse in Madagascar, in
Biotechnologie des symbioses racinaires en restauration écologique des écosystèmes dégradés à Madagascar, G. Sarasin (ed),
Archimede, Université Laval, pp. 64–78.
Smith, S.E. and Read, D.J. (2008) Mycorrhizal Symbiosis, 3rd ed, Academic Press, London, UK, 636 pp.
Setala, H. and McLean, M.A. (2004) Decomposition rate of organic substrates in relation to the species diversity of soil saprophytic
fungi, Oecologia, Vol. 139(1), pp. 98–107.

Taner, M.F., Trudel, P. and Perrault, G. (1986) Géochimie de la biotite associée à certains gisements d’or de Val d’Or, Malartic et
Chibougamau, Québec, Canadian Mineralogist, Vol. 24, pp. 761–774.
Tuor, U., Winterhalter, K. and Fiechter, A. (1995) Enzymes of white-rot fungi involved in lignin degradation and ecological
determinants for wood decay, Journal of Biotechnology, Vol. 41(1), pp. 1–17.
Van den Driessche, R. (1991) Mineral nutrition of conifer seedlings, CRC Press, Boca Raton, Florida, USA, 275 pp.
Vézina, M.M., Yonli, H., Campagnac, E., Baudrand, J., Sanon, K. and Khasa, D.P. (2012) Root symbionts: a tool for remediation of gold
mine tailings, in Proceedings 7th International Conference on Mine Closure, 25-27 September 2012, Perth, Australia, pp. 187–
196.

L.A. Barber O’Kane Consultants Inc., Canada

J. Bockstette Renewable Resources, University of Alberta, Canada
D.O. Christensen O’Kane Consultants Inc., Canada
L.K. Tallon O’Kane Consultants Inc., Canada
S.M. Landhäusser Renewable Resources, University of Alberta, Canada
Syncrude completed construction of a field-scale soil cover system trial at the Aurora North mine operations
in Fort McMurray, Alberta, Canada in 2012. Syncrude, university researchers, and consultants developed the
study design with an overall objective of determining appropriate soil cover system designs and capping
depth(s) for overburden reclamation at the Aurora North mine operation. Specifically, the study was designed
to address the uncertainty of reclaiming upland areas using coarse textured surficial soil materials that
contain embedded naturally occurring oil sand that is comprised of petroleum hydrocarbons over lean oil sand
overburden that also contains petroleum hydrocarbons. The trial consisted of 36 one hectare cells, made up
of 12 treatment options varying in capping material thicknesses and types. All treatments were constructed
in triplicate. The research cells were instrumented and planted with three native boreal forest tree species in
varying densities.
Key mechanisms and processes that influence cover system performance and tree growth include freeze/thaw
cycles and water retention characteristics. Instrumentation was installed to monitor water balance
parameters and groundwater, and to collect pore-water. Initial results indicate that peat surface cover system
materials had the lowest net percolation due to higher water storage capacity, and exhibited delayed soil
warming and freezing. Tree mortality was 3% on salvaged forest floor soil capping treatments and 4% on peat
capping treatments. Tree seedling growth after three growing seasons was up to 48% higher on treatments
capped with mineral dominated forest floor material compared with treatments capped with organic peat
material. The influence of lower subsoil material types and their arrangements on tree seedling performance
is still unclear. This is a long-term study and results will serve to identify appropriate cover system
configurations and guide future reclamation operations in oil sands mining.
Syncrude Canada Ltd. (Syncrude) developed a field-scale soil capping study at their Aurora North mine
operations in Fort McMurray, Alberta, Canada. The overall research objective of the Aurora Soil Capping
Study (ASCS) was to determine appropriate soil cover system designs and capping depth(s) for reclamation
using readily available surface materials, while understanding the risks and uncertainties associated with
overburden and soil reclamation materials containing naturally occurring oil sand (petroleum hydrocarbons).
Syncrude has undertaken the study to examine alternative soil cover system designs for reclamation of their
overburden at the Aurora North mine. Soil cover system designs must optimise the available reclamation
materials to provide the foundation for re-emergence of sustainable, productive ecosystems. A wide range
of land capability will be required within reclaimed areas and, as a result, reclamation treatments may change
from one area to the next within the final landform. Consequently, the differences in performance between
different surface and subsoil material types and their thickness and the influence of the underlying
overburden waste material, needs to be defined.

Effect of soil cover system design on cover system performance and early tree establishment L.A. Barber et al.
Indicators of cover system performance include net percolation, the amount of water that moves from the
base of the cover system to the waste material below, and health of vegetation, as indicated by growth and
mortality rates. This paper primarily examines the hydrological performance of soil cover system design
treatment alternatives. As soil water dynamics and vegetation health are interrelated, indirect indicators of
vegetation performance such as plant available water and soil temperature are also considered.
Oil sands mining is an open-pit operation. It involves the removal of surficial geologic materials within the
mine footprint that are suitable for use later in reclamation, which primarily includes discrete salvage of
upland surface topsoil, subsoil materials, and peat materials from lowland bogs and fens. Following
reclamation soil salvage, overburden is removed to expose the oil sand ore body. The overburden is
transported to a designated disposal area (generally above-grade or out of pit) where it is landform graded
to a defined elevation and grade specifications and capped with previously salvaged soil reclamation
materials.
The appropriate soil reclamation cover system design and capping depth is dependent on the underlying
overburden quality and risk of environmental exposure to constituents of concern. For Syncrude’s Aurora
North mine, there are a number of risks and uncertainties associated with overburden reclamation. The
primary issue is one of naturally occurring petroleum hydrocarbons in both the overburden and soil
reclamation materials. Naturally occurring hydrocarbons are generally present in the entire matrix of the
overburden, and are present as discrete bands or chunks of varying size and proportion within the soil
reclamation matrix. The removal, disruption, and spreading of these materials to a new setting poses a risk
to the closure landscape, and the appropriate soil cover system design and thickness to alleviate any risks if
present is also uncertain.
A second issue Syncrude faces with overburden reclamation at the Aurora North mine is the large abundance
of coarse textured (sandy loam to sand) mineral reclamation materials. Although Syncrude has over 30 years
of reclamation experience in the region, the Aurora North mine coarse textured surficial geological materials
differ from previous mining locations. The appropriate use of available soil reclamation materials to re-
establish similar vegetation present prior to the disturbance has not been determined.
The ASCS was constructed to address the risks and uncertainties with overburden reclamation at the Aurora
North mine. The objective of the study was to evaluate the environmental risks and provide an appropriate
soil cover system design and depth to mitigate these risks. This paper focuses on cover system design
performance. The ASCS is a large-scale, replicated multi-disciplinary study that encompasses key disciplines
such as soil physics, soil chemistry, geochemistry, hydrology, plant growth, and soil microbiology.
Twelve treatment options (Figure 1) were constructed in triplicate to facilitate scientifically rigorous
comparisons, resulting in a total of 36 cells (Figure 2). Surface layer materials included peat (a material of
high organic content, salvaged from bogs and fens with no underlying mineral material), LFH (forest floor
material which includes all of the L-F-H, A horizon plus a portion of the B horizon of an upland natural soil),
and a split-plot treatment with no cover soil additions, with half consisting entirely of a Bm soil salvage and
the other a subsoil salvage of approximately 15 to 200 cm. Instrumentation for measurement of groundwater
and various parameters of the water balance was installed in the cells (Figure 2). In each treatment cell four
25 × 25 m tree planting plots were established and assigned to one of the four vegetation treatments; single
species plots with Populus tremuloides Michx., Pinus banksiana Lamb. and Picea glauca (Moench) Voss and
a plot with an even mix of the three species. All plots were planted in the spring of 2012 with one-year-old
containerised planting stock at a 1 × 1 m spacing (equivalent to a density of 10,000 seedlings per hectare
(sph)). Average tree height per year and mortality rate was used as performance indicator. Therefore,
seedling height and mortality was measured at the end of each growing season (2012 to 2014). Seedling
mortality and seedling height after three growing seasons (2012 to 2014) are presented.

Analytical water balances were calculated for cells on the perimeter of the ASCS to quantify the volume of
water percolating through the cover systems in the 2012 to 2013 water year (1 November 2012 to 31
October 2013). Water balances use inputs that are field measured, calculated, or estimated from the residual
to solve the water balance equation (1) on a daily basis during frost-free periods. In this way, the water

dynamics of the various cover system treatments, and the hydrology of the system as a whole, can be
characterised.
PPT = RO/S + AET + NP + dS (1)
Where:
PPT = precipitation (rainfall plus snow water equivalent (SWE)),
RO/S = runoff and sublimation,
AET = actual evapotranspiration,
NP = net percolation, and
dS = change in water storage.
A meteorological station was erected at the ASCS on Cell 22 to measure site-specific climatic parameters
(Figure 2). The station included instrumentation to monitor air temperature, relative humidity, net radiation,
wind speed and direction, rainfall, snow depth, and air pressure.
Potential evapotranspiration (PE) was estimated using the modified Penman-Monteith method
(Vanderborght et al., 2010) and site meteorological data collected by the meteorological station. Actual
evapotranspiration was calculated based on climate data and rates of PE, while AET/PE ratios were based
upon soil saturation levels, available water holding capacity, and adjusted to match the calculated storage to
the measured storage in the water balance.
Vadose zone water dynamics in the cover systems and overburden at the ASCS were monitored by Campbell
Scientific (CS) model 616-L volumetric water content sensors and CS229-L matric suction sensors. Data from
these sensors was used to determine the measured change in storage for the water balance. Pairs of soil
water sensors were installed by hand in all layers of cover materials and in overburden, with a focus on
material interfaces.
Soil temperatures were measured using CS229-L matric suction sensors. Soil temperature data were used to
determine the frost-free period, and assist with the interpretation of water content data. In addition, the
season when vegetation on the test cells is active is heavily influenced by soil temperatures.
Net percolation through the cover system treatments was monitored by large-scale lysimeters (Figure 2). The
large-scale lysimeters were installed in the upper overburden profile of 12 test cells on the perimeter of the
study area. Smaller scale Gee drain gauges were installed to assist in the characterisation of water flow
dynamics. Water balance net percolation was estimated using lysimeter and Gee gauge data, and the change
in water storage at the base of the cover system.
Runoff was not directly measured at the site. Runoff and sublimation from the snowpack were estimated
from the water balance.
Two cells with surface treatment types of 20 cm LFH (Cell 18) and 30 cm peat (Cell 30) were selected for
presentation of results. Trends for research cells with large scale lysimeters located on the perimeter of the
study area are also discussed. Cell 18 and Cell 30 were selected for presentation as they have the same subsoil
treatment, and have the most common surface material thickness.
Cells with peat as a surface material did not freeze as quickly as cells with LFH as a surface material. Freezing
was more gradual for Cell 30 (Figure 3) than for Cell 18 (Figure 4), which quickly froze to approximately 50
cm depth in both years.

The number of days that soil temperatures at 10 cm depth exceeded 5°C was determined for the four major
surface treatment types for 2013 and 2014 growing seasons (Figure 5). The criterion of 5°C at 10 cm depth
was used as a rough indicator for when vegetation becomes active (Novak, 2005). The cells with 20 cm LFH
at the surface had a greater number of days with active vegetation than cells with 30 cm of surface peat. Peat
is an excellent thermal insulator and has a high resistance to temperature change. Resistance to temperature
change is manifested as lower mean temperatures and smaller temperature amplitudes in the material
underlying the peat. Conversely, the sandy textured LFH has a very high capacity for transmitting energy,
resulting in faster, deeper freezing in the winter, and higher subsoil temperatures in the summer. If soil-water
is present to meet vegetation demand, increased soil temperature will likely translate into increased tree
growth.
Peat
Blended BC
Overburden
LFH
Blended BC
Overburden

190
2013
2014
Number of days 10 cm depth >5 C 180
170
160
150
30 cm Peat 10 cm Peat 20 cm LFH Centre Pit BM
(subsoil)
Surface Material
Water balances were completed for all research cells with large scale lysimeters of the study. Water balances
for the 2012 to 2013 water year are provided as examples. Calculated and measured storages matched well
for Cell 30 (Figure 6) and Cell 18 (Figure 7) for the 2012 to 2013 water year. Runoff and sublimation were
higher for Cell 30, as well as for the majority of cells with peat as a surface material. Rates of AET were similar
for Cell 18 and Cell 30, and ranged from 61% to 76% for all estimated water balances. Rates of AET did not
have a discernible pattern based on material type. Rainfall for the year was above average, with two major
rainfall events resulting in the majority of net percolation events through the cover systems.
to

The rate of net percolation will affect plant available water and groundwater recharge rates. The net
percolation rate for Cell 30 for the 2012 to 2013 water year, at 12% of annual precipitation, was lower than
the rate for Cell 18, at 25%. On average, cells with 30 cm of peat as a surface material had lower net
percolation rates (14%), than cells with LFH as a surface material (23%) (Figure 8). This is attributed to the
higher water storage capacity of peat, which allows for a greater period of time over which water can be
removed from the soil surface via AET. In addition, it is possible that textural discontinuity between peat and
subsurface materials inhibits drainage from peat to subsurface materials.
Overall seedling mortality after three growing seasons was low for all species (Figure 9). When all species are
combined, cover material did not affect seedling mortality (p=0.2). However, analysis of variance (ANOVA)
indicated that mortality was significantly higher in the 30 cm peat compared to the 10 cm peat cover
(p=0.019).

The three tree species showed similar responses in total height (growth over three growing seasons) to the
cover soil materials; however responses of P. tremuloides and P. banksiana to capping materials appear to
be more pronounced than in P. glauca. All three species grew better in the treatments which were capped
with LFH compared with peat materials (across both thicknesses). For example after three growing seasons
the average height of P. tremuloides seedlings was 125 cm in the LFH material compared to 90 cm in the peat
(p=0.012). For P. banksiana the difference between LFH and peat was 28 cm (p<0.001) and in P. glauca the
difference was 10 cm in height (p<0.001). At 30 cm peat capping depth P. banksiana seedling height was
reduced by 20 cm (p<0.001) and 25 cm in P. tremuloides (p=0.04) compared to a peat thickness of 10 cm,
while P. glauca did not show a significant reduction in height (p=0.19). Populus tremuloides seedlings growing
in the selectively salvaged Bm material performed better than when planted in subsoil (p=0.03), while P.
banksiana and P. glauca did not show a difference in growth (p=0.29; p=0.77, respectively).
P. glauca
P. banksiana
140
P. tremuloides
100
Height (cm)
60
20
Peat 30 Peat 10 LFH 20 LFH 10 Bm Subsoil

The influence of cover system treatments on soil temperature and water content were measured at an oil
sands reclamation research study referred to as the Aurora Soil Capping Study. Net percolation rates for the
2012 to 2013 monitoring year for treatments with 30 cm of peat at the surface were 14% of annual
precipitation, which was lower than treatments with LFH (23% of annual precipitation) at the surface. The
greater soil-water content of the treatments with peat at the surface is attributed to peat having a higher
water storage capacity. In addition, textural discontinuity between peat and underlying sandy subsurface
materials potentially inhibits drainage. Peat also had an influence on soil temperature. Due to the ability of
peat to insulate the soil from large changes in temperature, the number of active growing days was lower in
the peat cover treatments relative to the LFH cover treatment. Although peat may improve vegetation
performance in terms of water storage, it has the potential to reduce effective growing season length and
might also limit nutrient availability. Indications of these limitations are reflected in the response of seedlings
heights to a change in thickness of the peat cap, while the thickness of a salvaged mineral soil dominated LFH
was not a factor in seedlings performance. Overall seedlings growing in LFH had a much better performance
over the first three growing seasons. The relative advantages and disadvantages of peat over a range of wet-
dry conditions and whether they persist as the site matures may shape the preferred reclamation soil cover
system design strategy.
The initial monitoring of the dynamics of the overburden disposal area at the ASCS have shown the
interactions of the vegetation, cover materials, and climate are complex and further study is required to
determine the preferred cover system design(s) for closure of the area. Data collected so far was sufficient
to provide a comparison of cover system performance and identify key trends that will affect long-term
performance such as net percolation, and the length of the vegetation growing season.
Novak, M.D. (2005) Soil temperature, in J.L Hatfield and J.M. Baker (eds). Micrometeorology in agricultural systems.
Vanderborgh, J., Graf, A., Steenpass, C., Scharnagl, B., Prolingheuer, N., Herbst, M., Franssen, H.H. and Vereecken, H. (2010) Within
field variability of bare soil evaporation derived from eddy covariance measurements special section: patterns, Vadose Zone


P. Audet EDI Environmental Dynamics Inc., Canada

B.D. Pinno Canadian Forest Service, Canada
E. Thiffault Université Laval, Canada
Boreal forests in northern Alberta have a growing anthropogenic footprint due to a rapidly expanding oil
sands mining industry. While land reclamation is a necessary aspect of responsible industrial development,
these activities nearly always affect higher order landscape components such as the broader landform, its
hydrology and biogeochemistry, which present significant obstacles when attempting to reinstate near
natural landscapes. The fact that some highly disturbed ecosystems following oil sands mining have been
carefully rehabilitated and “certified reclaimed” suggests that successful land reclamation is possible. Yet, the
number of sites still “under reclamation” far exceeds the number that has been reclaimed, indicating that this
path is neither smooth nor direct.
Focusing on the oil sands mining landscape of northern Alberta, this analysis seeks to situate current and
anticipated challenges affecting the reclamation of boreal forest following oil sands mining by describing: (a)
how Alberta’s regulatory criteria (e.g., Equivalent Land Capability) shape reclamation practices and targeted
end-goals, and (b) how these approaches embody latest trends and priorities in the area of restoration
ecology. Echoing wider opinions regarding the management of degraded landscapes, a significant
development across the field of restoration ecology is the acknowledgement that highly assertive
disturbances, such as mining, can (and often do) cause irreversible effects to natural landscapes leading to
the emergence of novel ecosystems. And so, land reclamation and conservation frameworks that target the
return of the post-disturbance landscape to its pre-disturbance condition may not always be possible or even
practical. We are then challenged to identify potential barriers and opportunities for reclamation practices in
these post-disturbance landscapes that do not compromise higher standards of environmental stewardship.
Canada’s oil sands – a bitumen reserve equivalent to ~1.6 trillion barrels of oil within the Peace River, Cold
Lake, and Athabasca River regions of northern Alberta (Figure 1) – are a major resource within Canada’s
energy sector and account for 13% of global oil reserves (Alberta Energy, 2013; Natural Resources Canada,
2013). Various federal, provincial, and industry stakeholders collaborate to develop these resources with the
responsibility of reclaiming post-disturbance landscapes to a reasonable level of ecological fidelity following
mining. For the most part, key environmental policy issues have been well documented (Gosselin et al., 2010;
Schindler, 2010; Grant et al., 2013). However, in an era of global environmental change, it remains to be
determined how best to define sustainable end-goals for highly disturbed ecosystems within current land
reclamation frameworks (Harris et al., 2006). In this regard, a significant development across the field of
restoration ecology is the acknowledgement that certain disturbances such as mining can (and often do)
cause irreversible effects to natural landscapes leading to the emergence of novel ecosystems (Jackson and
Hobbs, 2009; Bridgewater et al., 2011). And so, land reclamation frameworks that target the return of post-
disturbance landscapes to their pre-disturbance condition may not always be possible or even practical
(Doley et al., 2012; Doley and Audet, 2013; 2015; Hilderbrand et al., 2005). Ultimately, we need to define
what we want future reclaimed ecosystems to look like in terms of productivity, sustainability, and in
anticipation of future environmental change (Millar et al., 2007).

Land reclamation in Canada’s mining sandbox: do novel ecosystems require novel approaches? P. Audet, B.D. Pinno and E. Thiffault
The Canadian boreal forest is an important area for the sustainable production of wood products and for
communities supported by these activities (Maynard et al., 2013). That being said, over 4,700 km2 has been
leased for oil sands surface mining and, as of 2013, active operations have resulted in a disturbance footprint
of more than 700 km2 (Dyer et al., 2011; Grant et al., 2013; Alberta Energy, 2013; 2014) which is equivalent
to 1 hectare per ~100,000 barrels. The rapid expansion of localised oil sands mining activities imposes
considerable modifications to many aspects of the natural landscape including:
1. The destruction of the natural landform and its indigenous biodiversity due to clearing of
aboveground vegetation and stripping of cover soils.
2. The production of large volumes of dispersed overburden and tailings materials (having potentially
adverse chemical properties such as high salinity).
3. The alteration of soils and hydrological patterns
NB. All these effects bear close similarity to those caused by bauxite and heavy mineral sand surface-
mining activities in Australia (Gardner and Bell, 2007; Smith and Nichols, 2011; Tibbet, 2010), and
broad similarity to coal and mineral strip-mining in North America and elsewhere worldwide (Cooke
and Johnson, 2002; Holl, 2002; Zipper et al., 2011).
This imposes significant changes to the affected environment, especially higher order components (listed in
Table 1) such as the broader landform, its hydrology and biogeochemistry, which present significant obstacles
for land reclamation.
Although the majority of the reclamation for oil sands mines will occur after 2035, only 0.1% of land disturbed
by oil sands mining have presently been “certified reclaimed” and returned as public land. Whereas 10% is
still “under reclamation” and 90% is being disturbed directly by or ahead of mining activities (Alberta Energy,
2013; 2014). Within this context, we are challenged to determine how best to address these disturbed
ecosystems to achieve sustainable economic development while also promoting suitable ecological
outcomes to the highest levels of ecological integrity (Hobbs et al., 2013). Therefore, this analysis seeks to
situate current and anticipated challenges affecting the reclamation of boreal forest in Canada’s mining
“sandbox” by describing how regulatory criteria shape reclamation practices and targeted end-goals, and
how these approaches should embody the latest priorities in the area of restoration ecology.

Factor Attribute Physical Property
Abiotic 1. Geology* a. Consolidation
b. Origin
c. Mineralogy
d. Texture
2. Landscape a. Form I. Landform Class
II. Landform Pattern
III. Landform Element
IV. Quantitative Attributes  Relief
 Slope
 Texture
b. Function I. Surface  Erosion
 Deposition
II. Profile  pH
 EC
 Nutrient Status
 Water Status
Biotic a. Structure I. Height

3. Vegetation
II. Cover
III. Biomass
b. Composition I. Species Richness
II. Diversity
III. Structural Class
c. Conditions
4. Soil Biota a. Abundance
b. Diversity
c. Abundance
5. Fauna
d. Diversity
*Encompassing the geology and biochemistry of the parent material including, e.g., tailings and spoils.


Having the largest known crude bitumen deposits worldwide, Alberta’s oil sands lie within the boreal plain
ecozone. This region is characterised by upland forests on gently undulating plains interspersed with vast
wetlands (Webber and Flannigan, 1997) where common tree species include: trembling aspen (Populus
tremuloides), balsam poplar (Po. balsamifera), white spruce (Picea glauca), black spruce (Pi. mariana), jack
pine (Pinus banksiana), balsam fir (Abies balsamea), and tamarack (Larix laricina) (Beckingham and Archibald,
1996). Biodiversity patterns and biogeochemical cycles are shaped by the region’s subhumid and subarctic
climate, its predominant brunisol and luvisol soils, and the occurrence and distribution of natural (e.g., fire)
and anthropogenic disturbances (e.g., logging) (Brandt et al., 2013). More recently, this area has experienced
a growing anthropogenic footprint due to mining (Gosselin et al., 2010; Schindler, 2010; Grant et al., 2013).
Oil production from bitumen extraction was initiated in 1967 and expanded significantly from 1978 to 2003
(Alberta Government, 2009). Today, mining of oil sands is among the most significant drivers of change across
the region (Gosselin et al., 2010; Grant et al., 2013) and production is expected to more than double (from
~1.9 to 4.8 million barrels per day) over the next two decades (Canadian Association of Petroleum Producers,
2014) resulting in an increasingly large disturbance footprint. Unlike wildfire and logging, these mining
activities nearly always affect physical landform attributes causing significant challenges to land reclamation
(Doley and Audet, 2013; 2015). Given these circumstances, a particular dilemma is ensuring the integrity of
essential landscape attributes of the post-disturbance environment that may support suitable ecological
function.
A precursor to most large-scale mining operations is the stakeholders’ agreement to design and implement
reclamation practices to return affected lands to a reasonable degree of ecological fidelity. This is often set
to the standard of adjacent or pre-disturbance environmental conditions and applies equally to countries
such as Canada, the USA, and Australia that have mature resource-based economies and well developed
environmental protection legislation (e.g., USA Surface Mining Reclamation Act, 1977; Alberta Government,
1999; Australia ─ Department of Industry Resources and Tourism, 2006). According to existing legislative
requirements (Alberta Government, 1999), the particular objective of land reclamation in Alberta is to return
the specified land to an Equivalent Land Capability (ELC) where the reclaimed land should support post-
disturbance land-uses that are similar (or within a stipulated percentage range) to pre-disturbance
conditions, but not necessarily identical to it (Conservation and Reclamation Reg. s. 1e). At the level of
environmental regulation and permitting, Alberta’s ELC framework is used as a land capability ranking system
to assess the status and productivity of a wide range of landscapes and the species that characterise them.
This includes assessments of a combination of surface and sub-surface soil characteristics (water infiltration,
structure, organic matter, and nutrient holding capacities), landscape factors (topography, exposure, slope),
soil conditions (texture, moisture, erodibility), and predominant vegetation that may be interpreted as a final
ecological target community (Oil Sands Vegetation Reclamation Committee, 1998; Fung and Macyk, 2000;
Alberta Energy and Water, 2012).
In the case of oil sands mining in northern Alberta, where the targeted outcome is designated for the re-
instatement of forested land-use ELC (Figure 2), reclaimed ecosystems should support the production of
forest goods and services (both in quality and quantity) consistent to those prior to disturbance (Alberta
Government, 1999). From an economic perspective, this can be interpreted as establishing safe, stable, and
non-polluting forests for commercial forestry and other uses within the natural range of ecosites found in
central mixed-wood forests. Whereas from an ecological perspective, this means providing wildlife habitat
capabilities similar to pre-disturbance conditions that should include adequate levels of native species
composition (i.e., richness comparable to adjacent communities), ecological structure (i.e., equivalent
canopy, mid-story and understory cover), and ecological function (i.e., adequate resilience against pest and
fire disturbances) (Williamson and Edwards, 2014). Although the final ecosystems may not have to be
identical to pre-disturbance conditions, forested land-use ELC criteria still have a relatively narrow scope of
interpretation and require sufficient ecological similarity as found within the boreal plains ecozone. And so,

land reclamation practitioners must carefully delineate legal boundaries, describe characteristics and
properties of the reclaimed landscape (e.g., topography, hydrology, predominant soils and vegetation, and
land capability), document conservation and reclamation procedures, and finally monitor the improvements
anticipated in post-disturbance site conditions.
Criteria set by ELC provide a clear starting point for setting ecological targets encompassing similar attributes
of the pre-disturbance environment. However, reliably determining whether long-term ecological
trajectories are developing toward “natural” analogues can be challenging. Where sites may require
additional management inputs, deriving intervention approaches based on these trajectories can be equally
ambiguous and possibly lead to protracted costs of reclamation (Jackson and Hobbs, 2009). The general
approach of “looking to the past” to assess the development of future systems has been widely criticised for
both its narrow and its static assessment of recovering ecosystems (Harris et al., 2006; Choi, 2007; Hobbs
and Cramer, 2008; Hobbs and Suding, 2009), particularly those systems with complex natural disturbance
and succession regimes (e.g., fire, insect epidemics, drought). Given the inevitability of a rapidly growing
industry, a central question for the sustainable management of boreal forest in the Alberta’s oil sands region
(as elsewhere) is: “Can today’s land reclamation approaches account for future ecological issues and
anticipate future environmental management challenges?”
Climate change and atmospheric deposition now appear to represent major drivers of ecological change
across the boreal forest (Laxton et al., 2012; Lindner et al., 2002; Webber and Flannigan, 1997; Price et al.,
2013; Savard et al., 2014). For example, ongoing changes in climate across the boreal forest are pushing
environmental conditions beyond the range that has been experienced in the past centuries. Meanwhile,
atmospheric emissions of acid precursors (e.g., SO2 and NOx) have increased ~2-fold in the oil sands region
during the last three decades corresponding to mining and oil refining infrastructure development (Golder
Associates, 2003; Hazewinkel et al., 2008). In addition to the remarkable size and severity of mining
disturbance, these global change phenomena may further influence the types of ecosystems that can be re-
established at a given mine site. Yet, such considerations are currently missing from mined land reclamation
frameworks, along with the necessary socio-economic mechanisms that may afford flexibility for future
ecological interventions.
In an era of climate change (Harris et al., 2006), traditional forms of land management that target idealised
post-disturbance landscapes based on historical or pre-disturbance ecological criteria may not always be
appropriate for mine sites due to the significant and possibly irreversible differences between the pre- and
post-mining environments. This is especially true if the objective is to optimise both economic development
and environmental stewardship (Chapin et al., 2010, 2011; Doley et al., 2012; Doley and Audet, 2013; 2015).
It could then be argued that what is lacking in this approach is the scientific framework for determining what
we want our future forests to look like and the actual types of ecosystems that may develop among highly
disturbed landscapes in a global change context.
Land reclamation requires thoughtful assessment of both the pre- and post-disturbance environments
(Clewell and Aronson, 2006). This should be followed by careful improvement of key landscape components
(such as those listed in Table 1) being mindful of the successional development of recovering ecosystems
(Grant, 2006; 2009). In the context of large-scale mined land reclamation, greater management resources
tend to be allocated toward heavy earthworks for reconstruction of stable landforms and hydrological
characteristics to provide the essential structural and functional components of the desired ecosystem.
Where climate, geology, and landform reconstruction are not major obstacles to rehabilitation – for example,
with an environmental starting point having predictable climate patterning, intact geology, and landscape
attributes similar to those of the pre-disturbance conditions (Cooke and Johnson, 2002; Gardner and Bell,
2007; Holl, 2002; Smith and Nichols, 2011; Tibbet, 2012; Zipper et al., 2011) – Macdonald et al. (2012)
rightfully contend that self-sustaining boreal forest ecosystems can be achieved with foremost attention to

