UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Integration of sustainability into the mine design process Odell, Carol Jane 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2004-0584.pdf [ 27.44MB ]
JSON: 831-1.0081057.json
JSON-LD: 831-1.0081057-ld.json
RDF/XML (Pretty): 831-1.0081057-rdf.xml
RDF/JSON: 831-1.0081057-rdf.json
Turtle: 831-1.0081057-turtle.txt
N-Triples: 831-1.0081057-rdf-ntriples.txt
Original Record: 831-1.0081057-source.json
Full Text

Full Text

INTEGRATION OF SUSTAINABILITY INTO THE DESIGN PROCESS By: CAROL JANE ODELL B. Sc. (Hons) Geological Sciences 1989. A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIRMENTS FOR THE DEGREE OF MASTER DF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Mining Engineering) We accept this thesis as conforming tojhe required standard UNIVERSITY OF BRITISH COLUMBIA August 2004 © Carol J. Odell 2004 THE UNIVERSITY OF BRITISH COLUMBIA FACULTY OF G R A D U A T E STUDIES Library Authorization In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Name of Author (please print)Carol J Odell J Q - t l A - S^epV- l O O <f-Date (dd/mm/yyyy) 07/09/04 Title of Thesis: Integration of sustainability into the mine design process Degree: Master of Applied Science Year: 2004 Department of The University of British Columbia Vancouver, B C Canada Mining Engineering ABSTRACT With the growing worldwide interest in corporate social responsibility and sustainable development, the mining industry is under increasing pressure to design, operate and close mining operations in accordance with the principles of sustainable societies This thesis investigates the hypothesis that a holistic mine design process, which integrates social, environmental and institutional criteria on an equal footing with the geological, engineering and economic criteria traditionally considered, can improve the sustainability outcomes o f a mine. The research generated a composite case study using borehole data and a geological model from a North American porphyry copper deposit and environmental, social and institutional data from the Peruvian Andes. The guidelines for the incorporation of sustainability concerns into engineering practice produced by the Association o f Professional Engineers and Geoscientists o f Brit ish Columbia ( A P E G B C ) , were then applied to the case study to assess 4 mine design scenarios for the mineral deposit at the scoping level using the multi-criteria assessment methodology. In all areas: economic, engineering, geological, environmental , social and institutional; profiling tools were located which enabled the production o f holistic baseline data. Economic and environmental predictive tools appeared adequate for holistic design with minor modifications. Social predictive tools produced useful insights into impacts requiring further analysis. These tools require participatory assessment, and can be improved by using tools which assess specific community characteristics. Institutional predictive tools were found to be the least adequate, although this is an area which is undergoing rapid development. Overall the research provided useful insights into sustainability issues for the case study and alternative designs which might increase benefits and mitigate impacts. It also demonstrated the importance of the decision structuring step to producing useful modeling outcomes. The results suggested that certain sustainability issues may be better incorporated into corporate level strategy rather than project level design, although both corporate and project level attention to sustainability are critical to producing outcomes. The prototype tools developed showed considerable promise for integrating sustainability concerns into the mine design process, which with further refinement and practical application could develop into integrated decision support tools for holistic mine design. i i ACKNOWLEDGEMENTS This thesis would not have been possible without the collaboration, support and input o f a multitude of fellow students, professors, industry professionals, mining activists, friends and others who contributed in many different ways to enriching my life and supporting my work. In this acknowledgement I hope to include everyone. Anyone who I missed, you know who you are and your contributions are greatly appreciated. Many individuals from mining and consulting companies provided valuable contributions to this research. Steve Botts, Rosa Ocana and Neyer Cerna o f the Antamina mining company in Peru facilitated field visits, shared their experiences and aided me in developing my understanding o f the impacts o f mining on Peruvian Andean communities. Sandy Laird , Ian Pond, Kei th Fergusson, and others at Placer Dome generously imparted much o f their wisdom concerning surface mine design. Joe Ringwald and Alistair Kent will ingly shared their many years o f achievements and struggles as mining engineers and their incredible enthusiasm for innovative improved mine design process with me, as we worked together to catalyze the mine design process o f tomorrow. Joe receives special thanks for agreeing to sit on my defense committee. Shannon Shaw and Andy Robertson provided valuable insights into both acid rock drainage and participative decision analysis processes, Robin Gregory aided my understanding o f the role o f decision analysis in improving communication and deliberation and Rabel Burdge and N i c k Taylor generously enabled me to assist at a workshop that was critical to the completion o f this thesis. Others who share the roles o f industry professionals and fellow students and guides in this research include Simon Handelsman, Diana Sollner, Sandy Sveinson, Gelson Batista and Anoush Bozorgebrahimi. I am grateful to Simon, Diana, Sandy and Gelson for answering questions and providing comments on versions o f this research and my heartfelt thanks go to Anoush for sharing his data with me and for helping to develop the initial research concept. Professors who inspired this research include Drs Ma lco lm Scoble and Marcello Veiga whose leadership in the development o f a Sustainability Working Group ( S W G ) provided me with a network o f talented academics and wonderful human beings with whom to share research ideas, too much beer and some wonderful friendships during my Masters studies. Their well thought out and challenging discussions, classes, patience and ongoing support through sometimes trying personal situations enabled this research to reach completion and their willingness to spend many hours reading through long drafts greatly improved its academic quality and integrity. D r Susan i i i Nesbitt inspired me with her enthusiasm and dedication to implementing sustainability in the engineering professions, her role model as a female professor in the technical professions, and her ability to provide discerning thought provoking comments supportively. I am grateful for the many hours she spent poring over drafts o f this thesis and providing insights. Dr. Mar io Mor in receives my heartfelt thanks for stepping into the fray at the last minute and agreeing to evaluate this thesis, when my defence date coincided with another committee member's unforeseen trip abroad. To my fellow students in the S W G - Y o u guys are awesome! M y Masters studies would not have been the same without you. Jennifer Hinton always managed to find time for me, whether it was through kindness to me and my family or through insightful comments on my writing. She is a great scholar and I appreciate her impressive depth and breadth of knowledge. Ginger Gibson encouraged and supported me during some o f the hardest times shared her wisdom concerning social issues in mining, her experiences in Peru and has my sincerest respect for always taking a stand for what she believes to be right. Mar ia Claudia Sandoval provided an inspiring model o f how sustainability concerns could be incorporated into mining and now demonstrates the possibility o f achieving this in the field. Silvana Costa always managed to make everything fun while retaining her commitment and integrity. Carolina Si lva knows so much about so many things and showed me the power o f perseverance in reaching goals. She also didn't k i l l me when I was late getting her books back and has a great wiggle! Sonia Aredes is a kindred spirit on a journey to a better place, I appreciate her courage, wisdom and company. Alis ta i r MacDonald worked with me on pioneering the S W G concept in the early days, Andrew Thrift provided thoughtful comments on my thesis drafts and continues to impress me with how much can be accomplished in a short time and A J Gunson, brings a wealth o f knowledge about politics and institutional issues and his ability to calmly and concisely articulate and facilitate conflicts was much appreciated on many occasions. M i k e McPhie inspires me to dream big and to value that which is truly important. Many others who work for improved sustainability performance in the mining industry have participated in the S W G and in workshops on sustainability in mining I have attended through the S W G . Through this work I have had the privilege o f meeting Ian Thomson, Tony Hodges, Susan Joyce, Rick K i l l am , A l l a n Young , Laura Baretto, Ben Bradshaw, Nancy Gibson, Gary MacDonald, and many other inspiring individuals who enriched my understanding o f mining and sustainability. Others in the department who have helped me on my journey include Javier Garcia who shared his perspectives on Peru, Enrique Rubio who shared his unique perspective on Latin American iv issues, Hasssan Ghaffari for his optimism and encouragement, Joe Hunter for assisting me in locating data and for provision of "reality checks" which strengthened my research. I am also grateful to Rowena Rae who was the first to blaze the sustainability path in the department, Don Hallbom for teaching me the power o f my own strength and o f trusting myself and Andrew Bamber for his directness and for his insights on mining in developing countries. M y research was aided financially by a C y and Emerald Keyes Scholarship and two Kitsualt Community Scholarships from the U B C mining department, teaching assistantships in Mine Management (B i l l Stanley) and Professional Engineering Practice (Malcolm Scoble and Scott Dunbar) and research assistantships with Marcel lo Veiga and Malco lm Scoble. In addition I was fortunate to receive three bursaries from the U B C Aimer Mater Society. I am grateful to all of these people for enabling my research. Outside the academic and mining worlds, Cindy Milner has been my unfailing and staunchest supporter who has never given up on me despite everything. I am grateful to Cindy for so many things not least o f which are making sure nothing important fell through the cracks, dealing with the picky details of reference editing and being an inspiration and guide in my life. K e l l y Martin, Loreleg Mur i l lo , L u c y Stone, A l i c i a Aredes, Andrew Edgar, Dav id Christopher, Rod Gaucher, Barbara Hinton, Andrea, Bryan and K e v i n Mi lne r have not only ensured that my children have been happy, healthy, disciplined and surrounded by friends as I toiled (and sometimes as I played) but have also become cherished friends in the process. Monique Koningstein is another friend who has journeyed with me for many years and whose support and joie de vivre I much appreciate. Joaquin Alvarez provided me with the opportunity to spend three years immersed in the culture of the Andes: an opportunity for which I w i l l always be grateful. Finally I am grateful to my children Brian, Susana, Cushto and Kevyn who ensured that I balanced my work with family, who were patient with me when dinner was late because I became engrossed in my work, who provide me with so much love and j oy and who are the reason why I continue to dream about and work towards a better future. This thesis is dedicated to my maternal grandmother, Elspeth Mary Aver i l l nee Marshall who dared to travel, pioneer, found a profession and live a life o f integrity in times when women rarely did these things. In her wisdom she shared with me not only what worked for her in her life but also what did not and it is through her example that I have learned to work for what is right and not to accept second best. v TABLE OF CONTENTS A B S T R A C T ii A C K N O W L E D G E M E N T S i i i T A B L E O F C O N T E N T S vi T A B L E O F F I G U R E S x v i i i L I S T O F T A B L E S xx G L O S S A R Y x x i i i C H A P T E R 1 I N T R O D U C T I O N 2 1.1 Statement o f the Problem 2 1.2 Outline of the Work 3 1.3 Significance and Contribution o f the Research 5 1.3.2 Significance to the Discipline o f Min ing Engineering 6 1.3.3 Significance to the Peruvian Context '. 6 1.4 Originality and Innovation 7 1.5 Organization o f the Thesis 7 C H A P T E R 2 L I T E R A T U R E R E V I E W 9 2.1 Introduction 9 2.2 Sustainability and Min ing 9 2.2.1 Sustainability 9 Social Issues in Sustainability 9 Advances in Sustainability Research 11 Governance and Sustainable Development 12 2.2.2 M i n i n g Engineering 14 The Historical Development o f M i n i n g Engineering 15 Mine Design Past 15 vi Mine Design Present 17 Min ing Practice through the lens o f Sustainability 20 Applications o f Sustainability Principles and Tools to Min ing : The Best Cases 22 Rationale for Integrating Sustainability Directly into M i n i n g Engineering ....25 2.2.3 Sustainability and Engineering Approaches 28 The A P E G B C Process: Multi-criteria analysis 28 2.3 Mul t i Criteria Analysis 30 2.3.1 The Decision Analysis Process 31 M C A in the spectrum of Decision Appraisal 32 Description o f M C A 33 Decision Context and Decision Structuring 33 Model ing : 36 Quantitative Analysis 36 Qualitative Analysis 37 Outputs and Use o f Results 38 2.4 Literature Review Summary 39 C H A P T E R 3 T H E C A S E S T U D Y A N D R E S E A R C H P R O C E S S 40 3.1 Introduction 40 3.2 Overview of the Case Study 40 3.2.1. Governance and Legal Structure 41 Legislation governing mining 41 Peruvian Governance 42 3.2.1 Community Profile 43 Spatial Distribution 43 Social Organization 44 v i i Attitudes, Beliefs, and Values 45 Lifestyles 45 Health 46 Economy 47 3.2.2 Environment 48 3.2.3 Geology 49 3.2.4 Engineering 49 3.2.5 Summary of Case Study 50 3.3 AppI ication of the A P E G B C Approach to the Case Study 51 3.3.1 Decision Context and Decision Structuring 52 Decision Context 52 Decision Choice 52 Decision Goals 53 Identification of Alternatives 53 Refining Objectives: Model ing the problem structure 54 3.3.2 Model ing of Indicators 56 3.3.3 Treatment of uncertainty, preferences and level o f analysis 58 Quantitative Analysis 58 Qualitative Analysis 59 3.4 Criteria Model ing 59 3.4.1 Economic Criteria 60 Project Economics 60 Assumptions for N P V calculations 60 Financing Considerations 61 Project Risks 62 Costs o f Sustainability Measures 63 v i i i Summary o f Economic Model 63 Socio-economic Distribution 63 Assumptions in the Model 64 N o mine scenario 66 Summary o f Socio-Economic Mode l 67 3.4.2 Social Criteria 67 S1A Methodology 68 Summary o f Community Well-being Model 71 3.4.3 Institutional Criteria 72 SI A Variables 73 Integration of Institutional Aspects into Socio-Economic Model 73 3.4.4 Environmental Criteria 73 Greenhouse Gas Production 74 Waste Generation Impacts 75 Tailings Generation 76 Waste Rock Impacts 76 Summary o f Environmental Model ing 76 3.4.5 Construction o f M C A Matrix 77 C H A P T E R 4. R E S U L T S 78 4.1 Economic Impacts: Results o f the Economic Models 78 4.1.1 Project Economics 78 Economic Performance 78 Risks and financing 79 Costs of Sustainability Measures 80 Weaknesses in the Economic Modeling Data 81 4.1.2 Distribution of Socio-Economic Benefits 81 ix Comparison o f Total Revenues Accruing to Each Stakeholder Group 82 Socially Discounted Distributions 83 Temporal Distribution of Revenues among Stakeholder Groups 84 Distribution o f Benefits by Revenue Source 86 Community versus Individual/Family Payments 87 Strengths and Weaknesses in the Socio-Economic Data 88 Summary o f Socio-Economic Results 89 4.2 Social Model ing: Results o f Community Wellbeing Mode l 90 4.2.1 Construction Impacts Summary for Construction Towns 91 4.2.2 Social Impacts for the Transition to Operations 92 4.2.3 Mine Closure Impacts 94 4.2.4 Summary o f Community Wellbeing Model ing 96 4.2.5 Weaknesses in the S1A modeling 96 4.3 Institutional and Governance Model ing 96 4.3.1 Institutional Social Assessment Variables 97 Variable # 15 Presence o f an outside agency 97 Variable # 16 Inter-organizational Cooperation 97 Variable #8 Alteration in Size and Structure of Government : 98 Trust in Political and Social Institutions (Freundenberg, 1999) 98 4.3.2 Institutional Impacts on Socio-Economic Projections 98 4.3.3 Summary of Institutional Model ing 99 4.4 Environmental Model ing 100 4.4.1 Greenhouse Gas Production 100 4.4.2 Waste Production Impacts 103 Tailings 103 Waste Rock 103 x 4.4.3 Additional Environmental Criteria 104 Geochemical Considerations 105 4.43.2 Ecosystem Resilience 105 Water Management Considerations 106 Intergenerational Equity 106 Life Cycle Considerations 107 4.5 Performance Matrix and Decision-Analysis 108 4.5.1 Quantitative Data Processing o f the Mult i -Cri ter ia Performance Matr ix 110 4.5.2 Qualitative Analysis o f the Multi-Criteria Performance Matr ix 113 4.5.4 Summary of the application o f M C A to the case study 116 C H A P T E R 5. D I S C U S S I O N A N D C O N C L U S I O N S 117 5.1 Summary o f the Research 117 5.2 Discussion 118 5.2.1 Insights from building and profiling the case study 119 5.2.2 Insights from applying the prototype mine design tools to the case study 119 Decision and criteria structuring 119 Insights from the criteria modeling 120 Insights from the multi-criteria matrix and data processing 123 Reliability of Results 123 5.2.3 Insights into sustainable mining at the case study 124 5.2.4 Assessment of the prototype holistic mine design package 126 Constraints on the Process 126 Effectiveness in representing sustainability concerns 127 5.2.5 Future Applications 128 1. Participatory Involvement 128 2. Interdisciplinary Teams 129 3. Encourage and coordinate with Regional Development Plans 129 4. Use Strategic Sustainability to Guide Process 130 5.3 Conclusions 131 5.4 Future Directions for Research 133 5.5 Concluding Remarks 134 R E F E R E N C E S 137 A P P E N D I X I: Contents of Min ing Scoping Study 160 A P P E N D I X II: Details o f Secondary Criteria Selection Process 163 II. 1 Selection Methodology 163 A P P E N D I X III. Overview of the Case Study 166 AI I I . l Governance and Legal Structure 166 AIII. 1.1 Macroeconomic Issues 166 AII I . l .2 M i n i n g Legislation 166 AIII. 1.3 M i n i n g Taxation 167 AII I . l .6 Other Organizations Involved in M i n i n g 171 AIII.2 Community Profile 171 AIII.2.1 Spatial Distribution o f People 172 AIII.2.2 Social Organization 173 AIII.2.2.1 Loca l governments 173 A11I.2.2.2 Community Organizations 174 AIII.2.2.3 Gender in Andean Culture 174 AIII.2.3 Lifestyles 175 AIII.2.3.1 Farming and the Land 175 AIII.2.4 Attitudes, beliefs and values 177 AIII.2.4.1 Indigenous Peoples in the Andes 178 AIII.2.5 Health 178 xii AIII.2.6 Economy 1 8 0 AIII.2.6.1 M i n i n g in the Andes • 180 AIII.3 Environmental Mode l 181 AIII.3.1 Ecosystem 181 AIII.3.2 Livestock 182 AIII.3.3 Erosion I 8 2 AII1.3.4 Ecosystem Services 183 AIII.3.5 Climate 183 AIII.4 Geology, Geochemistry and Geological (Block) Model 184 AIII .4.1 Geology I 8 4 AIII.4.2 Geochemistry I 8 5 AIII.4.3 3 Dimensional Block Model 186 AIII.5 Engineering Mode l I 8 8 AIII.5.1 M i n i n g Method Determination I 8 8 AIII.5.2 Processing Method 1 8 9 AIII.5.2 Cut-Off Grade I 9 0 AIII.5.3 Mine Life 191 AIII.5.4 Other M i n i n g Parameters 191 A P P E N D I X I V . Stakeholder Preferences 194 A P P E N D I X V . Detail o f Calculations: Project Economics 196 A V . l . C a p i t a l Costs 196 A V . 2 Infrastructure and Labour Costs 196 A V . 3 Operating Costs and Revenues 198 A V . 4 Discount Rate I 9 8 A V . 5 Project Risks I 9 9 A V . 6 . 1 Market Risk: Commodity Prices I 9 9 x i i i A V . 6 . 2 Deposit Risk: Geological Uncertainties 199 A V . 7 Costs of Sustainability Measures 200 A P P E N D I X V I . Detailed Calculations: Socio-Economics 201 A V I . l Model ing Taxation Levels and Canon Redistribution 201 A V I . 2 Modeling development spending by companies 202 A V I . 3 Modeling Direct Employment 203 A V I . 3.1 Salaries 204 A V I . 3 . 2 Residence and ski l l levels of employees 204 A V I . 3 . 3 Model ing Mine Construction and Closure 206 A V I . 4 Model ing Contracting and Procurement 208 AVI .4 .1 Model ing of preferential procurement opportunities 209 A V I . 5 Compensation for Land 209 A V I . 6 Extrapolating the Pre-Mine Scenario 210 A V I . 7 Post M i n e Situation 212 A V I . 8 Synthesizing the Socio-Economic Mode l 213 A P P E N D I X VI I . Detailed Calculations: Social Criteria 214 A V I I . l Identification o f Stakeholders 214 AVII .3 Construction 216 AVII .3 .1 Population Impacts 217 Variable #2. Influx and Outflux of Workers 217 Variable #5. Assessing the Dissimilarity in Age , Gender, Racial and Ethnic Composition. 218 Variable # 6. Project Attitudes and Responses 218 Variable #24. Perceptions of Public Health and Safety 219 AVII .3 .2 Community and Cultural Impacts 219 Variable #11. Living/Family wage 219 xiv Variable # 12. Enhanced economic inequities 219 Variable #13. Changes in employment equity for minorities 220 Variable # 14. Changes in occupational opportunities 220 Variable #18. Change in commercial/industrial focus o f the area 221 Variable # 22. Alteration in family structure 221 Variable # 26 Change in Community Infrastructure 221 A V I I . 4 Transition to Operational Stability 222 Variables # 2 and 3. Population Change and Influx and Out flux of Workers 225 Variable # 4. Relocation o f Individuals and Families 225 Variable # 5. Dissimilarity in age, gender, racial or ethnic composition 225 Variable # 6. Formation o f attitudes towards the project and # 7. Interest group activity226 Variable #10. Industrial Diversification : 226 Impact variable #11. Living/Family wage 227 Variable # 12. Enhanced economic inequities 227 Variable #13. Assessing change in employment equity o f minority groups 228 Variable #14. Changes in occupational opportunities 228 AVII .4 .3 Transition Variables 229 Variable #17. Assessing the introduction of new social classes 229 Variable #18. Change in commercial/industrial focus o f the area 230 Variable # 20. Disruption in daily l iving and movement patterns 230 Variable #21. Dissimilarity in Religious and Cultural Practices 231 Variable #22. Alteration in Family Structure 231 Variable #23. Disruptions in Social Networks 231 Variable #26 Changes in Infrastructure 231 A V I I . 