the function and stability of soils. However, due to the size and severity of mining disturbances, post-mining
landscapes often contain significantly altered landforms and soils along with non-natural landscape
attributes (e.g., waste heaps and tailings impoundments). Consequently, even carefully managed post-
mining landscapes may bear little physicochemical and biological similarity to the pre-disturbance
environment (Koch and Hobbs, 2007). These disparities may result in different ecological determinants
influencing the range of ecosystems that may be established at a given site, which leads to the question of
whether the targets dictated by the requirements for ecological stewardship (such as ELC) are attainable or
even desirable.
Given the rise of no-analogue environments and continued environmental change worldwide (Harris et al.,
2006; Jackson and Hobbs, 2009; Hobbs et al., 2009), an increasing body of literature suggests that many
landscapes now contain new and possibly even irreversibly different assemblies of abiotic and biotic system
components (Perring et al., 2013) – so called hybrid and novel ecosystems (Hobbs et al., 2006; 2009; 2013;
Williams and Jackson, 2007). If the post-disturbance environmental conditions bear little resemblance to the
pre-disturbance ones, it stands to reason that achieving the return of a sufficient number of landscape
components to their pre-disturbance level of ecological integrity may not be economically viable or even
ecologically appropriate (Choi, 2007; Perring et al., 2014). Therefore, it is becoming widely acknowledged
that various factors may preclude restoring ecosystems to their pre-disturbance conditions (Groffman et al.,
2006; Perring et al., 2013; Shackleford et al., 2013), especially mine sites. Although comprehensive
assessment of the degree to which post-disturbance ecosystems are natural or novel takes time – preferably
through analysis of large datasets (Lindenmayer and Likens, 2009; Lovett et al., 2007) – there are already
many indications that the oil sands landscapes contain novel attributes. For example, Syncrude Canada Ltd.’s
~1km2 (104 ha) Gateway Hill site: the first and (thus far) only certified reclaimed landscape in the mineable
oil sands.
Originally characterised as a low-lying boreal wetland, the site has undergone significant changes compared
to the broader natural landscape due to the stockpiling of mineral overburden from adjacent mining
activities. Although the site was never directly mined, the most readily apparent novel features are the
ecosystem’s transformation from a wetland into a relatively dry and hilly landscape approximately 40 m
above its pre-disturbance elevation. From 1978 to 1981, steps were taken to reclaim the landscape by
replacing peat-based topsoil directly across the post-disturbance landscape and planting both native and
non-native trees and shrubs in hopes of developing a self-sustaining ecosystem. The ecology of northern
Alberta’s oil sands region has distinct features among boreal upland versus wetland environments that reflect
a fine balance between landform, hydrological features, and nutrient cycling. Overall, the altered topography
and integration of non-native species have resulted in different biological composition than that found within
the pre-disturbance environment, such that the reclaimed landscape would likely qualify as a novel
ecosystem. Then again, the site still reflects a stable and productive ecosystem with soil evolution, as well as
the notable inclusion of landscape features that promote public awareness (e.g., the establishment of nature
trails and an open-air information centre). Clearly, the anthropogenic heritage of this landscape has
significantly altered the final level of biodiversity and productivity in relation to any historical antecedent.
Yet, conditions for certification of the reclaimed landscape (granted in 2008) were still met through
negotiations between stakeholders, regulators, research scientists and local people, and the final landscape
continues to reflect a safe, stable, non-polluting, and productive environment. Therefore, these outcomes
should serve as a benchmark in mine site land reclamation, being mindful that future practices should seek
continuous improvement.
Acknowledgment of novel ecosystems in the mining sector can appear problematic since it may imply a
permanent loss of historic ecological fidelity when establishing land reclamation goals – and academic
debates have teetered over their broader implications in the fields of restoration ecology and conservation

biology (cf. Aronson et al., 2014; Hobbs et al., 2009; 2014; Murcia et al., 2014). Still, the natural-novel
ecosystems paradigm most recently proposed by Hobbs et al. (2013) provides a more realistic depiction of
the barriers to ecological restoration (including all associated management inputs) necessary for achieving a
given environmental target in light of various biological obstacles and socio-economic boundaries
(Bridgewater et al., 2013; Perring et al., 2013; 2014). Notably, within a hybrid or novel ecosystem, it is possible
to have abiotic and biotic conditions that are functionally similar to those found within the broader regional
ecosystem, and these ecosystems may represent the best that can be derived following certain industrial
disturbances. Some may view this reality in a negative light since there is a permanent change in ecosystem
structure (Grant et al., 2013; Murcia et al., 2015; Aronson et al., 2014). Then again, many others now contend
that hybrid and novel ecosystems are an unavoidable consequence of pervading anthropogenic effects on
natural ecosystems and that beneficial reclamation outcomes can still be derived in spite of these challenges
(Hobbs et al., 2014; Perring et al., 2013; 2014).
Of course, the designation of acceptable hybrid and novel ecosystems should not be used by developers,
regulators or policy makers to compromise standards of ecological integrity or conservation. Yet, existing
policies and regulations need to account for the fact that many industrial activities cause irreversible changes
to landscapes leading to novel ecosystems (Doley and Audet, 2013; 2015). Pragmatically, we can apply our
growing ecological understanding of natural versus novel ecosystems for the purposes of adaptive
management of mine sites without necessarily needing new approaches or investigative frameworks. A
possible advancement for setting more appropriate land reclamation targets would be to move away from
establishing reclamation criteria based entirely on conditions of the pre-disturbance environment (Choi,
2007; 2008; Hulvey et al., 2013; Shackleford et al., 2013). Instead, reclamation targets should be established
based on current and/or anticipated conditions of the post-disturbance environment. Meanwhile,
management regimes should be granted the flexibility to adapt and/or modify their approaches based on
ongoing trajectories to determine the range of ecosystems and potential interventions that are optimally
supported by these conditions. Ideally, the latter approach would lead to reclaimed ecosystems that provide
the greatest ecological resilience under both current and future climatic conditions and other changing
environmental drivers, no matter whether these ecosystems are ultimately deemed natural, hybrid, or novel.
Reclamation of highly disturbed post-mining landscapes may not provide realistic opportunities for the
redevelopment of “historical” ecological processes and recovery. In fact, the post-mining ecosystems may be
quite different from those that occupied the site prior to disturbance, and they may be difficult to anticipate
at the planning stage of a mining operation due to the size, complexity, and severity of disturbance and the
implications of changing environmental conditions. However, that does not mean that the ecosystem
services provided by these new landscapes are neither valuable nor desirable. There are redeemable
attributes to existing land reclamation approaches, so long as policy and regulatory frameworks afford them
the flexibility to adjust ecological targets and allocate necessary management resources, should they be
required. This underpins the importance of careful planning at the earliest stages of land reclamation, and
the necessity to accommodate a capacity to adjust goals according to changing conditions and best practices.
This paper contains content from a recent review by Audet, Pinno and Thiffault (2015) that will be published
in the Canadian Journal of Forest Research. For additional information about this, please refer to:
https://sites.google.com/site/drpatrickaudet/.
Alberta Government (1999) Conservation and reclamation information letter guidelines for reclamation to forest vegetation in the
Athabasca oil sands region, C and R/IL/99–1, Government of Alberta, Edmonton, AB.
Alberta Government (2009) Environmental management of Alberta’s oil sands, Government of Alberta, Edmonton, AB.
Alberta Energy (2013) Alberta’s oil sands: January Fact-Sheet, Government of Alberta, Edmonton, AB.

Alberta Energy (2014) Talk about oil sands, November Fact-Sheet, Government of Alberta, Edmonton, AB.
Alberta Environment and Water (2012) Best management practices for conservation of reclamation materials in the mineable oil
sands region of Alberta, “Best Management Practices Task Group of the Reclamation Working Group” of the Cumulative
Environmental Management Association, Fort McMurray, AB. March 9, 2011.
Aronson, J., Murcia, C., Kattan, G.H., Moreno-Mateos, D., Dixon, K. and Simberloff, D. (2014) The road to confusion is paved with
novel ecosystem labels: a reply to Hobbs et al., Trends in Ecology & Evolution, Vol. 29, pp. 646–647.
Audet, P., Pinno, B.D. and Thiffault, E. (2015) Reclamation of boreal forest following oil sands mining: anticipating novel challenges
among novel ecosystems, Canadian Journal of Forest Research, DOI: 10.1139/cjfr-2014-0330.
Australia ─ Department of Industry Resources and Tourism (2006) Mine closure and completion: leading practice sustainable
development program for the mining industry, Government of Australia, Canberra, ACT, Australia.
Beckingham, J.D. and Archibald, J.H. (1996) Field guide to ecosites of northern Alberta, Canadian Forest Service, Edmonton, AB.
Bergeron, Y., Leduc, A., Harvey, B.D. and Gauthier, S. (2002) Natural fire regime: a guide for sustainable management of the Canadian
boreal forest, Silva Fennica, Vol. 36, pp. 81–95.
Brandt, J.P. (2009) The extent of the North American boreal zone, Environmental Reviews, Vol. 17, pp. 101–161.
Brandt, J.P., Flannigan, M.D., Maynard, D.G., Thompson, I.D. and Volney, W.J.A. (2013) An introduction to Canada’s boreal zone:
ecosystem processes, health, sustainability, and environmental issues, Environmental Reviews, Vol. 21, pp. 207–226.
Bridgewater, P., Higgs, E.S., Hobbs, R.J. and Jackson, S.T. (2011) Engaging with novel ecosystems. Frontiers in Ecology & Environment,
Vol. 9, pp. 423.
Bridgewater, P. and Yung, L. (2013) The policy context: building laws and rules that embrace novelty, in Novel ecosystems: intervening
in the new ecological world order, R.J. Hobbs, E.S. Higgs, and C.M. Hall (eds), John Wiley & Sons, Chichester, UK, pp. 272–283.
Canadian Association of Petroleum Producers (2014) Crude oil: forecast, markets and transportation, June 2014 Report, Canadian
Association of Petroleum Producers, Calgary, AB, Canada.
Chapin, F.S., Carpenter, S.R., Kofinas, G.P., Folke, C., Abel, N., Clark, W.C., Olsson, P., Stafford Smith, M., Walker, B., Young, O.R.,
Berkes, F., Biggs, R., Grove, J.M., Naylor, R.L., Pinkerton, E., Steffen, W. and Swanson, F.J. (2010) Ecosystem stewardship:
sustainability strategies for a rapidly changing planet, Trends in Ecology & Evolution, Vol. 25, pp. 241-249.
Chapin, F.S. Power, M.E., Pickett, S.T.A., Freitag, A.E., Reynolds, J.A., Jackson, R.B., Lodge, D.M., Duke, C., Collins, S.L., Power, A.G.
and Bartuska, A. (2011) Earth stewardship: science for action to sustain the human-earth system, Ecosphere, Vol. 2, Article
89.
Chen, I.C., Hill, J.K.J., Ohlemüller, R., Roy, D.B., and Thomas, C.D. (2011) Rapid range shifts of species associated with high levels of
climate warming, Science, Vol. 333, pp. 1024–1026.
Choi, Y.D. (2007) Restoration ecology to the future: a call for a new paradigm, Restoration Ecology, Vol. 15, pp. 351–353.
Clewell, A.F. and Aronson, J. (2006) Motivations for the restoration of ecosystems, Conservation Biology, Vol. 20, pp.420–428.
Cooke, J.A. and Johnson, M.S. (2002) Ecological restoration of land with particular reference to the mining of metals and industrial
minerals: a review of theory and practice, Environmental Reviews, Vol. 10, pp. 41–71.
Doley, D. and Audet, P. (2013) Adopting novel ecosystems as suitable rehabilitation alternatives for former mine sites, Ecological
Processes, Vol. 2, Article 21.
Doley, D. and Audet, P. (2015) Identifying natural and novel ecosystem goals for rehabilitation of postmining landscapes, in
Responsible Mining, M. Jarvie-Eggart (ed), Society for Mining Metallurgy & Exploration, Englewood, CO, USA, pp. 609–638.
Doley, D., Audet, P. and Mulligan, D.R. (2012) Examining the Australian context for post-mined land rehabilitation: reconciling a
paradigm for the development of natural and novel ecosystems among post-disturbance landscapes, Agriculture, Ecosystems
& Environment, Vol. 163, pp. 85–93.
Dyer, S., Grant, J., Huot, M. and Droitsch, D. (2011) Solving the puzzle: environmental responsibility in oil sands development, April
2011 Report, The Pembina Institute, Drayton Valley, AB, Canada.
Fung, M.Y.P. and T.M. Macyk (2000) Reclamation of oil sands mining areas, in Reclamation of drastically disturbed lands, R.I.
Barnhisel, R.G. Darmody, and W.L. Daniels (eds), Agronomy Monograph Vol. 41, American Society of Agronomy, Crop Science
Society of America, and Soil Science Society of America, Madison, WI, USA, pp. 755–774.
Gardner, J.H. and Bell, D.T. (2007) Bauxite mining restoration by Alcoa World Alumina Australia in Western Australia: social, political,
historical, and environmental contexts, Restoration Ecology, Vol. 15, Suppl. S3-S10.
Golder Associates (2003) Evaluation of historic and future acid deposition effects on soils in the Athabasca oil sands region, Final
report submitted to “NOx–SOx Management Working Group” September 2003, Edmonton, AB, Canada.
Gosselin, P., Hrudey, S.E., Naeth, M.A., Plourde, A., Therrien, R., van der Kraak, G. and Xu, Z. (2010) The Royal Society of Canada
Expert Panel: Environmental and health impacts of Canada’s oil sands industry, The Royal Society of Canada, The Academies
of Arts, Humanities and Sciences of Canada, Ottawa, ON, Canada.
Grant, C.D. (2006) State-and-transition successional model for bauxite-mining rehabilitation in the jarrah forest of Western Australia,
Grant, C.D. (2009) State-and-transition models for mining restoration in Australia, in New models for ecosystem dynamics and
restoration, R.J. Hobbs and K.N. Suding (eds), Island Press, Washington, DC, USA, pp. 280–294.
Grant, J., Angen, E. and Dyer, S. (2013) Forecasting the impacts of oilsands expansion: measuring the land disturbance, air quality,
water use, greenhouse gas emissions, and tailings production associated with each barrel of bitumen production, June 2013
Report, The Pembina Institute, Drayton Valley, AB, Canada.
Groffman, P.M., Baron, J.S., Blett, T., Gold, A.J., Goodman, I., Gunderson, L.H., Levinson, B.M., Palmer, M.A., Paerl, H.W., Peterson,
G.D., Poff, N.L., Rejeski, D.W., Reynolds, J.F., Turner, M.G., Weathers, K.C. and Wiens, J. (2006) Ecological thresholds: the key

to successful environmental management or an important concept with no practical application? Ecosystems, Vol. 9, pp. 1–
13.
Harris, J.A., Hobbs, R.J., Higgs, E.S. and Aronson, J. (2006) Ecological restoration and global climate change, Restoration Ecology, Vol.
14, pp. 170–176.
Hazewinkel, R.O., Alexander, P., Wolfe, S.P., Curtis, C. and Hadley, K. (2008) Have atmospheric emissions from the Athabasca oil sands
impacted lakes in northeastern Alberta, Canada? Canadian Journal of Fisheries & Aquatic Sciences, Vol. 65, pp. 1554–1567.
Hilderbrand, R.H., Watts, A.C. and Randle, A.M. (2005) The myths of restoration ecology, Ecology & Society, Vol. 10, Article 19.
Hobbs, R.J. and Cramer, V.A. (2008) Restoration ecology: interventionist approaches for restoring and maintaining ecosystem
function in the face of rapid environmental change, Annual Reviews in Environmental Research, Vol. 33, pp. 39–61.
Hobbs, R.J., Arico, S., Aronson, J., Baron, J.S., Bridgewater, P., Cramer, V.A., Epstein, P.R., Ewel, J.J., Klink, C.A., Lugo, A.E., Norton, D.,
Ojima, D., Richardson, D.M., Sanderson, E.W., Valladares, F., Vila, M., Zamora, R. and Zobel, M. (2006) Novel ecosystems:
intervening in the new ecological world order, Global Ecology & Biogeography, Vol. 15, pp. 1–7.
Hobbs, R.J., Higgs, E.S. and Hall, C.M. (2013) Novel ecosystems: intervening in the new ecological world order, John Wiley & Sons,
Chichester, UK.
Hobbs, R.J., Higgs, E. and Harris, J.A. (2009) Novel ecosystems: implications for conservation and restoration, Trends in Ecology &
Evolution, Vol. 24, pp. 599–605.
Hobbs, R.J., Higgs, E. and Harris, J.A. (2014) Novel ecosystems: concept or inconvenient reality? A response to Murcia et al., Trends
in Ecology & Evolution, Vol. 29, pp. 645─646.
Holl, K.D. (2002) Long-term vegetation recovery on reclaimed coal surface mines in the eastern USA, Journal of Applied Ecology, Vol.
39, pp. 960–970.
Hobbs, R.J. and Suding, K.N. (2009) New models for ecosystem dynamics and restoration, Island Press, Washington DC, USA.
Hulvey, K.B., Standish, R.J., Hallett, L.M., Starzomski, B.M., Murphy, S.D., Nelson, C.R., Gardener, M.R., Kennedy, P.L., Seastedt, T.R.
and Suding, K.N. (2013) Incorporating novel ecosystems into management frameworks, in Novel ecosystems: intervening in
the new ecological world order, Hobbs, R.J., Higgs, E.S. and Hall, C.M. (eds), John Wiley & Sons, Chichester, UK, pp. 157–171.
Jackson, S.T. and Hobbs, R.J. (2009) Ecological restoration in the light of ecological history, Science, Vol. 325, pp. 567–569.
Koch, J.M. and Hobbs, R.J. (2007) Synthesis: is Alcoa successfully restoring a Jarrah forest ecosystem after bauxite mining in Western
Australia, Restoration Ecology, Vol. 15, Suppl. S137–S144.
Laxton, D.L., Watmough, S.A. and Aherne, J. (2012) Nitrogen cycling in Pinus banksiana and Populus tremuloides stands in the
Athabasca Oil Sands Region, Alberta, Canada, Water Air Soil & Pollution, Vol. 223, pp. 1–13.
Lindenmayer, D.B. and Likens, G.E. (2009) Adaptive monitoring: a new paradigm for long-term research and monitoring, Trends in
Ecology & Evolution, Vol. 24, pp. 482–486.
Lindner, M., Sohngen B., Joyce, L.A., Price, D.T., Bernier, P.Y. and Karjalainen, T. (2002) Integrated forestry assessments for climate
change impacts, Forest Ecology & Management, Vol. 162, pp. 117–136.
Lovett, G.M., Burns, D.A., Driscoll, C.T., Jenkins, J.C., Mitchell, M.J., Rustad, L., Shanley, J.B., Likens, G.E. and Haeuber, R. (2007) Who
needs environmental monitoring? Frontiers in Ecology & Environment, Vol. 5, pp. 253–260.
Macdonald, E., Quideau, S. and Landhausser, S (2012) Rebuilding boreal forest after industrial disturbance, in Restoration and
reclamation of boreal ecosystems: attaining sustainable development, D.H. Vitt and J.S. Bhatti (eds), Cambridge University
Press, Cambridge, UK, pp.123–160.
Maynard, D.G., Pare, D., Thiffault, E., Lafleur, B., Hogg, K.E. and Kishchuk, B. (2013) How do natural disturbances and human activities
affect soils and tree nutrition and growth in the Canadian boreal forest? Environmental Reviews, Vol. 22, pp. 1–18.
Millar, C.I., Stephenson, N.L. and Stephens, S.L. (2007) Climate change and forests of the future: managing in the face of uncertainty,
Ecological Applications, Vol. 17, pp. 2145–2151.
Murcia, C., Aronson, J., Kattan, G.H., Moreno-Mateos, D., Dixon, K. and Simberloff, D. (2014) A critique of the “novel ecosystem”
concept, Trends in Ecology & Evolution, Vol. 29, pp. 548–553.
Natural Resources Canada (2013) Oil sands: a strategic resource for Canada, North America and the global market, February 2013
Fact Sheet, Government of Canada, Ottawa, ON, Canada.
Oil Sands Vegetation Reclamation Committee (1998) Guidelines for reclamation to forest vegetation in the Athabasca oil sands region,
Government of Alberta, Edmonton, AB, Canada.
Perring, M.P., Standish, R.L. and Hobbs, R.J. (2013) Incorporating novelty and novel ecosystems into restoration planning and practice
in the 21st century, Ecological Processes, Vol. 2, Article 18.
Perring, M.P., Audet, P. and Lamb, D. (2014) Novel ecosystems in ecological restoration and rehabilitation: innovative planning or
lowering the bar? Ecological Processes, Vol. 3, Article 8.
Price, D.T., Alfaro, R.I, Brown, K.J., Flannigan, M.D., Fleming, R.A., Hogg, E.H., Girardin, M.P., Lakusta, T., Johnston, M., McKenney,
D.W., Pedlar, J.H., Stratton, T., Sturrock, R.N., Thompson, I.D., Trofymow, J.A. and Venier, L.A. (2013) Anticipating the
consequences of climate change for Canada’s boreal forest ecosystems, Environmental Reviews, Vol. 21, pp. 322–365.
Savard, M.M., Begin, C. and Marion, J. (2014) Modelling carbon isotopes in spruce trees reproduces air quality changes due to oil
sands operations, Ecological Indicators, Vol. 45, pp. 1–8.
Schindler, D. (2010) Tar sands need solid science, Nature, Vol. 468, pp. 499–501.
Seastedt, T.R., Hobbs, R.J. and Suding, K.N. (2008) Management of novel ecosystems: are novel approaches required?, Frontiers in
Ecology & Environment, Vol. 6, pp. 547–553.
Shackelford, N., Hobbs, R.J., Burgar, J.M., Erickson, T.E., Fontaine, J.B., Laliberté, E.L., Ramalho, C.E., Perring, M.P. and Standish, R.J.
(2013) Primed for change: developing ecological restoration for the 21st century, Restoration Ecology, Vol. 21, pp. 297–304.

Smith, P. and Nichols, O. (2011) Development and assessment of rehabilitation criteria for high dune mineral sand mining on North
Stradbroke Island, Proceedings of the Royal Society of Queensland, Vol. 117, pp. 347–363.
Tibbett, M. (2010) Large-scale mine site restoration of Australian eucalypt forests after bauxite mining: soil management and
ecosystem development, in Ecology of industrial pollution, Batty, L.C., Hallberg, K.B., Kevin, B., Jarvis, A. (eds.), Cambridge
University Press, UK, pp. 309–326.
Tongway, D.J. and Ludwig, J.A. (2011) Restoring disturbed landscapes: putting principles into practice, Island Press, Washington, DC,
USA.
USA Surface Mining Reclamation Act (1977) Office of Surface Mining, Dept. 432 Interior, Government of the United States of America,
Washington, DC, USA.
Webber, M.G. and Flannigan, M.D. (1997) Canadian boreal forest ecosystem structure and function in a changing climate: impact on
fire regimes, Environmental Reviews, Vol. 5, pp. 145–166.
Williams, J.W. and Jackson, S.T. (2007) Novel climates, no-analogue communities and ecological surprises, Frontiers in Ecology &
Environment, Vol. 5, pp. 475–482.
Williamson, T.B. and Edwards, J.E. (2014) Adapting sustainable forest management to climate change: criteria and indicators in a
changing climate, “Climate Change Task Force” of the Canadian Council of Forest Ministers, Ottawa, ON, Canada.
Zipper, C.E., Burger, J.A., Skousen, J.G., Angel, P.N., Barton, C.D., Davis, V. and Franklin, J.A. (2011) Restoring forests and associated
ecosystem services on Appalachian coal surface mines, Environmental Management, Vol. 47, pp. 751–765.


D. Polster Polster Environmental Services Ltd., Canada
Mine reclamation has evolved significantly since the first Technical and Research Committee on Reclamation
symposium in 1977. Early efforts based on experience in the agricultural and forestry sectors have
encountered problems. Dense covers of seeded agronomic grasses and legumes can prevent woody species
growth, while forests planted in compacted, low nutrient waste rock generally fail to thrive. Understanding
the ecology of natural recovery processes provides a suite of solutions where these natural processes do the
“work” of restoring the site. Pioneering species have evolved to rebuild productive soils on drastically
disturbed sites (landslides, etc.). Later successional species are designed to establish in the sheltered cover
provided by these pioneering plants. Natural successional processes ensure that disturbed areas are
vegetated by the appropriate species arranged in a suitable successional trajectory. Wet sites follow wetland
trajectories, while dry sites follow upland recovery patterns. Identification of the filters or constraints that are
limiting recovery on drastically disturbed sites is the first step in defining recovery strategies. A common filter
found in many mine sites is severe compaction from years of use by heavy equipment. Simply ripping with a
large bulldozer does not adequately address compaction and fails to create the micro-site diversity needed
for vegetation establishment. Slopes on most waste dumps are at the angle of repose and will not revegetate
naturally. Similarly, the size sorting of materials on waste dumps means that coarse textured materials end
up on the lower slopes. The lack of moisture holding capacity with these materials means that vegetation
establishment is limited. Mining wastes are generally lacking nutrients that are essential for plant growth. In
some cases chemical extremes (e.g. acid rock drainage) prevent vegetation establishment and growth. Non-
native, invasive species (weeds) can overwhelm recovery processes. Excessive herbivory due to un-natural
ungulate population levels can limit recovery. Where seed sources are relatively close (500 m or less) recovery
happens naturally and new forests arise on the disturbed sites without intervention. Understanding how
natural systems address these filters and then using these solutions to overcome the filters associated with
mining sites allows these natural processes to reclaim the disturbances associated with large mines. Allowing
the reclaimed sites to take advantage of these recovery systems will reduced costs and provide improved
results. It is hypothesised that strategies based on the natural recovery processes will be more effective than
traditional reclamation systems. This paper describes the methods employed in establishing natural processes
on drastically disturbed sites. Examples from Western Canada are used to illustrate specific points and results.
Mine reclamation has grown from the fields of agriculture (including soil science) and forestry (Polster, 1989).
More recently, the field of restoration ecology has offered strategies for the reclamation of large
disturbances (Tongway and Ludwig, 2011). However, these continue to be based on historical treatments.
Understanding how natural recovery processes operate allows new systems for mine reclamation to be
established and researched (Polster, 2013). These systems are based on the idea that the most effective
recovery systems are those that assist the natural recovery processes. Ecological restoration is defined as the
process of assisting the recovery of an ecosystem that has been degraded, damaged or destroyed (SERI,
2004). The trick then becomes one of determining what things can be done to assist the recovery and what
things should not be done to avoid preventing recovery. This is much like a small cut on your hand – there
are things you can do to help your hand heal, put on antiseptic, a bandage and keep it clean, and there are
things that you can do that will prevent or hinder the recovery of your cut hand, i.e. shovel out the chicken
coop and get the cut infected. However, we cannot force our hand to heal, and neither can we force an
ecosystem to recover.