5 . Transition to Closure 232 AVII .5 .1 Population Variables 233 xv Variable # 1. Population 233 AVII .5 .2 Community Variables • 2 3 3 Variable #10. Industrial Diversification 233 Variable #11. L iv ing Wage Jobs -234 Variable # 12. Changes in employment equity 234 Variable # 17 Disappearance o f social classes 235 AVII.5.3 Individual Family impacts 235 Variable # 26. Changes in Community Infrastructure 235 A P P E N D I X VIII . Detailed Calculations: Institutions 236 AVIII.1.1 Variable # 15. Outside Agency: 236 A VIII . 1.2 Variable #16. Inter-organizational cooperation 236 A VIII . 1.3 Variable #8. Alteration o f size and structure in local government 236 A VIII . 1.4 Variable #9 Planning and Zoning Activi ty 237 A VIII . 1.5 Freundenburg's Variables 237 AVIII .2 Integration of Institutional Aspects into Socio-Economic Mode l 237 A P P E N D I X I X Detailed Calculations: Environment 239 A I X . l Greenhouse Gas Emissions 239 AIX.1 .1 Methodology 239 A I X . l .2 Scoping / Boundaries o f Operations 240 A I X . l . 3 Assumptions 241 A I X . l . 4 Calculation o f Individual Emissions 243 AIX1.4.1 Emissions from Concrete (Cement) 243 A I X . l . 4 . 3 Transportation 245 Construction Transportation 245 Operations Transportation 246 Trucks and Shovels 247 xvi A I X . 1.4.4 Electricity Consumption 249 A I X . 2 Impacts on Waste Production 250 AIX.2 .1 Additional Tailings Production 250 A1X.2.2 Min ing o f Additional Waste Rock 251 A P P E N D I X X . Construction webbing and chaining diagram 253 xvii TABLE OF FIGURES Figure 1-1 Research Concept 4 Figure 2-1 Mine Design Past 17 Figure 2.2 Mine Design Present 18 Figure 2-3 Placer Dome Mine Design Process (Placer Dome, 1999) 20 Figure 2-4 The 7 questions to sustainability from Hodges (2003) 27 Figure 2-5 Decision Analysis Process from Clemen & Rei l ly (2000) 31 Figure 2-6 Incorporation of new data into the model 35 Figure 3-1 The areas of influence of the Antamina mine after Botts (2003) 44 Figure 3-2 Employment profile for Andean settlements data PWE1 (2004) - 1992 data 47 Figure 3-3 Influence diagram for sustainable outcomes at the case study 57 Figure 3-4 Risk in the Minerals Industry 62 Figure 3-5 List of SIA variables from Burdge (2004) 71 Figure 4-1 Payback Time for Scenarios A - C ($1.10 Copper, Discounted at 15%) 79 Figure 4-2 Annual Average Copper Prices From 1955 to present from Perez, 2004. 80 Figure 4-3 Revenue Distribution Discounted at 15% 82 Figure 4-4 Local revenues for training and non training scenarios (15% discounted). 83 Figure 4-5 Local revenues for training and non-training scenarios (5% discounted). 84 Figure 4-6 Cash Flows for Scenario B (35 years) 85 Figure 4-7 Revenue Sources at the National Level (15% Discounted) 86 Figure 4-8 Revenue Sources at the Local Level (5% Discounting) 87 Figure 4-9 Temporal salary and development investment for local community.. . 89 Figure 4-10 Revenues to Local Communities accounting for leakage 98 Figure 4-11 Greenhouse Gas modeling for mining scenarios 101 Figure 4-12 Annual Greenhouse Gas Emissions for the Scenarios 102 Figure 5-1 A Potential Future Mine Design Process 128 Figure AIII-1 Grade Tonnage Curve for the case study deposit 187 Figure AIII-2 F low diagram of copper processing (Chevron Phillips Chemicals, 2001). 189 Figure AIII-3 Formula for Cut off Grade 190 Figure AIII-4 Taylor mine life formula (Hustralid and Kutcha, 1995) 191 Figure AIII-5 M i n e layout for the case study 192 Figure AVI -1 Investment in Rural Development by the Yanacoccha Mine (Yanacoccha, 2002) 203 Figure A V I - 2 Native and Northern Hire rates for Saskatchewan Uranium Mines, (Parsons and Barsi , 2001) 205 Figure A V I - 3 Employment curves for the case study: scenarios A , B and C 206 Figure A V I - 4 Manpower requirements at the Antamina mine (Antamina, 1998) 207 Figure AVII -1 Stakeholders after Cooney (2001) 214 Figure A V I I - 2 K e y construction social impacts 216 Figure ALX-1 Greenhouse Gas emissions scoping diagram for the case 241 Figure A I X - 2 Extrapolation of Fuel Consumption Data for 380 Tonne Truck 248 Figure A I X - 3 Extrapolation of Fuel Consumption for 47 and 68 cubic yard shovels 248 Figure A I X - 4 Geometry for Ramp Widening from Bozorgebtrahimi (2003) 252 Figure A X - 1 Construction webbing and chaining diagram 253 x ix LIST OF TABLES Table 3-1 Blank M C A Matrix 52 Table 3-2 Engineering assumptions for the 3 mining scenarios 54 Table 3-3 Criteria, indicators and measures 55 Table 4-1 Financial Outcomes for the M i n i n g Scenarios 78 Table 4-2 Costs of Training Programs 81 Table 4-3 Summary o f Socio-Economic Distribution Data 90 Table 4-4 Results o f SIA for construction impacts at construction towns 91 Table 4-5 Summary o f Social Impacts in the Transition to Operational Stability 93 Table 4-6. Summary o f Social Impacts o f Closure 95 Table 4-7 Summary o f Institutional Model ing 100 Table 4-8. Tailings Production Impacts o f the Scenarios 103 Table 4-9 Extra Waste Production Results - Waste Rock 104 Table 4-10. Addit ional Criteria Identified through Sustainability Frameworks 104 Table 4-11 Mult i -Cri ter ia Performance matrix for Scenarios for the case study 109 Table 4-12 Summary o f M C A Performance matrix - weighted evenly I l l Table 4-13 Summary o f M C A Performance Matr ix - sustainability weighting 112 Table 4-15 Trade-offs concerning employment in the mining operation 113 Table 4-16 Possible solution to trade-offs matrix for employment/training 114 Table 4-17 Trade-offs concerning construction impacts on women and families 114 Table 4-18 Trade-offs concerning relocation 115 Table 4-19 Trade-offs for environmental issues 115 Table AI-1 Contents for Engineering Studies (Pincock, Al l en and Holt , 2002) 160 Table AII-1 Results o f Thematic Analysis o f Comprehensive Sustainability Criteria. 163 xx Table AIII-1 World Bank Governance Indicators for Peru 169 Table AIII-2 Ecosystems in the Peruvian Andes from Zoomers (1998) 176 Table AIII-3 Cumulative tonnages at different cut off grades for the case study deposit .. 188 Table A I V - 1 Stakeholder M i n i n g Issues in Peru (Glave & Kuramoto 2002) 193 Table A V - 1 Capital Costs for Scenarios at the case study ( A l l figures in $US) 197 Table A V I - 1 Effective Taxation and M i n i n g Canon Revenues 201 Table AV1-2 Compensation for Land 210 Table A V I - 3 Modeling the Value of Subsistence Sheep pasturing 212 Table AV1I-1 Construction worker populations 217 Table AV1I-2 Dissimilarity in Age , gender and racial composition 218 Table AVII -3 L iv ing wage jobs from construction 219 Table A V I I - 4 Enhanced Economic Inequities for construction 219 Table A V I I - 5 Changes in Employment for Minorities 220 Table AV1I-6 Change in Occupational Opportunities in Construction Towns 220 Table A V I I - 7 Family structure o f construction workforce and local community . . . . 221 Table A V I I - 8 Changes in community infrastructure needs 222 Table A V I I - 9 Social Impact Variables for Transition to Operational Stability 224 Table AVII -10 Population Impacts for Transition to Operational Stability 225 Table AVII-11 Demographic comparison for construction 226 Table AVI I -12 Demographic comparison for Operations 226 Table AVII -13 Percentage o f Campesino community members oriented to mining... 227 Table AVII -14 Years to achievement of shift towards mining at local town 227 Tab l eAVI I -15 Reduction in Ci ty unemployment levels from mine 227 Table AV1I-16 Changes in Employment Equity 228 Table AVII -17 Change in occupational opportunities 229 xxi Table AVII-18 Time to significant occupational opportunities 229 Table AVII -19 Calculations for N e w Social Classes in the Local Town 229 Table AVII -20 Disruptions in Dai ly L i v i n g 230 Table AVII-21 Disruptions in Networks 231 Table AVI1-22 Infrastructure Needs for Local City 232 Table AVI1-23 Social Impact Variables at mine closure 233 Table AVII -24 Population changes at mine closure 234 Table AVII -25 Loss of L iv ing Wage at M i n e Closure 234 Table AVII -26 Demographics o f Job Loss 235 Table AVII -27 Infrastructure needs change at closure 235 Table AVIII-1 Approximate haul road widths for various truck sizes 252 xxii GLOSSARY Terms Mining Engineering - discipline dealing with the application o f scientific knowledge to the practical problems o f obtaining minerals and mineral extracts o f value to society from naturally occurring deposits. Mine Design - process o f determining the criteria and targets for a mining operation and an overarching strategy for achieving the targets. Mine Planning - process of determining how a mine should be operated, scheduled and managed in order to meet design and /or production targets. Baseline Data /Profile - basic information gathered before a project commences which includes key information for projection o f impacts and for later comparison. Value - economic measure o f the worth o f something Values - Principles, standards or qualities considered worthwhile or desirable. Criterion (Criteria) - standard(s) used to judge a decision. Design Criteria — specific set of criteria targets that a mine design aims to achieve. Measure - quantified standard for the comparison o f criteria. Indicator - an aspect o f project performance to be measured. Variable - aspect o f a mining project which is prone to change and can be measured. Scenarios - alternative descriptions of the future, which focus on the forces driving change and the critical uncertainties leading to different possible future outcomes. Methodology - body o f practices, procedures, and rules used by those who work in a discipline or engage in an inquiry; a set of working methods. Campesino Community - settlement type unique to Peru officially created during land reforms in the early 1970s, although many communities have much earlier origins. Campesino Communities x x i i i typically consist of 50 to 500 families who own communal land and whose livelihood is gained through agriculture and subsistence activities. Canon - tax redistribution mechanism mandated in Peruvian law whereby 15% of the taxation revenues are distributed to local and regional governments for spending on infrastructure for poverty alleviation. Living Wage - amount o f money required for a family o f four to live above the poverty line in the community or region where they reside. Acronyms MCA - multi-criteria analysis FDIC- International Federation o f Consulting Engineers MMSD - Min ing Minerals and Sustainable Development Project IIED - International Institute for Environment and Development USD - International Institute for Sustainable Development i V P F - N e t Present Value IRR - Internal Rate o f Return NGO - N o n Governmental Organization CSO - Community Service Organization APEGBC-The Association o f Professional Engineers and Geoscientists o f British Columbia FIDIC - International Federation of Consulting Engineers CBA - Cost Benefit Analysis EIA - Environmental Impact Assessment PAHO- Pan American Health Organization MEM- Peruvian Ministry o f Energy and Mines DGM - General Min ing Directorate (Peruvian Minis try o f Mines) INC- Peruvian National Institute of Culture INRENA - Peruvian Ministry o f Natural Resources xxiv INEI- Peruvian National Institute o f Statistics SUNAT- Peruvian Revenue Agency CONAM-Peruvian National Environmental Agency SNMPE-The Peruvian National Min ing , Petroleum and Energy Association IIMP- The Peruvian Institute o f Min ing Engineers (Professional Association) CONACAMI- Peruvian association of communities impacted by mining WRI- Wor ld Resources Institute WBCSD - World Business Counci l for Sustainable Development XXV "We can't solve problems by using the same kind of thinking we used when we created them." Attributed to Albert Einstein (1879-1955). "A problem well put is half solved" attributed to John Dewey (1859-1952). xxvi CHAPTER 1 INTRODUCTION 1.1 Statement of the Problem M i n i n g has historically supported the emergence of civilization through fostering settlement, producing essential raw materials, and creating infrastructure, and economic growth. In its modern-day incarnation, mining increasingly occurs in isolated rural areas, where it is still heralded as a vehicle for community and regional development. A l l too often, however, mining has failed to achieve its full potential to catalyze sustainable development. Frequently, the development that has accompanied mining operations has been characterized by instability ("boom and bust" cycles), social conflicts and revenues which have failed to generate long term benefits or have accrued to distant investors and central governments. Min ing has tended to leave "ghost town" legacies and local communities have inherited contaminated water and land at the end o f operations (Ross, 2001). In a world concerned with fulfilling the fundamental prerequisites for sustainable societies, described by Francis (1999) as ecological sustainability, economic vitality and social equity, these outcomes are increasingly criticized and the mining industry has been under considerable pressure to improve social and environmental outcomes. Industry leaders endorse mining which contributes to sustainable societies (Peeling, 2004). A valuable model for mine development which meets the criteria for sustainable societies is the "sustainable mining community model" o f Veiga et al. (2001). In this model, the community realizes "a net benefit from the introduction o f mining that lasts through the closure o f the mine and beyond" (Veiga et al. , 2001 p. 191). The natural capital o f the mineral resource is converted into financial capital for investors, wages for miners and the utility value o f the services that the commodities provide, in a similar manner to a traditional mine. However, in the sustainable mining community model, attention is paid to minimizing long term social and environmental costs and maximizing gains. Part of the revenue from the mining operation is invested in physical infrastructure and capacity building for local and regional communities and in the development o f effective institutions and informal social structures. Significant negative long term environmental legacies are also avoided. When the mine closes, the local and regional communities have developed skills, institutions and economic activities which wi l l sustain them into the future. They can thrive with these benefits and without the burden of an environmental disaster. A s humanity's understanding o f ecosystems, societies and cultures improves, an integrated understanding of the implications of technological development on ecosystems, community 2 health and wellbeing and the interface with organizational institutions is increasingly possible. This knowledge can be applied predictively to design mines with a better likelihood of creating sustainable mining communities. The mining industry has learned (often through bitter experience) that the prevention of problems is a far more cost effective strategy for handling environmental and social problems than end of pipe measures for environmental problems and "fire fighting" in situations of social conflict (I1ED, 2002). Temporary and early closure of a number of mining projects due to community protests, related to unsatisfactory social and environmental performance of operations, have cost mining companies millions of dollars (Placer Dome, 2002). The unforeseen consequences and incidents resulting from inadequate anticipatory design have demonstrated the wisdom of a proactive approach over a reactive one. The industry lacks experience with a more holistic mine design process. It seeks new tools and more interdisciplinary capacity to account for the relationships between the policies and practices employed by mining companies, the health and wellbeing of the surrounding community and the regulatory environment in which mining takes place. However, despite undesirable outcomes in the past, the mine design process continues to focus on technical mining and financial considerations with environmental and social objectives considered later in the mine design sequence, unfortunately more in the form of impact mitigation. The tendency is therefore to treat social and environmental concerns as inflexible elements of the mine which require mitigation, rather than as holistic design criteria which should be integrated from the outset to produce sustainable results. 1.2 Outline of the Work Mine design1 typically aims to meet a set of criteria, which are traditionally technical (geological and engineering) and financial. It seeks to create a conceptual understanding of how the deposit should be mined, processed and transported and the economic implications of these considerations. The research presented here investigates the hypothesis that a holistic mine design process, which integrates social, environmental and institutional criteria on an equal footing with the geological, engineering and economic criteria, can improve the sustainability outcomes of the mine. Specifically the research aimed to: 1 Mine design is the initial portion of the process of designing a mine. It is distinguished from mine planning which concerns how the different aspects of the mine will be operated and scheduled. 3 1. Build a holistic baseline model of a case study mineral deposit. 2. Create a prototype holistic design assessment process. 3. Apply the prototype process to the case study in a pilot test. 4. Reflect on the insights about the case study gained through the design process. 5. Assess the effectiveness of the holistic assessment with respect to facilitating sustainable mining development. In order to achieve these aims a composite case study was developed using a stochastic ore body model from a North American mineral deposit and situating the deposit in an analogous geological setting in the Peruvian Andes. This South American region was selected both because it is a significant area for Canadian mining investment (MacDonald, 2002) and because of the researcher's familiarity with Peru. The ore body grades were tripled so that the mineral deposit was comparable to deposits currently in production in the Andes. Models were also created of the environmental, socio-cultural and institutional settings of the mineral deposit. Different mining scenarios were then created for the ore body using surface mining technologies. Despite being based on a composite case study, the research aimed to provide a close parallel to reality by using block model data from a real geological resource and following through standard mine design decisions in the engineering and economic portions of the design process model. In Figure 1-1, the ring of circles represents the different aspects of the mining project to be profiled, modeled and extrapolated in order to gain an integrated understanding of the potential Figure 1-1. Holistic Mine Design Concept. i m P a c t s o f t h e proposed mine design. The central circle indicates that each of these aspects are incorporated into the initial mine design. The inclusion of all of 4 these aspects during the scoping analysis is seen as a step towards more holistic, interdisciplinary and proactive mine planning, which has the potential to develop more responsible mines which contribute to sustainable societies. The approach is in contrast to current approaches which include a limited assessment of environmental and social criteria before a decision is made to proceed with the project (typically at feasibility (+5% to - 3 % accuracy). A t this point a linear set o f environmental and sometimes social impact assessment processes and approvals is triggered, which assess the potential impacts o f the completed design and suggest areas requiring impact mitigation. The prototype holistic mine assessment process developed in the research facilitated a multi-criteria assessment o f four different case study scenarios generated at the scoping level (i.e. capital and operating cost estimation with accuracy o f ± 3 0 % ) in the mine design process. Together with the engineering and geological models, the socio-cultural, institutional and environmental profiles o f the project were used as a basis for modeling the economic, socio-economic, community wellbeing, institutional and environmental aspects o f each scenario explored. The scenarios were limited to one decision point in the scoping stage o f the mine design process. The decision chosen was the selection o f a mine life for the deposit, which in turn dictated the scale o f operations and hence the size of equipment selected, necessary infrastructure, workforce employed and many other items. Overall the thesis presents initial building blocks for a potential future mine design process. This early work aims to provide insights and stimulus for other researchers and mining practitioners. 1.3 Significance and Contribution of the Research The rationale underpinning the research is that the integration o f social and environmental considerations into a more interdisciplinary and integrated mine design process w i l l result in mining operations which have a better likelihood of fulfilling the fundamental prerequisites for sustainable societies. Including social and environmental considerations on an equal footing with technical and economic considerations is proposed as a step toward ensuring that mining operations deliver on their long heralded potential to catalyze sustainable regional development as well as to guarantee that mines with significant negative social and environmental impacts remain undeveloped. The data produced may also support the generation o f regional strategies to offset the boom and bust cycles associated with mining as an industry dependent on fluctuating commodity prices. A s Long & Fai l ing (2002a) state, "it is probably a truism that practically all technology-related environmental and social problems arise as unintended consequences o f actions aimed at meeting some other need" (p.6). Providing decision-makers with quantified 5 information concerning the potential impacts and benefits of a proposed action at an early stage should at least limit these unintended consequences while providing the potential to enhance benefits and mitigate impacts. This research provides a tentative prototype for the future mine design process. Since this prototype process is applied to a case study, it is not limited to a conceptual overview. Rather it attempts to provide practical insights into potential alternative approaches toward mine design and into the utility o f the specific tools and methodologies that are part of the process. Interdisciplinary cooperation is critical to the approach which, in its fully developed form, is envisaged as involving a group o f individuals from varied disciplinary backgrounds and with depth and breadth o f expertise working together in an integrated participatory 2 planning process. 