Effective strategies for the reclamation of large mines D. Polster
Understanding the processes of ecosystem degradation can provide clues for treatments that can be used to
assist in the recovery. Hobbs and Suding (2009) suggest that degradation can be divided along two lines,
abiotic and biotic. Abiotic degradation is defined as those degrading processes that are associated with the
physical aspects of the ecosystem. A large bulldozer, rock drills and dynamite, loaders and haul trucks are
abiotic degrading elements. Grazing elk or invasive species are biotic degrading agents. These degrading
forces are discussed in Section 2 of this paper.
Degrading processes can be thought of as filters (Temperton et al., 2004) to recovery. That is, they may
prevent the recovery of a full suite of species, but allow some species to establish. For instance a compacted
waste rock dump surface (abiotic filter) may prevent the growth of a healthy forest, but allow weedy species
such as Mullein (Verbascum thapsus L.) to grow freely. Similarly, hungry elk or deer (biotic filter) may prevent
the growth of Western Red Cedar (Thuja plicata Donn ex D. Don) but may allow Red Alder (Alnus rubra Bong.)
to thrive. Some species are filtered out while others can establish.
Natural processes have evolved systems for dealing with ecological filters. By looking at how these processes
operate ideas for the restoration of drastically disturbed sites can be generated. For instance; soils and plant
communities can develop on glacially compacted basal tills with the accumulation of organic matter and the
loosening of the soils by weathering and root systems of pioneering species, particularly nitrogen fixing
woody species (e.g. Alnus spp.). Eventually, over a long time, soils on these sites can support forests. Natural
processes such as trees blowing over in storms with large root masses coming up with the stump can move
non-mobile nutrients such as potassium and phosphorus to the surface for subsequent use by growing plants.
By following the systems that natural processes have evolved for the treatment of disturbed sites solutions
to anthropogenic disturbances can be identified.
This paper is organised to present the common filters that are found at most mine sites and the natural
solutions to these filters. The paper provides strategies for implementing these solutions on large industrial
disturbances, either as post-mining treatments or as ongoing processes during active mining. The paper
presents strategies for the establishment of suitable vegetation covers that will build soils and create self-
sustaining ecosystems on the disturbed ground. Examples are drawn from the author’s experience. The
hypothesis, that natural processes provide effective restoration strategies for large disturbances is presented
through a series of case histories.
Polster (2013) listed eight abiotic and six biotic filters to recovery. The mining industry typically encounters
the following abiotic constraints to recovery; compaction (haul roads and dump platforms), steep slopes
(dump slopes), adverse texture (toe of waste rock dumps), adverse chemical properties (pH extremes), low
nutrient status (most mining wastes), soil temperature extremes (dark coal wastes), excessive erosion (many
sites), and lack of micro-sites (smooth surfaces). Some of the biotic filters are also found at mines; excess
herbivory (from high deer and elk populations), competition (from seeded agronomic grasses and legumes),
propagule availability (with large disturbances); phytotoxic exudates (from weedy species like Knapweed),
facilitation and species interactions (adverse ecological interactions). Many of these filters may act together
to limit recovery so for instance, steep slopes and adverse textures may be found on waste rock dump slopes.
Similarly, treatments that can ameliorate one filter may also help with another. These are discussed below.
Compaction is probably the most common filter found at mines. Spreading materials with bulldozers creates
compaction in the soils and prevents healthy growth of vegetation. Using an excavator to prepare the surface
of the compacted area can be an effective way of alleviating compaction. The excavator can make rough and
loose configurations to open holes on the site, dumping the material that is generated from the holes in
mounds between the holes. The excavator, using a digging bucket (not a clean-up bucket), takes a large
bucket full of material and places it to the left of the hole that was just opened; half a bucket width from the
hole so it is half in and half out of the hole. Rocks are fine to include as these tend to create additional
heterogeneity. A second hole is then excavated half a bucket width to the right of the first hole. Material
from this hole is then placed between the first and second holes. A third hole is now opened half a bucket

width to the right of the second hole, with the excavated soil placed between the second and third holes.
Care should be taken when excavating the holes to shatter the material between the holes as the hole is dug.
The process of making holes and dumping soil is continued until the reasonable operating swing of the
excavator is reached. The excavator then backs up the width of a hole and repeats this process, being sure
to line up the holes in the new row with the space between the holes (mounds) on the previous row. Rough
and loose surface configurations can be used on recontoured dump slopes to control erosion and provide
suitable sites for vegetation establishment. Making a former coastal hydroelectric dam site rough and loose
and scattering woody debris resulted in the natural establishment of an average of 5,412 alder plants the
first year after treatment (50 plots, SD = 57.08 seedlings) while in the second year an average of 8,550
seedlings were found (repeated sampling of same plots, SD = 94.50 seedlings) with 67 other species. Cost at
a large northern mine for making sites rough and loose were found to be $700/ha.
Steep slopes are common at many mines. Waste rock dumps are constructed by dumping wastes around or
over the rim of the dumps and then pushing these materials over the edge of the dump slope, forming angle-
of-repose slopes. The natural analogue of waste rock dumps are talus slopes (Polster and Bell, 1980). These
slopes tend to revegetate very slowly as most of the material on the slope is too coarse to support plants.
The materials at the top of the slope are fine textured and will support plant growth, but this area is
continually bombarded by materials from above. In the middle of the slope, the materials are too coarse to
hold moisture and therefore will not support vegetation. At the bottom of the slopes, the spaces between
the large rocks slowly fill with bryophytes and organic matter and eventually will support the growth of higher
plants. The slow erosion of the fine textured materials from the top over the middle portion of the slope will
eventually allow vegetation to establish. Eventually, the areas of adverse texture are addressed by either
organic matter collecting between the large rocks or by fines washing down from above. This process can be
greatly expedited at mine waste dumps by pushing the fines from the top of the slope over the face and
down to the toe of the slope.
The low nutrient status of most mining wastes can limit the growth of some plants. However, pioneering
species, including those that are associated with nitrogen fixing bacteria have evolved to grow in areas of low
nutrients. The use of early pioneering species as the principle revegetation species on mining sites (Polster,
1989) eliminates the need for fertilisers while providing a vegetation cover that will build healthy soils on the
relatively inert materials left from mining.
Adverse chemical properties of waste materials (tailings and waste rock) at some mines require that these
materials be isolated from the environment. Acid rock drainage and metal leaching is a problem at some
mines that requires that cover systems be established (O’Kane et al., 2001). The key to the design of waste
rock covers is to ensure that there is ample clean material on top of the material that is used to seal the
wastes to allow vegetation to grow freely. The idea that woody species can be prevented from growing on a
thin cover thereby reducing the immediate cost of the cover is false. There is a risk that the integrity of the
cover will be breached by the eventual growth woody species in these areas and the costs of not providing a
proper cover will be passed on to the future mining industry either as a direct cost or as the loss of social
license and the refusal of society to permit the opening of new mines. Most woody species in western North
America need at least two meters of growth medium to support woody vegetation. This is on top of the
materials that are used to seal the wastes.
Dark substrates are a common problem at many coal mines. In the hot summer sun these materials may
reach temperatures that are too extreme for plant growth. Excess erosion and a lack of microsites are other
common problems at many mines. Making the surface rough and loose and covering the surface with 100
m3/ha of woody debris (Vinge and Pyper, 2012) can control erosion and create ideal sites for vegetation
establishment while providing a long term nutrient source for the new forest. Woody debris can provide an
important habitat for a variety of animals as well as plants (Craig et al., 2014).
Biotic filters are not as common at mines as the abiotic filters. However, competition by seeded agronomic
grasses and legumes can severely restrict woody species growth (Polster, 2010). Seeded grasses and legumes
have historically been part of the mine reclamation tool kit (Ziemkiewicz, 1977). However, long-term studies

at mines throughout the Western Canada have shown that seeding mining wastes with agronomic grasses
and legumes provides habitat for small mammals that eliminate woody species (Green, 1982). Making sites
rough and loose, adding woody debris and seeding or planting in pioneering woody species avoids the
problems of seeded agronomic grasses and legumes.
Following the natural recovery processes that have evolved to address naturally disturbed sites is the most
effective strategy for treatment of mining disturbances. By working with the natural systems that have
developed to restore natural disturbances such as landslides, volcanic eruptions, glaciation and other major
disturbances, the restoration program can take advantage of the work that is done by these natural recovery
processes. At a dam site where the dam and related infrastructure (penstock) was removed, leaving large
disturbed areas the site was treated with woody debris and the rough and loose technique and left. Now,
two years after the treatment there are over 8,550 Red Alder (a pioneering species that fixes nitrogen)
established plants per hectare along with 67 other species including Western Hemlock and Douglas-fir. All of
these have established without any planting work by humans. By taking advantage of the erosion controlling
attributes of the rough and loose ground covered with woody debris (a waste material that would have had
to be disposed of) the project saved hundreds of thousands of dollars of planting work as well as the costs
associated with the slash disposal. Rough and loose surface configurations cost about $700/ha of machine
time (4 hours/ha) to install.
Where large distances are involved between the seed producing pioneering species and the areas needing
restoration, propagules (seeds, plants, cuttings, etc.) of the pioneering species need to be provided. Seeding
with pioneering woody species such as alder can be a very effective way to cover large areas at low cost.
Pioneering species such as alder produce abundant seeds so collection of seeds locally is generally easy. The
seeds of poplars and willows are wind-blown and may not retain viability for long. Establishing these species
from cuttings can be very effective. The important aspect of this process is to use the pioneering species for
the region, even if eventually a late successional forest is the desired outcome. Later successional species will
establish under the protection of the pioneering species.
The growth of later successional forest species is enhanced by the presence of pioneering species. Pioneering
species are often associated with nitrogen fixation as this tends to be a limited element in early successional
ecosystems. Lichens may be the first plants to establish on recent lava flows and other areas where nitrogen
is lacking (Henriksson and Simu, 1971). In forest ecosystems, alder plays an important role in creating
conditions that will encourage the future forest (Sanborn et al., 1997; Peterson et al. 1996). The canopy of
alder allows the conifers underneath to grow significantly better than without the canopy (Polster and
Dubois, 2007). The canopy of alder keeps the understory cooler and moister on hot summer days as the
transpirational loss of moisture from the photosynthesising alder is evaporated from the stomata of the
alder, taking the latent heat of vaporisation from the surrounding air thus allowing the conifers to maintain
photosynthetic activity during the hot days of summer. In the winter, the deciduous pioneering species lose
their leaves, allowing the winter sun to drive photosynthesis in the evergreen conifers. Following the natural
successional patterns in the establishment of vegetation on mine sites allows the vegetation to take
advantage of the improved growth associated with the patterns.
Erosion is a common problem at many mines. Elaborate (and costly) sediment ponds and erosion control
measures are employed to limit the discharge of sediment to local water bodies. Understanding how natural
systems manage erosion can provide clues for the management of erosion at mine sites. Erosion is an ongoing
natural process; however when bare sites are exposed to the erosive elements (water, wind, frost, etc.)
increases in erosion can occur. Soil loss through erosion has been studied for many years. The universal soil
loss equation (Wischmeier and Smith, 1965) provides a listing of the factors that go into erosion and can
provide clues to managing erosion.
Compaction is one of the main causes of erosion. Water flowing down a compacted haul road can pick up
sediment and transport it down the slope. Slope length and steepness are other factors as is the soil texture

(cohesionless fine silts are the most highly erosive soils). The soil surface configuration is also important. All
of these factors can be managed by using the rough and loose surface treatment. Rough and loose ground
encourages water to soak in rather than run across the surface. Coarse woody debris (100 m 3/ha) also
controls the movement of water across the surface. Slowing the flow of water in ditches can remove the
erosive energy of the water, again promoting the movement of surface water into the groundwater system.
It is useful to note that in most natural ecosystems water soaks into the soils and does not flow across the
surface. Springs and seeps occur where this water exits the groundwater system. Creating situations where
surface waters have the opportunity to soak into the soil is one of the most effective erosion control
strategies.
Competition can be a problem at mine sites where seeded grasses and legumes have been applied. In most
cases, natural successional processes do not include a cover of herbaceous species as dense as the seeded
cover found at many mines. This seeded cover can prevent the recovery of the pre-disturbance ecosystems
at the mines where they have been established. Using seeded grasses and legumes to control erosion is not
a good idea as it precludes the ongoing vegetation development on the site.
Natural processes can provide effective strategies for the restoration of mines and other large disturbances.
Treatments such as making the surfaces rough and loose and scattering woody debris on the rough and loose
surface can assist with the control of erosion, reduction of compaction, providing north and south facing
slopes (to avoid high temperatures associated with dark substrates) and creating conditions that foster the
establishment of vegetation. By following the natural successional patterns in the establishment of
vegetation the right species will establish at the right time in the right place. By creating a diversity of habitats
through the rough and loose surfaces a diversity of species will establish. Species that require moister
locations might establish in the hollows between the hills while the species that favour drier conditions could
be found on the hills. There are species that grow on rotting wood. These would be found on the scattered
woody debris. Similarly, species that require passing through the gut of a bird to germinate (scarified) will be
found near the woody debris areas where there are perches for birds that eat those seeds. By creating a
diversity of habitats a diversity of species will establish and make the established habitats their home.
Creation of heterogeneous reclaimed areas will provide diverse ecosystems that will help to build resilience
in the former mine site area. This will ensure that the established ecosystems are self-sustaining since there
will always be a progression of appropriate species on the sites. By using pioneering species to initiate the
recovery processes (as natural systems do) the right subsequent species will show up in the right locations.
The ability of natural recovery processes to create the conditions for the establishment of the species that
are best suited for the conditions of the site being vegetated should not be overlooked in the development
of restoration strategies for drastically disturbed sites. The ability of natural systems to recover from
disturbances such as fires, floods, landslides and lava flows can be applied to the recovery from mining
disturbances.
Natural systems can be directed to recover in specific ways to foster human uses of the land. For instance, if
a reclaimed mine is to be returned to a forestry use, looking at how forests recover from natural disturbances
is an excellent first step in defining the appropriate restoration strategy for the site. On Vancouver Island,
the Quinsam Coal Mine is located in one of the best climatic areas for growing Douglas-fir. The end land use
for the mining disturbances was determined to be forestry. Test plots that were established in 1982 and
reevaluated in 2007 showed that the growth of Douglas-fir trees with a canopy of Red Alder was significantly
better than in the adjacent cut blocks where the alder was discouraged (Polster and DuBois, 2007). A strategy
of growing a 50:50 mix of Douglas-fir and Red Alder at a rate of 2,400 stems /ha (normal stocking rate in the
region is about 700 stems/ha of Douglas-fir) was selected to create a profitable reclamation program. The
Douglas-fir and Red Alder will be allowed to grow for about 25 years at which time the stems will be about
30 cm in diameter. At that point, the Douglas-fir that has been growing below the canopy of Red Alder will
start to get above the alder and will over the next 10 to 15 years shade out the alder. However, the Douglas-
fir trees will be at a density of 1,200 stems/ha and will start to compete with each other. Both the Douglas-
fir and Red Alder will have straight, unbranched stems because of the dense stocking. All of the alder will be
harvested at year 25 and used for flooring and furniture making while half of the Douglas-fir will be harvested

and sold as veneer logs (for making plywood). The remaining Douglas fir will be allowed to grow for another
125 years so that the benefits of an “old growth” forest can be enjoyed. The stems of these trees will be knot
free and will command a high price for flooring and other high value uses. By mimicking the natural recovery
processes with the alder providing the habitat for the young fir trees, the benefits of improved growth can
be translated into profit for the mining company.
Natural processes have been rebuilding damaged ecosystems for millions of years. These processes provide
the best available model for the restoration of large mining disturbances. By addressing the filters that would
prevent recovery (compaction, steep slopes, adverse textures, etc.) and providing the pioneering species
(either naturally, if seed sources are available, or by planting) that start the recovery processes, the
establishment of sustainable restored ecosystems on the mining disturbances can be assured. The natural
processes that are at the foundation of this approach are available for free by creating the right conditions
for their development. Natural successional processes will ensure an appropriate vegetation cover on the
degraded sites, building soils and replacing species as the ecosystems mature. By using pioneering species to
start the process the high cost of topsoil replacement is not needed. Pioneering species will build suitable
soils directly on waste rock as long as the soil textures are suitable. These same pioneering species will create
the habitat needed for later successional species (generally conifers in most of boreal Canada). These species
will generally establish naturally, although in some cases where seed sources are distant from the restoration
sites, planting may be needed. Providing the context for natural processes to operate is the primary objective
of the restoration practitioner. Generally fancy seed mixes, elaborate fertiliser regimes or soil replacements
(except for prairie restoration) are not needed. The natural recovery processes can do the work.
Craig, M.D., Grigg, A.H., Hobbs, R.J. and Hardy, G.E. St. J. (2014) Does coarse woody debris density and volume influence the terrestrial
vertebrate community in restored bauxite mines? Forest Ecology and Management, Vol. 318, pp. 142−150.
Green, J.E. (1982) Control of vegetation damage by small rodents on reclaimed land, in Proceedings 6th Annual British Columbia Mine
Reclamation Symposium, Vernon, BC, Technical and Research Committee on Reclamation, Ministry of Energy Mines and
Petroleum Resources, and The Mining Association of British Columbia, Victoria, BC.
Henriksson, E. and Simu, B. (1971) Nitrogen fixation by lichens, Oikos. Vol. 22(1), pp. 119-121.
Hobbs, R.J. and Suding, K.N. (2009) New models for ecosystem dynamics and restoration, Washington, DC: Island Press. 352 pp.
O’Kane, M., Januszewski, S. and Dirom, G. (2001) Waste rock cover system field trials at the Myra Falls operations – a summary of
three years of performance monitoring, in Proceedings 25th annual British Columbia Mine Reclamation Symposium, 24-27
September, Campbell River, BC, Technical and Research Committee on Reclamation, BC Ministry of Energy, Mines and
Petroleum Resources, Victoria, BC.
Peterson, E.B., Ahrens, G.R. and Peterson, N.M. (1996) Red alder managers’ handbook for British Columbia, Canadian Forestry
Services and BC Ministry of Forests, FRDA II Report No. 240, Victoria, BC.
Polster, D.F. (1989) Successional reclamation in Western Canada: new light on an old subject, paper presented at the Canadian Land
Reclamation Association and American Society for Surface Mining and Reclamation Conference, 27-31 August, Calgary,
Alberta.
Polster, D.F. (2010) Long term reclamation monitoring of vegetative covers at the Island Copper Mine, Port Hardy, BC, paper
presented at the BC Technical and Research Committee on Reclamation, BC Mine Reclamation Symposium and the Canadian
Land Reclamation Association Meeting, 20-23 September, Courtenay, BC.
Polster, D.F. (2013) Processes and functions: a new approach for mine reclamation, paper presented at the BC Technical and Research
Committee on Reclamation, BC Mine Reclamation Symposium, 16-19 September, Vancouver, BC.
Polster, D.F. and Bell, M.A.M. (1980) Vegetation of talus slopes on the Liard Plateau, British Columbia, Phytocoenologia, Vol. 8(1).,
pp. 1-12.
Polster, D.F. and Dubois, C. (2007) Quinsam Coal Mine reclamation 25 years of reclamation experience, in Proceedings of the 31st
Annual British Columbia Mine Reclamation Symposium, 17-20 September, Squamish, BC, Technical and Research Committee
on Reclamation, BC Ministry of Energy, Mines and Petroleum Resources, Victoria, BC.
Sanborn, P. Brockley, R. and Preston, C. (1997) Ecological role of Sitka alder in a young lodgepole pine stand, Forest Research Note
#PG-10, British Columbia Ministry of Forests, Prince George, BC.
SERI (2004) The SER Primer on Ecological Restoration. Version 2. October, 2004. Science and Policy Working Group, October, 2004.
Society for Ecological Restoration International. Washington DC, viewed 12 January 2015 (http://www.ser.org/resources/
resources-detail-view/ser-international-primer-on-ecological-restoration).

Temperton, Vicky M., Hobbs, Richard J., Nuttle, Tim and Halle, Stefan (2004) Assembly rules and restoration ecology, Washington,
DC: Island Press. 439 pp.
Tongway, D.J. and Ludwig, J.A. (2011) Restoring disturbed landscapes: putting principles into practice, Washington, DC: Island Press.
189 pp.
Vinge, T. and Pyper, M. (2012) Managing woody materials on industrial sites: meeting economic, ecological and forest health goals
through a collaborative approach, Department of Renewable Resources, University of Alberta, Edmonton, Alberta. 32 pp.
Wischmeier, W.H. and Smith, D.D. (1965) Predicting rainfall-erosion losses from cropland east of the Rocky Mountains, Agr. Handbook
No. 282, Washington, DC: US Govt. Printing Office. 47 pp.
Ziemkiewicz, P.F. (1977) A comprehensive reclamation research program on coal mining disturbed lands, Proceedings of the 1 st
Annual British Columbia Mine Reclamation Symposium, 16-18 March, Vernon, BC.


E. Petelina Saskatchewan Research Council, Canada

D. Sanscartier Saskatchewan Research Council, Canada
S. MacWilliam Saskatchewan Research Council, Canada
R. Ridsdale University of Saskatchewan, Canada
Rehabilitation of mine sites, if effective, has obvious benefits and supports sustainable development goals
such as reducing risk to human health and the environment, restoring ecological services and preventing the
spread of pollution. However, these activities may be associated with negative impacts such as emission of
air pollutants and greenhouse gases, occupational risks, noise, traffic and use of limited financial resources.
The consideration of sustainability concerns in the design of mine closure is not a new concept, but the use of
decision-support frameworks to guide the selection of a rehabilitation alternative offering the most
sustainable outcome is still not widespread. Sustainability appraisal/assessment (SA) is a practical tool that
can be applied for this purpose. This reflects a shift away from a traditional rehabilitation approach that
focuses primarily on cost, regulatory acceptance, duration and technical feasibility.
This paper’s goal is threefold: (i) to provide an overview of an SA approach applied to assess the
environmental, economic and social benefits, and limitations of soil organic amendment options to promote
revegetation for a northern mine closure, (ii) to share lessons learned during this exercise and (iii) to propose
improvements to the approach applied.
The SA approach was based on a tiered framework developed by the UK Sustainable Remediation Forum for
remediation of contaminated land and consists of: (i) a preliminary qualitative screening of options’
sustainability, (ii) a semi-quantitative assessment based on a multi-criteria analysis of stakeholder opinion
and (iii) quantitative analyses, including a life cycle assessment and a life cycle costing analysis. In addition to
identifying a preferred revegetation option, several lessons were collected during the study, and
improvements for the conceptual SA framework were suggested. Some lessons learned were: the tool helped
to identify sustainability gaps and ways to improve the rehabilitation alternative, the term “sustainable” is
relative and depends on stakeholders’ viewpoints, stakeholder engagement is essential and trade-offs
between sustainability criteria are inevitable. Our experience showed that the SA is a powerful tool that can
ensure the application of sustainable rehabilitation practices for site closure and should be integrated from
the very beginning to optimise outcomes.
Rehabilitation of mine sites, if effective, has obvious benefits and supports sustainable development goals
such as reducing the risk to human health and the environment, restoring ecological services and preventing
the spread of pollution. However, these activities may be associated with negative impacts such as emission
of air pollutants and greenhouse gases, occupational risks, noise, traffic and use of limited financial resources.
Sustainability appraisal/assessment (SA) can be used to consider these impacts and guide the decision to
deliver sustainable rehabilitation of the site.
As Stoeglehner and Neugebauer (2013) stated, “SA is a relatively young domain with evolving, but little
practice.” In the mining sector, frameworks for assessing and reporting mining sustainability tend to examine
the entire mining industry or the entire life cycle of a mine (from exploration to post-closure) and are not
specifically tailored to examine the rehabilitation of mine sites (Fonseca et al., 2013). In Canada, there is no
formal practice for SA, but “over the past 40 years, a series of individual assessments have explored and

Sustainability appraisal for mine closure E. Petelina, D. Sanscartier, S. MacWilliam and R. Ridsdale
demonstrated some of the possibilities of sustainability evaluation” for a range of large industrial projects
(mines, forestry, pipelines, etc.) (Gibson, 2013a).
Although sustainability aspects are being considered for the design and implementation of mine site
rehabilitation plans (e.g. Isabel, 2014; Borealis, 2014), to the authors’ knowledge, few broad and detailed SAs
have been applied to the design of mine site rehabilitation.
In this study, we applied the Sustainable Remediation UK methodology (SURF-UK; Bardos et al., 2011), an SA
framework developed for the remediation of contaminated sites, to the rehabilitation of a legacy mine site
case study. The practical decision-support framework applied can be used to guide the selection of a
rehabilitation alternative offering the most sustainable outcome (i.e. optimal balance between the three
sustainability aspects: economy, society, and environment — details are provided in the method section),
which reflects a shift from a traditional rehabilitation approach that typically focuses on cost, regulatory
acceptance, duration and technical feasibility. The overall aim of this exercise was to gain, and add to the
existing, knowledge about effective operationalisation of SA frameworks. This paper has three goals:
1. Provide an overview of an approach applied to a case study and summarise the main findings of
the case study;
2. Share lessons learned during this exercise;
3. Propose improvements to the approach applied.
The Gunnar Project is used as a case study for this purpose. The project aimed to remediate an abandoned
uranium mine site in northern Saskatchewan, Canada. The site includes unconfined uranium tailings with
gamma radiation exceeding acceptable levels and thus poses a risk to human health and the environment.
To reduce gamma radiation to acceptable levels, an engineered vegetative cover is to be installed on top of
the tailings (SRC, 2013). This study examines options of soil organic amendments to promote revegetation of
the soil cover. Four amendment options are compared: no amendment (natural recovery with erosion
control), peat (a common amendment), biochar (a novel soil amendment) acquired commercially and biochar
produced at the site using local woody material. Biochar is a coal-like material produced by thermal
treatment of woody biomass. The soil amendment selection is based on greenhouse and field studies
conducted for the Gunnar Project (Petelina, 2013, 2014).
This section summarises the appraisal method applied during this study. The case study and four revegetation
options are then described. Additional description of the methods and case study can be found in Petelina et
al. (2014).
The SA framework applied in this study (SURF-UK) was initially developed for the remediation of
contaminated sites. We applied it to the rehabilitation of a legacy mine site.
This study was an internal pilot project for the Saskatchewan Research Council. The multi-disciplinary team
(the authors) comprised a revegetation and remediation specialist, an environmental engineer, a life cycle
assessment analyst and a graduate student in social geography.
The case study was a challenging project, as its implementation required significant resources and affected
both the local communities and the environment. (For the purpose of this report, the term “implementation
of revegetation option” means the range of activities required to establish a desired plant community on the
Gunnar tailings; depending on the option, the implementation may include soil amelioration, seeding, weed
and tree control, and monitoring.)

Miscellaneous
The complexity of the project made it a good candidate for assessment using a sustainability framework. The
objective of this assessment was to identify a viable option for soil organic amendments to promote
revegetation with an optimal balance between the three sustainability aspects: economy, society and
environment.
Pluralism (i.e. the consideration of views from different groups and individuals) is central to successful
decision making related to sustainability (Bond et al., 2013). As such, SA frameworks should ensure
stakeholders’ participation in order to capture different views, values and beliefs. As the project was an
internal pilot study, full-scale stakeholder involvement was not conducted because of budgetary constraints.
However, exercises were conducted to mimic stakeholders’ engagement for the purpose of generating
knowledge and lessons learned on the approach (details provided in Section 2.1.3).
The SURF-UK framework is a tiered approach (Figure 1), and consists of three levels of analyses: qualitative,
semi-quantitative and quantitative. As presented in Figure 1, the approach begins with a qualitative analysis.
If a decision can be made at that point, the assessment ends, the decision is recorded and the remediation is
implemented. One would carry on to the next tier only if data gaps remained and the information gathered
in the lower tier(s) could not support sound decision making. The tiered approach allows for adjusting
assessment scope while moving from one tier to another, with a corresponding increase in complexity, level
of effort and cost. The benefit of a qualitative approach is that it is cost-effective and relatively simple to
perform. Quantitative analyses require more resources, but can provide additional insights. For the purpose
of this internal pilot study, all three tiers were carried out as follows:
 Tier I: A preliminary qualitative screening analysis to assess the relative sustainability of the
options
 Tier II: A semi-quantitative assessment using multi-criteria analysis (MCA) to evaluate the relative
importance of the proposed sustainability criteria and study stakeholders’ preferences for the
revegetation approach
 Tier III: Quantitative assessment of environmental impacts and project cost estimated using
screening life cycle analysis (LCA) and life cycle costing (LCC)
The authors carried out the preliminary qualitative screening assessment of the options. The multi-
disciplinary team composition ensured a balance of technical, environmental, social and economic expertise
during the option screening. The team reviewed reports and existing literature, including a socio-economic
baseline study, public consultation reports, technical reports on the case studies and reports on the
technologies considered (e.g. SRC, 2013; Schackley et al., 2011). Interviews with parties relevant to the study
were also conducted (e.g. Gunnar Project management team, engineers, biochar producers, consultants).
The collected data were used to identify benefits and limitations for each option. After the team’s discussion
of option benefits and limitations, the options were screened through a set of environmental, social and
economic criteria, and the relative sustainability of each option was determined. First, all the options were
rated against each criterion (Table 1), then each option was ranked in relation to environmental, social or
economic aspects and, finally, the preferred option was identified. This approach also allowed the
identification of trade-offs associated within each option. The 19 evaluation criteria (Table 1) were selected
for their relevance with the project.

Environmental Social Economic

Biodiversity footprint Occupational risks Project cost
Air quality Site aesthetic Project risks
Energy consumption Land use Economic opportunities
Greenhouse gases Public safety Province revenue
Carbon sinks Community perception Job opportunities
Waste generation Community involvement Job diversity
Technical feasibility
Stakeholders’ opinions were studied using an MCA approach, which is commonly used to examine a variety
of options against a wide range of environmental, social and economic criteria. MCA can help ensure sound
decisions are made when multiple stakeholders with diverging and/or conflicting perspectives are involved
(e.g. White and Noble, 2012).
Expert Choice® was selected as an MCA tool. It allows accurate results to be obtained even when complex
sustainability issues are under consideration (e.g. White and Noble, 2012). Furthermore, the Web-based tool
alleviates some logistical issues around travel, financing and managing stakeholder consultation.
Typically, survey participants would be selected from across a range of stakeholder groups affected by, and
who could influence, a project. In this pilot project, the survey participants were five internal employees from
different departments (the authors did not participate in the survey). The selected employees acted as proxy

Miscellaneous
stakeholders and reflected different perspectives of five commonly interested parties in such projects: local
communities, proponents, environmental consultants, finance specialists and technical specialists.
Prior to participating in the survey, stakeholders were provided with documentation outlining the project
objectives, the options being evaluated and the survey methodology. All five participants completed 100%
of the assessment. The MCA was carried out by evaluating the four revegetation options based on the nine
Tier II criteria (Table 1). Reading the supporting documentation and answering the survey took participants
two to four hours.
Quantitative assessment consisted of a screening LCA and a screening LCC analysis. LCA is a method of
examining the potential environmental effects of a product or process across its life span. LCC is employed
to examine the total cost of systems over their entire life cycle. Screening LCA and LCC studies were
conducted for only three of the four options (Options 2–4) based on the results of Tier I and II analyses and
because of budgetary and timing constraints. Screening studies were limited in scope and detail when
compared to full-scale analyses and were intended to provide information for internal use, or to direct future
research.
Only readily available data were used in screening studies. The data were selected to be as specific as possible
to the case study, but constraints required the use of several assumptions. The following main activities were
considered: acquiring/producing organic soil amendment, application of organic soil amendment and
transport of personnel and materials. It was determined that only the relative impacts or costs of the options
would be assessed, so similar activities among the options (e.g. seeding and fertilising) were excluded from
the analyses because these activities are associated with similar impacts. Biochar was assumed to be
produced with slow pyrolysis systems (commercial and mobile) based on Roberts et al. (2010), while peat
was harvested in central Saskatchewan and transported to site. Details on the limitations of the tools and
their implications on results can be found in Petelina et al. (2014).
SimaPro (version 7.3.2) (a specialised LCA software program) was used for the LCA, and the IMPACT 2002+
midpoint method was employed to examine five environmental metrics (terrestrial acidification, respiratory
inorganic, non-renewable energy, mineral extraction and global warming). The LCC was carried out using an
Excel spreadsheet model. The price of peat was based on the retail price of bailed peat ($C 300 per dry tonne).
We assumed a price of $CAD 1,000 per dry tonne for commercial biochar study based on reported price
ranges (Kulyk, 2012; Brunjes, 2012). The capital cost and operation and maintenance cost of the mobile unit
were based on Shackley et al. (2011) using a method from Roberts et al. (2010).
As stated by Fonseca et al. (2013), SAs often follow “a siloed approach by considering the various indicators
in isolation.” In order to improve the methodology and consider indicators in a more interrelated way
(which the tiered approach helps accomplish), we evaluated the preferred option against Gibson’s (2013b)
“core imperatives for sustainability assessment” at the end of the three tiers of the assessment. Gibson
(2013b) argues that, to be considered sustainable, the preferred option(s) must deliver on six core
imperatives:
1. “Aim to reverse the prevailing (unsustainable) trends.” This means that the mitigation of adverse
effects, as commonly done, may not be sufficient to achieve sustainability, as mitigation tends to
reduce the effects but does not reverse them.
2. “Integrate attention to all of the key intertwined factors that affect sustainability.” For example,
interrelations exist between ecological integrity and equity among stakeholders. Change in
ecological integrity may impact the local aboriginal communities more than other stakeholders.
3. “Seek mutually reinforcing gains.” This imperative promotes the identification of win-win-win
options, and synergetic ways to reinforce ecology, economy and society.