1.3.2 Significance to the Discipline of Mining Engineering The research is significant because the current practice of mining engineering considers the impacts of mine design decisions on the environmental and social systems in a very limited way in the design process, rather than as integral elements o f design. Decisions are based primarily on financial considerations. The industry is, however, starting to realize that imposing decisions upon local communities, that are made based on financial criteria, can have very costly implications when the impacts of these decisions on local people and environments are not considered, predicted, discussed and/or mitigated in a participatory manner (I1ED, 2001). Prominent leaders in the mining industry have begun to talk o f the importance of gaining and maintaining a "social license to operate" (Peeling, 2003). A s this realization grows within the industry, tools need to be developed to support a mine design decision-making process which considers holistic design criteria. This thesis explores one potential format for the development o f these tools, applies it to a composite case study, and provides insight into the challenges and opportunities involved in the approach. 1.3.3 Significance to the Peruvian Context The research is also significant because o f the Peruvian context o f the case study. Peru is a country which has seen a noteworthy increase in mining and exploration activity; from 10 mi l l ion hectares in mining concessions in 1991 to 34 mil l ion at the end o f 2000 (Damonte et al., 2002; Glave & Kuramoto, 2002; Pasco Font et al., 2001). Although Peru is a country with a long history 2 Participatory refers to the active involvement of representatives of stakeholders / communities of interest for the mining project (Kent et al., 2004). 6 of mining activity, the magnitude of increased exploration is unprecedented and has resulted in considerable conflict between mining companies, local communities, non-governmental organizations and government institutions (Damonte et al., 2002). Several projects have been abandoned at the exploration stage or because permitting was significantly delayed as a result o f community protests (TeckCominco, 2002). The demonstrated ability o f local communities in Peru to thwart potential mining projects, even at the exploration stage, denotes a need on the part o f mineral exploration and development companies to address community concerns from the earliest stages of project development. A case study o f integration of potential community concerns into the mine design process, including an analysis of institutional constraints, provides valuable guidance respecting the obstacles and opportunities for sustainable development in the Peruvian context. In addition, several of the tools developed in the research, while specific to Peru, also have the potential to be applied in other countries with appropriate modifications. 1.4 Originality and Innovation This research is original because it experiments with the application o f multi-criteria analysis to economic, social, institutional and environmental variables in order to more holistically assess the potential for a mining project to produce sustainable outcomes for various stakeholders. M u l t i -criteria analysis has been applied to mining operations to site tailings facilities and plan reclamation (Robertson & Shaw, 1998, 1999). The methodology has also been applied to site selection, power generation method, waste cleanup and storage and pollution control decisions for electric utilities (Clemen & Rei l ly , 2001) and to the agricultural industry (El-Swaify & Yakowitz , 1998). A detailed review of the literature has not revealed an application o f multi-criteria analysis to a mine design. 1.5 Organization of the Thesis The thesis is organized into 7 chapters, including: Chapter One - Introduction. This provides a brief justification o f the research, identifies its purpose and objectives, and outlines the information contained in the thesis. Chapter Two - Literature Review. This chapter reviews the literature concerning sustainability, mine design engineering, and sustainability approaches to engineering focusing on multi-criteria analysis 7 Chapter Three - The Case Study and Research Process. This chapter consists o f a summary o f the case study, a description o f the decision path used in applying multi-criteria analysis in the research and a summary of the criteria modeling applied. Details of the case study and the modeling calculations are contained in Appendices. This chapter comprises the specific research aims 1 through 3. Chapter Four - Results. This chapter details the results o f the criteria modeling and assembles these results into an exploratory initial multi-criteria analysis. This chapter encompasses the results o f research aim 3. Chapter Five - Discussion and Conclusions. This chapter discusses insights acquired from: • Bui ld ing the holistic baseline model o f the case study mineral deposit, • A p p l y i n g the prototype holistic design assessment process to the case study, • Reflecting on the insights about the case study gained through the design process • Assessing the effectiveness o f the holistic assessment with respect to facilitating sustainable mining development. • Considering potential future applications o f the holistic mine design approach. A s such it includes the specific research objectives 4 and 5. The chapter concludes with a discussion o f the research hypothesis and specific aims and an identification o f gaps in the literature and opportunities for future study. 8 CHAPTER 2 LITERATURE REVIEW 2.1 Introduction This chapter examines the conceptual frameworks defining the research. The first section is an overview o f sustainability and focuses on the principles and tools of sustainability along with their strategic implementation in business situations. The second section, titled sustainability and mining, begins with a brief historical overview of the evolution of mining engineering. The section continues with an assessment o f the synergies and tensions between mining and the concept of sustainability and identifies current best practice. A final section of the chapter focuses on approaches which have been suggested for the incorporation o f sustainability concerns into engineering practice. The majority o f this final section consists o f a review o f the multi-criteria analysis methodology. 2.2 Sustainability and Mining 2.2.1 Sustainability Sustainability is a term that gained currency in the late 1980s and early 1990s following the publication of the Brundtland Report and the convening o f the Earth Summit in 1992. The most commonly cited definition of sustainability is "development that meets the needs o f the present without compromising the ability o f future generations to meet their own needs" (Brundtland, 1987). A second broad definition of sustainability was stated by the Canadian National Round Table on Environmental and the Economy as "parallel care and respect for the ecosystem and for the people wi th in" (Cormick et al. , 1996). A definition that is particularly applicable to engineering is described by A P E G B C in their sustainability approach as "design for Human and Ecosystem Wellbeing" (Hodges 2003, (Draft)). Social Issues in Sustainability One o f the key findings o f the Brundtland report was that poverty was a key cause o f unsustainable development and that attention to poverty eradication and equity were essential i f sustainable development was to be achieved (Brundtland, 1987). The social changes resulting from development have been qualified as direct (resulting from social interactions o f the project) or indirect (resulting from changes in the biophysical environment) (Slootweg et al. , 2001). Therefore, by focusing on both environmental and social aspects, environmental sociologists have 9 described 'ecological synthesis' as a desirable goal for society, where ecological disruption is minimized by controls over both production and consumption o f goods and resources are managed for sustained yields (Carton and Dunlap, 1978; Snaiberg, 1975). A t the same time they recognize that most current development falls either into an economic model, where growth is maximized and ecological impacts are largely ignored or else into a managed scarcity synthesis where only the most obvious consequences o f resource use are managed (Carton & Dunlap, 1978; Taylor et al. 2004). When the unwanted social consequences o f development are not explicitly addressed in proposals, adverse impacts are inevitable and are particularly evident in impoverished sectors (Taylor et al. 2004). This observation has led to the development of a growing body of research which focuses on factors which seem to confer resilience or vulnerability on communities in the face o f rapid development and change. In general, these theories recognize that the vulnerability o f a community to adverse impacts resulting from development activity (or another external force 3) is dependent both upon the severity o f the environmental changes brought about and upon the specific characteristics o f communities which enable or impede the adaptation to new circumstances (Kasperson & Kasperson, 2003 4). These theories, which are denominated "vulnerability theories", inform the sustainable livelihoods development model of the Brit ish Department for International Development (DF1D, 2000). In the sustainable livelihoods model, livelihoods, defined as "the capabilities, assets (including both material and social resources) and activities required for a means of l iving . . . are seen as sustainable when people can cope with and recover from stresses and shocks and maintain or enhance their capabilities and assets both now and in the future, while not undermining the natural resource base." (DFID, 2000). The sustainable livelihoods and vulnerabilities approaches show promise for improving knowledge on the intersection between development activities and community responses and impacts. 3 Much of the research on vulnerability has been directed to understanding vulnerability and resilience in the face of "natural" disasters. 4 The vulnerability theories vary in their foci and in their intended purposes and include: capacities and vulnerabilities analysis (Anderson and Woodrow, 1998), which focuses on physical and material resources, the relationships among people and the community's view of its ability to create change as critical determinants of vulnerability; the pressure and release model of vulnerability (Blaikie et al. 1994) which focuses on the intersection between existing patterns of vulnerability (through power distributions and risk and hazard taking on a daily basis); and the access to resources vulnerability theory, which sees livelihood strategies as the key to gaining insight into coping strategies for rapid change (Blaikie et al. 1994). The model sees the selection of livelihoods by an individual, family or social grouping as a result of a variety of social and environmental factors. Those with better access to information, cash, means of production, equipment and social networks are seen as less vulnerable to sudden changes and able to recover more rapidly. 10 Advances in Sustainability Research In the decade following the emergence o f sustainability as a major world political issue, research into sustainability increased dramatically. The research has focused on a number o f issues: • Documenting the current situation of ecosystems. This includes research into ecosystem functions (e.g. climate regulation and the toxic cleanup capacities o f sinks) as well as the monitoring of key indicators of human and ecosystem wellbeing (e.g. rate o f deforestation, C O 2 levels in the atmosphere and numbers of people l iving in extreme poverty. (Brown et al., 1994; Straskraba,. 1994).) • Principles and approaches for sustainability. The Brundtland Report introduced the principles of inter-generational and intra-generational equity (Brundtland, 1987). Other important contributions to the canon o f sustainability concepts from academics, N G O s and multi-stakeholder panels include the Bellagio Principles ( U S D , 1996), Robinson et al.'s (1990) principles defining a sustainable society, discussions o f how to value non-market elements through environmental economics (Barbier & Pearce, 2000; Daly, 1990) and principles for sustainable development which are included in the Natural Step (Nattrass & Altomare, 1998), Natural Capitalism (Hawken et al. , 1999) and Industrial Ecology (Garner & Keoleian, 1995; Graedel & Allenby, 1994). • Tools for full cost accounting. This research develops and applies tools aimed at quantifying improved sustainability performance. The tools include life cycle analysis 5 (Heijungs et al., 1992; Ayres, 2002) which illuminates the materials and energy costs o f bringing goods to market and disposing o f them at end of life, ecological foot printing 6 (Wackernagel & Rees, 1996) and Factor X 7 (Schmidt-Bleek, 1999) which measure materials and energy intensity, the Green Bui ld ing Challenge tools (Larsson & Cole, 2001) and total costs accounting (Long & Fai l ing, 2002b), which aims to quantify many 5 Life cycle analysis is the examination of everything that happens in the manufacture, use, and disposal of a product, from the time the raw materials are taken from the earth to the time the product is thrown away and is added to the ecosystem. The basic idea of L C A is to identify and evaluate all the environmental impacts of a given product.(www.greenliving.org/readmore/definitions.html) 6 The ecological footprint is the total ecosystem area that is essential to the maintenance of a given human settlement (International Institute for Sustainable Development glossary) www.iisd.org/didigest/glossary 7 Factor X is a material reduction approach espoused in the Carnoules Declaration of October 1994. The scientists who signed the Carnoules declaration argue that material utilization should be reduced by at least 50% on a worldwide basis preferably by a factor of 10 to balance population increases. (United nations University, 2004) http://www.ias.unu.edu/ecology/g_economy/factorx.htm 11 of the hidden costs of projects which are typically ignored because they are not easily identified or are difficult to calculate. • Strategic implementation of sustainability. This research addresses strategic issues in business with regard to sustainability, the business case for sustainable development and case studies of the implementation of sustainability in business and includes Natural Capitalism, the Natural Step, Earth Systems Engineering as wel l as the work o f the World Business Counci l for Sustainable Development (Allenby, 2001; Nattrass & Altomare, 1999; Hawken et al., 1999.; Schmidheiny, 1992). Together these different branches o f sustainability research and implementation provide a synergistic and complementary toolkit for sustainable development which is constantly improving as more is learned (Robert, 2002). Governance and Sustainable Development In the decade following the Earth Summit, perhaps the most important lesson learned concerning sustainability implementation was the essential role that governance and institutions play in the outcomes o f sustainable development initiatives (Culverwell et al., 2004). Chapter X I of the Institutional Framework for Sustainable Development o f the Plan o f Implementation from the World Summit on Sustainable Development (2002) states: Good governance is essential for sustainable development. Sound economic policies, solid democratic institutions responsive to the needs of the people and improved infrastructure are the basis for sustained economic growth, poverty eradication, and employment creation. Freedom, peace and security, domestic stability, respect for human rights, including the right to development, and the rule o f law, gender equality, market-oriented policies, and an overall commitment to just and democratic societies are also essential and mutually reinforcing (p. 64 #138). Governance is the mechanism by which power is exercised and mediated through institutions. Institutions are defined as the rules which shape behaviour and can be either formal rules such as laws, constitutions and regulations or informal such as tribal hierarchies, family rules or mafia systems o f control (Turkewitz & Sutch, 2003). Institutional structures impact the outcomes of development projects at many levels. A t the regional level, the existence o f integrated policy and planning strategies greatly facilitates successful project implementation. Therefore regional governments can play an important role in mediating between disparate national and regional goals (Taylor et al. 2004). Policies can either promote sustainable development or provide what 12 have been termed "perverse subsidies" or incentives for unsustainable development focused on short term goals (Pearce & Barbier, 2000). A t the national government level, coordination between central government agencies and the existence o f an integrated framework o f social, economic, environmental and natural resource policies is required to support sustainable development (Taylor et al., 2004). Government bureaucracies also wield significant power in redefining laws through their exercise o f discretion in the application of laws (Burdge et al., 2004). In addition to formal governance structures, informal community organizations and networks also play an important role in determining project outcomes. For instance, the distribution o f power at the community level determines the level o f knowledge, response and reaction that w i l l be created by the proposed project (Burdge et al., 2004). These networks of horizontal connections, which lead to mutual commitment and trust, and enable people and their institutions to function effectively, have been called social capital (Putnam, 1995). According to the Wor ld Bank, development needs to both strengthen institutions and enhance the social capital on which they depend in order to produce human wellbeing (World Bank, 2003). The Wor ld Bank identifies corruption as "the single greatest obstacle to economic and social development" (World Bank, 2003b). They state that "corruption undermines development by distorting the rule o f law and weakening the institutional foundation on which economic growth depends" (p. l ) . Corruption is both a symptom and a cause o f poor governance. Corruption at the state level can be at the executive level in the enactment o f unfair laws, or at the administrative level with the corrupt implementation o f policies (Turkewitz & Sutch, 2003). The enactment o f unfair laws is frequently associated with the 'agency capture" model where industries increasingly control the agencies which police them, resulting in the concealment and selective interpretation o f information and the subversion o f legislation away from its original intended purpose (Burdge, 2004) 8 . Corruption has also been observed to occur at the local level, where it is In this sense the institutional features of a mining company are critical determinants of sustainability outcomes. Weick (2001) recognizes that "deep structures" or "institutional gaps" which prevent rational response to legitimate environmental and social concerns and are a key cause of inertia in companies. This inertia has produced models which position companies on a continuum between bureaucratically rigid organizations which respond to challenges by discrediting sources, and learning organizations which increase over time in their capacity to consider and generate new information, as well as their capacity to interpret and utilize this knowledge (Hoffman and Ventresca, 2002; Senge, 1997). Companies contain varied and often competing interests and can be forced to change in response to changes in the external environment. As Burdge states "how [companies] respond [to change] depends on the nature of their management structures, the values of their senior managers, views of stockholders" (Burdge, 2004 p. 62). 13 exacerbated by low education levels and a lack o f awareness of laws (Alcazar & Wachtenheim, 2000). 2.2.2 Mining Engineering M i n i n g engineering encompasses the entire process o f the design and planning of a mine-mill operation (Mcintosh, 2003). The process begins during exploration. Initially, economic geologists, geological engineers and mining engineers make preliminary calculations on an emerging mineral deposit geometry, location and grade. Surface outcrops, geophysical data and any previous mining excavations are used to design an exploration dril l ing program which samples lithology, mineralogy, economic mineralization values and potentially many other parameters (Barnes, 1980). The drill-hole data are input into stochastic modeling programs to produce probabilistic maps o f ore grade distribution for the mineral inventory (Barnes, 1980, Placer Dome, 1999). These data are used to build a series of initial assumptions about the appropriate method(s) for mining the deposit, appropriate method(s) for a mineral recovery process and costs for infrastructure, mining, processing, transporting products to market and mine closure (Placer Dome, 1999; University o f British Columbia M i n i n g Department, 2001). Some estimates are usually included to account for mitigation o f environmental issues, but social issues are largely ignored in early mine design except in respect to salary levels and labour law (Buchanan et al. , 2003; Placer Dome, 1999). Initial mining engineering processes are intuitive and based on knowledge gained through experience o f what has worked in the past and how much has been spent at other operating mines (Pincock, A l l e n and Holt , 2002). Scoping requirements are outlined in Appendix I. A s exploration progresses and evidence mounts that an economically viable deposit has been located, successive iterations o f the mine design are made as more detailed data are developed for the engineering calculations. The mine passes through a scoping level analysis (with capital and operating costs estimated to within ± 3 0 % ) and into pre-feasibility analysis (with capital and operating costs estimated to within ±25%) and finally reaches the level o f a bankable feasibility study (with capital and operating costs estimated to within +15% and -10%) where sufficient data has been gathered to make a final determination o f whether the deposit can be financed and is profitable to mine (Pincock, A l l e n & Holt, 2002). A t this stage the potentially viable resource is 14 classified as a proven or probable mineral reserve9. After a decision to proceed has been made, detailed engineering is prepared for construction with accuracy within +5% to -3%. A t any point during the process leading up to and even following a production commitment, a decision can be made not to continue. Less than 0.1% of exploration projects progress to become a mine, a fact that limits the willingness o f exploration executives to invest in community and environmental issues (Rio Tinto Zinc, 2003). Each new step involves investment and the decision to continue balances expected potential gains against the costs and risks of continuing (Placer Dome, 1999). For an open pit mine 1 0 a number o f steps must be completed to reach scoping: key mining parameters must be defined", the reserve must be optimized and the final pit outline determined, the pit must be scheduled, an effective method o f mineral processing must be determined and costed, materials handling decisions taken, waste disposal planned and finally environmental protection and reclamation designed (Pincock, A l l e n & Holt , 200. There is scope for the introduction o f social risk assessment and the costing o f mitigation measures during this process, as the pre-feasibility risk analysis is expected to include a fatal flaws analysis and a preliminary assessment o f the environmental, country (political), permitting and technological risks presented by the project, along with a more detailed assessment o f geological and engineering risk. The Historical Development of Mining Engineering Mine Design Past M i n i n g has played an important role in human history since the earliest times. Indeed, in the so called cradle o f civilization in the Mediterranean region, historic periods are defined by the most sophisticated metalworking achieved: Chalcolithic period (copper) 4500 to 3500 B C . (Golden et 9 In order to achieve this classification the resource must be assessed by a Qualified Person, i.e., an engineer or geologist, member of a professional association, with at least 5 years experience relevant to the nature of the project. There are also explicit legal requirements for the reporting of mineral resources and reserves in Canada defined under National Instrument 43-101 - Standards of Disclosure for Mineral Projects, Form 43-101. This legislation was introduced in 2001 (Postle, 2001), in response to the Bre-X gold mine stock swindle and other fraudulent claims of reserves by some junior mineral exploration companies which had eroded investor trust in the industry. In Canada, a preliminary feasibility study is "the minimum pre-requisite for conversion of Mineral Resources to Mineral Reserves" (Postle, 2001). An assessment must be made of environmental and social risks which may impact the economics of the project. 