4. “Minimise trade-offs.” Following the previous imperative, the preferred option should provide the
best compromise between benefits and limitations for all three pillars of sustainability (i.e.
economy, society and environment).
5. “Respect the context.” SAs are site- and context-specific and should consider the people and
places involved.
6. “Be open and broadly engaging.” As discussed, pluralism is a key component of sustainability
(Bond et al., 2013). These assessments must be more than technical exercise and must involve the
public in meaningful ways to ensure public acceptance of the decisions.
The above approach was used to assess the following four options for soil organic amendments to promote
revegetation of engineered tailings cover for Gunnar Project in northern Saskatchewan (only three options
were assessed during Tier III). The site includes unconfined uranium tailings, with gamma radiation exceeding
acceptable levels, and thus poses a risk to human health and the environment. To reduce gamma radiation
to acceptable levels and decrease contaminant loading to the environment, approximately one metre of
engineered vegetative cover is to be installed on the top of the tailing deposits (SRC, 2013).
 Option 1 (natural restoration) consisted of a passive restoration approach (i.e. creating favorable
conditions for native recovery of the site). This approach included installation of wind controls
(e.g. wind breakers) to protect the engineered covers from blowing and to create conditions
favourable for natural vegetation recovery. Owing to harsh environmental conditions and poor
growing media, it is expected that the project revegetation goal might take a few decades to
reach (Petelina, 2014). The goal is the establishment of a native self-sustaining grass-legume
community with 60–80% ground cover. As natural recovery would be driven not only by native
herbaceous species, but also by woody and exotic species, tree suppression and active weed
management would be required until a dense herbaceous cover is developed.
 Option 2 (peat) consisted of a conventional proactive revegetation methodology including soil
amended with peat followed by fertiliser application and native plant seeding. Based on
greenhouse and field tests, this approach could achieve the revegetation goal within two or three
years without additional maintenance (Petelina, 2013).
 Option 3 (commercial biochar) was similar to Option 2, with the only difference being that peat
was substituted with biochar obtained from a commercial supplier. It was assumed that biochar
was similar to peat in its ability to promote plant growth (based on Petelina, 2013), so all other
features of Option 3 are the same as for Option 2.
 Option 4 (local biochar) was similar to Option 3, but biochar was assumed to be produced with a
mobile unit installed onsite using local wood wastes as feedstock. The other features of Option 4
are the same as for Option 3.
Under northern Saskatchewan conditions, revegetation might demand a lot of resources with a high risk of
failure owing to factors such as site remoteness, hard access to the sites, poor soil properties, short growing
season, harsh winter conditions and limited supply of native plant seeds adapted to northern conditions.
Therefore, a thorough consideration of potential revegetation options is essential for the project’s success.
Additional information on the case study can be found in Petelina et al. (2014).

Miscellaneous
Table 2 provides an overview of limitations and benefits for the options revealed at each stage of the SA.
Petelina et al. (2014) provides a detailed discussion of results obtained at each tier. The following summarises
the overall results from the options analysis:
 Tier I — qualitative screening: Options 1 and 3 provided the least preferred approaches, while
Option 4 was evaluated as the option offering the most sustainability gains. However, biochar
production and application for Option 4 were less proven than peat harvesting and application,
and thus had an increased risk of failure. As such, Option 2 could be considered a more practical
solution, though opportunities for economic development and community involvement would be
lost.
 Tier II — MCA: Stakeholders selected Option 4 as the preferred option. Options 1 and 3 were
identified as the least desirable options, which supports the validity of the Tier I results.
 Tier III — quantitative analyses: It was not immediately clear from the quantitative analysis results
which option was preferred. Option 4 was found to be preferable from the environmental point
of view, while Option 2 was the least costly. Option 4 appeared to strike a balance between cost
and environmental performance.
The overall results show that Option 4 is likely to be the most sustainable revegetation option for the Gunnar
Project. This option is technically feasible and the most socially accepted (i.e. preferred by stakeholders). It
also provides some environmental improvements (e.g. carbon sink) and economic opportunities (local
development and testing of a new technology). On the other hand, this option trades high cost and
uncertainties associated with a new technology for social and environmental benefits.
If decision makers deem Option 4’s trade-offs unacceptable, the decision should be made in favour of Option
2. This option involves a proven and cheaper technology than Option 4, but it trades off the environmental
aspect of sustainability, as peat harvesting and application are associated with higher environmental impacts
than biochar production and provide no environmental gains. The stakeholder survey indicated Option 2 is
the second preferred option. Therefore, it can be considered socially acceptable within the context of the
analysis (i.e. an internal pilot study).
The other two options (i.e. Options 1 and 3) are not advised as sustainable options. Option 1 trades a high
risk of project failure and lack of social benefits for the lowest cost and environmental impact. Option 3 is
both the option least preferred by stakeholders and the most expensive.

Option 1 (natural Option 3 (commercial

Tier Option 2 (peat) Option 4 (local biochar)
restoration) biochar)
In spite of some benefits This option was believed This option was believed This option may be more cost
Tier I: Preliminary qualitative screening
(i.e. low cost, low to be more costly and to be more costly than the effective than Options 2 and 3,
occupational risks, low complicated than Option previous two and but it was thought to bear
emissions), this option 1, yet more reliable and associated with significant significant challenges, risks and
was assessed as requiring allowing achievement of challenges and numerous uncertainties. A successful
a long time to achieve the the desired results uncertainties. On the implementation of the option
project goals and was within a shorter period. other hand, successful may have a positive impact on
related to long-term risks, implementation of the local ecosystems and local
which could also have option may not only have communities’ project
negative impacts on local a positive impact on local perceptions, and may introduce
ecosystems and local ecosystems and local alternative methods of land
communities’ project communities’ project reclamation. It may also
perceptions. perceptions, but may also provide long-term
create a precedent for a opportunities for local
large-scale use of biochar communities.
in land reclamation.
Survey respondents Survey responses Respondents believed Respondents selected this
believed that this option showed that this option that this option may be option as providing the most
could provide significant may be the most the most beneficial in benefits for community
benefits in relation to air beneficial in terms of terms of biodiversity, but involvement, project risk
quality, greenhouse gas future land use may have poor reduction and local economy. It
Tier II: Multi-criteria analysis
emissions, occupational opportunities. It performance for also promised a higher

risks and project cost. significantly reduces the community involvement, performance in greenhouse gas
However, its project risks and project cost, economic emission reduction, biodiversity
implementation may supports biodiversity opportunities and air and future land use. On the
compromise biodiversity goals, yet is associated quality. It was thought to other hand, it was believed to
and community with limited community be associated with higher have the lowest rating for air
involvement, may pose involvement and fewer project and occupational quality and project cost. Its
higher project risks and economic opportunities risks, and lower benefits occupational risks survey score
may have a lower impact for the region. In in terms of future land was close to that of Option 3.
on economic comparison with Option use, than Option 2.
opportunities and future 1, Option 2 was thought
land use. to compromise
greenhouse gas
emissions, occupational
safety and project cost.
n/a: This option was This option resulted in This option had the This option showed the best
excluded from the the greatest potential lowest potential balance among all
Tier III: Quantitative analyses
quantitative analysis environmental impacts greenhouse gas impacts environmental impacts. It had a
because of timing and in all impact categories, because of biochar’s relatively low potential impact
budgetary constraints yet it was the least costly potential to sequester on air quality, terrestrial
option. atmospheric carbon. This acidification/nutrification, non-
option also had the renewable energy and mineral
lowest impact for non- extraction, and resulted in a
renewable energy reduction of greenhouse gas
consumption owing to emissions. It is slightly more
potential usage of biochar costly than Option 2.
combustible co-products.
On the other hand, its
total cost was the highest.

Miscellaneous
As argued by Gibson (2013b), the selected option should deliver on six “core imperatives” to be considered
sustainable. Evaluation of the SA approach and selected option against Gibson’s imperatives showed the
following:
1. “Aim to reverse the prevailing (unsustainable) trends”: Option 4 prevents a number of
unsustainable trends prevailing in conventional revegetation practices, such as disturbance
of wetlands due to peat harvesting, low community involvement in the project and absence
of local economy opportunities, and saving resources without considering biodiversity gains.
2. “Integrate attention to all of the key intertwined factors that affect sustainability”: An
assessment of interaction between different factors is a challenging task and was not
formally carried out other than by examining a number of sustainability criteria and
consulting stakeholders.
3. “Seek mutually reinforcing gains”: Option 4 provides a number of mutually reinforcing gains.
For example, it promotes local biodiversity, which will result in increased land use
opportunities for future generations; it creates job opportunities for local people; and it can
boost the local economy by applying new technology for other projects.
4. “Minimise trade-offs”: The results of the MCA and quantitative analysis showed that Option
4 provides the best compromises between different stakeholder values, as well as
environmental and economic gains.
5. “Respect the context”: Option 4 was selected specifically for the Gunnar Project context,
location and site-specificities. The sustainability criteria were applied in the context of the
key issues at the local, regional and global levels. The project stakeholders’ involvement in
the SA also ensured that the selected option would meet sustainability requirements in the
local context.
6. “Be open and broadly engaging”: The internal pilot survey engaged a wide range of
stakeholders. Option 4 provides the best balance between the values and views of
stakeholders representing different groups, such as technical professionals, financial experts
and aboriginal people, so broad engagement of the public and interested parties was
achieved within the context of the study (i.e. an internal pilot study).
Our analysis showed that the preferred option (Option 4) selected through three tiers of SA satisfies five of
the six sustainability core imperatives. We therefore deem Option 4 to be sustainable.
In addition to the revegetation option evaluation, our team compiled lessons learned from the application of
the SA. Lessons learned are listed by topic below.
Tier I — Preliminary qualitative analysis aiming to identify advantages and disadvantages of each remediation
option and assess relative sustainability of the options:
 Tier I was beneficial for acquiring mutual understanding of the appraisal process, case study and
options being compared.
 It should involve key stakeholders to ensure pluralism from the very beginning.
 Option ranking against each criterion was challenging, so additional research should be conducted
to simplify the method.
Tier II — MCAs performed to survey stakeholders about their opinions regarding the relative importance of
proposed sustainability criteria and stakeholders’ preferences for revegetation options:
 The tool used for the survey design and data processing was suitable to collect diverging views
from stakeholders and facilitate decision making.

 Application of the Web-based software simplified the survey process for both researchers and
participants and can be used for large-scale projects.
 Survey participants found the survey support documentation helpful. However, this
documentation may also be a source of bias.
 Stakeholder selection for MCA should ensure a broad representation of all groups of interested
parties.
 In order to limit surveys to a reasonable duration, only a limited number of criteria should be
applied; therefore, using criteria optimisation is crucial for MCA success.
 Participants should be informed about the MCA outcome, as participant feedback on the MCA
outcome can provide additional insights for decision making.
Tier III — Quantitative assessments of environmental impacts and costs of revegetation options:
 Even with the several assumptions used in Tier III, the assessment provided valuable information
to support decision making. It also helped in identifying the areas that needed further
examination in analysis, design and project planning.
 The LCC helped identify the economically viable options.
 Quantitative screening analyses may be best conducted before the MCA for more informative
stakeholder opinion survey (i.e., prior to Tier II).
The overall SA process:
 The term “sustainable” is relative and depends on stakeholders’ viewpoints, so the SA goal is not
to find the only possible option, but to improve the understanding of the possible trade-offs
associated with each option and clarify the ways to achieve a balance between the project goals.
 Early engagement of different groups of stakeholders is essential to the success of the SA and to
stakeholders’ acceptance of its results.
 The framework helped to identify sustainability gaps and ways to improve project performance in
terms of sustainability.
 Trade-offs among sustainability criteria are inevitable.
 Evaluation of the selected option(s) against Gibson’s core imperatives (Gibson, 2013b) is an
important aspect of the SA, which allows going beyond a siloed assessment. The SA should be
designed with these imperatives in mind.
 Assessment of intertwined factors is challenging but should be incorporated into the SA approach.
The lessons learned were used to develop a roadmap for SA projects that provides improvements to the
SURF-UK framework applied in the current study (Bardos et al., 2011). We believe that this roadmap enriches
theoretical SA frameworks with practical aspects, ensuring SA efficiency and clarifying the SA process flow.
Figure 2 presents the roadmap as a flowchart. It is important to note that this is an iterative approach.
Findings and discussion during the assessment may lead to revisions of previous steps. Steps 1 to 7, 9, 11 and
14 are carried out by the core team. The core team may consist of (but is not limited to) the project
proponent, site manager/owner and consultant(s). Stakeholders are involved in all other steps.
The details of the steps are as follows:
 Step 1: Identify the need for an SA.
 Step 2: Agree on SA objectives.

Miscellaneous
 Step 3: Review potential remediation options and agree on which options are to be considered in
the course of the SA.
 Step 4: Develop clear descriptions of the options to be evaluated.
 Step 5: Identify the SA approach, including, for example, tiers to be implemented (e.g. budgetary
constraints may limit the assessment to Tier I, which may influence the level of detail of Tier I) and
sequence of studies to be performed (e.g. LCA and LCC screening can be carried out as part of Tier
II to provide stakeholders engaged in the assessment with more reliable data regarding
environmental and economic benefits of the remediation options).
 Step 6: Develop the design for each study identified at Step 5.
 Step 7: Develop a stakeholder engagement plan to ensure involvement of all interested and
affected parties.
 Step 8: Communicate with stakeholders to provide them with the SA concept and objectives, and
clarify expectations about their participation in the SA and acquire corresponding commitments.
 Step 9: Identify criteria to be used for the SA and ensure that each criterion is clearly defined and
linked with the six core imperatives for SA.
 Step 10: Conduct an introductory session with the stakeholders to agree on the SA approach to:
Introduce the options and identify the advantages, disadvantages and uncertainties associated
with each; introduce the SA approach (including core imperatives); and review the SA objectives
and criteria to ensure they reflect stakeholders’ values and concerns regarding the project and
that the parties have a common understanding of the terminology.
 Step 11: Update the options’ description and SA approach (i.e. the outcomes of Steps 4 to 6),
including but not limited to the objectives, criteria and study design, in line with the stakeholders’
feedback.
 Step 12: Implement the SA. Upon completion of each tier, arrange a stakeholder meeting to
review the results and evaluate a preferred option against the six core imperatives. Determine
whether sufficient information exists for each sustainability aspect, or if additional data are
required. The team carries on to the next tier only if the information collected cannot support
sound decision making. If gaps are identified during the previous tier, the design of following
studies should be adjusted accordingly.
 Step 13: Upon SA completion, conduct a wrap-up meeting with stakeholders to present the SA
results and acquire feedback.
 Step 14: Report the SA outcome.
SA is a practical tool that can be applied to consider and incorporate sustainability concerns in the design of
mine closure. There are existing SA frameworks and tools for the mining industry, but few are specifically
tailored to examine the rehabilitation of mine sites. In this paper, we applied an SA framework developed for
the remediation of contaminated sites to the rehabilitation of a legacy mine site case study. Practical
improvements to the SA approach were proposed in the form of a roadmap. The improvements were based
on the experience acquired and lessons learned during the case study. Some of the overall key insights gained
by the project team included the relative nature of sustainability, the inevitability of trade-offs among
sustainability criteria and the importance of incorporating the six core imperatives into the SA in order to
avoid a siloed analysis. Our experience showed that the SA is a powerful tool that can ensure the application
of sustainable rehabilitation practices for site closure, and should be integrated from the very beginning of a
remediation project to optimise outcomes.

We acknowledge the Saskatchewan Research Council executive team for financial support of this project. We
thank Expert Choice® for providing a free trial of the MCA software as well as technical support. Finally, we
thank our five colleagues for acting as stakeholder representatives for the survey.
Bardos, R.P., Bakker, L.M.M., Slenders, H.L.A. and Nathanail, C.P. (2011) Sustainability and Remediation, in Dealing with contaminated
sites – From theory towards practical application, F.A. Swartjes (ed), Springer, New York, USA, pp. 889-948.
Bond A., Morrison-Saunders, A. and Howitt, R. (2013) Conclusions, in Sustainability Assessment – Pluralism, practice and progress, A.
Bond, A. Morrison-Saunders and R. Howitt (eds), Routledge, New York, USA, pp.263-270

Miscellaneous
Borealis (2014) Mine site rehabilitation: Key elements to consider, white paper, viewed February 6, 2015, http://www.boreal-
is.com/white-papers/, last updated March 2014.
Brunjes, L. (2012) Biochar industry status update, in Proceedings 2012 US Biochar Conference, July 29-August 1, 2012, Sonoma State
University, Rohnert Park, CA, viewed Dec 10, 2013, http://2012.biochar.us.com/362/biochar-industry-status-update.
Contaminated Land: Applications in Real Environments
CL:AIRE (2010) A framework for assessing the sustainability of soil and groundwater remediation, SURF-UK, London, United Kingdom.
Fonseca, A., McAlllister, M.L. and Fitzpatrick, P. (2013) Measuring what? A comparative anatomy of five mining sustainability
frameworks, Mineral Engineering, Vol. 46–47, pp. 180–186.
Gibson, R.B. (2013a) Sustainability assessment in Canada, in Sustainability assessment — pluralism, practice and progress, A. Bond,
A. Morrison-Saunders and R. Howitt (eds), Routledge, New York, USA, pp. 167-183
Gibson, R.B. (2013b) Why sustainability assessment?, in Sustainability assessment — pluralism, practice and progress, A. Bond, A.
Morrison-Saunders and R. Howitt (eds), Routledge, New York, USA, pp. 3-17.
Kulyk, N. (2012) Cost-benefit analysis of the biochar application in the U.S. cereal crop cultivation, Paper 12, Center for Public Policy
Administration Capstone, University of Massachusetts, Amherst, USA.
Petelina, E. (2014) Restoration of native plant cover after uranium mining: A case study from northern Saskatchewan, Canada,
Master’s Dissertation, University of Saskatchewan, Saskatoon, Canada.
Petelina, E., Sanscartier, D., MacWilliam, S. and Ridsdale, R. (2014) Environmental, social, and economic benefits of Biochar
application for land reclamation purposes, in Proceedings, 38th Annual Mine Reclamation Symposium, 22–25 September
2014, Prince George, Canada, pp. 170-182
Roberts, K.G., Gloy, B.A., Joseph, S., Scott, N.R. and Lehmann, J. (2010) Life cycle assessment of biochar systems: Estimating the
energetic, economic, and climate change potential, Journal of Environmental Science and Technology, Vol. 44, pp. 827–833.
Shackley, S., Hammond, J., Gaunt, J. and Ibarrola, R. (2011) The feasibility and costs of biochar deployment in the UK, Carbon
Management, Vol. 2(3), pp. 335–356.
Stoeglehner, G. and Neugebauer, G. (2013) Integrating sustainability assessment into planning: Benefits and challenges, in
Sustainability Assessment — pluralism, practice and progress, A. Bond, A. Morrison-Saunders and R. Howitt (eds), Routledge,
New York, USA, pp.245-262
Saskatchewan Research Council (SRC) (2013) Gunnar site remediation project, Environmental impact statement, SRC publication,
Saskatoon, Canada, 920 p.
White, L., and Noble, B. (2012) Strategic environmental assessment in the electricity sector: An application to electricity supply
planning, Saskatchewan, Canada, Impact Assessment and Project Appraisal, Vol. 30(4), pp. 284–295.
Isabel, D. (2014) Conception et mise en oeuvre du programme de restauration du site minier Siscoe, Oral presentation, Canadian
Land Reclamation Association National Conference, 22-25 September 2014, Mont Tremblant, Canada.
Petelina, E. (2013) Gunnar revegetation research, Field trials, stage II, interim report. SRC Publication No. 12194-460-10B12.


D. Harrison Golder Associates Ltd., Canada

D. Crockett Golder Associates Ltd., Canada
L. Eykamp Teck Resources Ltd., Canada
Surface mining has the potential to impact both natural and cultural landscapes. Hence, reclamation and
closure planning is an integral part of the mine development process — one that presents an opportunity for
increasingly effective and sustainable solutions. Reclamation and closure planning, including landscape
design, can address post-mine scenarios as they relate to a range of environmental and social objectives. This
study examines the role of landscape design and visualisation techniques to support an integrated
reclamation planning process that increases landscape performance and better communicates technical
information related to post-mine scenarios. Teck’s Elkview operation provides a case study of progressive
integrated landscape design and visualisation techniques in closure and reclamation planning. This study
examines landscape design efforts that help create more naturalistic landforms and their potential benefits
related to environmental management and building strong community relationships.
Regulators and stakeholders are placing increasing demands on mining companies to develop post-mining
landscapes that demonstrate a more natural appearance. Post-mine landform design can meet this objective
while concurrently providing other benefits such as enhanced wildlife habitat. This study explores methods
for developing diverse landforms and revegetation patterns that blend with adjacent natural landscape
character; these effectively integrate drainage systems to control surface flows, reduce erosion, minimise
long-term maintenance, and create micro-topography that supports wetlands and tree, shrub, and forb cover
that promotes biodiversity.
While the contribution of visual aesthetics to landscape performance is difficult to evaluate, the use of data-
driven three-dimensional (3-D) visualisations can support improved public acceptance of post-mine landform
and revegetation design options by illustrating design intentions and reclamation strategies. Visualisations
play a role in communicating technical project design information, supporting community involvement by
identifying interests and priorities, and building awareness and understanding of post-mine reclamation
scenarios. Advanced visualisation techniques are able to adapt to changes in project design and play a role
much larger than graphic support.
Proactive management during mine closure is important to the discharge of environmental and social
responsibility in the mining industry and has an important influence on the landscape. Leading industry
practice is to commence closure planning well before mine closure is anticipated; this not only enables
improved environmental and social outcomes but also helps guide mining operations towards post-mining
landscapes that avoid expensive earthwork activities after closure (Government of Western Australia, 2011).
Mine closure in British Columbia is guided by the Health, Safety and Reclamation Code for Mines in British
Columbia (British Columbia, 2008). Closure plans for mines are required to allow for progressively reclaiming
areas and must focus on the development of geotechnically and hydrologically stable landscapes. An
important consideration in closure planning is the development of post-closure landforms that address visual
impacts caused by mining operations. One of the challenges for mining companies is to communicate
potential visual changes to the landscape, which result from mining activity over the life of the mine

Teck’s Elkview Operations: An integrated approach to conceptual closure & reclamation planning D. Harrison, D. Crockett, and L. Eykamp
operation, to regulators and the public in order to identify reclamation and landform design options that can
return a post-mining landscape to an acceptable level of natural appearance.
The township of Sparwood, with a population of approximately 3,667 (British Columbia, 2011), is located in
southeastern British Columbia, within the Elk Valley region of the Southern Canadian Rocky Mountains
(Figure 1). This scenic region is characterised by the wide Elk River valley, forested slopes, and rocky ridgelines
at higher elevations. The area has a long history of industrial use, including forestry and coal mining. Mining
at the Elkview mine site, adjacent to Sparwood, commenced in the late nineteenth century and progressed
in 1969 to the first large-scale open-pit operation in the Elk Valley. Teck Resources Limited (Teck) today
continues open-pit coal mining at its Elkview Operations (Elkview), which is one of five active steelmaking
coal mining and processing operations owned and operated by Teck in the Elk Valley.
Elkview is located approximately 2.5 km east of downtown Sparwood, and mining occurs on three north-
south trending ridgelines. Baldy Ridge is the closest ridgeline to Sparwood and the most visually prominent,
forming the eastern skyline of the community. To sustain mining at Elkview, Teck is proposing to extend
mining operations by developing the Baldy Ridge Extension project. This project will see mining continue
along the remainder of Baldy Ridge, resulting in the lowering of the ridgeline and development of waste rock
spoils in the mined-out pits; mining operations will reach very close to Sparwood. Permitting of the project
is proceeding via the British Columbia (B.C.) government-led regulatory process, which prescribes regulatory
and public consultation as part of the approval process.
While mining has a long and economically positive association within Sparwood and the Elk Valley, local
residents have voiced concern over anticipated long-term visual changes to the existing landscape associated
with the mining of Baldy Ridge during operations and after closure. In addition to community concerns over
visual changes to the local landscape, B.C. provincial regulators responsible for reclamation have
communicated a desire to have Elkview’s post-mining landscape resemble natural land forms.
Teck engaged an interdisciplinary team of landscape design and visualisation specialists from Golder
Associates to develop a reclaimed landform concept that addresses community and regulatory concerns

Miscellaneous
regarding visual aesthetics and also maintains existing environmental requirements and closure plan
objectives. The team was tasked with providing recommendations to Teck’s engineering and closure planning
team to consider ways to achieve these objectives in a cost effective and operationally efficient manner. To
aid in communicating the proposed visual changes of the project through the permitting process and to
demonstrate the post-closure landscape, Teck recognised that professionally developed and detailed
landscape visualisations of operation and post-closure landforms were needed. Such images needed to show
both a sequential presentation of visual changes to Baldy Ridge over the mining period, and, more
importantly, the anticipated final landform and reclaimed post-mining landscape.
Waste rock spoils are typically strongly engineered geometric forms designed with geotechnical and
hydrologic stability and operational efficiency as primary concerns. The formation of waste rock spoils are
typically characterised by large, uniform angled slopes with level tops and strong edges visible at their crest
and along bench lines. The slopes, composed of angular mineral rock and fines, can be darker in colour, more
uniform in texture, and slow to develop tree or shrub cover in comparison with adjacent natural landforms.
The result is the introduction of prominent landforms that contrast with and appear unnatural in character
relative to the adjacent unaltered landscape, which may include undulating, rounded, and weathered terrain
such as avalanche chutes, ravines, variable clusters of mixed deciduous and coniferous trees, shrubs, open
grasslands, and rock outcrops (Figure 2).
To facilitate feedback from the public on predicted project-related visual changes, Teck began by engaging
the community of Sparwood through presentations and poster displays at community consultation events in
2014; these included simulated images of an initial design concept of Baldy Ridge from key local viewing
locations. Simulated visual images of the Baldy Ridge project were developed by visualisation specialists using
landscape modelling software based on digital terrain data provided by Teck’s engineers. Initial designs were
created to address current regulatory and operational requirements for landform design, which focused on
stability and efficient waste rock movement, and demonstrated the strong engineering features described
previously.
Public consultation events successfully generated feedback that identified specific community concerns
related to visual aesthetics based on these initial visualisations. Figure 3 presents a simulation image used

for community consultation and illustrates the character of the initial design visible from Highway 3, near the
entrance to Sparwood.
To address the community’s visual aesthetic concerns, the team’s landscape architect initiated a preliminary
visual review of adjacent and similar landform typologies, starting with photographs and graphic images that
explore potential end states for the waste rock spoils closest to Sparwood. The images were used to explore
ways to diminish the strong, anthropogenic character of the engineered landforms by interrupting the large,
uniform masses of the slopes with ravines or draws to simulate mountains or hillsides formed through natural
environmental processes. Teck engineers were provided with recommendations to vary slope gradients, with
slopes often in the 20 degree range rather than the angle of repose. Engineering methods to achieve diverse
landforms included varying the width of the benches to create a more scalloped edge, which would form
dump fans to simulate ravines and draws. Rounding the shoulders of the bench and pit crests would help to
provide a more convincing weathered appearance. Adding large mounds of spoil material and organic soils
to the top of plateaus and benches, to simulate weathered knolls and hummocky features, would help to
create additional variations in slope and aspect with the potential for associated soil and moisture benefits.
Figure 4 illustrates an initial concept design developed to mimic natural landform and vegetation.
The three-dimensional (3-D) landscape modelling technique uses geographically specific information to
develop recognisable, simulated graphic images. There are a diverse range of technologies and techniques

Miscellaneous
available to develop and present these geographic visualisations, ranging from tools like Google Earth to
more advanced landscape simulation software capable of nearly photorealistic levels of detail. These
powerful data exploration tools can help represent the composition of landscape elements based on data-
driven modelling information. Data-driven landscape models provide a high degree of spatial accuracy in
their representations, which allows predictions of post-closure landscape simulations to be both defensible
and reliable illustrations of closure designs (MacEachren, 2004).
While landscape modelling and visualisations can present accurate and recognisable landscapes, they also
allow for modelling of alternative landscape design scenarios. These scenarios may be geographic in nature
and demonstrate alternative closure designs and/or they may be temporal and demonstrate changes over
time. This approach is useful for characterising post-closure designs to assess potential alternatives for
landscape performance. Where designs are not seen to be effective, landscape modelling enables specialists
to adapt the design or portions of the design to improve its overall effectiveness and then to reassess its
ability to meet objectives such as reducing visual impacts.
As outlined above, portions of the project are highly visible at sensitive viewing locations within the
community of Sparwood and along Highway 3, a popular tourist route between British Columbia and Alberta.
The concerns of regulators and the public over the potential visual impacts of current landform practices
place an increasing demand on mining companies like Teck to communicate post-closure landscape designs
and demonstrate that they understand and respect local stakeholder’s values in the process of gaining social
acceptance for a project. The use of 3-D landscape modelling and visualisation in community consultation
supports opportunities to gather constructive input that identifies community interests and priorities and
builds awareness and understanding of post-closure landscape aesthetics.
Technical complexity is a characteristic of developing and communicating reclamation and mine closure
plans. The design process involves a large amount of detailed information and studies from a variety of
technical disciplines. Proponents rely on this approach to provide sound and transparent strategies; however,
communicating the technical detail used to define mine closure plans and their implications for community
values such as visual aesthetics to the broad range of stakeholders involved may present a barrier to a clear
understanding of the post-closure environment. Designing for geophysical and hydrological stability is
important, but what does that look like? This can make it difficult to gather information from stakeholders,
and it may result in adverse reactions to a project’s legitimacy (Garcia, 2008). Community consultation in the
closure planning process is an opportunity to explain the objectives and technical scope of a design.
Landscape modelling can support this process by presenting technical information accurately and in a
recognisable graphic format, allowing for a higher degree of understanding and a higher potential for
consent. Advanced landscape modelling capable of producing highly realistic simulation images can provide
easily understood representations of post-closure design aesthetics. Conveying subtle and detailed
landscape information that is easily recognisable to stakeholders who are familiar with these landscapes is
critical to local perceptions of the land, and this has been found to enhance the ability to achieve fair and
meaningful discussion on local values during the consultation process (Lewis and Sheppard, 2006). For the
Baldy Ridge Extension project, the use of geographic visualisation tools to depict landform design and
reclamation has been valuable in developing awareness and involvement from stakeholders and identifying
issues and options.
Teck’s mining engineers at Elkview worked collaboratively with reclamation and landform experts to facilitate
changes to the initial Baldy Ridge waste rock spoil concept design based on community feedback. Landform
design principles were interpreted and applied to mine design data resulting in a revised 3-D landscape model
and simulated images. Available information on the conceptual reclamation design related to vegetation and
land cover components and their prescriptive locations was also incorporated into the landscape model.
Using geographic allocation methods and vegetation composition parameters within the 3-D landscape
modelling environment, specialists added the conceptual reclamation vegetation plan to the new landform
design; this provided a more complete visual representation of Baldy Ridge’s evolution during operation and
post-closure phases and demonstrated the visual effects of progressive reclamation and closure. The changes

focused on developing a final landform that better met the community’s and regulators’ desire for a post-
mining landscape that resembles the natural landscape.
Updated images of the revised mine design were used in posters presented at consultation events in 2015 to
communicate the redesign of the project to the community. Figure 5 illustrates an updated concept design.
This iterative process allowed Teck to demonstrate to the public the changes made to the initial designs.
These visual images form part of the permitting application for the project and are still subject to government
review and approval.
Addressing concerns related to visual aesthetics in the context of Teck’s goal to develop mining areas
transformed by human activity into functioning ecological landscapes provides an opportunity to address
other ecological and social objectives. As part of its Sustainability Strategy for Biodiversity, Teck has made
the goal to achieve a Net Positive Impact (NPI) on biodiversity. In pursuit of this goal, Teck’s closure planning
aims to reestablish ecological landscapes that become self-sustaining over time. To address this need, it must
consider varied aspects of natural systems related to landform design. These include creating microclimates,
varied vegetation communities, and related wildlife habitats; considering soils and the dynamics of local
hydrologic regimes such as surface erosion; and reestablishing natural ecological process. Landscapes are
rarely static but develop over time; hence, incorporating naturally evolving processes into landform design
systems is expected to aid in reestablishing natural systems and a naturalised post-closure landform
aesthetic. Reclamation work at Teck’s operations to date has incorporated trials of machine roughening of
finished terrain surfaces, which encourage microclimate variations that have the potential to retain fines and
moisture and improve plant success. While the roughened surfaces will likely not contribute significantly to
the landform as it is seen from a distance, on-site observations suggest the practice is important for reducing
plant mortality and encouraging the different visual effects that come from natural vegetation patterns and
textures.
Achieving an acceptable natural landform aesthetic in post-closure phases of mine remains a topic of interest.
In the development of the Baldy Ridge reclamation design, the landform aesthetic is addressed by simulating
adjacent natural landforms and vegetation patterns to the extent possible so that the resulting mine-related
landforms contrast minimally with the existing landscape context. While achieving pre-mining conditions is
not possible, the design reflects an approach where form follows function; the efforts to create an acceptable
visual aesthetic reflect the parallel efforts to address multiple technical constraints and ecological objectives.
Understanding post-closure landscape design objectives will allow mine operators to initiate measures during
mining operations and at early stages of progressive reclamation that could potentially result in reduced
maintenance cost and improved long-term landscape performance while satisfying the public’s desire for a
suitable landscape aesthetic.