10 The case study is of an open pit mine and thus from this point on only open pit mine development is considered in the thesis. 11 Scale of operation, Mine life, Pit wall slopes - based on rock strength, structure and presence of groundwater; bench height and cut off grade - determined by costs and equipment selection. 15 al., 2001), Bronze Age 3300 to 700 BC. and Iron Age 800BC to 1100 A D 1 2 (van Andel, 1998). The Bronze Age is characterized as the era of the emergence of cities and characterizes the move from agricultural to urban societies (Mazar, 1990) and as urbanization increased mining gained in importance. During the Roman Empire (27 BC to 476 AD) mining was a mortally dangerous occupation most frequently carried out by slaves (New Internationalist, 2001). There is also archeological evidence of significant environmental damage resulting from mining operations during the Roman period, which appears not to have been a concern (Schmidt et al., 2001). Knowledge of early metallurgical processes is limited to archaeological studies and occasional references in classical literature (Cohen Duncan, 1999). The earliest detailed description of exploration, mining and mineral processing technologies is De Re Metallica written by Agricola, a physician and a farmer in Germany in 1556 (Agricola, 1556). Agricola described geology and prospecting techniques, mining and mineral processing techniques for various minerals, economics and also practices to prevent occupational illnesses. In Peru, metallurgy has origins in the "Formative Period" (1000 BC to 250AD) of Peruvian history. The oldest copper piece found is believed to date from 50AD (Procobre Peru, 2003). Gold, silver and copper (and associated alloy forming metals) were important during the Inca empire (1200-1572) with gold and silver mining being of most significance under Spanish rule from 1572 onwards. Under both Incan and Spanish empires, mining was characterized by forced labour, high death rates, and environmental contamination (Barron, 2000). Several mines exist, in the Andes, where mining occurred over 100s of years (Ossio, undated) and at some sites in Europe mining may have occurred for over a thousand years (Ayres et al., 2001). Early miners tended to extract higher grade vein deposits and only recently has it been possible to mine the huge tonnages necessary to extract low grade disseminated ores. Thus available technologies for mining and labour force constraints have been important considerations throughout mining history. At the end of the 19th Century a large injection of cash and the employment of trained mining engineers enabled mines to become highly profitable worldwide (Avery, 1974). Until this time, mines were designed based on experience and empirical rules of thumb. In 1867 the Columbia School of mines began to train mining graduates and the first comprehensive mining volume 1 2 The dates for each period vary by location as the new technologies for metalworking were adopted i: different regions. 16 Figure 2-1 Mine Design Past Geologic Data Base Mine Design Cost since De Re Metallica, the Peele, M i n i n g Engineers handbook was published in 1912 (Sorgenfrei, 2000). In the early 1900s the largest mines produced 100s of tons per day, by the 1930s production had increased to 1000s of tons per day, in the 1960s, 10,000s o f tons and by 2000, 100,000s of tons (Robertson & Shaw, 2004). Because the economic system externalized most environmental and social costs o f a mine, these were not considered in the design and, i f addressed, were handled through mitigation measures once the design was complete (International Institute for Environment and Development, 2002). Mines developed before the later part o f the twentieth century frequently o f past mining activity (Danielson, 2002). The mine design process o f the past is shown in Figure 2.1 and represents the typical process used in the 1960s. The figure shows that the mine design process was cycl ical , beginning with a geological model which was used to derive an initial mine design (Dohm, 1979; Placer Dome, 2002). The mine was then subjected to an initial financial analysis. The results o f the mine model and the financial analysis led to the refinement o f the geological model. Further exploration dril l ing, assaying or bench scale testing o f samples could be ordered and an improved geological model produced. The improved geological model was then used to refine the engineering model, which was subsequently subjected to a second round o f financial analysis. The iterative process continued until enough data had been produced to make a decision regarding whether the deposit could be profitably mined. In the past sustainability issues did not feature in the mine planning process. The tools used in mine design were engineering design and costing/financial analysis (Dohm, 1979). M i n e Design Present Prior to the second half o f the twentieth century mine environmental impacts were considered only when direct impacts on large populations were observed. One o f the earliest recorded mine clean-ups was in the Eastern U S A in the 1930s. The clean-up was o f coal mines and cost $10 left legacies of acid rock drainage, ghost towns, human rights abuses and social conflict in areas 17 mill ion dollars (Robertson & Shaw, 2004). The 1960s saw the birth o f the c iv i l rights and environmental movements. Rachel Carson's Silent Spring, published in 1962 was groundbreaking in its exposure of the risks associated with technological development. In terms o f environmental engineering, mine waste engineering was formalized in the 1960s (Michaelson, 1979; Shaw & Robertson, 2004). Increased public awareness o f environmental impacts throughout the second half o f the twentieth century led the mining industry to focus on its environmental performance (Pfeidler, 1973). Between the 1960s and the 1980s formalized mine permitting and environmental impact assessment processes were developed (Shaw & Robertson, 2004). M i n i n g engineers were still required to design the most short-term cost effective mine, but were also expected to protect the environment (Pfeidler 1973). During this period, for example, high density sludge treatment for acid rock drainage and scrubbers for smelter emissions were introduced (Horswil l , 2000). Simulation models emerged during this period, as wel l as effective software packages for mine optimization and design, permitting increased sophistication and more iterations of design (Handelsman & Fahrni, 1971). Risk assessment tools became incorporated into mine design to address some o f the major uncertainties involved in mining and to better delimit the probability and consequences of hazards and the potential costs involved (Oboni, 2001). The trend o f increasing recognition o f environmental issues continued through the 1990s and into the new millennium. In the 1990s, formalized closure planning was introduced (Shaw & Robertson, 2004). The focus on sustainability in this period widened to include social and institutional issues in addition to environmental and financial ones (1IED, 2002, Culverwell et al . , 2004). The concept of the sustainable mining community was developed as one which could "realize a net benefit from the introduction o f mining that lasts through the closure o f the mine and beyond" (Veiga, et al., 2001). "Responsible" mine design should maximize this net benefit while minimizing and mitigating all impacts. Finally, in the late 1990s and early years o f the new millennium the Figure 2.2 Mine Design Present 18 mining industry began to recognize the importance of the participation of stakeholders in mine permitting and the concept of gaining a "social license to operate" was recognized by industry leaders (Joyce & Thomson, 1997; I i E D , 2002). There has been a trend towards hiring social and environmental experts to identify and aid in the mitigation o f impacts (TeckCominco, 2001). However, critics still complain that many technical, scoping, pre-feasibility, feasibility and impact assessment reports present a biased and overly optimistic view of the environmental and social impacts o f prospective mining operations (Glave & Kuramoto, 2002; I IED, 2002). While some o f these criticisms appear valid, others may be based on political interests and a misunderstanding of scientific information. Figure 2-2 shows the mine design process o f the present. The mine planning process is shown as iterative beginning with geological exploration, and then developing an engineering model which is subjected to financial analysis before refining the models and repeating the cycle. The mine design process o f the present differs from that o f the past in two key ways. Firstly, mining engineers rely on a more sophisticated toolkit. Rather than relying on simple deterministic tools, advances in computer technologies mean that stochastic modeling tools are increasingly available allowing engineers and business analysts to carry out risk analysis and simulation to achieve a clearer picture o f the risks and potentials o f a project (Sturgul, 2000). Secondly, the increasing emphasis on sustainability issues has resulted in the inclusion o f both environmental and social impact analysis in mine design. These issues tend to be included as retrospective assessments of designs at key junctures of the design process (for example, environmental impact assessment and stakeholder consultation are typically mandated for permitting processes). Social and environmental issues are not generally integrated into the mine design process (Dies, 2002; International Institute for Environment and Development, 2002). A n example o f a current mine design process is Placer Dome's practice (Figure 2-3). Placer Dome is one o f the world's leading gold producers and is recognized by some as a leader in responsible min ing 1 3 . Figure 2-3 shows that the main iterative cycle at Placer Dome is between geological, engineering and financial considerations with sustainability concerns addressed indirectly in the financial portion o f the analysis. There appears to be a good correlation between the theoretical model (Figure 2-2) and the actual approach employed (Figure 2-3). 1 3 Placer Dome is included in the Dow Jones Sustainability Group Index (Thomsen, 2001), however, the Mineral Policy Centre gave the company a sustainability rating of "incomplete" citing its consistent failure to describe concrete measurable sustainability goals and measure results (www://mineral policy.org). 19 Geologic Initial Assumptions-,; Geotechnical Groundwater Optimization I Scheduling I Metallurgical Equipment Maintenance Financial Costing Sustainability Figure 2-3 Placer Dome Mine Design Process (Placer Dome, 1999). Mining Practice through the lens of Sustainability There is considerable debate regarding the future role of mining in a sustainable society. Several large mining companies have embraced the vision of a "Sustainable M i n i n g Community" (Veiga et al., 2001) that maximizes the long term benefits while attempting to minimize the negative impacts to local communities. On the ground application o f this model has been sporadically successful and typically ad hoc. The model is frequently invoked late in the mining process, when companies recognize that they must resolve conflicts with local communities (International Institute for Environment and Development, 2002). Best practice guidelines have been adopted to promote proactive attention to sustainability concerns. Current best practice guidelines applicable to the Canadian context range from the Environmental Excellence in Exploration Guide (E3) ( P D A C , 2003), aimed at junior and exploration companies, the M i n i n g and Minerals Sustainable Development, ( M M S D ) guidelines (International Institute for Environment and Development, 2002), which were developed for major companies active in multinational jurisdictions and Towards Sustainable M i n i n g ( T S M ) (Peeling, 2004), which is aimed at Canadian companies with operating mines. Similar initiatives exist in other jurisdictions including Australia (International Institute for Environment and Development, 2002). The guidelines typically encompass both the social and environmental 20 aspects o f sustainability. Thus in addition to considering community relationships they consider strategies for pollution reduction, improving energy and materials efficiency, waste management and biodiversity impacts ( I IED, 2002). M i n i n g companies have devoted considerable resources to improving environmental performance both in response to the increasing stringency o f regulations and because improved materials and energy efficiency is an important cost reduction strategy (IIED, 2002; Veiga et al., 2001). More recently the attention o f the industry has been focused on social factors, with the aim of obtaining a social license to operate (Thomson & Joyce, 2001) and on the mitigation o f social risk. Gordon Peeling expressed this perspective in a presentation on mining and communities given at the 2003 Conference o f the Prospectors and Developers Association o f Canada stating that "aligning our priorities and values with our communities o f interest, through a dialogue process w i l l lead to improved performance which w i l l then allow us to improve our reputation and earn an improved judgment of the industry." More progressive companies have started to move toward a coordinated strategy with international and national agencies and aim to incorporate mining into sustainable development plans at the regional level (Kalimantan Gold Corporation, 2003). There is considerable debate about the ability o f mining projects, particularly large mines in remote areas o f developing countries, to contribute to sustainable development and poverty alleviation and several authors have identified examples where mining countries and regions display higher indices o f poverty than less resource rich areas (Auty, 1993; Ross, 2001). While the mining industry focuses on improving current practice, much o f the sustainability literature focuses on higher level concerns, including whether mining has a place in sustainable society (Robert et al., 2002). Because mining uses non-renewable resources, it is seen as an unsustainable activity and therefore, some theorists suggest that sustainability mandates a significant reduction in mining activity over the long term and in some commodities a near elimination of mining (Daly, 1990; Nattrass & Altomare, 2001). The industry notes that it produces commodities in response to society's demand for metals and materials and that the timing o f any changes depends on changes in these demands as much, i f not more than on changes in mining. Indeed these demands are fuelled by population growth and infrastructure development in the developing world and the denial o f an improved standard o f l iving to these people is argued to have ethical implications (IIED, 2001). Metals are also being promoted as "the new-renewables" because of their ability to be recycled "almost without l imit" (Norgate & Rankin, 2002 p. 103). A s stakeholders debate the timing and drivers o f change in the metals industries, sustainability analysts concur with the need to reduce the negative impacts o f mining 21 operations in the interim, but debate the extent to which this is possible and the rate at which changes can occur (Danielson, 2002). Progress on the implementation of sustainability principles into mining operations has been hampered by inertia within the industry (Szablowski, 2003). There are "strong feelings....throughout the industry that it has been the victim o f unfair and misinformed criticism (Wilson, 2000). In addition, low metals prices during the last seven years have increased the focus on minimizing production costs to remain competitive, making mining executives more conservative in their approach to sustainability. Funding for innovative projects has been limited to those that w i l l produce a very rapid return on investments. This same focus on short term financial outcomes has led, in some cases, to a focus on producing shareholder value at the expense of addressing other equally important corporate values (Smith, 2004). Many sustainability initiatives produce handsome returns, for example, a power saving initiative at the Ekati mine saved 6% o f energy costs at the mine and aims to increase this saving to 12% ( B H P Bi l l i ton , 2003). However, the period o f capacity building and research that is usually necessary to build awareness (World Business Council for Sustainable Development, 1999) can deter initiatives. These types o f programs have been cut back in order to reduce costs in many companies. Finally, the domination o f mining and consulting companies by those trained in the physical sciences (Burdge and Opreyszek, 1994) has limited the capacity that these companies have to address sustainability issues. The companies lack experience and skills in the collaborative approaches and social science disciplines that are critical to achieving sustainable project outcomes. A t present, most guidance for future mining comes from examples of best practice. The Mining , Minerals and Sustainable Development Final Report (International Institute for Environment and Development, 2002) demonstrates this aspect wel l , by providing many examples of best practice and highlighting ongoing concerns. However, the guidance provided to engineers and companies for integrating best case examples into practice is limited. The report provided an assessment framework to evaluate the sustainability o f a project (Task 2 Workgroup, 2002) but does not comment on appropriate processes and structures for achieving sustainable project outcomes. The following section examines best practice examples. Applications of Sustainability Principles and Tools to Mining: The Best Cases There is an emerging consensus that in order to fulfill the parameters o f sustainability, business operations must be assessed using a variety of different criteria corresponding to each of the four dimensions o f sustainability: Economic, Social, Environmental and Governance (Nattrass & 22 Altomare, 1999; Barbier & Pearce, 2000; Hawken et al. , 2000). The M M S D process addressed these four dimensions of sustainability in the following ways: • Within the economic sphere, mining must provide the minerals needed by human society for its welfare and wellbeing through an efficient use o f resources while internalizing environmental and social costs. • On a social level, mines are expected to ensure that the costs and benefits o f their operations are fairly distributed and they must respect the "fundamental rights o f human beings". The benefits that are provided to local communities are to be sustainable over time and are to consider both market and non-market effects. Finally, mining must ensure that today's use o f mineral resources does not jeopardize the availability o f mineral resources for future generations. • On the environmental front, a sustainable mining operation w i l l operate within ecological limits and protect critical natural capital. It w i l l use the precautionary principal where 1 impacts are unknown or uncertain. It w i l l minimize waste along the entire supply chain and it w i l l be a responsible steward o f the environment, including taking responsibility for past damage. • Finally, in the governance sphere, the mining company must be accountable for its decisions, which should be made based on a comprehensive and reliable analysis. It should provide relevant and accurate information to stakeholders. It should promote representative democracy and participatory decision-making, ensuring that decisions are made at the appropriate government level and it should encourage a fair and clear system o f rules and incentives for enterprise, avoiding excessive concentration o f power through appropriate checks and balances. (International Institute for Environment and Development, 2002 (Part 1: A Framework for Change, p 24.)) Clearly this is an onerous list o f requirements to be incorporated in mine design. In addition, the application is required to be integrated, so that for example the fair distribution o f costs and benefits does not compromise the application o f waste minimizing technologies. The best case examples o f the implementation o f sustainability in mining can be subdivided into four groups each exemplifying an underlying approach: efficiency approaches, community approaches, strategic