Miscellaneous
The mine operator may also balance potentially competing land use objectives with other land uses and land
capability interests that may be of value to the community, especially at Elkview where the mine is located
immediately adjacent to Sparwood. Land may be considered both a liability and an asset in the post-closure
context. While current mine closure requirements focus on geotechnically stable landforms, hydrologic
function, and long-term ecological integrity, consideration of other environmental services and land uses that
support local economic diversity (for example, commercial, industrial, agricultural, or recreational uses) are
strong determinants for the local community. Listening and responding to community aspirations to the
extent possible will help build trust and social acceptance of the project.
An emerging aspect in reclamation and mine closure planning is the development of post-closure landforms
that reduce visual impacts caused by mining operations. Community concerns over visual changes to the
local landscape and recent comments from regulators have communicated a desire for post-mining
landscapes to resemble natural land forms. The challenge for mining companies is to design reclamation and
landform environments that succeed in reducing visual impacts from mining operation while meeting existing
environmental requirements, ongoing mine operations requirements, and closure plan objectives.
To support landscape design and aid in communicating predicted visual impacts related to the Baldy Ridge
Extension project, Teck recognised that detailed, data-driven landscape visualisations of operation and post-
closure environments should be created. It used 3-D landscape modelling and visualisations to improve public
understanding and community involvement by illustrating design intentions, communicating technical design
information, and facilitating the identification of interests and priorities for the post-mine reclamation
landscape. Images were developed from a key viewing location near the community of Sparwood for initial
open-house events in 2014; these demonstrated potential visual changes to Baldy Ridge landscape features
over the mining period and illustrated the anticipated reclaimed post-mining landscape. Feedback identifying
specific community concerns related to visual impacts that, along with a design review by Golder’s landscape
architect, informed Teck’s mining engineers’ efforts in redesigning the landform.
Updated visualisations were prepared based on new landscape model data, and these were presented at
subsequent open-house events in 2015. The revised mine design has demonstrated the potential for
reclamation and landform design to broadly support environmental and social responsibility by addressing
aesthetic concerns, providing opportunities for increased ecological landscape performance, and supporting
an enhanced community and regulatory engagement process.
British Columbia (2008) Ministry of Energy, Mines and Petroleum Resources, Mining and Minerals Division, Health, safety and
reclamation code for mines in British Columbia, viewed 15 April 2015, http://www.empr.gov.bc.ca/mining/health
andsafety/documents/hsrc2008.pdf.
British Columbia (2011) BC Stats, 2011 Census Total Population Results, viewed April 2015,
http://www.bcstats.gov.bc.ca/StatisticsBySubject/Census/2011Census/PopulationHousing/CensusSubdivisions.aspx.
Garcia, D.H. (2008) Overview of international mine closure guidelines, in Proceedings 2008 Meeting of the American Institute of
Professional Geologists, Arizona Hydrological Society, and 3rd International Professional Geology Conference, 20–24
September, Flagstaff, Arizona, pp. 113-130.
Government of Western Australia (2011) Department of Mines and Petroleum and Environmental Protection Agency, Guidelines for
preparing mine closure plans, http://www.dmp.wa.gov.au/documents/Mine_Closure%282%29.pdf.
Lewis, J.L. and Sheppard, S.R.J. (2006) Culture and communications: Can landscape visualization improve forest management
consultation with indigenous communities? Landscape and Urban Planning, Vol. 77, pp. 291–313.
MacEachren, A.M. (2004) Geovisualization for knowledge construction and decision support, IEEE Computer Graphics and
Applications, Vol. 24(1), pp.13–17.
Przeczek, L. (2014) Conceptual reclamation plan for the Elkview Operations property at Sparwood, BC, PRYZM Environmental Ltd.


B.H. Cranston Paragon Soil and Environmental Consulting Inc., Canada

L.P. Waterman Paragon Soil and Environmental Consulting Inc., Canada
Regulatory approvals for oil sands mining in Alberta, Canada, require that disturbed land be reclaimed to a
self-sustaining, locally common boreal forest ecosystem; biodiversity is also a key stakeholder concern. Not
all ecosites are readily reclaimed, yet conceptual reclamation plans often make assumptions that discount
this challenge. To this end, we have developed a ranking system that uses measurable values to determine
the level of difficulty to establish target vegetation on upland and transitional reclaimed sites. The ecosite
phase reclamation targets of the Athabasca Oil Sands (Alberta, Canada) were used as a model system.
Six metrics were used to determine the relative difficulty to reclaim each upland and transitional ecosite
phase: time required for target vegetation establishment, level of management required, difficulty associated
with achieving the requisite edaphic (soil moisture and nutrients) regime, sensitivity to changing soil
chemistry, utility of direct soil placement and erosion potential. Each metric was given a weighted value (1
through 4) according to its overall influence on the difficulty of reclaiming each ecosite phase. The weighted
values of metrics deemed difficult, neutral and easy for a particular ecosite phase were multiplied by −1, 0
and +1, respectively. The values for all six metrics were then summed for each ecosite phase, yielding an
overall value for the difficulty to reclaim each phase. Values below −2 were deemed to be difficult, −2 to +2
were deemed neutral and values exceeding +2 were deemed not difficult (“easy”) relative to the innate
challenges of reclaiming anything on the scale of mineable oil sand leases.
This method of quantifying the level of difficulty in establishing target vegetation communities can be applied
to any post-disturbance reclamation project. Its adoption would have broad-scale implications on activities
from reclamation planning to revegetation practices and monitoring techniques.
According to the Environmental Protection and Enhancement Act (EPEA) of Alberta, Canada, oil sands
companies are required to submit a long-term, conceptual planning document (life of mine closure plan) that
extends to the end of mine life. This plan outlines the technical measures and planning programs to be
undertaken for the decommissioning, remediation and reclamation of a mine and plant site, with the
ultimate goal of reclamation certification and return of land to the Crown (Pickard et al., 2013). Each operator
must outline the extent (number of hectares) of reclamation planned for each year of project duration,
allowing for identification of any mismatches between early and late reclamation. Ensuring that all site types,
landforms, soils and regions are included in early reclamation increases learning and allows processes and
practices to be adapted for maximum efficiency in later years. Adaptive management allows information to
be fed back into the planning and design process so that future reclaimed areas will meet the intended
objectives. Years in which more reclamation is planned than can practically be completed may also be
identified.
While life of mine planning may identify pitfalls with respect to reclamation extent, it does not specifically
account for the relative difficulty and required effort to re-establish various vegetation communities. To this
end, we have developed a ranking system that uses measurable values to determine the level of difficulty of
establishing target vegetation on reclaimed sites, using the ecosite phase reclamation targets of the
Athabasca Oil Sands (Alberta, Canada) as a model system. As of 2013, only ~7% of the total area disturbed

Assessing the level of difficulty of vegetation establishment on reclaimed sites B.H. Cranston and L.P. Waterman
within the Athabasca Oil Sands has been reclaimed (Pickard et al. 2013). Approximately 32% is expected to
be reclaimed by 2035, but the majority of reclamation will occur after 2035 (Pickard et al., 2013).
The ecosite phase targets of Athabasca Oil Sands reclamation were used to demonstrate the application of
the difficulty matrix. Ecosites and their associated phases are mappable units dependent on the edaphic
conditions (soil moisture and nutrient levels) at the site, and are not tied directly to landforms or vegetation
communities (Beckingham and Archibald, 1996). For the ecosites under consideration in this paper, moisture
regimes range from xeric to hygric, with very poor to very rich nutrient regimes. Each ecosite phase is
described briefly in Table 1, and examples are depicted in Figure 1.
Moisture Nutrient Key overstory

Ecosite Key understory species
regime regime species
a1 Subxeric–xeric Poor Jack pine Blueberry, bearberry, bog cranberry, cladina
Submesic– Bog cranberry, blueberry, green alder,
b1 Poor Jack pine, aspen
subxeric bearberry, Labrador tea
Submesic– Aspen, white Bearberry, Labrador tea, blueberry, green
b2 Poor
subxeric birch alder
Submesic– Aspen, white Blueberry, bearberry, bog cranberry, prickly
b3 Poor
subxeric spruce rose
Submesic– White spruce, jack Bearberry, blueberry, green alder, white
b4 Poor
subxeric pine spruce, bog cranberry
Jack pine, black Labrador tea, bog cranberry, black spruce,
c1 Mesic–submesic Poor
spruce blueberry
Aspen, balsam Prickly rose, low-bush cranberry, Canada
d1 Mesic Medium
poplar buffaloberry, twin-flower
Aspen, white Low-bush cranberry, prickly rose, twin-
d2 Mesic Medium
spruce flower
White spruce, Twin-flower, low-bush cranberry, balsam fir,
d3 Mesic–subhygric Medium
balsam fir prickly rose
Aspen, balsam Prickly rose, dogwood, low-bush cranberry,
e1 Subhygric Rich
poplar bracted honeysuckle
White spruce, Dogwood, low-bush cranberry, prickly rose,
e2 Subhygric Rich
aspen bracted honeysuckle
White spruce, Low-bush cranberry, willow, prickly rose,
f2 Hygric Rich
white birch dogwood, horsetail
Prickly rose, low-bush cranberry, currant,
f3 Hygric Rich White spruce
horsetail, marsh reed grass
Black spruce, jack Labrador tea, black spruce, bog cranberry,
g1 Subhygric Poor
pine blueberry
Medium– White spruce, Labrador tea, bog cranberry, willow, prickly
h1 Hygric
Rich black spruce rose, twin-flower, horsetail

Miscellaneous
To date there has been limited consideration of the practicability of reclaiming a comparatively full spectrum
of upland and transitional ecosites. The research and technology transfer required to reclaim upland areas is
relatively advanced compared to that for wetlands. Nonetheless, field experience and an examination of

some industry closure plans submitted by area operators suggest that closer examination of upland
reclamation challenges is warranted.
Six metrics were used to determine the relative difficulty to reclaim each upland and transitional ecosite
phase:
 the time required for a particular ecosite phase to establish;
 the level of management required for ecosite phase vegetation to establish and be maintained;
 the difficulty associated with achieving the edaphic (soil moisture and nutrients) regime;
 the sensitivity of each ecosite phase to changing soil chemistry;
 the utility of direct soil placement; and
 the erosion potential.
Each metric was given a weighted value according to its overall influence on the difficulty of reclaiming each
ecosite phase. The weighted values of metrics deemed difficult, neutral and easy for a particular ecosite
phase were multiplied by −1, 0 and +1, respectively. The values for all six metrics were then summed for each
ecosite phase, yielding an overall value for the difficulty to reclaim each phase. Values below −2 were deemed
to be difficult, −2 to +2 were deemed neutral, and values exceeding +2 were deemed not difficult relative to
the innate challenges of reclaiming anything on the scale of mineable oil sand leases. The six metrics are
described below.
The time required for a particular ecosite phase to become established is partly dependent on the
management level needed. The longer it takes for a vegetation community to develop, the more erosion and
invasive species mitigation will be necessary.
In the past, vegetation communities were generally assumed to be established once they reached a “stable
state.” However, a new understanding of community succession following disturbance suggests that
succession rarely leads to a stable, unchanging configuration. This is because unpredictable changes in
weather, hydrology, seasonal changes of light and temperature, and internal mechanisms in populations
continuously affect the structure and function of natural systems (Connel and Slatyer, 1977; Dokulil et al.,
2006). In a review by Connel and Slatyer (1977) no examples of a community of sexually reproducing
individuals were found in which the average species composition had reached steady-state equilibrium. On
this lack of evidence, they concluded that, in general, succession never stops. Therefore, achieving a stable
state is a poor metric for community establishment. To determine the time required for each ecosite phase
to become established, all phases were considered to be successional.
Two factors were considered when determining time to establish: (1) when are the successional trajectories
initiated that will likely lead to the specified community configurations, and (2) at what time in succession is
that configuration typically reached? Most seedlings will have established within 10 years following the initial
disturbance, but it is the changing dominance of species within the community that will determine which
ecosite phase has become established at any particular point in time. Individual growth rates for species
characteristic of target ecosite phases are therefore a critical factor for this metric.
The individual growth rates of target species for each ecosite phase vary significantly; established
communities may not perfectly represent target species assemblages. By taking this variation into account,
it is possible to rank the ecosite phases relative to one another. Some ecosite phases (a1, b1–2, c1) are
characterised by an understory dominated or co-dominated by lichen species. These lichens are slow to
disperse and grow, sometimes taking 50+ years to establish. Wet-rich ecosite phases (e, f, h1), by comparison,
with a predominantly fast-growing, herbaceous understory, should become established relatively quickly. To
quantify these differences, the site index at 50 years is supplied for each ecosite phase.

Miscellaneous
There were several considerations behind the determination of management level for each ecosite phase.
Mainly, the amount of erosion mitigation required, control of undesirable plant invasion, and fertilisation
were evaluated. Weed management requirements were anticipated for all ecosites and were therefore not
directly accounted for in the matrix. The primary methods of erosion control include proactive practices such
as rough soil mounding, coarse woody debris (CWD) placement and the planting of nurse crops (Burger et
al., 2009; Paquin and Brinker, 2011; Naeth and Brown, 2014). Nurse crops are not suggested for ecosite
phases that have low productivity and target vegetation that is slow to establish, such as a1. Nurse crops can
outcompete target vegetation, preventing it from establishing in such cases (Padilla and Pugnaire, 2006).
Those ecosites that are unable to support nurse crops are therefore prone to invasion from weeds and non-
native and non-target plant species, likely requiring moderate to aggressive management. In some cases,
even target species (e.g. aspen) may need to be managed to ensure that they do not begin to dominate the
site. This is not to say that initial ecosite targets will never be adjusted to allow for emergence of unplanned
yet desirable species, but if biodiversity somewhat comparable to pre-development conditions is sought,
some areas will require management to achieve the original targets.
Finally, the need for fertilisation was evaluated. Fertilisation should only be used if necessary; fertiliser type,
timing of application and the amount applied should be carefully considered (Alberta Environment and
Water, 2012). Applying fertiliser directly after soil placement or before planting can encourage the growth of
weeds and other herbaceous plants to the detriment of slow-growing target plants, which can be easily
outcompeted (Preston et al., 1990). Likewise, excess fertilisers not taken up by the targeted plants tend to
encourage the growth of herbaceous competitors (Preston et al., 1990). Fertilisation programs should rather
be based on timely soil and foliar analyses.
Along with establishing the edaphic regime, management level has the highest weighting relative to the other
metrics. The amount of management required to establish and maintain ecosite vegetation communities is
directly related to the difficulty of reclaiming each phase, and thus the cost.
Along with vegetation, the edaphic regime is a key factor determining ecosite phase classification. For this
reason, it has been assigned the highest weighting value of 4, an equivalent weighting to management level.
Some edaphic conditions are known to be difficult to achieve, particularly those of “extreme” upland ecosite
phases, such as the dry-poor (a1), and wet-rich (e, f) phases. The c1 and g1 ecosite phases are also challenging
to recreate; as soil moisture levels rise, it is difficult to maintain the poor to very poor nutrient levels required.
This explains why c1 and g1 ecosites commonly occur in relatively narrow bands in the transition between
upland and lowland, often adjacent to each other. Oil sand mine closure plans commonly target substantially
larger swaths of c1 and g1 ecosites than may be feasible to achieve.
Some plant species (i.e. Labrador Tea and bog cranberry) are tolerant of acidic conditions and can be indictors
of plant communities that are relatively tolerant to changes in soil chemistry. Moreover, ecosite phases that
have very specific, difficult-to-achieve edaphic requirements are predicted to be intolerant to changes in soil
chemistry, much as specialist species are intolerant to change (Dukes and Mooney, 1999). No ecosite phase
was rated as positive for this metric (very tolerant).
This metric was assigned a low weighting, primarily because it is also captured to some degree in the
“difficulty to achieve edaphic regime” metric, which was assigned a higher weighting.
Direct soil placement is widely recognised as a useful tool for reclamation (Bell, 2001), not only because of
the cost-savings associated with moving the soil only once (relative to moving it to stockpiles and placing it

at a later date), but also because it encourages more rapid establishment of target vegetation — assuming
the ecosite at the source is aligned with that of the target location. Avoiding lengthy stockpiling of soils allows
viable source seed and other propagules (root/rhizome segments, etc.) to be retained in the soils (Bell, 2001).
Once placed, these seeds/propagules can germinate and root, allowing target vegetation to establish as
quickly as possible after soil placement. This natural revegetation can then be supplemented by planting as
needed; however, planting requirements will likely be reduced, providing further savings.
Direct soil placement will be particularly effective for ecosite phases with low productivity vegetation, which
is slow to establish and has low dispersive ability. No ecosite phase was predicted to be negatively impacted
by direct soil placement practices. This metric was assigned a moderate weighting value of 2.
Erosion potential includes both wind and water erosion. Wind erosion is a particular issue for dry ecosite
phases, those in which vegetation is slow to establish and those in which nurse crop planting is not
recommended. For these phases, soil is not stabilised and can easily be picked up by the wind.
Water erosion can be a problem for any ecosite, but is most severe before planting. Establishing a nurse crop
has been proven to be an effective method of erosion control and soil stabilisation for those ecosites with
highly productive, herbaceous vegetation (Burger et al., 2009). However, nurse crops have the potential to
attract wildlife, and thus their use must be carefully considered when near active mine areas. Other methods
of erosion control proven effective include rough soil mounding and CWD placement (Paquin and Brinker,
2011; Naeth and Brown, 2014). Both of these methods involve the creation of microtopographic variability,
resulting in microclimatic pockets of favourable growing conditions.
The erosion potential metric was given a moderate weighting because it is also captured, to a degree, in the
management level metric above.
The rating system presented is not intended to suggest that reclamation of any ecosite is easy; rather, the
system provides a relative classification. Careful management of reclamation resources and thoughtful
planning will always be required, regardless of the estimated level of difficulty. The difficulty matrix is
presented in Figure 2.
Ecosite operational constraints

Time to Difficulty to
Sensitivity to Utility of
establish Management achieve Erosion Difficulty
changing direct soil
(site index level edaphic potential to
chemistry placement
[m]) regime reclaim*
Weight 1 4 4 1 2 3
High: CWD for Neutral: High:
50+ years High: very
erosion, no High: xeric tolerant of introduces
slow
Site index: fertilisation, and acidity, lichens,
establishment −10
Lichen, jack pine
Pj = 13.4 ± competition nutrient tolerance to mosses,

of ground
0.3 must be poor alkalinity ericoid
cover
managed unknown mycorrhizae
a1
Strategy: Establish on coarse, rapidly draining soil that is relatively level. Typically found on topographic
high points, but in reclaimed situations erosion risk necessitates more level area targets. Invasion and
competition are high because of slow understory establishment. Any fertilisation will cause target
species to be outcompeted (mosses/lichens in particular). Instead of vegetative erosion control, CWD
and soil mounding is suggested. Lichens and dwarf shrubs are slow to establish. Competition from aspen
will need to be controlled.

Miscellaneous

chemistry placement
Weight 1 4 4 1 2 3
~35 years Neutral: no Low: Neutral:
Blueberry, jack pine, aspen
High:
fertiliser, use develop a tolerant of
Site index: introduces High: slow
CWD for range of some
Pj = 14.3 ± lichens, establishment
0.5
erosion, less moisture acidity,
mosses, of ground 2
competition and tolerance to
Aw = 15.8 ± ericoid cover
b1 management nutrient alkalinity
0.5 mycorrhizae
than in a regimes unknown
Strategy: Establish on coarse, well-draining soil. Invasion and competition are strong, but less so than
for a1. Fertilisation will promote the establishment of herbaceous competitors, outcompeting target
groundcover species. Lichens and dwarf shrubs are slow to establish. Special care and additional
monitoring are suggested if using alkaline soils.
Low:
Blueberry, aspen (white birch)
~35 years Neutral: no Neutral:

develops High:
fertiliser, use tolerant of
Site index: on a range introduces High: slow
CWD for some
Aw = 15.8 ± of lichens, establishment
0.6
erosion, less
moisture
acidity,
mosses, of ground 2
competition tolerance to
Bw = 17.5 ± and ericoid cover
b2 management alkalinity
0.7 nutrient mycorrhizae
than in a unknown
regimes
Strategy: Establish on coarse, well-draining soil. Invasion and competition are strong, but less so than
for a1. Fertilisation will promote the establishment of herbaceous competitors, outcompeting target
groundcover species. Lichens and dwarf shrubs are slow to establish. Special care and additional
monitoring are suggested if using alkaline soils.
Low:
Blueberry, aspen, white spruce
~30 years Neutral: no Neutral:

develops High:
fertiliser, use tolerant of
Site Index: on a range introduces High: slow
CWD for some
Aw = 15.8 ± of lichens, establishment
0.7
erosion, less
moisture
acidity,
mosses, of ground 3
competition tolerance to
Sw = 17.5 ± and ericoid cover
b3 management alkalinity
0.7 nutrient mycorrhizae
than in a unknown
regimes
Strategy: Establish on coarse, fairly well-draining soil, moisture trending towards submesic–mesic.
Vegetation invasion and competition are strong, but less so than for a1. Fertilisation will promote the
establishment of herbaceous competitors, outcompeting target groundcover species. Lichens and dwarf
shrubs are slow to establish. Special care and additional monitoring are suggested if using alkaline soils.
Bluebe
spruce
Neutral: no Low: can Neutral: High:

white
Pine
jack
rry,
High: slow
b4 3
,
~30 years fertiliser, use trend tolerant of introduces

establishment
CWD for toward d some lichens,


chemistry placement
Weight 1 4 4 1 2 3
erosion, less when acidity, mosses, of ground
Site index:
competition moisture tolerance to ericoid cover
Sw = 17.5 ±
management conditions alkalinity mycorrhizae
0.7
than in a permit unknown
Pj = 14.3 ±
0.5
Strategy: Establish on coarse, fairly well-draining soil, moisture trending towards mesic. Invasion and
competition are strong, but less so than for a1. Fertilisation will promote the establishment of
herbaceous competitors, outcompeting target groundcover species. Lichens and dwarf shrubs are slow
to establish. Competition from aspen will need to be controlled. Special care and additional monitoring
are suggested if using alkaline soils.
~30 years Neutral: no High: Neutral:

fertiliser, use mesic High: introduces
jack pine, black spruce
Site index: High: slow

CWD for moisture Sensitive to native
Pj = 14.3 ± establishment
−8
Labrador tea,
erosion, less with low changes in propagules,

0.4 of ground
competition nutrients, soil less
Sb = 11.5 ± cover
c1 management specific conditions important
0.6 than in a soil types than for a/b
Strategy: Requires very specific edaphic regimes and is difficult to reclaim. Established on medium-fine
soils with poor nutrient conditions. Fertilisation will alter the edaphic conditions and push the
vegetation towards d. Soil mounding and CWD are suggested as alternative erosion control measures.
More likely to occur as unplanned inclusions.
Neutral: can Neutral:

Low-bush cranberry, aspen
~30 years Low: fertiliser Low: mesic Neutral:

tolerate a introduces
Site index: ok once moisture ground cover
range of native
Aw = 18.2 ± established, and quicker to
0.2 competition medium-
conditions propagules,
establish, 8
(submesic– less
Pb = 17.3 ± management to-rich more
d1 subhygric, important
0.6 not required nutrients herbaceous
poor–rich) than for a/b
Strategy: Can establish on a range of soil types (medium-to-fine texture, or peat-mineral mix) and
edaphic conditions. Moderately productive. Fertiliser can be added once woody species have
established. Soil mounding and CWD can be employed as additional erosion control methods.
Low: fertiliser Low: mesic Neutral:

white spruce
Neutral: can Neutral:

cranberry,
~30 years
Low-bush
ok once moisture ground cover

aspen,
Site index: established, and quicker to
d2
Aw = 18.2 ± competition medium-
range of native
establish, 8
0.2 management to-rich more
(submesic– less
not required nutrients herbaceous

Miscellaneous

chemistry placement
Weight 1 4 4 1 2 3
Sw = 16.8 ± subhygric, important
0.2 poor–rich) than for a/b
~30 years Neutral: can Neutral:

Low: fertiliser Low: mesic Neutral:
Low-bush cranberry,
Site index: ok once moisture ground cover

range of native
established, and quicker to
white spruce
Sw = 16.8 ±
0.2 competition medium-
establish, 8
(submesic– less
d3 Fb = 14 ± management to-rich more
subhygric, important
1.1 not required nutrients herbaceous
poor–rich) than for a/b
Dogwood balsam, poplar, aspen
High:
~25 years High: may Neutral:
specific,
require introduces Low: grasses
Site index: challengin Sensitive to
fertilisation native and forbs
Aw = 21.4 g soil changes in
± 0.4
program to
moisture soil
propagules, characteristic, −5
obtain less shrubs quick
Pb = 19.7 ± conditions conditions
e1 edaphic important to establish
0.6 (subhygric,
regime than for a/b
rich)
Strategy: Can only establish under very specific edaphic conditions. Requires subhygric soils rich in
nutrients. Understory is competitive and quick to establish. A nurse cover can be used for erosion
control if necessary. Soil mounding and CWD are also suggested for erosion control. This ecosite prefers
low topographic positions with imperfectly drained soils.
Dogwood
, balsam,
~25 years High: may High: Sensitive to Neutral:

poplar,
spruce
Low: grasses
white
require specific, changes in introduces

e2 Site index: fertilisation challengin soil native
and forbs −5
Sw = 17.8 ± characteristic,
program to g soil conditions propagules,


chemistry placement
Weight 1 4 4 1 2 3
0.3 obtain moisture less shrubs quick
Aw = 21.4 ± edaphic conditions important to establish
0.4 regime (subhygric, than for a/b
rich)
Strategy: Can only establish under very specific edaphic conditions. Requires subhygric soils rich in
nutrients. Understory is competitive and quick to establish. A nurse cover crop can be used for erosion
control if necessary. Soil mounding and CWD are also suggested for erosion control. This ecosite prefers
low topographic positions with imperfectly drained soils.
High:
Horsetail, balsam, poplar,
~25 years High: may Neutral:

specific,
require introduces Low: wet soils,
Site index: challengin Sensitive to
white spruce
fertilisation native blanket of

Pb = 17.8 ± g soil changes in
1.8
program to
moisture soil
propagules, horsetails and −5
f2 obtain less herbaceous
Sw = 16.4 ± conditions conditions
edaphic important cover
0.3 (subhygric,
regime than for a/b
rich)
Strategy: Flooding and seepage enhances nutrient supply in f ecosites. Additional fertilisation may be
required. Once established, horsetails will provide some erosion control. CWD is also suggested
High:
High: may Neutral:
Horsetail, white spruce
specific,
~25 years require introduces Low: wet soils,
challengin Sensitive to
fertilisation native blanket of
Site index: g soil changes in
Sw = 16.4 ±
program to
moisture soil
propagules, horsetails and −5
f3 obtain less herbaceous
0.3 conditions conditions
edaphic important cover
(subhygric,
regime than for a/b
rich)
Strategy: Flooding and seepage enhances nutrient supply in f ecosites. Additional fertilisation may be
required. Once established, horsetails will provide some erosion control. CWD is also suggested
High:
spruce, jack pine
~35 years
subhygric black
Neutral: no Neutral: few

Labrador tea,
specific,
Site index: fertiliser, use Sensitive to High: forbs and
challengin
Sb = 9.9 ± CWD for changes in introduces grasses, wet
g1
0.7 erosion, less
g soil
soil native soil reduces −4
conditions
Pj = 11.7 ± competition conditions propagules erosion
(subhygric,
0.4 than in a, c potential
poor)

Miscellaneous

chemistry placement
Weight 1 4 4 1 2 3
Strategy: Difficult to reproduce because of the specific soil types (finer textured, poorly drained)
required. High risk of site modification if operations are conducted on unfrozen soil because of the high
soil water content. Operations should be limited to the winter season. CWD and mounding can be used
to deter vehicle and equipment traffic.
Neutral: may Neutral:
white spruce, black spruce
~30 years Neutral:

Labrador tea, horsetail,
require introduces Neutral:

Site index: moisture Neutral:
fertiliser native blanket of
Sw = 12.9 ± regime vegetation
h1
1.0
program,
easier to tolerant to
propagules, horsetails, 0
erosion less shrubs slower
Sb = 9.5 ± recreate some acidity
control less important to establish
0.7 than g1
important than for a/b
Strategy: Limited to topographic depressions with finer, poorly drained soils. Vegetation is more
difficult to recreate than in g1.
*less than −2 = high; −2 to +2 = neutral; more than +2 = low
This method of quantifying the level of difficulty of establishing target vegetation communities can be applied
to any post-disturbance reclamation project. Its implementation may have broad-scale implications for
activities from reclamation planning to revegetation practices and the monitoring techniques required. Using
the matrix, we have identified trends in the extent (ha) of difficult ecosite phase revegetation over the course
of 50 years of planned reclamation at an oil sands project. A distinct peak of “difficult” revegetation activities
was identified during the final years of reclamation and monitoring. These activities can now be rescheduled
when practicable to take full advantage of monitoring and adaptive management during earlier stages of
reclamation. Research can be adjusted as needed to focus on developing successful conditions for reclaiming
the more challenging ecosites. Moreover, when applied to planned or placed soil prescriptions, specific
limitations to vegetation establishment may be identified and potentially mediated.
Alberta Environment and Water (2012) Best management practices for conservation of reclamation materials in the mineable oil
sands region of Alberta, prepared by MacKenzie, D. for the Terrestrial Subgroup, Best Management Practices Task Group of
the Reclamation Working Group of the Cumulative Environmental Management Association, 9 March 2011, Fort McMurray,
Canada.
Beckingham, J.D. and Archibald, J.H. (1996) Field guide to ecosites of northern Alberta, Natural Resources Canada, Canadian Forest
Service, Northwest Region, Edmonton, Canada.
Bell, L.C. (2001) Establishment of native ecosystems after mining: Australian experience across diverse biogeographic zones,
Ecological Engineering, Vol. 17, pp. 179–186.
Burger, J., Davis, V., Franklin, J., Zipper, C., Skousen, J., Barton, C. and Angel, P. (2009) Tree-compatible ground covers for reforestation
and erosion control, Forest Reclamation Advisory No. 6., pp. 1–6.
Connel J.H. and Slatyer, R.O. (1977) Mechanisms of succession in natural communities and their role in community stability and
organization, The American Naturalist, Vol. 11, pp. 1119–1144.
Dokulil M.T., Donabaum, K. and Pall, K. (2006) Alternative stable states in floodplain ecosystems, Ecohydrology and Hydrobiology,
Vol. 6, pp. 37–42.

Dukes, J.S. and Mooney, H.A. (1999) Does global change increase the success of biological invaders? Trends in Ecology & Evolution,
Vol. 14, pp. 135–139.
Naeth, M.A. and Brown, R.L. (2014) Woody debris amendment enhances reclamation after oil sands mining in Alberta, Canada,
Padilla, F.M. and Pugnaire, F.I. (2006) The role of nurse plants in the restoration of degraded environments, Frontiers in Ecology and
the Environment, Vol. 4, pp. 196–202.
Paquin, L.D. and Brinker, C. (2011) Soil salvage and placement: Breaking new ground at Teck’s Cheviot open pit coal mine. In
Proceedings of the 2011 Mine Closure Conference, Lake Louise, AB, September 18-21, 2011.
Pickard, D., Hall, A., Murray, C., Frid, L., Schwarz, C. and Ochoski, N. (2013) Long-term plot network: Effectiveness monitoring program,
prepared for the Plot Network Task Group, Terrestrial Subgroup, Reclamation Working Group, Cumulative Environmental
Management Association (CEMA), Fort McMurray, Canada.
Preston, C.D., Marshall, V.G., McCullough, K. and Mead, D.J. (1990) Fate of 15N-labelled fertilizer applied on snow at two forest sites
in British Columbia, Canadian Journal of Forest Research, Vol. 20, pp. 1583–1592.

T.A. Gray Tetra Tech Inc., USA

R.E. Gray DiGioia, Gray and Associates, USA
W.P. Balaz Jr. Tetra Tech Inc., USA
The second edition of the Society for Mining, Metallurgy and Exploration Inc.’s (SME) Mining Engineering
Handbook was published in 1992. Chapter 8.7, entitled “Mine Closure, Sealing and Abandonment” covered
regulatory requirements (in the United States); mine closure practices for surface mines, underground mines
and mine facility removal; sealing of underground mine openings; and abandonment, including post-mining
liabilities. The authors of this paper discuss how the practices described in 1992 have changed over the past
23 years
A brief summary of the 1992 chapter is provided for historical context prior to presenting the issues with mine
closure practices identified by regulatory agencies. Innovative mine closure practices and design concepts are
presented to illustrate new approaches to addressing mine closure issues. Central tenets of the design
approach include understanding the influence of site conditions on the design and having sufficient data to
characterise and quantify the site materials used for mine closure.
The second edition of the Society for Mining, Metallurgy and Exploration Inc.’s (SME) Mining Engineering
Handbook was published in 1992. Chapter 8.7, entitled “Mine Closure, Sealing and Abandonment” covered
regulatory requirements (in the United States); mine closure practices for surface mines, underground mines
and mine facility removal; sealing of underground mine openings; and abandonment, including post-mining
liabilities. Several practices have changed the design of mine closure plans since the handbook (Gray and
Gray, 1992) was written in 1992. This paper discusses many of the important changed practices and includes
several novel design approaches. The authors’ intention, however, is not to cover all closure, sealing and
abandonment issues in any detail. Detailed design methods can be obtained by reviewing the references.
This section provides a summary of Chapter 8.7, “Mine Closure, Sealing and Abandonment” in the 1992 SME
Mining Engineering Handbook (Gray and Gray, 1992).
When a mine permanently stops operating owing to economic conditions, the depletion of reserves or any
other reason, the following activities generally occur:
1. Underground mine openings are sealed.
2. Surface facilities are removed.
3. Surface mines and the surface areas of underground mines are reclaimed.
Only after these activities are completed can the mining company abandon the site. However, even after
the mine is abandoned, specific post-mining liabilities rest with the mining company.
The following definitions are used in Chapter 8.7:

Update on mine closure, sealing and abandonment practices T.A. Gray, R.E. Gray and W.P. Balaz, Jr.
 Closure: the act of closing or the condition of being closed, such as the closing of a mine. There are
different degrees of closure: permanent, temporary and semi-permanent.
 Sealing: the securing of mine entries, drifts, adits, slopes, shafts and boreholes with suitable
materials to protect against fires, gas and water emissions and for the safety of the public
(Foreman, 1971; Thrush, 1968).
 Abandonment: the act of abandoning and relinquishment of a mining claim or intention to mine;
a voluntary surrender of the claim or mine to the next party.
In the United States, federal, state and local governments have issued regulations that must be followed
during the closure of a mine. The federal agencies that regulate mine closure activities are the Mine Safety
and Health Administration (MSHA), Environmental Protection Agency (EPA), Office of Surface Mining
Reclamation and Enforcement (OSMRE), US Geological Survey (USGS), Bureau of Land Management (BLM),
US Forest Service (USFS), National Park Service (NPS) and Bureau of Indian Affairs (BIA). MSHA regulations
are applicable for health and safety standards and apply to all coal and non-coal mines. EPA’s water
discharge effluent limitations apply also to all coal and non-coal mines. OSMRE reclamation regulations
apply only to non-anthracite coal mines.
Individual states have different rules and regulations that must be followed during the closure of mines.
Local government may enact rules covering topics such as (1) zoning (which could affect the selected post-
mining land uses) and (2) land value assessment (for taxation). Chapter 8.7 contains a brief synopsis of the
applicable federal regulations in Tables 8.7.1, 8.7.2 and 8.7.3.
The closure of all new surface mines must be planned prior to the beginning of operation. A reclamation
plan must be included with all new permit submittals. This planning is also important in evaluating the
economic justification for the mining operation. A substantial amount of money can be saved if the most
appropriate reclamation or closure procedures are thoroughly considered and carried out as planned.
Reclamation planning and closure methods vary among types of surface mines. Chapter 8.7 provides
information on closure of open pit mines, quarries, surface coal mines and placer mines.
In accordance with mine reclamation regulations, all associated facilities must be removed unless they serve
a useful purpose in the post-mining land use. These facilities include buildings, material handling systems
(conveyors, rail lines, transfer stations, storage bins, docks, etc.), electrical lines, transformers, substations,
pipelines, roadways, drainage ponds and drainage channels. Waste disposal areas must be reclaimed, and
any hazardous material must be removed from the site and disposed of at a hazardous waste disposal
facility. Chapter 8.7 contains a sample demolition specification for removal of a mine facility; a waste disposal
mine, utilities and roadways, drainage facilities and removal of hazardous materials are also discussed.
Prior to the closure of a mine, the operator should decide whether some equipment (parts, supplies, etc.)
could be economically removed from the mine and sold or used elsewhere by the operator. An inventory
of available equipment and materials should be made. A hydrology analysis is required to determine if a
hydraulic head will be placed on the mine seal after closure.
During the closure of an underground mine, the shutdown of the mine’s ventilation system must be well
planned, particularly in gassy mines with water conditions. The ventilation shutdown plan needs to be

Miscellaneous
reviewed on a mine-by-mine basis. The engineer must plan for the systematic shutdown of the system to
keep all of the active work areas ventilated and safe from harmful or explosive gases
Revegetation of the surface areas of the mine must be accomplished in accordance with the mine’s permit.
Mining regulations now require sealing to be performed during closure. According to Utah’s shaft
abandonment guidelines (Utah Department of Natural Resources, 1987), the following parameters should
be considered in sealing shafts.
1. Eliminate any danger to the health and safety of the general public.
2. Control release of hazardous, acid/toxic-forming materials or gases to the atmosphere.
3. Control the movement of underground water or hydrologic communication.
Only after the mine has been sealed and reclaimed can the mining company abandon the site. Items that
must be considered prior to the mining company’s legally abandoning the site include post-mining land use,
maintenance of the site, recovery of reclamation bond and post-mining liability.
While the regulations have not changed significantly since 1992, regulatory agencies have observed
reoccurring issues that affect mine closure success. Understanding the root causes of the issues provides an
opportunity to identify potential mitigations to prevent future issues. The authors have identified reoccurring
issues associated with mine closure based on phone conversations (held during January 2015) with personnel
from the Colorado Division of Reclamation, Mining and Safety (DRMS) and the Utah Division of Oil, Gas and
Mining (DOGM); these are discussed below.
Mine opening backfill failures have occurred frequently. Subsidence depressions or openings created by
surface water piping into underground openings via cracks or failure of the mine opening roof appear to be
the primary cause of these failures. Engineering design improvements are needed.
Post-mining drainage is still found to contain constituents above applicable water quality standards and
create impacts on the environment that require treatment, sometimes perpetual. This has continued to occur
frequently, particularly in the area of acid, highly saline and high metal (Zn, Fe, etc.) containing rocks. Typical
discharge sources are mine openings, waste rock and tailings.
Conflicts between different regulatory agencies have caused uncertainties to designers, particular regarding
stormwater and stream water quality. As an example, state permitting programs regulate sediment control
and storm water drainage under the Surface Mining Control and Reclamation Act of 1977 (SMCRA), while
departments of environmental quality or public health and environment regulate the National Pollutant
Discharge Elimination System (NPDES) permits and stream water quality. Reviewing the effects of differing
regulatory agency criteria on mine closure design success is beyond this paper’s intent
The closure of re-mined abandoned mine land (AML) sites can trigger the perpetual treatment liability of a
contaminated point source and hinder or eliminate mine development, even if an overall improvement to
the environment was expected. Some states have re-mining rules that permit this practice by documenting
pre-existing contaminated conditions and comparing them to post-mining conditions.
Post-closure failures of geotechnical and hydrologic designs may require performance-based rather than
prescriptive design standards, particularly regarding slope and drainage channel failures. Streambeds and

underlying bedrock that have been disturbed by mining require stringent engineering design incorporating
both hydrologic and geomorphologic approaches.
Failure of reclamation bonds to cover post-mining liabilities has continued. Bond forfeiture has frequently
occurred with small mine operators that have inadequate legacy reclamation bonds and/or insufficient
operating funds to sustain operations. Another factor resulting in bond forfeitures is mining low-grade ore
deposits far from markets or mills/smelters.
Establishing vegetation on disturbed areas can also be difficult, especially in desert environments, and has
been found to be a recurring problem. The primary issues remain low rainfall, poor soil quality, inadequate
growth medium, failure to treat noxious weeds and harsh weather. Improvements to design standards are
needed.
This section presents updated mine closure practices that have been successful and addresses many of the
issues discussed in the previous section.
Underground mines require access to the surface using adits, slopes, stopes, shafts or a combination of these.
The following definitions are used for this paper.
 Adit: an entrance to an underground mine that is horizontal or near horizontal, by which the mine
can be entered, drained of water and ventilated, and minerals extracted.
 Slope: an entry driven at an angle from the surface to the target mining horizon, such as a coal
seam, trona or potash bed or ore body level. Slope angles can be either down (decline) or up
(incline) from horizontal.
 Stope: the surface expression of an underground mining method that extracts ore and leaves
openings underground that when close to the surface allow ground to cave into openings. Stopes
can be long and deep with unstable ground near the surface.
 Shaft: a long, narrow, typically vertical hole that gives access or provides ventilation to a mine.
The Colorado Division of Minerals and Geology uses three approaches for safeguarding shafts, stopes and
adits. They design a barrier to keep people away from the mine, seals to prevent people from entering the
mine and plugs to eliminate the hazard (Colorado Division of Minerals and Geology, 2002).
Updates to mine opening sealing practices include practices to minimise or eliminate settling or crack
formation around mine openings, improved designs of hydraulic seals to limit post-mine discharges, the use
of bat gates to provide habitat and the use of polyurethane foam (PUF) for a plug.
Mine opened recently have typically been designed with the goal of having no polluted water discharge from
the mine after permanent closure. If this can be technically accomplished, the mining company is able to
abandon the mine site without any long-term treatment liability. In some instances, hydraulic mine seals
could be used to contain water within the mine. If all discharges cannot be eliminated or the pollutants
mitigated, then treatment becomes the only option.
In most jurisdictions, regulatory agencies are imposing more stringent effluent standards. The EPA currently
imposes daily load limits for various metals and other pollutants based on studies that determine the total
maximum daily loads (TMDLs) of receiving stream segments. The TMDL is a calculation of the maximum
amount of a pollutant a water body can receive and still safely meet water quality standards under Section
303(d) of the Clean Water Act (US EPA, 2014). In the past, an effluent standard for the discharge quality was
used as the standard for an acceptable discharge. This standard has changed treatment practices in several

Miscellaneous
areas, particularly coal mining regions, to include treatment for high total dissolved solids, selenium and
other metals. This has necessitated the use of additional active treatment methods such as reverse osmosis,
nanofiltration (ultrafiltration, microfiltration) and, in some cases, crystallisation and evaporation to meet
discharge standards. Those in need of an active treatment acid neutralisation system can still refer to the
EPA’s Design Manual for Neutralisation of Acid Mine Water, published in 1983 (US EPA, 1983). For newer
treatment technologies refer to the EPA’s Reference Guide for Treatment Technologies for Mine-Influenced
Water (US EPA, 2014).
Passive treatment technology is another technology that has developed recently. Passive treatment of mine-
influenced water uses chemical and biological processes that decrease metal concentration and neutralise
acidity (Hedin, et. al., 1994). According to Ford (2003) passive treatment systems differ from active treatment
systems since they do not require electrical power, mechanical equipment, hazardous chemicals, buildings
or daily operation and maintenance. However, passive treatment is not appropriate or suitable in all cases.
Hedin (1994) includes a flowchart to help determine whether a mine discharge can be passively treated. Very
low pH, high acidity, and high metals or sulphate concentrations limit the effectiveness of passive treatment.
Hedin et al.’s (1994) report also includes sizing criteria for use in designing passive treatment systems.
Several mining companies have begun utilising water from their closed mines. CONSOL Energy, Inc. recently
announced that it had launched a water division to reuse treated mine water and pursue its use as a water
source for industrial consumption (Bloom, 2012). As CONSOL is also an oil and gas exploration and production
company, CONSOL Energy has used its mine water for hydraulic fracturing of nearby gas wells. According to
Rassenfoss (2013), Seneca Resources has also used abandoned mine drainage for gas well hydraulic
fracturing. The Pennsylvania Department of Environmental Protection (PADEP) created a white paper in 2012
with the purpose of promoting mine water use for natural gas extraction (PADEP, 2012). In addition to use
of mine waters for natural gas extraction, Mike Korb of PADEP reported the successful use of mine pool water
in a geothermal energy system to heat the campus of Mary Wood University in Scranton, Pennsylvania, USA
(Korb, 2012). In South Africa, Anglo American received a World Coal Association award for environmental
excellence in 2013 for its eMalahleni water reclamation plant. The plant treats 25–30 million litres of water
per day to supply water for both its mining operation and the eMalahleni community, meeting 12% of the
city’s water needs (World Coal Organization, 2014).
Mines abandoned prior to 1977, when the SMCRA was approved, do not have the same long-term liability
issues of those mines closed after the enactment of SMCRA. Many of those abandoned mines still contain
valuable resources. Some of the coal mining states in the USA have recognised the value of having these
abandoned mines re-mined in a manner that improves their environmental condition. PADEP prepared a
reclamation and re-mining incentives report in 2013 to report on incentives to restore previously un-
reclaimed mines (PADEP, 2013). Between 1991 and the end of 2013, PADEP issued 253 contracts to reclaim
lands with an approximate reclamation value of US$ 16.9 million. These mining and reclamation contracts
were issued at no cost to Pennsylvania residents. In addition, Pennsylvania has a permitting program through
which documentation of pre-existing discharges in the permitting process can be used to limit post-mining
discharge treatment liability after re-mining.
Addressing geotechnical and hydrological design standards is a critical aspect of closure plans for either an
existing or proposed waste rock storage area (WRSA). An existing WRSA means a facility not yet meeting final
reclamation guidelines at an operating or abandoned mine. This section discusses waste rock management
guidelines to address these issues. Design of a WRSA requires coordination with mine planning engineers and
environmental staff to obtain the necessary data to select a site and design the WRSA, along with a good
understanding of applicable guidelines and regulations. The flowchart presented in Figure 1 is based on a
waste rock design guide prepared for a confidential client in Western Australia in response to the new mining
regulations enacted in 2012. The design guide approach focuses on understanding site conditions to develop

sustainable, natural looking, stable and reclaimed landforms and provides an opportunity to correct some of
the issues observed by regulators as described in Section 3.
The WRSA design typically defines:
 final rehabilitated slopes, including any benches and diversion drains required;
 WRDA storage capacity and footprint area for reclaimed slopes;
 growth medium volumes required to reclaim the WRSA;
 any impacts on existing drainage patterns;
 any impacts on existing or proposed infrastructure and pits;
 proximity to tenement or lease boundaries;
 proximity to other sensitive receptors; and
 proximity to heritage and ethnographic sites.
A critical aspect of design is an understanding of site conditions, including constraints, life of mine plan,
annual waste rock volumes and waste rock characteristics. To the extent practical, the design should work
with the site conditions to blend into existing ground to minimise unnatural looking waste piles. Using the
final reclaimed slope design to define the WRDA footprint provides an opportunity to salvage growth medium
material under the WRDA and to develop a stacking plan using bench widths and intermediate slopes
between benches to minimise the rough grading required to meet the final reclaimed slope configuration.
Waste rock characteristics are used to develop a stacking plan placing undesirable waste toward the centre
of the placement area to minimise potential environmental impacts.
Drainage control features can be used to retain water for wildlife or other post-mining uses or to remove
surface runoff from the WRDA. Typically, regulations define a design storm for detention ponds and drains.
Depending on location, high-intensity, short duration storms generate higher than design storm flows that
can cause failures of drains. Hydrologic models should be developed to define these high-intensity, short
duration storm flows. These flows can be used to verify the robustness of drain design based on regulatory
design storms. Drain design should be revised as needed to provide additional capacity, thereby preventing
future drain failure and reducing long-term maintenance costs.
A WRSA design flow chart illustrates the design process.
A new edition of the US MSHA Engineering and Design Manual for Coal Refuse Disposal Facilities
(D’Appolonia, 2009) was published in 2009 and recommends that all coal refuse disposal plans include
abandonment and post-mining land use requirements. These plans should incorporate locating soil and
topsoil stockpiles near the refuse site and on stable ground for use during reclamation. Final reclamation
designs should be prepared so that impoundments are eliminated, surface drainage re-established, coal
refuse stabilised and settlement of the coal refuse over time accounted for.
Many regulatory agencies provide reclamation guidance and design manuals that are useful for reclamation
design and planning. Two such sources are the Utah DOGM’s Practical Guide to Reclamation in Utah (Wright,
2003) and the Colorado Division of Reclamation, Mining & Safety (DRMS’s), formerly known as Division of
Minerals and Geology Best Practices in Abandoned Mine Land Reclamation: The Remediation of Past Mining
Activities (Colorado Division of Minerals and Geology, 2002). Reclamation practices in cold climates can be
found in Mine Site Reclamation Guidelines for the Northwest Territories, published by Indian and Northern
Affairs Canada (INAC) in 2007 (INAC, 2007).

Miscellaneous
Adit and slope sealing plans need to be based on a good understanding of the site conditions, regulatory
concerns regarding post-closure surface expression and the identification of the root cause(s) of mine
opening closure failures. Innovative or out-of-the-box thinking is sometimes required to develop a sealing
plan that mitigates the root causes of closure failure. The authors provide two examples to illustrate these
closure practices. The first example is an Illinois coal mine closure performed in 1994 that involved the sealing
of two vertical shafts and a slope. Historically, slopes with arch support have experienced crack formation
behind surface bulkheads, and shafts have experienced subsidence around the sealed opening. The slope
had steel arches on 1.5 m centres, in addition to roof bolts for roof support. The final sealing plan, illustrated
schematically in Figure 2, involved excavating the slope and removing the steel arches to the point that a 9.1
m (30 feet) brow was created above the slope to surface. Waste rock on site was used to backfill the slope
opening, creating a bulkhead. The trench excavated for the arch removal was backfilled using compacted
waste rock and excavated material to approximate finish grade. An additional 0.3 m of fill, topsoil and finer
waste rock was mounded over the excavated slope and brow area in case settlement occurred during the
bond release period.

The shafts, each 46 m deep, were backfilled and sealed using a layered design by dozing large coarse material
about one metre in diameter and broken concrete on the bottom of the shaft and decreasing the size of the
backfill in layers to within two metres of the surface. The last two metres of fill comprised 1.5 m of compacted
backfill material and 0.5 m of topsoil. An additional 1.5 m of topsoil was left over each shaft in case settlement
occurred during the bond release period. A field change occurred during the placement of the first backfill
lift. A combination of fine refuse, gravel and water was used to hydraulically fill in the voids between the
large rocks and concrete before finishing the shaft backfilling.
Final bond release was obtained in 2000, approximately five years after reclamation, when the land was being
farmed for corn production. No settlement was observed prior to the land being returned to agricultural use.
The authors contacted the Illinois Department of Mines (IDM) for this paper in January 2015. The abandoned
mine site was checked using aerial photographs on the IDM computers, and no sign of settlement was
observed. The outline of one shaft could be seen, but no settlement was noticeable.
A second example involved sealing two adits on a steeply pitching 20%, coal seam in Colorado. One adit had
been used for miner and material access, and the other had been used for ventilation. The portal seal plan
obtained from Colorado Division of Reclamation Mining and Safety Laserfiche® WebLink™ , illustrated in
Figure 3, comprised two cinder block seals separated by approximately 7.6 m and covered with polyurethane
spray foam around the perimeter of the seal. The outby seal was located approximately 7.6 m inby the portal.
The space between the seals was filled with available incombustible material. Waste material was placed in
the entry from the outby seal to the mouth of the adit using permissible diesel scoops. Backfill was pushed
up to the roof to minimise voids. The backfill at the portal was covered with topsoil and seeded. A 100 mm
steel drainpipe was extended from the inby side of the inby seal to the surface.

Miscellaneous
The success of these closures was attributable to the design going beyond the prescriptive regulation of
simply sealing the opening to eliminate the cause of seal failures through implementation of the following
elements:
 excavating and removing the steel arches and backfilling the trench with compacted fill;
 using fine backfill and water to hydraulically fill the void spaces within the large rock and concrete
layer at the bottom of the shafts;
 placing seals far enough inby the adit portal that cover was deep and provided ability to push
backfill against the roof as the portal entry was backfilled.
Waters can be safely impounded behind a hydraulic mine seal if the right conditions exist and proper design
practices are used. This approach can impound water within a mine and thus eliminate any post-mining
discharge. Mine seals were designed and installed in the mines surrounding Lake Arthur at Moraine State
Park near Butler, Pennsylvania, USA, in the 1960s (Foreman, Bullers and Hong, 1969) during construction of
the lake. They have successfully reduced the influence of mining-impacted water on the water quality of the
lake. The lake has been a thriving fishery for more than 40 years. Kirkwood and Wu (1995) authored an
excellent paper, “Technical consideration for the design and construction of mine seals to withstand
hydraulic head in underground mines,” emphasising evaluation of the strata surrounding the seal, as that has
caused failures from the deterioration of the strata over time. In addition to hydraulic mine seals designed
to contain water, mine seals to withstand methane explosions are required to comply with MSHA’s final rule
on sealing of abandoned areas (MSHA, 2008).
The decision to gate a mine opening depends on many factors and should be made in consultation with local
state and federal agencies, landowners, knowledgeable bat gate design organisations or personnel and legal
counsel. Experience has shown that a bad design or installation can cause harm to the species and/or habitat
being protected. Factors that contribute to constructing a gate include but may not be limited to:
1. The need to protect cultural remains;
2. Presence of rare or endangered species;
3. A cave or mine threatened by vandals, looters and trespassers;

4. Gate design compliant with current industry standards such as those from the American Cave
Conservation Association (ACCA) and/or Bat Conservation International, Inc. (BCI) (information on
gate design is available in the BCI report “Managing Abandoned Mines for Bats” [Sherwin et al.,
2009]);
5. No vigilant owner or manager living nearby;
6. Controls such as road gates, signs, surveillance, and so on failing to provide protection;
7. Experienced gate contractors/builders being available; and
8. A post-construction inspection and maintenance plan being in place.
The gate should be placed in an area that does not restrict airflow, which can cause changes in the
temperature, pressure and humidity levels deep into the mine. These changes can have consequences for
the ecosystem. If bats are present, gate placement should not impede bat flight.
Gates are typically constructed of metal. The basic design is a flat grid of bars vertically placed across the cave
or mine passage. Bars should be spaced to allow access of small bats and mammals but not allow human
entry. The bars are typically 100 mm angle iron with the apex oriented up. Bars are oriented horizontally with
vertical supports placed widely but not exceeding 4.6 m. Treated wood can sometimes be used in lieu of
angle iron. This basic gate design is widely used even when bats are not present. Figure 4 depicts a typical
bat gate.
Other gate designs include a half gate or fly-over gate, basic gate with window, chute gate and cupola or cage
gate. The cupola gate is used to protect vertical openings, while the other gates are used on horizontal
openings.
Construction should be scheduled during normal construction seasons, avoiding winter and rainy seasons to
the extent practical, and should consider species that inhabit the mine and be performed by experienced
contractors.
The mine or cave environment should be monitored for temperature, pressure, humidity and other
microclimate changes and biological inventories both before and after gate construction to create a yardstick
to measure gate effects on mine or cave environment and species. It is especially important if bats are present
to note bat acceptance of the gate and any behavioural changes, positive or negative, to the gate.
The Colorado Division of Minerals of Geology uses PUF plugs as a technique to plug a mine opening. They
place a column of PUF several feet down into a vertical opening followed by placing mine waste or common
fill material on top up to the level of the surrounding grade. The PUF plug is constructed by mixing two liquid
components that immediately expand into a solid mass (Colorado Division of Minerals and Geology, 2002).
PUF is inert when mixed and can be installed in situations with poor access. The disadvantages include