business approaches and governance approaches: Efficiency approaches. The minerals industry uses 4-7% of the global energy demand and thus has a corresponding responsibility for greenhouse gas emissions and global warming 23 (International Institute for Environment and Development, 2002). Energy use also accounts for a significant proportion o f costs at mining operations typically 5-10% o f mining costs and just over 25% of processing costs are spent on energy, and there is thus a considerable financial incentive to increase energy efficiency (Schumacher, 1999). A quarry in Georgia in the U S A was able to reduce electricity consumption by 16% by upgrading motor systems, a savings which had a payback time of just 2.4 years (US Department o f Energy, 1999). Community approaches. The "Sustainable M i n i n g Community" model of Veiga et al. (2001), advocates for mines which, "leave a legacy o f sustainability and well-being to the community, avoiding environmental degradation and social dislocation" (p. 191). They also note that "Unt i l community members themselves feel that they are partners in decisions that intimately affect their lives and the environment in which they live, little progress w i l l be made on the path to sustainability" (p.200). Among the leading initiatives in this respect are the impact benefit agreements negotiated between aboriginal communities in Northern Canada and mining companies at Ekati, Diavik and Voiseys Bay (International Institute for Environment and Development, 2002). The closure of the Sull ivan mine in Kimberley, Brit ish Columbia, Canada, is cited as a successful example o f the achievement of a sustainable post closure outcome, although critics suggest that this success relates more to a favourable location than to community involvement in closure planning (TeckCominco, 2001; Veiga et al., 2001). Strategic business approaches. Strategic approaches typically integrate a number o f tools for material and energy efficiency, social sustainability tools and principles to aid in raising awareness of sustainability in the context o f a long term visioning process. For example the Natural Step framework identifies 'backcasting' as a useful tool for conceptualizing a sustainable future in for the specific business based on a series of simple scientific principles (Nattrass & Altomare, 1999). Natural Capitalism is based on the premise that the current economic system undervalues natural and human capital. From this basis they have employed four key strategies within businesses to catalyze the transition to a more sustainable economy: Radical Resource Productivity, Biomimicry , a Service and F low Economy and Investment in Natural Capital (Hawken et al. 2000). SustainAbility has focused, since their conception in 1987, on the business case for sustainable development, and they carry out research aimed at demonstrating the business case for sustainable development in developing and developed economies (SustainAbility, 2001, 2002). The Natural Step framework has been applied to coal mining by B H P Bil l i ton in Australia to create a vision for the company o f contributing to sustainable development by developing cleaner coal based energy and steel production technologies and 24 increasing net social benefits from their operations as coal is gradually replaced by renewable energy sources (Nunn et al., 2001). Their strategy is thus essentially to combine efficiency and community approaches in the service of a long term vision of sustainability. SustainAbility (2002) demonstrated that C V R D in Brazil increased their profit margins in selling steel to European car manufacturers through achieving environmental certification. Governance initiatives. Sustainability initiatives have also been promoted within the mining industry through governance initiatives. These initiatives include national government legislative initiatives to mandate environmental and social sustainability at the country level (International Institute for Environment and Development, 2002). They also comprise the initiatives o f financial institutions, typically led by the World Bank, but usually followed closely by other lending institutions. For example the World Bank has issued standards on environmental and social performance as well as specific guidelines for forced resettlement resulting from development projects. The current Wor ld Bank guidelines are being reassessed following a review of the bank's investments in the Extractive Industries (Extractive Industries Review, 2004). Finally in terms o f governance issues, mining companies both individual and through industry associations have created policy guidelines and have endorsed voluntary initiatives addressing sustainability concerns ranging from the use o f diamond revenues to fund c iv i l war (The Kimberley Initiative), toxic releases to the environment (TeckCominco, 2002) and cyanide use (International Cyanide Management Institute, 2002). Whi le providing useful inspiration for mining executives who wish to see their industry as a pillar o f sustainable future development, the best case scenarios are more difficult to apply to mining engineering practice and decision making. A t present, mine design decision making still focuses on direct financial considerations. It was found that sustainability considerations, where they are included, tend to be reactive rather than proactive. That is they respond to a conflict with local populations, new environmental guidelines or new company policy. Engineers have completed the ground work for the more proactive inclusion o f these issues into the profession, which is discussed below. Rationale for Integrating Sustainability Directly into Mining Engineering L o n g and Fail ing (2001b) argue that many secondary impacts o f engineering projects occur as unintentional consequences and that a more complete investigation and reporting o f expected impacts to clients or decision makers has significant potential to improve project outcomes. In addition, many conflicts originate in misunderstandings and poor communication (Joyce & Macfarlane, 2001) so, providing an accurate description of possible impacts and benefits to a 25 community has the potential to prevent very costly misunderstandings at a later date, while having the additional benefit o f boosting credibility early (Gibson, 2001). Engineers are trained to produce predictive models based on scientific principles and on previous experience (rules o f thumb) (Mcintosh, 2003). According to FIDIC, the engineering professions represent a natural leader in sustainability because, "not only do consulting engineers possess the predictive tools to see these impending problems, they also possess the technological tools and the creativity to help solve them" (Van der Putte, 2001. p. 6). In mining, the integration of sustainability issues into mining engineering brings with it possibilities o f new ways to design mines that contribute more directly to human and ecosystem wellbeing. In order to achieve this, engineers w i l l need to undertake professional development in sustainability areas which are outside the traditional realms o f engineering practice and also work collaboratively with social and environmental specialists (Long & Fail ing, 2001). In some fields o f engineering the integration of sustainability into practice is already beginning to happen and much o f what has been learned is easily transferable to mining engineering. For example, utility companies are often publicly owned 1 4 and thus have to be accountable to the public for their expenditures. Consequently tools such as decision analysis have been applied extensively in utility planning. Ut i l i ty planning has many similarities with mine design: it frequently involves large projects in remote areas which have considerable social and environmental implications (Clemen & Rei l ly , 2000). Historical sociological analyses o f mining and other resource sectors such as forestry and fishing have produced concepts which are applicable to the mining industry 1 5 and improve the data base concerning the impacts and benefits 1 4 This is still true in Canada, however, there is a worldwide trend towards privatization. 1 5 Resource based communities have been studied in some detail by many researchers (Burdge, 2004b; Burdge et al., 2004; Freundenburg, 1998; Taylor & Fitzgerald, 1988). Establishment of a new industry involves large increases of labour and capital in a locality (construction). When the operation phase begins labour and capital are reduced and a period of demographic stability follows, usually accompanied by service improvements until the resource is exhausted, technology changes or markets change and plant closure occurs. Taylor et al. (1999) suggest that frequently resource communities experience a series of these "boom and bust" cycles associated for example with cycles in commodity prices. There are also important relationships between affluent and "developed" core economies and peripheral "underdeveloped" ones which play out in natural resource development, which are frequently both national and international. Many of the unwanted social and environmental costs have been externalized onto local communities and in a pattern that has been seen repeatedly world wide as well as in large resource based economies, many of the costs of production and of depletion have been borne by the peripheries (Taylor et al. 2004). Fitzgerald et al. (2002) note that the tendency is for spatial and temporal factors to isolate people employed in mining from their families and community. These groups develop their own distinct working culture which encourages them to either leave their community, or i f enough local people are employed in the mine for 26 of mine development. In addition, participatory resource management models are being piloted in forestry and utilities, which may be applicable to the mining sector (Gregory, 2000; Gregory, personal communication, 200416). A mining specific module of the APEGBC Sustainability Primer has been developed (Dies, 2002), which recommends strengthening the APEGBC approach by combining it with the "7 questions to Sustainability framework". This comprehensive framework focuses specifically on the mining industry and was developed by a multi-stakeholder group as part of the MMSD process (Task 2 Workgroup, 2002). The 7 questions framework considers the impacts of a mining operation in 7 categories which are shown schematically in Figure 2-4: engagement, people, environment, economy, traditional and non-market activities, institutional arrangements and governance and synthesis and continuous learning. For each category, the framework includes a question, an ideal answer, objectives, indicators and measures. Combining the 7 questions framework with the lessons learned from applying sustainability tools in other engineering disciplines and advances in environmental and social knowledge may Figure 2-4 The 7 questions to sustainability from Hodges, 2003. the l o c a l c o m m u n i t y to a c q u i r e charac te r i s t i c s o f the cu l ture o f the m i n i n g c o m p a n y . T h u s e m p l o y m e n t o f l o c a l p e o p l e has b o t h po ten t i a l impac t s a n d benef i t s , i n that it c a n mi t i ga t e the d i s s i m i l a r i t i e s i n e thn ic o r i g i n o f the w o r k f o r c e , but m a y a l so i n t r o d u c e n e w patterns o f b e h a v i o u r to l o c a l c o m m u n i t i e s a n d m i g r a t i o n to l a rger se t t l ements . 1 6 G r e g o r y (pe r sona l c o m m u n i c a t i o n ) notes that W e y e r h a u s e r and Interfor are d o i n g s eve ra l c o -m a n a g e m e n t p i l o t s i n fo res t ry and that B C H y d r o has c o m m u n i t y p a r t i c i p a t i o n i n r u n n i n g s o m e o f the i r f ac i l i t i e s (e .g. the A l o u e t t e D a m i n M a p l e R i d g e , B . C , C a n a d a ) . 27 provide important guidance to mining engineers. In addition, other professions involved in project assessment are recognizing the importance o f iterative, participatory project development for impact mitigation. For example, Joyce and MacFarlane (2001) identify a move afoot to bring impact assessment from "a static one shot technocratic assessment undertaken to gain project approval or financing" towards a "dynamic, ongoing process o f integrating knowledge on potential and real social [and environmental] impacts into decision-making and management practices" (p. 8). This demonstrates that integrating sustainability concerns into the mine design process in early stages is an action which corresponds to current trends in resource planning and management. 2.2.3 Sustainability and Engineering Approaches A number o f engineering associations worldwide have produced guidelines and primers addressing the integration o f sustainability principles into more general engineering practice (van der Putte, 2001; Institution of C i v i l Engineers, 2003; Long & Fail ing, 2002). A P E G B C proposes a seven guideline approach to sustainability for the incorporation of sustainability into engineering practice (Long & Failing, 2002). F IDIC suggest a similar process involving 2 steps for creating a sustainable business environment and a further four steps to promote sustainable business practice (van der Putte, 2001). Both processes propose an initial screening step to determine key impacts for further analysis. After cataloguing and prioritizing impacts, both A P E G B C and F I D I C suggest employing sustainability tools such as Life Cycle Analysis, Eco Efficiency tools, Total Costs Accounting and Environmental Economics to further analyze and quantify the relative impacts of alternative scenarios in the engineering design. The sustainability tools suggested by A P E G B C and FIDIC have considerable advantages for the pragmatic engineer acting perhaps in a setting that is not overly supportive o f sustainability approaches, although the F I D I C approach also emphasizes company approaches. The APEGBC Process: Multi-criteria analysis Specifically the A P E G B C sustainability approach involves moving through 7 guidelines which focus on 4 focus areas of sustainability (Long & Failing, 2002b). The first focus area is "Increasing Awareness o f Sustainability" (p.4) and there is one associated guideline: "Guideline 1: Develop and maintain a level o f understanding of the goals of, and issues related to sustainability" (p.4). This guideline addresses a primary barrier to the implementation o f sustainability identified by engineering stakeholders in workshops and interviews carried out by A P E G B C , which was awareness. The primer directs engineers to resources for self study and formal study in this area. 28 The second focus area is "Fu l ly Investigating the Impacts of Potential Act ions" (p.6) and consists o f three guidelines. Guideline 2: Take into account the individual and cumulative social, environmental and economic implications. Guideline 3: Take into account the short- and long-term consequences. Guideline 4: Take into account the direct and indirect consequences (Long & Failing, 2001. p.6). These guidelines address the fact that many environmental and social impacts o f engineering projects are unintended and arise from failing to account for the full impacts o f actions before they are carried out. Guideline 2 describes processes ranging from full environmental impact assessments required by most jurisdictions for large projects, to estimates o f the major implications o f the decision in each of the three areas: environmental, social and economic for less important decisions. The primer also emphasizes the need to address by comparing possible scenarios and of considering the "do nothing" alternative, which w i l l entail consequences of its own. Guideline 3 emphasizes the importance of considering the long term consequences of decision-making as wel l as the shorter term consequences, which are typically addressed in engineering costing. The standard use o f discounted cash flow analysis in engineering may provide little incentive to design for ease o f disassembly or recyclability or for other long term considerations. Increasingly, decision-makers and other communities o f interest for projects demand information concerning these longer term impacts and the guidelines suggest that engineers should consider ease o f decommissioning, the reversibility o f actions, the consumption o f nonrenewable resources, the longevity o f impacts, the long term impacts, and the option values 1 7 o f operations. Finally, in terms of the second focus area, Guideline 4 suggests taking account o f the indirect consequences of a project. L i fe cycle analysis and total costs analysis are two approaches recommended for ensuring that indirect costs are accounted for. Life cycle analysis focuses on the entire lifecycle of products including upstream and downstream linkages. Total costs analysis is a methodology for including social and environmental impacts in a project assessment. The treatment of risk and potential cumulative impacts created by a number o f projects operating concurrently within a region are also mentioned as important considerations under the rubric of fully investigating the impacts o f potential actions. 17 The option values concept requires the consideration of other possible uses that could be made of resources and which are precluded by deciding to undertake a particular project. 29 The third focus area is "evaluating alternatives" (p. 15) and Guideline 5 directs the engineer to "consider reasonable alternative concepts, designs and/or methodologies" (p. 15). This requires the articulation o f explicit objectives o f a decision. From this, either an informal evaluation o f alternatives can be carried out or a more formal multi-criteria analysis or multiple accounts evaluation can be performed. Alternatives can also be improved using brain-storming and other creativity enhancing techniques. Two further guidelines are included in the A P E G B C sustainability primer under a fourth focus area of consultation and partnerships: "Guideline 6: Seek appropriate expertise in areas where the Member's knowledge is inadequate." and "Guideline 7: Cooperate with colleagues, clients, employers, decision-makers and the public in the pursuit o f sustainability" (p. 14). These two guidelines emphasize the importance o f capacity building and collaborating in areas where expertise is lacking and also recognize that sustainability can only be achieved by working together with others, particularly communities of interest for an engineering project. 2.3 Multi Criteria Analysis The key methodology underpinning the A P E G B C process is Multi-Criteria Analysis ( M C A ) , which is also known as Mult iple Accounts Evaluation ( M A E ) and Mult iple Attribute Analysis ( M A A ) (Shaw & Robertson, 2002). M C A , M A E and M A A are elements o f the wider discipline of decision analysis. Decision analysis "provides structure and guidance for thinking systematically about hard decisions" (Clemen & Rei l ly , 2001, p. 2). B y structuring the decision to be made, decision analysis aids decision-makers in addressing the complexity o f decisions which have impacts in a number o f different areas. The approach can also help to identify important sources of uncertainty, assist in balancing objectives and support the assessment (or weighing) of a decision from different perspectives (Keeney & Raiffa, 1976, 1992). Overall the decision analysis process involves thinking long and hard about problems in a structured way and serves to "complement, augment and generally work alongside the decision-maker in exemplifying the nature o f the problem (Bunn, 1984 p. 8). Personal judgment and intuition are important subjective elements o f good decisions and are typically included in the decision-analysis approach through explicit weighting procedures (Clemen & Rei l ly , 2001). Research into decision making has concentrated in three major areas: descriptive research characterizes how decisions are made in practice by individuals, groups and organizations; normative research focuses on how decisions should be made to achieve the best rational outcomes on a consistent basis (Von Neumann & Morgenstern, 1947; Savage 1954); and prescriptive research recognizes imperfections in intuitive decision making processes identified in 30 mm Identify the decision situation and understand ohiectives 5" v * "Identify fa,. Decompose and Model the problem \, 1.-Model of problem structure 2. Model of uncertainty 3 Model of preferences. Choose the best Alternative Implement Chosen Alternative Figure 2-5. Generic Decision Analysis Process from Clemen & Reilly, 2000. the descriptive data and prescribes approaches to assist decision makers to achieve results closer to the normative ideals (Keeney & Raiffa, 1976; Clemen & Reilly, 2001). It is the latter approach that is most relevant to mine development and therefore the following discussion of decision analysis and its application focuses on the prescriptive research. 2.3.1 The Decision Analysis Process The decision analysis process is typically an iterative process as shown in Figure 2-5. The process begins with the identification of the decision situation or context along with the objectives to be achieved. The process ends with the decision, although the outcome of the decision may not be known until much later (Clemen & Reilly, 2001). The answers to the decision analysis are fundamentally dependent upon the scope of the decision context chosen and the objectives to be met. For example, a company executive is likely to focus on the question of "Which project will produce the highest return on investment to our investors while minimizing financial risk?" while an environmentalist might focus the question differently and ask "Which option contributes best to sustainable long term strategies for meeting social needs for the services currently provided by 31 this metal?". Clearly, the wider focus of the environmentalist's question invites the deliberation o f more extensive scenarios than the project specific financial focus o f the business executive. In addition, the time frame o f the decision context is fundamental to the answer to the question (Clemen & Rei l ly , 2000). MCA in the spectrum of Decision Appraisal Decision analysis is an important technique for appraising decisions. Decision appraisal includes a number of different approaches to evaluating projects or policies. While decision appraisal has many possible uses, much o f the groundwork in its application has been gained from governments' assessments o f public spending initiatives (Clemen & Rei l ly , 2001). Merkhofer (1999) describes three "mega tools" for structuring, assessing, evaluating and comparing decision approaches: probabilistic risk assessment, cost benefit analysis and decision analysis. These "mega tools" each include a toolbox of individual techniques which can be applied over several or all steps of the decision-making process. Probabilistic risk assessment combines a series o f tools for assessing the probabilities and consequences (possible risk outcomes, magnitudes and timing) of uncertain future events. The focus o f probabilistic risk assessment is quantifying the future uncertainties to produce data adequate to support decision making (Merkhofer, 1999, Oboni, 2001). Cost benefit analysis is a decision appraisal mega tool based on financial analysis techniques and on monetary valuation (Dodgson et al., 2001). Cost benefit analysis compares alternatives to ascertain the least-cost way o f achieving a given objective. It includes measures of service quality and values externalities, " i f these are relatively straightforward to value in monetary terms" (Dodgson et al., 2001 p. 12) based on the willingness to pay or accept and hedonic pricing strategies (Barbier & Pearce, 2000). Decision analysis involves identifying alternatives to a decision, estimating outcomes and assigning probabilities and finally establishing the utility function 1 8 or desirability o f the package of probable and possible outcomes resulting from any option. It is the only mega tool which can be applied in situations where important costs and benefits cannot be valued. M C A is considered 18 Decision analysis builds on multi-attribute utility theory, which was developed by Neumann and Morgenstern (1947) and elaborated by Savage (1954). The theory defines the Subjective Expected Utility (SEU) of different potential futures to a decision maker taking into account all future states reasonably viewed as relevant. The theory considers the subjective probability of an outcome and its attractiveness (utility). The SEU is the sum of the probabilities of achieving the quantified utilities. The theory is the most widely accepted model for rational choice and is implicitly a M C A approach (Dodgson et al., 2001). 32 by Merkhofer (1999) to be a tool in the toolkit of decision analysis, although Dodgson et al. (2001) describe M C A and multi-criteria decision analysis ( M C D A ) as comprehensive decision analysis processes in their own right. Description of MCA The term multi-criteria analysis covers a range of distinct approaches. A l l o f the approaches include the decision analysis steps o f defining options and assessing each option with respect to a series o f explicit criteria. The main difference between different M C A approaches is the manner in which data are combined. More formal M C A methodologies use explicit weighting systems for the different criteria. They function "to select a single most preferred option, to rank options, to short-list a limited number of options for subsequent detailed appraisal, or simply to distinguish acceptable from unacceptable possibilities" (Dodgson et al. 2001, p. 16). K e y determinants in selecting an appropriate M C A approach are: • Decision type (finite or infinite options) • Time and financial resources available • Amount and availability o f supporting data • Analytical skills o f practitioners • Administrative culture and requirements of decision making body. A distinguishing feature of M C A is that a "performance matrix" or consequences table is derived for each scenario or option. In the matrix, options are compared in terms o f each of the criteria selected. In a basic form o f M C A this matrix is the final product o f the analysis. Decision makers are required to process the information provided in the matrix in an intuitive fashion. More analytically sophisticated M C A involve the conversion o f the raw matrix data to consistent numerical values (Dodgson et al., 2001) or using qualitative techniques to manipulate the data and achieve insights (Gregory, 2000; Hammond, et al., 1999). Decision Context and Decision Structuring Establishing the decision context combines identifying the boundaries for the decision with identifying objectives. A decision may, in effect, be a series of decisions and it w i l l be necessary to determine the extent to which the decisions can be viewed discretely or whether it is better to build one iterative model for the chain o f decisions (Clemen & Reil ly , 2001). There is debate between researchers over the relative value o f focusing on defining the problem and then 33 identifying appropriate objectives to consider in solving the problem (Clemen & Rei l ly , 2001; Dodgson, 2001) and the alternative o f developing a clear understanding of objectives and values and then looking for ways to achieve the objectives (Gregory, 2000, Hammond et al., 1999). Both groups concur that structuring the decision through decomposing it into its component parts is essential to successful decision analysis. Decision structuring can be aided through the use of decision trees, objectives hierarchies, means-ends networks and influence diagrams, all o f which provide a logical structure for the elements of concern to the decision maker (Clemen & Rei l ly , 2001). Decision trees are the simplest decision structuring tool, although researchers concur that they can become unwieldy i f they are too large and thus other tools may become indicated (Dodgson et al., 2001; Gregory, 2000). Within the context o f the A P E G B C Sustainability Primer, a "prioritization o f impacts process" is suggested for identifying criteria (Long & Fail ing, 2002). Two approaches are suggested for this process. The first involves making an exhaustive list of potential impacts, as would be done for an environmental / social impact study - the inventory o f impacts approach - and then simplifying the list by identifying priorities. The second involves gathering experts in all relevant aspects of the project together to brainstorm and prioritize criteria - the brainstorming approach. A P E G B C recommend the latter methodology as the most effective. Dodgson et al. (2001) recommend a brainstorming approach responding to the question, "What would distinguish between a good choice and a bad one in this decision?" (p. 27) and suggest that because stakeholder perspectives may be important, involving affected parties may be appropriate in some or all stages o f the M C A . Final ly, in terms o f criteria, Dodgson et al. (2001) suggest that the provisional set o f criteria established should be assessed in a number of ways: • For completeness. Have all important criteria been included? • For redundancy. Duplicates and criteria judged unimportant should be removed. • For operationality. Are the criteria clear and are adequate measures of the important dimensions of each impact provided? • Mutual independence of preferences. Linear additive M C A data processing requires that each criterion be independent o f outcomes in other criteria. • For double counting. Impacts should only be accounted for once. • For size. A n excessive number o f criteria increases financial and temporal requirements. It may also result in the M C A being difficult to communicate. 34 • For temporal considerations. Has attention been paid to impacts occurring over time? In addition to structuring decision criteria, it is important to structure the attributes o f a criterion which are to be measured, carefully (Gregory, 2000). One criterion may consist of a number of important attributes. For example, optimal employment is a criterion which may have importance in terms o f salaries, numbers o f positions, stability o f the jobs offered or the transferability o f the skills learned for different stakeholders. The essence o f defining attributes is to locate measures, which can describe the criterion in a simplified but meaningful form. Measures for performance can be natural scales, or subjective constructed measures. Monetary units or sizes o f area impacted are examples o f natural scales. They are easily and directly measured. Subjective or constructed measures are used for variables which are difficult to measure directly and must be assessed through an appropriate surrogate measure. Highly qualitative data may be assessed on a five point value scale assigning numerical values to qualitative assessments of performance; for example: poor, fair, average, good, excellent; 1,2,3,4,5. A n effective decision structuring and attribute identification process streamlines the modeling process by identifying the key information gaps necessary to understanding the decision outcomes. This can result in more effective use of scarce resources to develop an appropriate quantitative model to provide insight to guide the decision (Gregory, 2000). Once objectives and the decision context have been identified, the M C A process identifies alternatives or scenarios for comparison. A n iterative decision model is shown in Figure 2-6 demonstrating the level of flexibility that can be incorporated in a decision- model which is both iterative and modular. New Data/ ^ ^ "^ If" Uncertainties ^ \ f ^ ^ ^ Integrated Modular Model Figure 2-6. Incorporation of new data in an iterative modular model. The iterative modular approach is appealing because it enables transparency, selective iteration between the individual indicators modeled and because o f its flexibility to incorporate new data as it becomes available. A potential drawback o f this approach is that, mining in its social and 35 environmental context, contains considerably more interaction between the modular elements than the multi-criteria analysis approach allows. Complex systems theory (Waldrop, 1994) suggests that such additive processes may produce results that are very different from the integrated results actually produced by the complex systems they attempt to model. Model ing Once the decision has been structured and attributes identified it is usually necessary for experts to model the expected outcomes for the different scenarios in terms of the attributes identified for decision-making. In communicating the results o f this modeling it is important that an accurate assessment is made o f the degree o f uncertainty inherent within the modeling process and that this is communicated to the decision-maker(s). Formal models exist for the communication o f uncertainty in the form of probabilities and distributions. Some approaches to quantitative decision analysis incorporate mathematical models o f uncertainty directly into the data and this is expressed in the decision analysis tables or multi criteria matrix produced. Dodgson, et al., (2001) consider that M C A ends with the creation o f the matrix and that data processing beyond this point comprises M C D A . Whatever the process is to be called, once the data has been collected there are two main approaches to data processing for a multi-criteria matrix: quantitative and qualitative. Quantitative Analysis In the quantitative approaches, the numerical assessments o f the performance o f the criteria for each scenario can be combined mathematically. Data need to be scored so that they can be compared (Dodgson et al, 2001) and a weighting approach can be applied so that the values o f multiple stakeholders can be incorporated into the analysis or so that priorities between variables can be established (Gregory, 2000). Scores should be applied to the data in an overt way so that the rationale behind the scoring system is evident to all and scoring should be consistent so that better performance leads to a higher score on all variables (Dodgson, et al., 2001). The most common way to combine scores is by use of a simple weighted average. Linear additive processes are underpinned by the assumption that good performance on one criterion can in principle compensate for weaker performance on another (compensation). Non-compensatory techniques are limited and have been shown to be ineffective in quantitative comparison o f scenarios (Dodgson et al., 2001). Quantitative approaches are used to gain insight about differences in performance o f the alternatives. Linear additive weighting techniques also assume mutual independence o f criteria. This ensures that double-counting of impacts or benefits is absent from 36 the summative analysis. Min imum and maximum values can also be incorporated into the analysis to eliminate impossible or unfeasible alternatives (Dodgson et al., 2001). Where mutual independence cannot be established, other mathematical procedures are available, although these are more complex to apply. Keeney and Raiffa (1976) developed procedures, which not only account for potential interdependence between criteria, but also account formally for uncertainty in each of the variables. The result is a wel l regarded and effective process which is most commonly implemented by specialists on major projects. However, some authors believe that, in practice, it is often better to ignore interdependence and uncertainty "to allow a simpler and more transparent decision support . . . by a wider set o f users and for a larger set o f problems" (Dodgson et al. , 2001 p. 21). While it is possible to combine processes which account formally for uncertainty with simplified presentations for stakeholder deliberation, the costs o f the formal analysis may be high. Methods which combine limited representations o f uncertainty within a qualitative framework for trade-offs analysis have been suggested (Gregory, 2000; Hammond et al., 1999), as clear representations o f uncertainty in a transparent format at an affordable cost. Qualitative Analysis For qualitative processes, Keeney and Raiffa (1992) distinguish between an alternatives focused process and a values focused process. In an alternatives focused process the decision analysis concentrates on deciding between options, while in the values focused approach the aim of the analysis is to determine the best way o f achieving multiple value based goals (Gregory et al., 2000). In values focused decision analysis, the multi-criteria matrix is used as a basis for deriving improved alternative scenarios which combine benefits from the initial scenarios and minimize their impacts. The tool used for accomplishing the qualitative comparison o f alternatives is called a consequence table (Hammond et al., 1999) and is a smaller M C A matrix, which is used to structure comparisons between two different scenarios. The matrix contains a third column for an improved third alternative scenario, which attempts to maximize the benefits and minimize the costs from the initial scenarios. In this way trade-offs between the options are overtly identified and decision-makers or stakeholders can propose alternatives. According to Burdge (2004) alternatives identification should consider ways to accomplish the proposed action with fewer resources and fewer impacts on the human and physical environment. Taylor et al. (2004) explain that the formulation o f alternatives ideally occurs during a phase o f intensive consultation with stakeholders regarding the project. They also provide a list o f project characteristics which commonly contribute to alternative identification: 37 • Location o f area boundaries, infrastructure etc. • Operation (for example the type o f machinery to be employed, type of concentrate transportation to be utilized). • Location o f construction workforce, sources of raw materials. • Timing "Many negative effects can be mitigated given time alone." • Financial. (Building mitigation and management o f social effects into the project costs from the start may produce long term returns.) Outputs and Use of Results A common critique o f decision analysis approaches by decision makers is that they take away the ability to apply intuition in the decision context, making some decision-makers feel uncomfortable (Clemen & Reil ly , 2001). This critique, however, implies a misunderstanding o f the purpose o f the decision analysis process, which is explicitly to assist the decision maker to structure and gain insight into the complex decision rather than to make the decision itself. The decision maker's intuition has a crucial role to play. In a quantitative analysis, the tendency to select the highest scoring alternative, without consideration o f significance, uncertainties and biases, could be a drawback. A qualitative analysis w i l l tend to produce more overt information regarding the trade-offs associated with any proposed alternative. M C A Summary M C A is a decision analysis approach to structuring decisions, identifying key objectives and alternatives, identifying consequences and clarifying trade-offs. Structuring both the decision objectives and the specific attributes to be measured is crucial to effectively allocating resources for the consequence identification (modeling) phase of the analysis. Once consequences have been identified all criteria can be quantified and uncertainties incorporated into a multi-criteria performance matrix. There are two endpoint types o f data processing which can be applied to the multi-criteria performance matrix: the first is a quantitative approach to assessing fairly rigid alternatives which have been optimized prior to the modeling o f impacts; the second involves a more flexible approach where the process is driven by objectives and initial alternatives are assessed in order to provide insight into the iterative development o f flexible and evolving alternatives. Both processes recognize the importance of careful decision structuring but the data are evaluated using different quantitative or qualitative techniques. It is believed that the two endpoint approaches are probably suitable to different applications and that many processes w i l l 38 be hybrids occurring somewhere along the spectrum between the two endpoints (Hammond, 1999). 2.4 Literature Review Summary This chapter has considered the conceptual frameworks underlying mining engineering, and has examined how these concepts have changed over time. In addition the concept of sustainability has been examined and its applications and implications for mining have been discussed. Final ly an approach for incorporating sustainability into the mine design process has been described. From the discussion o f sustainability it is evident that the integration o f sustainability into mining engineering requires the use of tools that are able to incorporate values and weigh complex objectives to determine optimal outcomes. Research into how managers make complex decisions has resulted in the creation o f a family of decision analysis techniques which are designed to structure complex information and values to assist decision makers (Clemen & Rei l ly , 2000, Dodgson et al., 2002). One o f these techniques, M C A underpins the A P E G B C sustainability primer methodology and forms the basis o f the thesis research. The next chapter incorporates the insights from the literature reviews on sustainability in mining engineering and the A P E G B C multi-criteria analysis process into a description of the methodology applied to integrate sustainability into the mining engineering process for the case study. 39 CHAPTER 3 THE CASE STUDY AND RESEARCH PROCESS 3.1 Introduction This chapter outlines the assumptions and choices used in applying the A P E G B C Sustainability Primer methodology to structure the engineering decision making process and M C A for the case study. The chapter begins with an introduction to the case study and follows this with a description o f the application o f the research process undertaken. A n ideal situation for exploring the integration o f sustainability into the mine design process would involve analysis of real geological, environmental, social and institutional data from a geological resource with good potential to be transformed into a viable mine. The analytical process would follow this mine through to development and compare predictions with outcomes. M i n i n g projects in this late exploration / early feasibility stage are generally both shrouded in confidentiality and analyzed under tight time and financial constraints. Data of this type are rarely available for publication, which, o f course, is a requirement of academic work. In addition, the time scale for a thesis does not permit following a mine through from exploration to start-up and beyond. Retrospective data for operating mines are also subject to confidentiality concerns. Therefore, this research not based on a real mining project, thus eliminating confidentiality issues. Instead realistic data from more than one source was combined to form a case study that was then analyzed to generate four development scenarios. Multi-criteria analysis was employed to determine the preferred mining option. 3.2 Overview of the Case Study The case study mining deposit was based on borehole data from an actual porphyry copper deposit, chosen because o f the accessibility of data, the existence of an operational block model from which to calculate mining parameters and because o f the author's familiarity with environmental aspects o f the deposit through previous studies. It was decided to site the case study mine in the Peruvian Andes, because of the researcher's familiarity with the social and institutional frameworks relating to mining within Peru. Several years of l iv ing in this area also provided a useful insight into community and cultural issues. Peru experienced significant expansion in mining projects throughout the late 1990s and into the new millennium, which was 40 accompanied by a parallel increase in social conflict surrounding mining projects (Glave & Kuramoto, 2002; Pasco Font et al., 2001). This section provides an overview of the main economic, political, social and environmental factors pertaining to the case study. A more comprehensive profile of the case study is presented in Appendix III. The description o f the case study consists o f 5 sections: 1. Governance and Legal Structure, 2. Community Profile, 3. Environment, 4. Geology, and 5. Engineering. This type o f structure is proposed as the appropriate way to characterize a potential mineral development project. 