Miscellaneous
potential for vandalism if exposed. The unmixed chemicals are toxic, so exposure to fumes must be prevented
during construction. Exposed PUF supports combustion and degrades in sunlight, so a minimum of 600 mm
of cover is required. Proper installation procedures are critical to the success of this plug type.
Post-mining land uses are defined in accordance with regulations and stakeholder input. A stakeholder simply
defined is someone who can affect or is affected by a project. For projects in the United States, stakeholders
typically include federal government agencies – OSMRE, the US Fish and Wildlife Service (USFWS), USFS and
BLM. State (provincial in Canada) agencies; environmental groups; landowners; local, county and municipal
governments; and Indigenous peoples are also important stakeholders. Stakeholder input, particularly from
landowners and local governments, provides an opportunity to develop a more beneficial post-mining land
use concept. The authors experienced this when reclaiming a small surface mine in Colorado. Instead of
restoring a surface drainage, the reclamation plan was revised to create a long embankment that detained
water, creating grass pasture for the landowner’s cattle instead of sagebrush and sparse grass. The
embankment also prevented the county road from flooding.
Projects that are on Indigenous lands present cultural, religious, social and other challenges to projects.
Projects should establish a working relationship with affected peoples early in the project. Guidelines have
been developed to aid in establishing sustainable relationships with Aboriginal people. Two such guidelines
are the Aboriginal Engagement Guidebook (Association for Mineral Exploration British Columbia, 2014) and
Best Practices for Consultation and Accommodation (Meyers Norris Penny LLP, 2009).
Mine operators are required by state law and, where mining is conducted on federal land, federal law to
provide estimates of the cost of reclaiming the mine site and a financial guarantee to the state agency
regulating mining in the amount of the estimate. The estimates are based on the assumption that a third-
party government contractor will perform the reclamation work. The proceeds of a forfeited financial
guarantee are used by the agency to pay the contractor to do the work.
The regulatory basis for a reclamation bond is based on numerous state and federal laws. Information on
applicable laws is available from the appropriate state agency having jurisdiction for the mine location.
The use of standardised reclamation cost estimating programs is encouraged to provide a consistent basis
for estimating reclamation costs. These cost estimates are based on a third-party contractor performing the
work. These programs include reclamation activities normally found on a mine site but do not include highly
site-specific conditions such as process and mine water management and water treatment.
The Standardised Reclamation Cost Estimator (SRCE) (SRCE Software) is a reclamation cost estimating
program created through a collaborative effort by the US Department of the Interior, BLM, Nevada State
Office and Field Offices; and the Nevada Mining Association (NvMA).
SRCE can be used in other jurisdictions by updating unit costs and allows the addition of site-specific data not
included in the program. Reclamation costs can also be calculated using spreadsheets based on reclamation
cost estimating guidelines developed by individual states. These guidelines also include variable cost
information. Updating equipment and consumables such as fuel, oil, tires and labour with more current data
than the guideline costs provides a better estimate of reclamation costs.
The reclamation cost estimate is dependent on the reclamation strategy, available equipment and the
availability of a suitable growth medium. The use of current and accurate data such as material volumes, haul
distance, labour and supply costs along with infrastructure design drawings and a familiarity with site
conditions provides a reasonable reclamation cost estimate. For surface operations, concurrent reclamation
provides an opportunity to minimise reclamation liability at the end of the mine life. The completion of

reclamation during the mine’s operating period can also result in reduction of financial assurance obligations
or partial bond release.
In the past, revegetation practices relied on simple methods such as direct seeding of reclaimed areas using
standard seed mixes. Current revegetation practices incorporate the following:
 sampling of soils to determine type and amount of amendments that may be needed;
 matching vegetation to water holding capacity of soils;
 selecting quality plant materials, preferably from the same ecosystem and native to the area;
 seed mix selected to provide diversity;
 erosion controls in the grading plan; and
 seeding prior to periods when moisture typically occurs. In the west, this is in the fall prior to the
first snow.
A good reference for revegetation of mine sites is “An introduction to using native plants in restoration
projects” (Dorner, 2002).
A mining company’s financial responsibility for a mine site does not end when production ends. Robertson
and Shaw (2006) discuss transferring a mining company’s stakeholder interest to a third-party custodian in
their report entitled “Mine closure.” They recommend developing a post-mining sustainable use plan as part
of the closure planning process. Including any succeeding custodians in the development of this plan
increases the possibility of the future custodian accepting the transfer and the future risk of any physical or
chemical instability of the site. Establishing an endowment for residual care and a mechanism for an
independent technical review is also discussed. A technical review of the closure plan includes success
indicators, identifying any residual liabilities and establishing an amount for an endowment fund to provide
for perpetual care. For sites that necessitate long-term collection and treatment of mine-influenced water
and any other maintenance activities, such as for dams, ponds or vegetation, a bond to provide funding for
post-reclamation maintenance may be necessary. Robertson and Shaw (2006) include links to bonding
information sources.
The International Council on Mining and Metals published a report entitled “Financial assurance for mine
closure and reclamation” in 2005 (Miller and Phil, 2005). This report includes discussion of standards of
reclamation, standards of certainty, one-time reclamation versus long-term care, the need for and extent of
financial assurance and the various instruments and timing of financial assurance. Government oversight,
taxation and release of liability (or the exit ticket) are also addressed in the report.
The Commonwealth of Pennsylvania has had a long coal mining history and has had to deal with thousands
of abandoned mine discharges that degraded water quality in many streams. Mines closed since passage of
Pennsylvania’s Clean Water Law are required to treat any post-mining discharges for as long as the discharge
exists and to provide a sufficient bond so that the Commonwealth can continue the treatment in the event
that the permittee does not (PADEP, 2015). Other financial assurance mechanisms can be used as an
alternative to bonds. Pennsylvania has successfully established more than 50 trust fund agreements that,
when fully funded, will generate sufficient income to cover the cost of post-mining water treatment. Most of
these agreements include a consent order and agreement between the mining company and the PADEP and
a trust fund agreement between the mining company and a third-party trustee that manages the trust fund.

Miscellaneous
In addition to the previous issues, sustainability of mining sites has become an important issue, especially
during closure of a mine site. Sustainability is defined by a variety of organisations in slightly different ways.
The 1987 United Nations World Commission on Environment and Development (Jones and Boger, 2011)
defines sustainable development as a process of change that involves exploiting resources, direct
investments and using technological development and institutional changes to make future use of a mine
site consistent with present needs. An excellent resource for consideration of sustainability in mine closures
is “Post mining sustainable use plans vs. closure plans” by Robertson et al. (1998).
Dr Bjorn Stigson, former President of the World Business Council for Sustainable Development, may have
defined sustainability best in the context of the mining industry when he described sustainability as leaving
a positive legacy while exploiting the resource (Jones and Boger, 2011). Regardless of the definition used, it
is generally accepted that sustainable mine closure practices consider the social, economic and
environmental impacts of reclaimed mining lands.
Approximately 40 years ago, mining companies had to start planning for post-closure operations owing to
increasing public concern and regulations related to environmental degradation. Some view these post-
closure plans, including sustainability plans, as a hindrance to the economic viability of mining projects. Van
Der Veen and Strongman (2003) state that many investors are concerned that the actions required to ensure
lasting benefits from mining projects will raise costs and make projects non-competitive. However, a report
prepared by the International Finance Corporation, titled “Development value: The business case for
sustainability in emerging markets,” contends that sustainability measures can achieve higher sales and
reduced costs coupled with the obvious advantages of lower risks, enhanced reputation, improved
environmental practices and investments in social and economic development (Cowe, 2002).
A primary message of the Robertson et al. (1998) document is that mining companies should develop post-
mining sustainable use plans, rather than closure plans, at the onset of a mining project. Some of the key
concepts include establishing post-mining land-use plan goals (e.g. achieve reclamation, post-mining land-
use and custodial transfer in an economical, timely and secure manner), establishing a flexible closure path
with regulatory authorities and stakeholders, establishing success indicators for closure and custodial
transfer and determining a party to perform an independent technical review. The document sets forth a
framework for achieving sustainable use of closed mine lands and offers useful suggestions such as
encouraging progressive reclamation and custodial transfer using separate areas of a mine site (reclamation
units) in order to build trust with and encourage confidence in the successor custodian.
The authors wish to thank the following for their contributions and input to this paper: personnel from the
Colorado DRMS and Utah DOGM for their input into Section 3, and the IDM for the current status of the
Illinois mine closure example. The authors were supported by Matthew Kraus, Cynthia Haluszczak, Wendy
Kowalski, Henry Sauer, Kenneth Carey and Brook Wade.
Association for Mineral Exploration British Columbia (2014) Aboriginal engagement guidebook: a practical and principled approach
for mineral explorers, Association for Mineral Exploration British Columbia, Canada.
Bloom, E. (2012) Consol invests in Epiphany’s water treatment technology, Pittsburgh Post-Gazette, June 26, 2012.
Colorado Division of Minerals and Geology (2002) Best practices in abandoned mine land reclamation: the remediation of past mining
activities, USA.
Cowe, R. (2002) Development value: the business case for sustainability in emerging markets, SustainAbility Ltd. International Finance
Corporation, Washington, DC.
D'Appolonia Engineering (2009) Engineering and Design Manual - Coal Refuse Disposal Facilities, Second Edition, US Department of
Labor, Mine Safety and Health Administration, Pittsburgh, PA, 2009.
Dorner, J. (2002) An introduction to using native plants in restoration projects, Bureau of Land Management, United States
Department of the Interior, USA.

Ford, K.L (2003) Passive Treatment Systems for Acid Mine Drainage, US Bureau of Land Management, University of Nebraska, Lincoln,
NB.
Foreman, J.W. (1971) Deep mine sealing, in Proceedings Acid Mine Drainage Workshop, Ohio University, Athens, USA, pp. 19–45.
August 2–6.
Foreman, J., Bullers, I. and Hong, I. (1969) Moraine State Park: Deep mine sealing engineering study, PADER Operation Scarlift Report
SL105l, Gwin Engineers, Altoona, USA.
Gray, T.A. and Gray, R.E. (1992) Mine closure, sealing, and abandonment, SME Mining Engineering Handbook, 2nd Edition. Published
by Society for Mining, Metallurgy and Exploration, Littleton, CO
Harvey Gap Project (1985) Typical portal seal drawing 8 no scale, Colorado Division of Reclamation and Mining Safety Laserfiche®
WebLink™DRMS Permit File C1984063.
Hedin, R.S., Narin, R.W. and Klienmann, R.L.P. (1994) Passive treatment of coal mine drainage, Bureau of Mines Information Circular,
United States Department of the Interior, USA, Washington, DC.
Indian and Northern Affairs Canada (INAC) (2007) Mine site reclamation guidelines for the Northwest Territories, Indian and Northern
Affairs Canada, Ottawa, Canada.
Jones, H. and Boger, D.V. (2011) Sustainable development and tailings, in Proceedings Second International Future Mining
Conference, pp. 203–208, Sydney, Australia.
Kirkwood, D.T. and Wu, K.K. (1995) Technical considerations for the design and construction of mine seals to withstand hydraulic
heads in underground mines, 1995 SME Annual Meeting, Denver, CO.
Korb, M.C. (2012) Minepool geothermal in Pennsylvania, in Proceedings Pennsylvania Abandoned Mine Land Conference, USA, 2012
Pa AML Conference, State College, 2–4 August 2012.
Meyers Norris Penny LLP (2009) Best practices for consultation and accommodation, Meyers Norris Penny LLP, October 15, 2009,
British Columbia.
Miller, G.C. and Phil, D. (2005) Financial assurance for mine closure and reclamation, ICMM. International Council on Mining and
Metals, Ottawa, Canada, February 2005.
Mine Safety and Health Administration (MSHA) (2008) Final rule on sealing of abandoned areas, viewed http://www.sja.gov/REGS/
FEDREG/FINAL/2008 finl/08-1152.asp.
Pennsylvania Department of Environmental Protection (PADEP) (2012) White paper: utilization of mine influenced water for natural
gas extraction activities, USA, viewed http://www.portal.state.pa.us/portal/server.pt/community/mine_influenced_water_
for_natural_gas_extraction+activities/21390.
Pennsylvania Department of Environmental Protection (PADEP) (2013) Surface Mining Conservation and Reclamation Act 2013,
Reclamation and re-mining incentives report, PADEP Office of Active and Abandoned Mine Operations, USA, viewed
http://www.elibrary.dep.state.pa.us/dsweb/Get/Document-103903/5600-RE-DEP4249%20%202013%20Reclamation%20an
d%20Remining%20Incentive%20Report.pdf
Rassenfoss, S. (2013) Searching for opportunities in environmental trouble, Journal of Petroleum Technologies, December 2013, pp.
62–68.
Robertson, A.M., Devenny, D.W., and Shaw, S.C. (1998) Post mining sustainable use plans vs closure plans, British Columbia Mine
Reclamation Symposium, Vancouver.
Robertson, A. and Shaw. S. (2006) Mine closure, Infomine, Vancouver, Canada.
Sherwin, R.E., Altenbach, J.S., and Waldien, D.L. (2009) Managing abandoned mines for bats, Bat Conservation International, Austin,
TX.
Thrush, P.W. (ed) (1968) A dictionary of mining, mineral, and related terms, US Bureau of Mines, Superintendent of Documents,
Washington, DC, USA, 1,279 pp.
US Environmental Protection Agency (US EPA) (1983) Design manual: neutralization of acid mine drainage, EPA 600/2-83-001, US
Environmental Protection Agency, Washington, DC, USA, 247 pp.
US Environmental Protection Agency (US EPA) (2014) Reference guide to treatment technologies for mining-influenced water, EPA
542-R-14-001, U.S. Environmental Protection Agency, Washington, DC, USA.
Utah Department of Natural Resources (1987) Division of Oil, Gas, and Mining, Shaft abandonment guidelines, Utah Department of
Natural Resources, Salt Lake City, USA.
Van Der Veen and Strongman (2003) Sustainable development – The way forward for the mining industry, Mining Policy Division,
World Bank Group, Proceedings of the International Conference: Sustainability Development Indicators in the Mineral
Industries, Milos Island, Greece, pp. 17–23.
World Coal Organization (2014) Case study: Anglo American eMalahleni Water Reclamation Plan – Winner of WCA Award for
Excellence in Environmental Practice 2013. April, 2014. London, United Kingdon
Wright, M.A. (ed) (2003) The practical guide to reclamation, Utah Division of Oil, Gas, and Mining, USA, 163 pp. Salt Lake City, UT.

A.S. Orechwa Tetra Tech Inc., United States

D.C. Workman Tetra Tech Inc., United States
Mining activities that occurred during the 1950s and early 1960s at the North Cave Hills in Harding County,
South Dakota, are a significant part of South Dakota’s uranium mining history. Under the General Mining
Laws, per the Atomic Energy Act of 1946 and Public Law 357, unrestricted strip mining took place within the
North Cave Hills during this time. More than 1 square kilometre of mine-affected land has been approved by
the US Forest Service and the US Environmental Protection Agency as a non-time critical removal action under
the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) in 2007. Cleanup
levels for surface soil are established for total arsenic and radium-226 based on a site-specific risk assessment.
While numerous scientific studies have been conducted at the site, none of these has characterised the site at
the level of detail achieved using the sampling methods presented here. This case study presents the
methodology and results of two different sampling methods used to independently characterise the
distribution of heavy metals and radionuclide soil concentrations at an abandoned uranium mine. Statistical
evaluations used in sizing sampling grids, evaluation of the feasibility of utilising double sampling methods
versus simple random sampling, geostatistical mapping to establish cleanup boundaries, and an economic
evaluation of the final methods selected are presented in this paper.
A site-wide X-ray fluorescence (XRF) field survey involving collection of more than 800 measurements was
conducted on a defined square sampling grid pattern. A strong correlation was established (r = 0.95) between
in situ XRF arsenic measurements and laboratory-reported total arsenic concentrations in surface soils. This
correlation was the basis for development of a site-specific regression model used in estimating total arsenic
concentrations in surface soils. Additionally, a continuous mobile gamma radiation survey was conducted on
a defined square sampling grid pattern. A strong correlation (r = 0.96) was established between gamma
exposure rate and radium-226 concentrations in surface soils. This correlation was the basis for development
of a site-specific regression model used in estimating the radium-226 concentrations in surface soils. More
than 60,000 gamma exposure rate measurements were collected and used in the estimation of radium-226
concentration within 5,988 block-averaged 100-square-meter (m2) grids. Advanced geospatial analysis
techniques were applied to the arsenic and radium-226 data sets, and cleanup boundaries were estimated.
The surficial extent of soils with arsenic and radium-226 concentrations above the cleanup levels were
estimated at 14 percent (87,817 m2) for arsenic and 2.7 percent (16,682 m2) for radium-226 of the total study
area. An economic analysis compared the incurred cost of the case study with conventional soil sampling
methods that would achieve the same data quality objectives. A project savings factor of 4.4 was calculated
for the XRF field survey and 84 for the gamma survey. The data and information collected during this study
were used to develop site-wide removal action designs that will be implemented as part of the mine closure
process. It is concluded that there is significant economic benefit in using the sampling approaches presented
in this report to characterise mine waste.

A case study of cost effective detailed waste characterisation techniques for closure of abandoned uranium mines A.S. Orechwa and D.C. Workman
During the Cold War, many private companies operated uranium mines in the western US under contracts
with the US government, removing four million tons of ore that went into making nuclear weapons and fuel
and leaving a legacy of radiological and chemical hazards. Currently, 15,000 sites are associated with uranium
extraction in the US Environmental Protection Agency (EPA) Uranium Location Database (EPA, 2006). At many
sites, ore was extracted by strip mining. Protected under the General Mining Laws, including the Atomic
Energy Act of 1946 and Public Law 357, strip mining did not require environmental restoration or post-mining
responsibility (USFS, 2006). This case study focuses on the Riley Pass Abandoned Uranium Mines Site Tronox
Bluffs (Riley Pass) located in the North Cave Hills in South Dakota. Mining that occurred during the 1950s and
early 1960s at the North Cave Hills in Harding County, South Dakota, is a significant part of South Dakota’s
uranium mining history. The strip mining at Riley Pass involved removal of uranium-bearing lignite coal beds.
The elevated total arsenic and radium-226 (Ra-226) within soils and sediment in the mine areas pose a threat
to public health and environment. Since the mine areas are located on US Forest Service (USFS) land, cleanup
authority is delegated to the USFS. More than 100 hectares of mine-affected land was approved by USFS and
EPA as a non-time critical removal action under the Comprehensive Environmental Response, Compensation,
and Liability Act (CERCLA) in 2007. The primary objective for the removal action at the site is to attain a
degree of isolation, containment, and cleanup of hazardous substances that assure protection of public
health, safety, and welfare of the environment (USFS, 2007). Before an engineering removal design plan was
developed, a detailed characterisation study was required. The study presented in this paper was funded by
USFS with the intent to provide the information necessary for engineering removal design. Numerous
scientific studies have been conducted at Riley Pass, but none of these studies has characterised the mine
waste at the site to the level of detail exhibited in this study.
This case study focuses on the methodology and results of using cost-effective environmental sampling tools
to characterise the mine waste at the largest study area at Riley Pass, referred to as “Bluff B.” Bluff B
encompasses 60.7 hectares of mine-affected area. The established site-specific arsenic cleanup criterion is
142 milligrams per kilogram (mg/kg) and 30 picoCuries per gram (pCi/g) for Ra-226. These levels were based
on a risk assessment previously reported in an Engineering Evaluation/Cost Analysis (USFS, 2006). Arsenic
and Ra-226 concentrations in soil are not directly correlated with each other and therefore need to be
estimated using individual techniques. Given the level of detail needed and the large land area to assess, a
sampling program consisting of conventional characterisation methods such as soil sampling would be
extremely costly. The cost-effective sampling methods presented in this study involve (1) X-ray fluorescence
(XRF) field survey used to estimate arsenic surface soil concentrations, and (2) Gamma radiation survey
(gamma survey) used to estimate Ra-226 surface soil concentrations.
Statistical evaluations used in sizing the sampling grids, evaluation of the feasibility of utilising double
sampling methods versus simple random sampling, geostatistical mapping to establish cleanup boundaries,
and an economic evaluation of the final methods selected are presented in this paper. Using statistical hot
spot location techniques, the sampling grid sizes were calculated for both primary sampling methods (XRF
field survey and gamma survey). After the initial sampling requirements were established, a statistical
evaluation showed that a double sampling approach would be more cost effective for both sampling methods
and would provide an economic benefit to the project. Strong correlations were developed for both double
sampling methods, allowing for a cost-effective and definitive estimation of arsenic and Ra-226
concentrations in surface soil at the site. Geostatistical methods were applied to the data sets and final
removal boundaries were estimated, resulting cleanup areas of 87,817 m2 (14 percent of the site) for arsenic
and 16,682 m2 (2.7 percent) for Ra-226. Finally, a cost evaluation compares the actual incurred costs of the
project with the estimated costs of using conventional soil sampling to achieve the project data quality
objectives. A project savings factor of 4.4 and 84 were calculated for the estimation of arsenic by XRF field
survey and Ra-226 by gamma survey. Overall, the methods used in the study were proven cost effective and
provided a level of detail necessary to meet the project data quality objectives.

Miscellaneous
Hot spot location techniques involve systematic sampling from a grid of sampling points arranged in a
particular pattern. A “hot spot” refers to any sample that exceeds a cleanup standard for a specified
contaminant of concern. Using statistical hot spot location techniques outlined in Gilbert (1987) and EPA
(1989), the required sampling grid size can be calculated that corresponds to a probability of identifying a
hot spot of a given size and shape. In practice, environmental monitoring programs are seldom conducted
solely to accurately estimate the population mean; instead, a sampling design to determine spatial
characterisation of a particular contaminant is often the primary objective. Any samples that contain less
than the site-specific cleanup criterion (< 142 mg/kg for arsenic; < 30 pCi/g for Ra-226) are assumed to be
“clean” and no action is required. Furthermore, any surface soils exceeding the cleanup criterion (≥ 142
mg/kg for arsenic; ≥ 30 pCi/g for Ra-226) are considered contaminated and need to be excavated and isolated
to the meet the project objective. The grid spacing required for finding a specified hot spot of a pre-defined
size and shape with a specified level of certainty can be established using the following procedure:
1. Specify L, the length of the semi major of the smallest hot spot to detect. L is one-half the length of
the long axis of the ellipse.
2. Specify the expected shape (S) of the elliptical target (S = 1 for a circle), where S is the length of the
short axis of the ellipse divided by the length of the long axis of the ellipse. The target hot spot is
assumed.
3. Specify an acceptable probability, referred to as “consumer’s risk” (β) of not finding the hot spot.
4. Refer to Figure 1 below. The curves provided in this figure show the relationship between β and
L/G, where G is the required grid spacing.
Frequently, two or more techniques may be available for measuring the amount of contaminant in an
environmental sample (Gilbert, 1987). Double sampling involves collection of a large set of samples by an
inexpensive and less expensive rapid analytical method (“fallible” method) and a second smaller set of
samples collected and analysed by a more expensive and accurate method (“infallible”) of analysis. The
required grid spacing for sampling the surface soils must be relatively small to create a high probability that
smaller hot spots will be found; given the size of the site, the cost of sampling and analysis can be high. The

following condition must be met to assess whether double sampling is economically advantageous compared
with simple random sampling (Gilbert, 1987):
2
𝐶𝐴 (1 + √1 − 𝑟 2 )
𝑅= >
𝐶𝐼 𝑟2
Where:
𝑅 = cost ratio
𝐶𝐴 = cost of accurate (“infallible”) method
𝐶𝐼 = cost of more inaccurate (“fallible”) method
𝑟 = correlation coefficient between two methods
Double sampling will be cost effective if the linear correlation between measurements obtained by the two
techniques is sufficiently near unity and if the fallible method is substantially less costly than the more
accurate method (has a higher cost ratio). The optimum number of infallible and fallible samples that should
be collected can be calculated by the methods outlined in Gilbert (1987); however, the optimum number of
samples determined by the double sampling procedures were not collected since both techniques used in
this study rely on scientifically accepted procedures. For this particular study, the number of fallible and
infallible samples required for each method is selected based on the specific sampling procedure for those
methods. Since the optimum sample size requirements for the double sampling method are not achieved, it
is not guaranteed that the linear regression double sampling approach will yield a more precise, on average,
estimate of the mean than would be achieved by simple random sampling. Minimising the variance of the
estimated mean is not the primary objective of this study, however. The primary objective is to characterise
the spatial extent of arsenic and Ra-226 contamination using a reliable and cost-effective sampling approach.
The equation presented above is used to qualitatively assess whether double sampling should be used, based
on cost and estimation accuracy, compared with simple random sampling.
XRF field surveys are widely used in the field of environmental engineering as a non-destructive, cost–
effective, and rapid tool for screening soils or characterising hazardous waste sites or sites contaminated
with mine waste. The XRF is considered the “fallible” sampling technique because it is less expensive and
provides instantaneous results when compared with the “infallible” sampling technique, consisting of
conventional soil sampling and subsequent laboratory analysis. The XRF measurements taken directly from
on-site surface soils are referred to as in situ XRF measurements. The in situ XRF measurements were
collected throughout the site using a portable Niton XRF 900 Xlt spectrum analyser. The field portable XRF
procedures were performed in accordance with EPA Method 6200 – “Field Portable X-ray Fluorescence
Spectrometry for the Determination of Elemental Concentrations in Soil and Sediment“ (EPA, 2007). The Niton
XRF uses an X-ray tube to irradiate soil samples. The source X-rays excite electrons in the surface soil sample
(EPA, 2007), dislodging electrons from atomic shells and creating vacancies. The vacancies are filled by the
ray spectra as they cascade down to fill the inner shell vacancies. This process allows the XRF instrument to
identify elements present based on the unique spectra emitted and to estimate metal concentrations in soil
based on emitted flux. The Niton XRF collects data from 26 analytes; however, the contaminant of concern
for this particular site is arsenic. EPA method 6200 also specifies that one of every 20 XRF samples be
collected and submitted for laboratory analysis. The samples submitted to the laboratory are referred to as
confirmation samples. A site-specific linear correlation can then be made between the in situ XRF arsenic
measurements and the arsenic concentration reported in the confirmation soil samples. The data should be
log-transformed in cases where the data spread over more than one order of magnitude. The data can be
used as a screening-level tool if the correlation coefficient is greater than 0.7. If the correlation coefficient is

Miscellaneous
greater than 0.8, the correlation can be used to estimate that specific contaminant at a definitive level.
Stringent quality assurance and quality control (QA/QC) procedures are followed as outlined in EPA Method
6200.
Using gamma radiation to estimate radionuclides is a common approach at sites contaminated with
windblown uranium tailings (e.g. former uranium mills) and at abandoned uranium mines. The success of this
approach depends on whether radionuclides of interest have gamma emissions, potential contamination is
located on the ground surface and, most importantly, acquiring regulatory approval of the technique
(Albequist, 2000). All soils and rock exhibit differing levels of radioactivity, depending on varying levels of
naturally occurring potassium, uranium, thorium, and radium. On open ground, about two-thirds of the
measured gamma radiation dose comes from radionuclides contained in the top 15 centimetres (cm) of soil
(NRC, 1994). The objective of the continuous gamma radiation survey is to characterise the spatial
distribution of gamma radiation emanating from surface soils at the site. Using soil correlation methods, the
gamma data can then be used to predict the radionuclide concentrations in surface soils.
Field engineers used mobile backpack scanning systems consisting of 5-cm by 5-cm Ludlum 44-10 thallium
laced sodium iodide (NaI(Tl)) scintillation detectors and Ludlum 2350-1 data loggers coupled with Wide Area
Augmentation System (WAAS) enabled global positioning systems (GPS) connected to a field laptop. The
gamma exposure rate data were transmitted once per second to a portable computer and logged using
proprietary logging and mapping software (Tetra Tech, 2006). A detector height of 1 meter (m) above the
ground surface is considered standard practice for this application (EPA, 1999, and OSD, 2012) and was
used during this survey. All measurement data were automatically stored and processed with the
measurement location information to be mapped and analysed in real time. Real-time mapping allowed the
field engineer to maintain position on pre-determined scan transect lines and to identify any problems that
arose during the scanning efforts. NaI(Tl) detector systems exhibit energy-dependent response
characteristics normalised to the cesium-137 0.662 megaelectronvolt (MeV) photon. The Ludlum 2350-1 data
logger system employs a calibration factor to internally convert detector counts per minute (cpm) to gamma
exposure rate. The calculated exposure rate, directly proportional to the measured cpm, is transmitted by
the data logger to the scanning system portable PC. The system does not retain a record of count rate, but
count rate can be calculated using the instrument-specific calibration factor.
A strong correlation must exist between the two parameters for the gamma radiation survey to be an
effective tool to estimate Ra-226 concentrations in soil. Correlation sampling was performed in accordance
with the methods outlined in Johnson et al. (2006) and Whicker et al. (2008). After the gamma radiation
survey had been completed, field personnel selected 10-m x 10-m (100 square meters, or m2) correlation
plot locations. The plot locations were selected to represent the range of gamma radiation at the site and in
areas where homogeneity in the gamma field was observed. Composite soil samples from each plot were
submitted for laboratory analysis of Ra-226 by gamma spectrometry using Method E910.1. Gamma exposure
rate data was collected within the boundary of each correlation plot using the scanning systems used in the
gamma radiation survey. Tetra Tech incorporated stringent QA/QC protocols for this project.
The action memorandum (USFS, 2007) specifies that the mine areas are to be divided into appropriate sized
grids and a block averaging technique will be applied to the post-reclamation data after the site has been
reclaimed. The block averaging approach was applied to the characterisation data to maintain consistency
throughout the reclamation process. Block averaging used geostatistical methods. Geostatistical methods
are a powerful tool for mapping spatial data and providing interpolation between existing data points that
have been collected (EPA, 1989). Geostatistical methods are commonly used in geographic, geological, and
environmental sciences, as outlined in Journel and Huijbregts (1978), David (1977), and Verly et al. (1984).
Two geostatistical methods were utilised for this study: kriging and the inverse distance weighted (IDW)

method. Kriging is used to interpolate the denser gamma radiation point data, and IDW is used to interpolate
between the more scattered XRF point data.
The probability that a circular hot spot would be found was calculated using the methodology outlined in
Section 1.1. Field personnel performed additional judgment sampling after hot spots were located in the
field, further increasing the accuracy of mapping hot spots. Different data quality objectives were required
for the sampling program because the required cost and data collection efficiency were different between
the discrete XRF field survey and the continuous gamma radiation survey. The acceptable probability for the
XRF field survey grid size was determined to be 90 percent (β = 10 percent) certain of finding a circular hot
spot greater than or equal to 34 m (L = 17 m) in diameter. The resulting XRF field survey grid size (G) was then
calculated to be at 30 m increments, as shown in Figure 2. Similarly, the acceptable probability for the gamma
radiation survey grid size was 95 percent (β = 5 percent) certain of finding a circular hot spot greater than or
equal to 18 m (L = 9 m) in diameter. The resulting grid size for the XRF field survey was calculated to be at 15
m increments, as shown in Figure 3. The following assumptions are used to identify hot spots of
contamination using the sampling approach described earlier:
1. The target hot spot is assumed circular (S = 1 for a circle).
2. Sample grid is assumed square (such as G =30-m or G = 15-m).
3. A small proportion of the area being studied can actually be measured.
4. The areas considered “hot spots” are defined as any arsenic concentration in surface soil that is
greater than or equal to 142 mg/kg or surface soil Ra-226 concentration greater than or equal to
30 pCi/g.
5. A β assumed 0.05 indicates there will be a 5 percent chance of not finding a hot spot of a given size.
(β is defined as the acceptable probability that a hot spot will not be found.)
A summary of the double sampling parameter analysis for both methods is shown in Table 1. Figure 2 shows
the reference grid, XRF measurement locations, and selected confirmation samples. Figure 3 shows the raw
gamma surveys points collected during the study.
Parameter XRF Gamma
S 1 1
G (m) 30 15
β 0.10 0.05
L (m) 17 9

Miscellaneous

A sampling grid size of 30 m was used for sampling arsenic at the site. An evaluation following the methods
outlined in Section 1.2 was used to determine if using the in situ XRF method as a form of double sampling is
more cost effective than simple random sampling. The cost ratio (R) was calculated to be 7.11, which satisfied
the condition of exceeding the value of 1.91, as shown in Table 2. Therefore, use of double sampling is cost
effective.
Cost of lab sample Cost of XRF Correlation 𝑪𝑨 𝟐

𝑹= (𝟏 + √𝟏 − 𝒓𝟐 )
(CA) sample (CI) coefficient, r 𝑪𝑰
𝒓𝟐
US$121 US$17 0.95 7.11 1.91
A sampling grid size of 15 m was used for the gamma survey at the site. An evaluation following the methods
outlined in Section 1.2 was used to determine if using the mobile gamma radiation survey method to estimate
Ra-226 soil concentrations as a form of double sampling is more cost effective than simple random sampling.
The cost ratio was calculated to be 121, which satisfied the condition of exceeding 1.78, as shown in Table 3.
Therefore, use of double sampling is cost effective.
Cost of 𝑪𝑨 𝟐
Cost of lab sample Correlation 𝑹= (𝟏 + √𝟏 − 𝒓𝟐 )
gamma 𝑪𝑰
(CA) coefficient, r 𝒓𝟐
sample (CI)
US$181 US$1.50 0.96 121 1.78
In situ XRF sampling as a double sampling method was determined to be cost effective compared with simple
random sampling. Following EPA Method 6200, a strong correlation (r > 0.8) must be established between
the in situ XRF measurements and the arsenic concentrations measured in the laboratory to be used as a
definitive characterisation tool. A total of 804 in situ XRF arsenic measurements (fallible samples) were
collected from Bluff B and 69 confirmation samples (infallible samples) were collected from Bluff B and other
areas and submitted for laboratory analysis. The confirmatory samples were selected from the lower, middle,
and upper range of concentrations measured at the site. EPA laboratory analytical Method 6020A was used
for the analysis of total arsenic in the soil samples. The arsenic concentrations measured spanned four orders
of magnitude; as such, these data were log-transformed to standardise the variance proportional to the
magnitude of the measurement. Figure 4 shows the correlation between in situ XRF arsenic and laboratory-
reported arsenic at the site. The correlation can be used as a definitive level characterisation tool since the r
is greater than 0.8 and inferential statistics indicate the confirmatory data are statistically equivalent at a 99
percent confidence level.

Miscellaneous
10000
Regression
95% PI
R-Sq 89.7%
Lab Total Arsenic (mg/kg)
1000
100
10
10 100 1000 10000

In-situ XRF Total Arsenic (mg/kg)
A strong correlation (r = 0.95) exists between the in situ XRF arsenic measurements and the laboratory-
reported total arsenic concentrations. The following equation was used to convert the in situ XRF
measurements to laboratory equivalent arsenic concentrations at the site:
𝐿𝑎𝑏 𝐴𝑟𝑠𝑒𝑛𝑖𝑐 = 100.352+0.891𝑙𝑜𝑔10 (XRF Arsenic)

Where:
Lab arsenic = laboratory reported arsenic concentration in surface soil (mg/kg).
XRF Arsenic = XRF measured arsenic concentration in surface soil (mg/kg).
Gamma surveying as a double sampling method was determined to be cost effective compared with simple
random sampling. Following methods in Johnson et al. (2006) and Whicker et al. (2008), a correlation
between gamma radiation and Ra-226 concentration in soil was performed using 100 m2 correlation plots
throughout the site. A total of 22 soil correlation plots were surveyed for gamma radiation, and soil samples
were collected and submitted to a laboratory for analysis of Ra-226 using gamma spectrometry (Method
E.901.1). Since the gamma and radium concentrations measured spanned three orders of magnitude, the
data were log-transformed to standardise the variance proportional to the magnitude of the measurement.
Figure 5 shows the correlation between gamma exposure rate and Ra-226 concentration in soil at the site.

10000 Regression
95% PI
R-Sq 92.2%
1000
Ra-226 (pCi/g)
100
10
0.1
10 100 1000
Gamma Exposure Rate (µR/hr)
A strong correlation (r = 0.96) was established between gamma exposure rate and the Ra-226 concentration
in soil. The following equation was used to convert the gamma radiation measurements collected at the site
to laboratory equivalent Ra-226 soil concentrations:
Lab Ra − 226 = 10−1.979+1.835 𝑙𝑜𝑔10 (𝐺𝑎𝑚𝑚𝑎)
Where:
Lab Ra-226 = laboratory equivalent soil radium-226 concentration (pCi/g).
Gamma = Gamma exposure rate measurement (µR/hr)
The double sampling methods presented earlier were selected because of the cost effectiveness of the
techniques compared with conventional soil sampling methods. The cost ratio (R) between the more
expensive and accurate method (CA) compared with the cheaper, rapid method (CI) were established for both
techniques. The cost ratio does not represent the total costs associated with the project; rather, it only
compares the cost of the technique on a per-sample basis. Therefore, a project savings factor was also
evaluated to assess the economic benefit when applied to the actual project and compared with the case of
characterisation by conventional soil sampling. The project savings factor is defined as the cost of a project
using conventional sampling methods proportional to the cost of a project using double sampling methods.
A cost analysis calculated the unit cost for each sampling technique for both characterisation methods. The
analysis includes factors such as labour for sampling, field crew mobilisation and lodging, analytical
laboratory costs, equipment rental, and shipping costs. The unit costs for each technique of the sampling
methods are presented in Table 2 and Table 3. Table 4 shows the actual number of samples that were
collected in the case study for both techniques. The project savings factor is less than the cost factor because
it includes additional project costs associated with using both sampling techniques (CA and CI). The cost factor

Miscellaneous
for the XRF field survey method is 7.11 (Table 2) and the project savings factor is 4.42 (Table 4). The cost
factor for the gamma survey method is 121 (Table 3) and the project savings factor is 83.6 (Table 4).
Project
Survey # of CA # of CI Double sampling Conventional method
savings
method samples samples method cost (US$) (US$)
factor
XRF 69 804 $22,017 $97,284 4.42
gamma
22 5,988 $12,964 $1,083,828
survey 83.6
Soil contamination maps showing the spatial extent of arsenic and Ra-226 concentrations in soil at the site
were developed by applying geostatistical tools to the point data collected from the XRF field survey and the
gamma survey. The final removal boundaries were mapped for arsenic and Ra-226 above the soil cleanup
levels as shown in Figure 6 and Figure 7. The total soil cleanup area for arsenic removal is 87,817 m2. The
total soil clean-up area for Ra-226 removal is 16,682 m2. The arsenic removal boundaries contain 73 percent
of the Ra-226 removal boundaries. Therefore, 27 percent of the identified radium cleanup areas would
remain if only the arsenic contamination is removed, assuming the contamination is surficial.

Site-specific cleanup levels of arsenic and Ra-226 were established for removal action criteria at the Riley
Pass Abandoned Uranium Mine. The site reclamation is to be performed by stabilisation and isolation of
contaminated soils and sediment to effectively eliminate pathways for human health risks. Before the
removal action is implemented, it was essential to identify the spatial extent of contaminated material
needing removal. Two cost-effective environmental sampling methods were presented that utilise the
principles of double sampling to characterise the spatial extent of arsenic and Ra-226 contamination in
surface soil. The two methods presented included (1) XRF field surveys to characterise the arsenic
concentrations in soil, and (2) gamma radiation surveys to characterise the Ra-226 concentrations in soil.
For this case study, the objective of the sampling was to determine whether “hot spots” of contamination
exist and to characterise the spatial extent of these areas. Systematic sampling grids were sized for each
sampling method using statistical hot spot characterisation techniques. Sampling grid spacing of 30 m were
used for the XRF field survey and 15 m for the gamma radiation survey. It was determined that it is cost
effective for the XRF field survey and the gamma radiation survey to utilise these methods as tools for
characterising the site. A strong correlation (r = 0.95) was observed between the in situ XRF arsenic
measurements and corresponding laboratory-reported values; similarly, the correlation between gamma
exposure rate and Ra-226 soil levels was also strong (r = 0.96). Using the site-specific correlation models
developed for both methods, the arsenic and Ra-226 concentrations in soil were created for the site.
Geostatistical methods were applied to the data sets and removal boundaries were estimated, resulting in
cleanup areas totalling 87,817 m2 (14 percent of site) for arsenic and 16,682 m2 (2.7 percent of the site) for
Ra-226.
A detailed cost analysis evaluated the economic benefit of using double sampling methods over conventional
soil sampling methods. The unit costs for each sampling technique were determined for both sampling
methods. The cost ratio for the XRF field survey is 7.11, indicating that it is 7.11 times more expensive to
collect a soil sample using conventional techniques than it is to collect a single in situ XRF measurement. An

Miscellaneous
analysis shows that the project savings factor for the XRF field survey versus conventional soil sampling
methods was 4.42, indicating that it is 4.42 times less expensive to use an XRF field survey to characterise
the site than a conventional soil sampling method. The most substantial cost savings observed is the gamma
radiation survey method. The cost ratio and project savings ratio for the gamma radiation survey are 121 and
83.6. In other words, it is 83.6 times cheaper to use gamma radiation to estimate Ra-226 concentrations for
a site compared with a conventional soil sampling method that would achieve the same data quality
objectives. Characterising Ra-226 activity in soil across the site using conventional soil sampling techniques
would result in a total cost of more than US$1 million; alternatively, using a mobile gamma survey system
with a correlation to Ra-226 results in a characterisation cost of less than US$13,000. The significant
economic benefit to employing the methods presented in this case study rely on a thorough understanding
of project objectives, the ability to apply statistical evaluations to these objectives, and the use of reliable
field techniques and technology. When these elements are present, the project costs for site characterisation
can be minimised while meeting the project data quality objectives.
Abelquist, E.W. (2001) Decommissioning health physics: a handbook for MARSSIM users, Institute of Physics Publishing, 438 pp.
David M. (1977) Geostatistical ore reserve estimation, Philadelphia: Elsevier Scientific Publishing Company. 364 pp.
Gilbert R.O. (1987) Statistical methods for environmental pollution monitoring, New York: John Wiley & Sons, Inc. 320 pp.
Johnson J.A., Meyer, H.R. and Vidyasagar, M. (2006) Characterization of surface soils at a former uranium mill, Health Physics, Vol. 90
(Suppl 1): S29-S32, 2006.
Journel, A.G., Journel, A.G. and Huijbregts, C.J. (1978) Mining geostatistics, New York: Academic Press. 600 pp.
Tetra Tech Inc. (2006) Gamma data map viewer. Software. Fort Collins, Colorado.
US Environmental Protection Agency (EPA) (1989) Methods for evaluating the attainment of cleanup standards, Vol. 1: Soils and solid
media, Washington, DC: EPA. 153 pp.
US Environmental Protection Agency (EPA) (1992) Preparation of soil sampling protocols: sampling techniques and strategies,
EPA 600-R-92-128, July 1992, pp. 1–7.
US Environmental Protection Agency (EPA) (1999) Distribution of OSWER radiation risk assessment Q&A’s final guidance
(Memorandum) Luftig SD.
US Environmental Protection Agency (EPA) (2006). Uranium location database compilation, EPA 402-R-05-009, August 2006
Washington, DC: EPA
US Environmental Protection Agency (EPA) (2007) Field portable XRF spectrometry for the determination of elemental concentrations
in soil and sediment, Washington, DC: EPA.
US Forest Service, (USFS) (2006) Riley Pass Uranium Mines site, final engineering evaluation/cost analysis, US Department of
Agriculture/Forest Service Region 1, Custer National Forest, Sioux Ranger District, Harding County, South Dakota, November,
2006.
US Forest Service, (USFS) (2007) Riley Pass Uranium Mines site removal action, action memorandum, Custer National Forest, Sioux
Ranger District, Harding County, South Dakota, February, 2007.
US Nuclear Regulatory Commission, (NRC) (1994) NUREG-1501: background as a residual radioactivity criterion for decommissioning,
August 1994, prepared by A.M. Huffert, R.A. Meck and K.M. Miller.
Verly, G., David, M., Journel, A.G. and Marachal, A. (1984) Geostatistics for natural resources characterization, Parts 1 and 2,
Dordrecht: D. Reidel Publishing Company, 2,080 pp.
Whicker, R., Cartier, P., Cain, J., Milmine, K. and Griffin, M. (2008) Radiological site characterizations: gamma surveys, gamma/radium-
226 correlations, and related spatial analysis techniques, The Radiation Safety Journal S180, November 2008.


Adams, M.H. ...................................................................................................................................................327

Agustina, F. .....................................................................................................................................................817
Ahola, R...........................................................................................................................................................673
Ainsworth, G.L. ...............................................................................................................................................805
Allen, D.E. ............................................................................................................................................... 275, 951
Allen, G.C. .......................................................................................................................................................939
Ames, S.E. .......................................................................................................................................................497
Amizlev, I. .......................................................................................................................................................737
Anderson, T.R. ................................................................................................................................................517
Ardito, C. .........................................................................................................................................................301
Audet, P. .......................................................................................................................................................1081
Ayres, B.K. ............................................................................................................................................... 929, 939
Baker, R. ..........................................................................................................................................................599
Baker, T. ..............................................................................................................................................1023, 1035
Balaz, W.P., Jr. ..............................................................................................................................................1137
Baldwin, S.A. ...................................................................................................................................................106
Barber, L.A. ...........................................................................................................................................929, 1071
Barbour, S.L. .......................................................................................................................................1023, 1035
Barnekow, U. ..................................................................................................................................................251
Barros, D.V. .......................................................................................................................................................53
Bartle, S...........................................................................................................................................................561
Bennett, K.E. .....................................................................................................................................................71
Berthelot, D. ...................................................................................................................................................995
Bierman, D.F. ..................................................................................................................................................725
Blickwedel, R...................................................................................................................................................301
Bockstette, J..................................................................................................................................................1071
Botha, P.R. ........................................................................................................................................................92
Bowell, R. ........................................................................................................................................................413
Bradley, C. .......................................................................................................................................................965
Brand, K. .........................................................................................................................................................193
Breadmore, R.E. ..............................................................................................................................................753
Bright, D.A.......................................................................................................................................................461

Author Index
Brough, C. .......................................................................................................................................................413
Brown, C.B. .....................................................................................................................................................263
Bucar, D. .........................................................................................................................................................471
Butler, A.R. ......................................................................................................................................................517
Buttineau, P. ...................................................................................................................................................471
Carey, S. ..............................................................................................................................................1023, 1035
Castle, J.W. .....................................................................................................................................................545
Cayouette, J. ...................................................................................................................................................143
Chaparro, G.....................................................................................................................................................507
Charest, D. ....................................................................................................................................................1023
Charette, T. .....................................................................................................................................................787
Chartier, P. ......................................................................................................................................................877
Chirinda, T.......................................................................................................................................................697
Christensen, D.O. ..........................................................................................................................................1071
Christoffersen, L.M. .......................................................................................................................................275
Chute, A. .........................................................................................................................................................389
Clemens, M.D. ..................................................................................................................................................59
Consigny, A. ....................................................................................................................................................575
Corrigan, C. .....................................................................................................................................................449
Costa, S. ..........................................................................................................................................................797
Cottard, F. .......................................................................................................................................................877
Coulthard, T.J. .................................................................................................................................................207
Cowan, W.R. .....................................................................................................................................................14
Cranston, B.H. .......................................................................................................................................485, 1125
Crockett, D. ...................................................................................................................................................1117
Cunningham, K.E...............................................................................................................................................23
Dagva, M.B......................................................................................................................................................805
Davaatseren, Ts. .............................................................................................................................................805
Davies, M. .................................................................................................................................................. xv, 12
Defferrard, P.L. ...............................................................................................................................................917
Dey, M. ...........................................................................................................................................................413
Dickin, R.C. ......................................................................................................................................................339
Dickin, S.E. ......................................................................................................................................................339
Digby, C. ............................................................................................................................................................13
Dobrowolski, M.P. ..........................................................................................................................................829
Dombrowski, N. ..............................................................................................................................................389

Author Index
Doyon, J.-F. .....................................................................................................................................................143

Ebbs, S.............................................................................................................................................................837
Erdenetuya, O. ................................................................................................................................................805
Etches, K. ..........................................................................................................................................................52
Eykamp, L. .....................................................................................................................................................1117
Fernandez, S. ....................................................................................................................................................81
Ferster, E. ........................................................................................................................................................609
Flemming, D....................................................................................................................................................939
Flynn, M. .........................................................................................................................................................121
Franklin, W......................................................................................................................................................863
Fraser, L.H. ......................................................................................................................................................106
Gardiner, E.J......................................................................................................................................................14
Gardner, W.C. .................................................................................................................................................106
Garris, H.W. ....................................................................................................................................................106
Glaeser, L. .......................................................................................................................................................837
Goerke-Mallet, P.............................................................................................................................................711
Grant, C.D. ................................................................................................................................................ 92, 185
Gray, R.E. ......................................................................................................................................................1137
Gray, T.A. ......................................................................................................................................................1137
Groat, L. ..........................................................................................................................................................339
Guittonny-Larchevêque, M. ...........................................................................................................................851
Gullison, T. ......................................................................................................................................................863
Haakensen, M. ................................................................................................................................................545
Halbert, B.E. ....................................................................................................................................................901
Hancock, G.R...................................................................................................................................................207
Hango, D. ........................................................................................................................................................449
Hantler, L. .......................................................................................................................................................185
Harkin, C. .......................................................................................................................................................377
Harrison, D. ...................................................................................................................................................1117
Hart, M.J. ........................................................................................................................................................363
Hattingh, R. .....................................................................................................................................................829
Healey, N.J. .....................................................................................................................................................633
Hegemann, M. ................................................................................................................................................711
Henkel, L. ........................................................................................................................................................711
Henschel, M.D. ...............................................................................................................................................673
Hilts, S. ............................................................................................................................................................863

Author Index
Hine, A. ...........................................................................................................................................................737
Hockley, D.E. ....................................................................................................................................................... 3
Hockley, L.C......................................................................................................................................................... 3
Hogan, C.M. ......................................................................................................................................................14
Holder, L.K. .....................................................................................................................................................289
Holmes, R. .......................................................................................................................................................121
House, M. .......................................................................................................................................................837
Hovdebo, D.G. ..................................................................................................................................................23
Huffman, D.M. ................................................................................................................................................263
Humphries, R.N.....................................................................................................................................983, 1047
Humphries, S.A. ..............................................................................................................................................315
Ind, M.J. ..........................................................................................................................................................193
Jacob, J.C.........................................................................................................................................................353
Jeffrey, C.M. ....................................................................................................................................................763
Jensen, S.R. ....................................................................................................................................................275
Jessop, M.L. ....................................................................................................................................................725
Jia, K. ...............................................................................................................................................................339
Jones, R. ..........................................................................................................................................................289
Julien, M.R. .....................................................................................................................................................143
Kennard, D. .....................................................................................................................................................433
Khasa, D.P. ....................................................................................................................................................1061
Kirsch, P. .........................................................................................................................................................737
Knapp, R. ...........................................................................................................................................................37
Kofoed, V.O.....................................................................................................................................................725
Koop, G. ..........................................................................................................................................................375
Kos, P. .............................................................................................................................................................363
Kratochvil, D. ..................................................................................................................................................575
Kristoff, D.M. ............................................................................................................................................ 23, 633
Lacy, H.W.B. ......................................................................................................................................................71
Lafferty, G.J. ....................................................................................................................................................753
Lambert, J. ......................................................................................................................................................951
Landhäusser, S.M. ........................................................................................................................................1071
Lanteigne, S. ...................................................................................................................................................531
Larivière, M. ....................................................................................................................................................449
Le Loher, F. .....................................................................................................................................................877
Lehane, S.........................................................................................................................................................657

Author Index
Leskiw, L.A. .....................................................................................................................................................485

Limpitlaw, D. .....................................................................................................................................................13
Liskowich, M.W. .............................................................................................................................................275
Littlejohn, P.....................................................................................................................................................575
Lord, M.T. ............................................................................................................................................... 327, 917
Lower, N..........................................................................................................................................................621
Lowry, J.B.C. ....................................................................................................................................................207
Lucas, C.E. .......................................................................................................................................................901
Ludgate, I. .......................................................................................................................................................531
MacInnis, C. ....................................................................................................................................................433
MacWilliam, S. ..............................................................................................................................................1103
Maimela, F.A. ..................................................................................................................................................585
Marsland, R.....................................................................................................................................................621
Martin, J. .........................................................................................................................................................787
Mayer, U. ........................................................................................................................................................339
Mayich, D. .......................................................................................................................................................965
McCullough, C.D. ...........................................................................................................................................377
McParland, M.A. .............................................................................................................................................673
Meiers, G. .......................................................................................................................................................965
Melchers, C. ....................................................................................................................................................711
Merkel, G. .......................................................................................................................................................251
Metschies, T....................................................................................................................................................251
Mills, R. ...........................................................................................................................................................339
Montpellier, R. ................................................................................................................................................645
Murray, L. .......................................................................................................................................................193
Nadeau, M.B. ................................................................................................................................................1061
Nagare, R.M. ...................................................................................................................................................891
Nicholson, R.V.................................................................................................................................................561
Nielsen, B. .......................................................................................................................................................289
Noël, J. ............................................................................................................................................................143
Norlund, K. ......................................................................................................................................................995
Novy, L. ...........................................................................................................................................................375
Nurtjahya, E. ...................................................................................................................................................817
O’Kane, M. .................................................................................................................................. 965, 1023, 1035
Orechwa, A.S. ...............................................................................................................................................1151
Ortiz, C.C. ..........................................................................................................................................................53

Author Index
Pal, J. ...............................................................................................................................................................891
Palmer, M.W. ................................................................................................................................................1011
Park, Y.J. ..........................................................................................................................................................891
Paul, M. ...........................................................................................................................................................251
Pearce, S.R. .....................................................................................................................................................657
Pedlar-Hobbs, R. .............................................................................................................................................531
Pelletier, C.A. ............................................................................................................................................ 52, 995
Pérez, N...........................................................................................................................................................507
Petelina, E. ....................................................................................................................................................1103
Peterson, R.A. .................................................................................................................................................315
Petherbridge, W. ............................................................................................................................................375
Pinno, B.D. ....................................................................................................................................................1081
Piorkowski, G.S. ..............................................................................................................................................403
Pittet, V. ..........................................................................................................................................................545
Pizarro, J.K. .......................................................................................................................................................53
Polster, D. .....................................................................................................................................................1093
Power, B..........................................................................................................................................................599
Power, B.A. .....................................................................................................................................................633
Prelypchan, E. ...................................................................................................................................................52
Price, G............................................................................................................................................................403
Pugh, J.C..........................................................................................................................................................763
Quach, M. .......................................................................................................................................................621
Reid, C. ............................................................................................................................................................951
Rickson, J. .....................................................................................................................................................1047
Ridd, M. ..........................................................................................................................................................193
Ridsdale, R. ...................................................................................................................................................1103
Rodgers, J.H., Jr...............................................................................................................................................545
Rohde, T.K.......................................................................................................................................................917
Rowlands, S.A. ................................................................................................................................................829
Runciman, D....................................................................................................................................................413
Sanborn, M. ....................................................................................................................................................339
Sandys, N. .......................................................................................................................................................461
Sanscartier, D........................................................................................................................................951, 1103
Sanzana, E. ........................................................................................................................................................81
Sato, S. ............................................................................................................................................................673
Save, M. ..........................................................................................................................................................353

Author Index
Sawatsky, L. ...................................................................................................................................................... xv
Schmid, B.J. ............................................................................................................................................. 263, 929
Schwarz, A. .....................................................................................................................................................507
Sexton, A. ........................................................................................................................................................517
Sharp, T.R. ......................................................................................................................................................275
Shaw, S.................................................................................................................................................... 561, 939
Shea, J. ............................................................................................................................................................965
Shearman, T. J.................................................................................................................................................327
Shuri, F. ...........................................................................................................................................................289
Shurniak, R. ...................................................................................................................................................1035
Sigda, J.M. .......................................................................................................................................................301
Siwik, R.S. ..........................................................................................................................................................59
Slingerland, N..................................................................................................................................................235
Smith, C.........................................................................................................................................................1011
Spencer, J. .......................................................................................................................................................545
Steenhof, P.A. .................................................................................................................................................157
Stewart, G. ......................................................................................................................................................131
Stirling, J..........................................................................................................................................................561
Straker, J. ............................................................................................................................................1023, 1035
Sulatycky, T. ....................................................................................................................................................471
Tallon, L.K. ............................................................................................................................................939, 1071
Tashe, N. .........................................................................................................................................................403
Theberge, D.J. .................................................................................................................................................609
Theisen, M.S. ..................................................................................................................................................221
Thiffault, E.....................................................................................................................................................1081
Thompson, M..................................................................................................................................................193
Thompson, R. ................................................................................................................................................1047
Thorpe, M.B. ............................................................................................................................................. 38, 121
Tibbett, M. ....................................................................................................................................................1047
Tibbett, M. ......................................................................................................................................................983
Tiemann, C.D. .................................................................................................................................................681
Tremblay, G.A. ..................................................................................................................................................14
Turea, V...........................................................................................................................................................413
Unger, M.L. ............................................................................................................................................. 315, 763
Urquhart, M.E. ................................................................................................................................................169
Van Hamme, J.D. ............................................................................................................................................106

Author Index
Van Zyl, A. .....................................................................................................................................................1011

Varris, P.A. ........................................................................................................................................................38
Venhuis, M. .....................................................................................................................................................561
Vitt, D. .............................................................................................................................................................837
Vladimir, Kh. ...................................................................................................................................................805
Waterman, L.P. .............................................................................................................................................1125
Wealleans, M. .................................................................................................................................................681
Weber, P. ........................................................................................................................................................965
Webster, M.S. ........................................................................................................................................... 23, 901
Weeks, B. ..........................................................................................................................................................81
Wells, C. ..........................................................................................................................................................423
Wen, M.E. ............................................................................................................................................... 375, 995
Whitlock, D.S. ...............................................................................................................................................1047
Wilkens, M.A.....................................................................................................................................................53
Williams, C. .....................................................................................................................................................413
Williams, J.B. ...................................................................................................................................................363
Wilson, G.W. ...................................................................................................................................................235
Wirtz, R.J.M. ...................................................................................................................................................169
Wolff, G.W.R. ..................................................................................................................................................995
Workman, D.C. .............................................................................................................................................1151
Zvarivadza, T. .................................................................................................................................. 585, 697, 775

PROCEEDINGS of The 10th International Conference On Mine Closure PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PROCEEDINGS of The 10th International Conference On Mine Closure PDF

Uploaded by

Copyright:

Available Formats

June 1−3, 2015  Vancouver, Canada

Andy Fourie, The University of Western Australia, Australia

Production and Editorial: Joe Goodwill, InfoMine Inc.

For more information, visit our website at http://www.infomine.com/.

 Tailings and Mine Waste (2011, Vancouver, Canada)

Please see http://www.infomine.com/conferences/ for a list of upcoming conferences.

Members, BC TRCR Representatives

Marketing and Sales Manager

Les Sawatsky, P.Eng., M.Sc. Michael Davies, P.Eng., Ph.D.

D.E. Hockley SRK Consulting (Canada) Inc., Canada

Mine Closure 2015, Vancouver, Canada 3

4 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 5

6 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 7

8 Mine Closure 2015, Vancouver, Canada

Content type Example content

Mine Closure 2015, Vancouver, Canada 9

10 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 11

M. Davies Teck, Canada

12 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 13

E.J. Gardiner NOAMI, Canada

14 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 15

16 Mine Closure 2015, Vancouver, Canada

As stated earlier, a principal recommendation of Cowan et al. (2010) was that:

Mine Closure 2015, Vancouver, Canada 17

18 Mine Closure 2015, Vancouver, Canada

An effective long-term monitoring and maintenance program at closed/post-closure, orphaned/ abandoned

Mine Closure 2015, Vancouver, Canada 19

20 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 21

22 Mine Closure 2015, Vancouver, Canada

D.G. Hovdebo Kingsmere Resource Services Inc., Canada

Mine Closure 2015, Vancouver, Canada 23

24 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 25

26 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 27

28 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 29

30 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 31

32 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 33

34 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 35

2. Protect the environment;

36 Mine Closure 2015, Vancouver, Canada

R. Knapp SENES Consultants, Canada

Mine Closure 2015, Vancouver, Canada 37

P.A. Varris Golden Star Resources Ltd., Ghana

38 Mine Closure 2015, Vancouver, Canada

GSWL, Hwini Butre

Mine Closure 2015, Vancouver, Canada 39

40 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 41

42 Mine Closure 2015, Vancouver, Canada

Mine Closure 2015, Vancouver, Canada 43

 long-term (50+ years) stable pit walls;

44 Mine Closure 2015, Vancouver, Canada