3.2.1. Governance and Legal Structure Legislation governing mining Peru introduced legislation promoting foreign direct investment in mining during the 1990s (Glave & Kuramoto, 2002). Land tenure and mining concession laws were changed to guarantee land access, a taxation system which taxed profits and required no royalties was introduced, tax deductions for imported machinery were instituted and the right to repatriate profits enshrined in legislation (Otto et al., 2000). These changes led to a significant increase in exploration and were combined with the privatization o f several state owned mining operations (Damonte et al, 2002; Pasco Font et al., 2001). Min ing is significant to Peru in macro economic terms. It accounted for 48% of foreign earnings in 1998 (Peruvian Ministry o f Energy and Mines, 2001) with exchange revenues amounting to $2.7 bi l l ion in 2002 (Glave & Kuramoto, 2002). Internally the industry provides 13% o f taxation revenues and directly employs 3% o f Peruvians (Damonte et al. , 2002; Tolmos, 2001). Under the then existing taxation regime, Otto et al. (2000) calculated an effective taxation rate of 43% for a 209 mil l ion tonne copper mine 1 9 . Other salient features of the taxation system include a unique legislative instrument to ensure that some o f the benefits o f mining remain in the area surrounding the mine (Pasco Font et al., 2001) and a law mandating the distribution o f 8% of profits to workers at the mine up to a maximum of 18 monthly salaries per year (Damonte et al., 2002; Otto et al. , 2000). The legislation mandating distribution to local communities is known as "The Canon L a w " and mandates the return o f 50% of the income tax from a mining operation or 1 9 Because mining taxation depends on the scale of operations and metal price it is not possible to provide a generic taxation rate for mining without linking this to assumptions concerning the operation. For example, Otto et al. (2000) calculated that copper price increases and decreases of 10% resulted in effective taxation rates of 41% and 46% respectively for the same model mine. 41 approximately 15% o f the gross profits to local and regional governments (Otto et al., 2000). There have been considerable challenges over distributions o f canon revenues by the central government, although payments are now published electronically on an annual basis (Pasco Font et al., 2001; Slack, 2001). Environmental legislation was introduced in the late 1990s (Glave & Kuramoto, 2002) and the permitting process began to revolve around the environmental impact assessment (EIA) coordinated by the Commission for the Environment ( C O N A M ) under the auspices o f the Ministry of Mines ( M E M ) . The environmental legislation in mining is criticized for its lack o f independence from the M E M , limited capacity among Ministry officers for enforcement, lack o f independence on the part of consultants and weak guidance regarding consultation processes. A number o f high profile conflicts have arisen over proposed mines resulting from environmental and social concerns from neighbouring communities (Glave & Kuramoto, 2002; Pasco Font et al., 2001; Slack, 2001). Peruvian Governance Peru is recognized as a democratic country and holds elections every 4 to 5 years. The Wor ld Bank ranks Peru "poor" for political stability, below average for effectiveness o f government and rule o f law, "average" for control of corruption and "above average" for both accountability and for regulatory quality. Combined together, these indicators produce an integrated Wor ld Bank Governance Index o f 667, giving the country a global ranking o f 57 out o f a total of 139 countries (Kaufmann et al. , 2002). Matsuda (2002) identified the following specific areas o f concern in terms o f governance: 1. Lack of restraint over arbitrary executive decisions. 2. Weak political system, poorly informed voters and relatively recent introduction of democratic elections leading to poor representation o f broad societal interests. 3. Extreme centralization of decision-making where L i m a dominates Provinces, the President dominates Congress and even in district municipalities Mayors tend to have dictatorial powers. Transparency International (TI) give Peru a 'corruption perceptions index' o f 4.0 on a scale o f 0-10 where 10 indicates no corruption and 0 indicates extremely high corruption (Transparency International, 2003). 25% of respondents to a survey carried out in the country claimed personal experience of corruption and 67% of public officials were perceived to be corrupt. Overall responses indicated that there had been a slight decrease in corruption from 2000 levels in 2002 42 (Guerrero & Hoffbauer, 2002). A public expenditure tracking study (PETS) showed that leakage between central government and local levels o f government for the mining canon were negligible, accounting for no more than 1% of revenues. However, 50% of the recipient municipalities were unable to provide any data regarding how the mining canon revenues had been spent and a further 33% could account for only 70% of revenues (Alcazar & Wachtenheim, 2001). Alcazar and Wachtenheim (2001) also carried out a parallel review of the "Glass o f M i l k " social program. In this review it was clear that leakages occurred at both community and local levels, as wel l as in the formal government structures. These studies allowed the range o f leakage o f canon revenues to be determined as ranging from 30% to 70%. Rancanatini, (2003) found that companies indicated a willingness to pay amounts equivalent to an extra tax of 7.5% for the elimination o f corruption within the Peruvian legal system and that companies expect to pay bribes equivalent to the value of 17% of a contract in order to secure the contract. 3.2.1 Community Profile Spatial Distribution The extended Andean mountain range is over 2000 kilometres long and covers over 2 mi l l ion square kilometers. It includes land from 1500 to 6000 metres above sea level and includes a wide variety o f climates and ecological zones (Tapia & de la Torre, 1998). The Andes has a population of over 50 mil l ion people, most o f whom live in rural communities and practice agriculture for a l iving (Swinton et al. , 2001). The population is expected to increase by 33% in the 20 years leading up to 2020 (McDevitt 1999). Communities range from isolated communities in high mountain areas consisting of communities o f 10 to 100 families to cities, usually in fertile valleys which have populations of up to 200,000. Most people live in towns which range from 2,000 to 20,000 people which are generally situated in river valleys. It is estimated that by the year 2025 over 75% of the population w i l l be urban. (Swinton et al., 2001). The map shown in Figure 3-1 is a schematic diagram of the area of influence o f the Antamina mine (Botts, 2003). This figure and additional data from Pasco Font et al. (2001) provide data concerning the spatial impacts o f mining operations in Peru. The Antamina mine occupies land belonging to two Campesino communities and has three towns within its larger area o f influence, all o f which are likely to have kinship and economic linkages with the Campesino communities. The wider area o f influence o f the mine includes six towns and the regional capital city. Communities in the vicinity o f the port city, as wel l as communities residing along the access route are all considered to be impacted by the mine. 43 Figure 3-1. The area of influence of the Antamina mine after Botts, 2003. Social Organization Important social organizational units in the context o f Andean mining include "Campesino communities", towns and cities. Campesino communities are communal organizations which own land in remote areas for subsistence agriculture and pasturing (Glave & Kuramoto; Pasco Font et al. , 2001). They vary from ancient communities organized along traditional indigenous lines to the inhabitants of private estates which were turned over to the workers during the agricultural revolution in the late 1970s (Ossio, undated). These communities have strong kinship and livelihood relationships with local towns, although there is a distinct class structure where landowners and valley bottom dwellers have higher status and mountain people are the lowest class (Swinton et al., 2001). Local governments are present in the Peruvian Andes at district, provincial and departmental levels. Governments are democratically elected every 4-5 years (Rancanatini, 2003). Dominant 44 social organizations in towns are the Catholic Church and its social and environmental justice offices (Damonte et al. , 2002). In addition towns typically include a number o f women's and anti-poverty organizations which are focused on food distribution activities (Rancanatini, 2003). In cities there are stronger institutional networks with regional governments and national government agencies represented. Anti-poverty organizations have an important presence in poorer areas and non-governmental organizations have a presence. Kinship ties and the Church predominate in social networks and the majority of Peruvians work for small family enterprises (Damonte et al. , 2002). Attitudes, Beliefs, and Values Traditional Andean society revolved around the agricultural and pasturing cycles. Nature was venerated as the source o f sustenance and intricate irrigation and cropping systems were developed to maximize yields while maintaining fertile soils. The extreme climate was a particularly important element in determining work patterns (Larme, 1993). Male and female roles were differentiated but complementary with women concentrating on domestic activities seed and food preservation and men carrying out most pasturing activities (Tapia and de la Torre, 1998). The Spanish conquest brought the Catholic Church to the Andes and Catholicism predominates in the area having incorporated many traditional festivals (Larme, 1993). There is an ongoing transition from more traditional subsistence lifestyles to a wage economy. Television provides images o f different worlds and different attitudes and young people tend to idolize the American lifestyle, while older people see strength in traditional ways and tend to be more conservative. There is some agreement, however, that the majority o f Andean people qualify as indigenous people and many speak native languages and have livelihoods which have much in common with those o f their ancestors. Lifestyles Despite l iving in terrain where only 3-4% of the land is arable ( C I I F A D , 1998) small scale agriculture is the dominant livelihood in the Andes. A n agricultural system has developed where family groupings carry out a mixture o f pasturing and cultivating activities across a variety of climatic zones to reduce risk in a marginal environment and spread peak labour requirements over an extended period (Aramayo, 1998; Zoomers, 2002). Increasing populations and changing agricultural practices have led to overgrazing and families frequently have insufficient land to meet subsistence needs. The population pressure, as well as external influences and years of c iv i l war during the 1980s and early 1990s along with government political strategies have led to increases both in migratory labour and permanent migration to cities for work. M e n are more 45 oriented to paid labour and have traditionally participated in communal labour exchanges. Some estimates suggest that as much as 70% of agro-pastoral activities are now carried out by women (Tapia & de la Torre, 1998) and typically around 25 to 30% of mothers are single women (INEI, 2004). Towns are dominated by commercial and service activities but agriculture remains important and most families either live on farmland or own plots o f agricultural land. Elementary schools are present in most small villages however students must travel to towns for high school. Limited post secondary education is available in larger towns but is concentrated in cities. Villages have clinics and larger towns have local hospitals. Professionals congregate in cities and larger towns and include significant numbers of women in nursing, teaching, midwifery and other acceptable female professions. Health Almost 20 mil l ion people or 78%> of the Peruvian population live in poverty and nearly 12 mil l ion people live in extreme poverty (PNUD-Peru , 2002). Poverty and health indicators are worse in rural areas. Life expectancy ranges from 54.4 in the Andean region o f Huancavelica to 77.1 in suburbs o f L ima ( P A H O , 1998b). There have been dramatic decreases in fertility since the mid 1970s and the fertility rate in L i m a is now 2.5 children per woman. In Provincial cities and towns the fertility rate is 2.8 per woman and in rural areas it soars to 5.6 per woman (McDevit t , 1999). Communicable diseases are the leading cause o f death in Peru accounting for 27.5%> o f all deaths and 44%o of deaths among the poorest quintile of the population ( P A H O , 1998b). Acute respiratory infections accounted for 25.2% of deaths among the poorest group and intestinal infections were the next most important cause o f death. Respiratory diseases are related to cooking with straw, dried animal feces or occasionally wood and produce high fatality in combination with malnourishment. Intestinal infections are symptomatic o f the lack o f potable water available to many rural households. Peru has one of the highest maternal mortality rates in Latin America, which is related to the high rural birth rate and high levels o f births which are not attended by medically trained assistants (less than half of rural mothers receive any prenatal care (McDevitt , 1999)). Maternal mortality runs at 221 per 100,000 births for all Peruvian women but reaches 448 per 100,000 births in rural areas with hemorrhage and infection the leading killers ( P A H O , 1998b). Another key health issue in Peru is alcohol abuse. Yamamoto et al (1993) found alcohol abuse to be a problem for 35%) o f men in a suburb of L i m a . Access to modern health care is improving with a 61%> increase in primary care infrastructure reported between 1995 and 1999 (Martinez, 1999). However, 46 medicines and even clinic visits are frequently both distant and too costly for family budgets. In many rural areas there is a system of traditional medicine for many common ailments which combines herbal remedies with shamanic and psychosocial practices. Income equity is considered to be a leading indicator o f health within a nation. A recognized indicator o f income equity is the G in i coefficient, which measures the difference between observed income distributions in a country by quintile or decile and a completely equitable distribution where all people have equal earnings. From a potential range o f 0 (perfect equity) to 1 (complete inequity) observed Gin i coefficients fall within the range of 0.2-0.7. Peru had a G in i coefficient in 1996 o f 0.46 which placed it just above the worst quartile for income distribution (CIA Wor ld Fact book, 2003; Prescott Al l en , 2000). Economy 45.00% 40.00% 35.00% • | 30.00% "I 25.00% <£ 20.00% ° 15.00% ^ 10.00% 5.00% 0.00% <# & t £ J <F 2? Jr <cT ^ J3 (# Occupation C3 % of population •-Figure 3-2 Employment profile for Andean settlements data, Peruvian National Institute of Statistics (INEI, 2004) - 1992 data. The economy in Andean towns is dominated by agriculture, commerce and services. According to statistics from the Peruvian Institute of Statistics (INEI) up to 40% of the members o f the employed members o f Andean towns, cities and small rural communities work in unskilled services jobs such as cleaning, food service and vending. The next largest category is for 47 agricultural employment, which typically represents 15-20% of employment. Skil led service provision and labourers each occupy 10-15% of employment and professionals and technologists each account for 5-10% of employment. Sources suggest that up to 60%> of the Peruvian economy is informal and that many employed professionals earn less than a living wage (US Library of Congress, 1992). The average employment profile of Andean settlements is shown in Figure 3-2. Min ing has a long history in the Andes and provides an average o f 3%> of employment, although it is also an important source o f local revenues through canon taxation redistribution payments. The history o f mining in the Andes is described in more detail in Appendix III. 3.2.2 Environment The case study deposit has been located at high altitude in the Peruvian Andes (3,500-4,800 metres above sea level). This area forms what is known as the 'High Mountain ' or 'Puna' ecosystem which is defined as the area o f high plateaus and inter-montane basins, having an average height o f 4000 metres above sea level (Aramayo, 1998). Vegetation in these areas consists mainly o f grass species adapted to the harsh conditions. Soils are thin and o f volcanic origin on slopes, but are frequently rich in organic matter in river valleys and in glacial outwash. Domestic stock create significant pressure on the puna ecosystems with overgrazing o f llama, alpaca, sheep and occasionally cows due to population pressure, poverty and lack o f education (Swinton et al. , 2001). This leads to reduced soil fertility. Maximum daily rainfall is high in mountainous regions and where roads cross slopes drainage must be managed carefully to prevent slump failures. In addition, the high rainfalls impede the ability to control erosion and an inadequate data base impedes the ability to correctly predict the volumes and intensities of storm events (Mclnnis et al . , 2001). There is generally a long dry season followed by a short wet season from December to March . 75% o f rainfall falls during this wet season. Max imum rainfall is around 700 mm per annum. Maximum daytime temperatures are 20 degrees Celsius in summer and 13 degrees in winter. Night time temperatures are considerably lower, because o f lower levels of greenhouse gases at high altitudes, falling to around 1 degree Celsius in summer and -11 degrees in winter (Peruvian Minis try of Agriculture, 2003b; P U C C , 2002). In addition to the problems o f overgrazing and species loss, the melting o f glaciers is a particularly concerning trend. While scientists argue over whether the melting is related to global warming, 230 square kilometers o f the 850 existing in 1930 in the largest glacier field in Peru have now melted (Georges 2004). This impacts human activities because streams, which are essential for irrigation, are fed by glaciers and some no longer flow throughout the year. 48 3.2.3 Geology The case study deposit is typical o f low-grade porphyry deposits produced by subduction volcanism. In Peru this volcanism is related to the collision o f the Nazca and South American plates. The deposit underlies a river valley with 50 meters to 350 metres o f glacial tills. Economic mineralization is chaicopyrite, bornite, and molybdenite and there is also significant pyrite. The host rock mineralogy is alkaline series intrusive rocks and, mines hosted in similar host rocks in North America, have been observed to be non-acid generating over a time frame in excess of fifty years. The majority o f the buffering capacity o f the rock is in calcite and clays in fractures which also host the sulphide minerals and in calcic feldspars in the matrix o f the rock. The presence of clays and carbonate minerals in the veins suggests that early oxidation o f sulphide minerals w i l l be effectively buffered. A c i d generation is not expected to be a problem although neutral leaching of molybdenum is predicted to occur. Molybdenum has been documented to impact cows grazing on pastures with high molybdenum levels causing hair, fleece and bone abnormalities, impaired growth hormonal problems and low conception rates (Jones, 1994). It is expected that post mine pasturing w i l l be possible in a managed grazing regime similar to that studied at Highland Val ley copper mine in Canada (Gizikoff et al., 2001). Copper is highly toxic to fish (Oatchere et al, 2002). However, post mine copper mobility is expected to be limited at neutral p H and fish populations are not expected to be impacted. A sample file o f composite sample lengths was calculated from 143 diamond holes. The original dr i l l data was composited into 3755 intervals. The composited sample data were used to create a block model for the proposed area. The block model consisted o f 273,189 regular blocks each 15m x 15m x 15m. After studying the variography, vertical and horizontal search distances were determined and grades were estimated for each block in the model using the Ordinary Kr ig ing method by Dr. Anoush Bozorgebrahimi 2 0 . The grade tonnage curve shows a maximum ore grade o f 1.4% copper and relatively uniform grades with increasing quantities o f lower grades as would be expected for a porphyry copper deposit. 3.2.4 Engineering The deposit was determined to be suitable for open pit mining and standard copper floatation. From cost data for these methods a cut off grade o f 0.15% was determined for the deposit giving an overall ore tonnage o f 1.5 bill ion tonnes averaging 0.6% copper. A stripping ration o f 2:1 gave Using Surpac software. 49 a total tonnage o f material to be mined of 4.5 bi l l ion tonnes. The final pit outline had a footprint o f 3.3 square kilometres and was 830 metres deep. The energy requirement for crushing and grinding the ore was assumed from average data in the initial model 2 1 . A mine life was calculated using the Taylor formula (Hustrulid and Kutcha, 1995) and was determined to be 37.5 years, with a daily production rate o f approximately 350,000 tonnes. Once the mine life had been determined it was possible to make assumptions concerning daily tonnages mined, equipment selection and cost estimates. The data were also used to estimate the land area required for waste dumps and tailings facilities. Two waste dumps (one for overburden and one for hardrock) were conceptualized with a combined area of 14.4 square kilometres. The total land area required to be purchased by the mining company was estimated at 115 square kilometres. 3.2.5 Summary of Case Study From the baseline data a case study outline was produced: The project involves the development o f the case study copper deposit as an open pit mine with on site processing facilities and transportation o f concentrates to port facilities. The project involves the production o f an open pit 3.3 square kilometers in area and 830 metres deep, two waste dumps with areas of 9 square kilometers and 5.4 square kilometers which w i l l require reclamation and a tailings pond 7.5 kilometres in length, varying from 1 to 2.5 kilometres in width. Concentrates w i l l be piped or trucked 400 kilometres down the west flank o f the Andes to the coast where they w i l l be shipped to foreign smelting facilities. It is assumed that 100 kilometres o f access road w i l l be built along with improvements to other roads. Employee housing w i l l be constructed in the nearest regional capital city and workers w i l l stay at the mine site during work rotations. The total land area which w i l l be purchased by the mining company w i l l be 115 square kilometres and w i l l involve the relocation of 100 families from 2 'Campesino communities' and w i l l buy 115 square kilometers o f land from these communities and other private land owners. The mine site is currently dedicated to subsistence agriculture. M i n e offices w i l l be established in L i m a and in the regional capital and a community relations office wi l l be established in the nearest town. 2 1 However, analysis of mineral processing parameters for ores derived from the same intrusive systems analyzed by U B C students showed that the ore could be expected to have a bond work index of approximately 10.4 kWh. and to require grinding to 50% passing 150 microns for optimal ore liberation. The ore characteristics were assumed to be uniform across the deposit. This suggests that grinding energy requirements for the ore are likely to be 25.4 kWh/t and that this would account for 85% of power consumption giving an overall consumption of approximately 30kWh/t. The power required appears to have been underestimated in the original model. 50 Local communities w i l l house migrant populations o f construction workers during the building o f mine facilities and infrastructure. The mine w i l l also be responsible for providing service improvements in the regional capital to serve employees using partnerships where possible. 3.3 Application of the APEGBC Approach to the Case Study The A P E G B C Sustainability primer methodology was selected as the approach for the research. The rationale for selecting this approach is based on a literature review which revealed that A P E G has provided a set o f guidelines which is designed to be incorporated into individual engineering practice. These guidelines are flexible enough that they can be applied by engineers working in different engineering disciplines and with different levels o f engineering responsibility. In this way it was more appropriate to the exploratory research presented here than either the 7 Questions framework, which is seen more as an assessment process for overall mining operations or the F1DIC process, which is designed to be implemented by consulting companies. B y applying the A P E G B C guidelines, it was hoped that insight would be gained into practical methods for integrating best case examples, such as the types produced by M M S D , into the design and decision-making involved in mine engineering. Where they were compatible and helpful, aspects o f other approaches were incorporated into the A P E G B C methodology,. The A P E G B C guidelines begin with an awareness building step. In the research this step consisted o f undertaking the literature review presented in this thesis. The insights from sustainability and best case mining scenarios were incorporated into the later research. For the second step: "Ful ly investigate the impacts of potential actions". The guidelines included in the second focus area (2-4) were supplemented with concepts from the multi-criteria analysis literature to select criteria for modeling. The aim o f the criteria selection and modeling was to provide sufficient data to enable the research to progress to the third focus area o f the A P E G guidelines "evaluating alternatives", which was carried out using multi-criteria analysis and decision analysis techniques in conjunction with A P E G guideline 5. The final focus area o f the A P E G guidelines, "Consultation and partnerships" was applied in a limited way to the research due to the exploratory nature of the research and the virtual nature o f the case study. This aspect is discussed in more detail in the discussion in chapter 5. The following sections outline the rationale for selecting specific research choices using the decision analysis structure. 51 3.3.1 Decision Context and Decision Structuring Decision Context The decision selected for the research is: "Which mine design scenario for the case study produces the best outcomes in terms o f sustainability criteria?" The decision analysis considers the entire mine life through to the achievement of stable post closure outcomes based on recommendations from M M S D (IIED, 2002) and Veiga et al. , (2001). Because the decision context focused on 4 mine design scenarios the research initially used an alternative focused approach to the decision following Clemen and Rei l ly (2001) rather than Keeney's (1994) value focused approach. The expected outcome was a multi-criteria performance matrix, which would quantify or qualify the performance o f each scenario for each o f the chosen criteria. This matrix would then be analyzed to evaluate the overall performance o f the scenarios. A blank M C A performance matrix is shown in Table 3-1. Scenario A B C D Weight Indicator Measure A A , w A i A 2 W A 2 A 3 w A 3 . . . A n W An B B l w B i . . .B n w B n C CI W C i ...cn W C n D D l w D i ...D„ Won Overall £ (w A 1 xAi . .w A n XA n ) . . (w D 1 xDi . .w D n XD n ) Table 3-1 Blank MCA Matrix Decision Choice The decision chosen for evaluation using the sustainability approach was the mine life decision. This decision determines the tonnage o f ore to be mined daily. In turn it affects the equipment selected and the size o f the labour force. Because o f economies o f scale and a faster payback time on borrowed capital, it is generally considered to be more profitable to mine a deposit faster with larger machinery and a smaller workforce (Bozorgebrahimi, 2003). The frequent use o f relatively high discount rates by the industry also favours shorter term projects by apportioning little value to benefits accruing towards the end o f longer projects. Longer life mines may be more vulnerable to commodity price fluctuations and provide little timely cash flow and low job numbers to local regions. Short life mines may require a significantly larger capital investment, 52 which increases the financial risk o f the project. In addition, there may be hidden social and environmental costs o f large scale, short lifetime mining, which are not captured by the economic system, but which may also increase project financial r isk 2 2 . Where corruption is a concern, officials frequently focus on immediate benefits rather than having a long term sustainability focus. In all of these areas, social, institutional, economic and environmental, it is the hidden social and environmental and engineering costs that the M C A approach aims to make explicit, describe and, where possible, quantify, so that decision-makers may operate from a more complete data set. The M C A process can also guide iterations within the mine design by identifying priorities. In terms of its implications for M C A methodologies, the decision presented involves a finite number o f alternatives and thus was amenable to simple M C A approaches. Decision Goals For the research in this thesis the objectives chosen were those defined in the A P E G B C sustainability primer in guidelines 2 to 5: • Individual and cumulative social, environmental and economic implications: maximize benefits, minimize impacts. • Address both short and long-term consequences. • Address indirect as well as direct consequences. The overall objective was to achieve a balance between these considerations in determining which mine scenario was preferable. Identification of Alternatives Once the decision-context had been identified, alternative potential mine lives were generated using the 37.5 year life from the Taylor formula (See section as a median value and rounding for simplicity. This created 4 scenarios: • Scenario A . 15 year mine life • Scenario B . 35 year mine life 2 2 It has also been recognized that there appear to be limits to the economies of scale model and that at a certain point the long run average cost curves, that is the cost of producing a unit of product over time, cease to show improvement in cost per unit produced. In a study of economies of scale in open pit mining, Bozorgebrahimi (2003) suggests that perhaps the mining industry has reached this level. Cost improvements are also related to the level of skill of workforce, learning and technological advances. 53 • Scenario C. 55 year mine life • Scenario D. No mine option. The no mine option was included based on suggested practice in the APEGBC primer (Long & Failing, 2002). It also reflects the considerable opposition to mining worldwide, which frequently suggests that the no-mine option should be considered more carefully in developing mine development scenarios (International Institute for Environment and Development, 2002). Some engineering assumptions for the three mining scenarios are shown in Table 3-1. The mine models were created in association with research assessing the potential impacts of economies of scale on the future of mining operations (Bozorgebrahimi, 2004). A second part of the identification of alternatives was improving upon the original alternatives (Dodgson, et al, 2001). In the research, improving the alternatives was an iterative process throughout the modeling. As more information became available about the probable consequences of a scenario, it was possible to envisage potential mitigation strategies. Refining Objectives: Modeling the problem structure. Scenarios A B c Mine Life, Years 15 35 55 Daily Production, Tonnes 866,692 371,440 236,371 Open Pit Personnel 2,074 1,053 733 M i l l Personnel 255 198 173 Service Personnel 592 318 230 Administration Personnel 321 173 125 Total Personnel 3,242 1,741 1,262 Electric Power Demand, kWh/day 155,016 93,237 71,091 Drills required 23 10 6 Size of Shovel required Cubic Yards 68 47 38 Number of Shovels required 9 7 6 Truck Size tonnes 380 320 240 Number of Trucks 40 24 22 Mine site clearing, Acres 4,000 4,000 4,000 Table 3-2. Engineering assumptions for the 3 mining scenarios (data from Bozorgebrahimi, personal communication). For this research a brainstorming approach coordinated by the author with UBC researchers was used initially to identify key impacts to be modeled and this was followed at a later date by a screening approach. Initial discussions aimed at determining sustainability criteria which might 54 show variation between the options. Again, the development of alternatives prior to the development o f scenarios follows the alternatives focused approach to decision analysis (Keeney, 1994). Sustainability criteria and indicators were selected based on the A P E G B C sustainability primer guidelines and are shown in Table 3-3: Criteria APEGBC Guideline Considered and Indicators (dependent variables) Measures Good Project Economic Performance: #2 Long term economic: Project economics Net present value, internal rate of return, break even copper price, competitiveness, cost of sustainability measures #2. Short Term economic: Project Economics Pay back period. Maximize Social Benefits and Minimize Impacts: #2 & 3 Long and short term direct social and economic: Socio-economic distribution of benefits Revenues to foreign, national, regional and local stakeholders. Revenue sources #2 & 3 Long and short term and indirect social: Social wellbeing Social Impact Variables Minimize Environmental Impacts: #2,3 & 4. Long term, cumulative indirect environmental: Greenhouse Gas emissions23 Tonnes of carbon dioxide equivalent and potential cost. #2 short- medium term direct environmental impact: Waste impacts Tonnage of waste produced. Table 3-3 Criteria, indicators and measures. A n examination o f the overall economics of the scenarios was included in the analysis because a sustainable project must be economically viable. A n examination of the distribution o f socio-economic benefits from the case study mine was included in the research, because the perception that all o f the impacts of mines are borne by the local community and that all o f the profits accrue either to businessmen and professionals from distant cities or to foreigners, is a key cause of unrest surrounding large scale mining projects in developing countries (IIED, 2002). Social impacts were included, because many o f the unintended impacts of mining operations are non-economic social impacts (IIED, 2001). The environmental focus was on the production o f large 2 3 Greenhouse gas emissions are an environmental issue that is long term cumulative and global. While there is debate among scientists regarding the extent of the contribution of anthropogenic greenhouse gas emissions to the phenomenon of global warming (Lindzen, 1993), a group of eminent scientists who formed Working Group 1 of the Intergovernmental Panel on Climate Change (2001) concluded that humans are already causing climate change with current carbon dioxide levels unprecedented during the last 400,000 years. In light of concerns regarding the negative impacts of rapid change from global warming, in 1992, the United Nations Framework Convention on Climate Change (UNFCCC) established a goal of stabilizing atmospheric carbon dioxide levels at a level that would prevent dangerous interference with the climatic system (United National Framework Convention on Climate Change, 2003). 55 volumes of waste, and the high energy use by the minerals sector, which were selected mainly to comply with the APEGBC requirement to consider long and short term impacts. When the selected indicators were examined using the assessment criteria from Dodgson et al. (2001), there appeared to be gaps in terms of completeness. Redundancy was not seen to be a problem and the size of the study to be realized was considered to be optimal. Potential problems operationalizing the indicators into appropriate measures were assessed during the modeling stages. Potential problems of double counting and mutual dependence of preferences were mitigated during the analysis phases. Thus the only issue which appeared problematic at this stage was completeness. A second indicator selection process was carried out in order to ensure that the indicators used were capable of comprehensively describing the sustainability dimensions of the project and to ensure that no major impacts had been omitted. The examination of indicators used F1DIC screens (van der Putte, 2002), questions from the APEGBC guidelines (Long & Failing, 2002) and indicators from the Seven Questions to Sustainability (International Institute for Environment and Development, 2002). Some important issues which had been overlooked in the initial phase were introduced through this step including institutional constraints, materials efficiency, recycling, life cycle analysis and potential substitutions. A more detailed description of the second indicator selection process is provided in Appendix II. Because the research was in danger of becoming unwieldy, it was decided that the initially selected indicators would be the focus of the multi-criteria analysis, except for the institutional category, which would be subjected to more detailed analysis. For the other supplementary indicators subjective comments would be incorporated into the appropriate sections concerning expected trends in the secondary indicators. The research would therefore have a more exploratory and less determinative character. Measures and threshold levels for the indicators were selected as a part of the modeling process. 3.3.2 Modeling of Indicators Details of the methodologies used for measuring the individual indicators are included in section 3.4 of the current chapter in overview and in detail in Appendix V. For some of the indicators measures (attributes) followed directly from the definition and standard calculation methodologies. For example the net present value was considered to be a natural measurable attribute (Dodgson et al., 2001). Other indicators required the location of more innovative methodologies and produced either direct quantitative attributes or qualitative data which were 56 ranked on qualitative scales. For some measures, threshold levels corresponding to achievement o f the overriding criterion (eg positive N P V for good economic performance) were established and for others the direction o f improved performance was subjectively defined. Models were located in the literature for modeling differences in waste production between scenarios and greenhouse gas production (World Business Council for Sustainable Development and Wor ld Resources Institute, 2001a,b,c, 2004). The financial implications o f these calculations were incorporated into the financial analysis where there were direct costs. Measuring o f the socio-economic and social wellbeing and institutional implications o f the mine design required significant research to locate suitable comparative data from which to extrapolate measurable effects and, in the case o f institutional impacts, suitable predictive tools. A n influence diagram, shown in Figure 3-3 was also created to depict linkages between the chosen objectives and indicators to gain greater insight into the connections between indicators. Figure 3-3. Influence diagram for sustainable outcome at the case study (dotted lines represent influences). While it is recognized that the linkages chosen are subjective and might be better derived in a participatory setting, some initial insight can be gleaned from the diagram. The influence diagram showed that, the environmental impacts were more clearly linked to economic outcomes than the social impacts. From this diagram it was expected that social data would require more innovative modeling techniques in order to understand the magnitude of impacts. The social impacts were also linked in a complex fashion to other social impacts and it was therefore expected that there would be significant iteration between institutional, socio-economic, and community wel l being 57 models. The high level of interdependence between indicators suggested that the application of simple linear additive models to the data might introduce substantial double counting o f impacts. 3.3.3 Treatment of uncertainty, preferences and level of analysis Once measurable indicators had been modeled, the next steps in the decision analysis were the identification o f uncertainty and o f weighting. In practice, the final selection o f indicators for the criteria and consideration o f uncertainty for each indicator were carried out within the detailed modeling context o f each criterion (included in Appendices V - I X ) . Weighting was not modeled formally, but weights were included in the quantitative analysis as weights w A ) to w D n . In addition the research experimented with assigning equal weight to economic, social, environmental and institutional criteria simulating the balance suggested by a sustainability approach. A subjective analysis o f expected stakeholder preferences was utilized to assess priorities for the qualitative analysis o f the multi-criteria matrix (Appendix IV). Multi-Criteria Analysis Matrix Construction In order to evaluate the different scenarios following A P E G guideline #5, the indicators were assembled into a matrix which included all relevant measures for each criterion. The initial concern was to achieve a representation within the M C A matrix o f the richness of the data uncovered during the modeling process. Thus measures portraying the level o f risk associated with a particular criterion, its time variability and its summation over the lifetime o f the project were all included in the matrix. Quantitative Analysis The quantitative analysis applied to the data consisted first o f scoring the alternatives on a scale of 1 to 10 for each criterion. For many measures this was a simple task o f scoring numerical values in a way that provided a consistent data trend where 10 represented good performance and 0 represented poor performance and o f scaling the values on a 0-10 scale. For other measures the scoring was more subjective and involved estimating a "best case" value and assigning a 10 to this value. The results data in the criteria matrix were then scaled relative to this best case value on a 1-10 scale. Once measures had been scored, a careful deliberative process was applied to each criterion attribute considering the extent to which the information it provided was unique and the extent to which it overlapped or duplicated other measures. Economic performance for example had been expressed in terms of N P V and IRR and these represented a repetition of data. Duplicate or near duplicate measures were eliminated. 58 Once these two processes had been completed, an addition o f the scores was carried out using all and the results were examined in terms o f the relative performance of each scenario in terms o f economic, social, environmental and institutional aspects and in terms of overall performance. The weighting scenario set the sum of weights A | to A n at 0.25, B | to B n at 0.25, C\ to C n at 0.25 and D) to D n at 0.25 and assumed equal distribution of weights within each category A to D . This weighting scenarios was recognized to be artificial, however, it enabled some insight to be gained from the process in the absence of interdisciplinary or stakeholder weighting data. Qualitative Analysis The data were also processed using a qualitative approach through the use of consequence tables. In this exercise, scenarios which represented better and worse performance on a set o f related variables were compared in pairs. For any measure that produced better performance for one scenario, there was usually another measure which performed more poorly and it was common that the measures were related to different criteria and/or impacted different stakeholder groups. A n important first step in the qualitative analysis was therefore to attempt to determine the interests o f the different stakeholders. Because the approach taken to defining the problem was not based explicitly on stakeholder values this was challenging. Data from a survey o f Peruvian stakeholders in mining were used in an attempt to determine interests (Glave & Kuramoto, 2002), but the data were poorly suited to the function o f determining interests as they did not distinguish the attributes o f indicators being compared (discussed in Appendix IV) . Once interests had been determined, it was possible to identify trade-offs directly in the data and to use the consequence tables to suggest improved alternative scenarios. The inclusion of stakeholder interest data in the qualitative analysis was the only attempt in the research to address the consultation and participation focus of guidelines 6 and 7 o f the A P E G framework. 3.4 Criteria Modeling This chapter concludes with an overview of the modeling o f the economic, social and environmental criteria identified in the M C A . More detailed descriptions including the rationale underpinning decisions and the data sources used are located in Appendices V - I X . The six criteria selected (as described in for detailed modeling were: good project economics, equitable socio-economic distribution, maximize social benefits and minimize impacts, optimize institutional impacts, and minimize environmental impacts. The modeling aimed to produce data which were capable o f comparing the four scenarios in terms o f the criteria selected. 59 3.4.1 Economic Criteria The analysis of economic criteria incorporated both the modeling o f the mine economics and the modeling of the socio-economic distribution o f revenues from the project. Project Economics The economic analysis sought to determine: • The impacts on N P V o f the different scenarios. • The impacts on financing of the different scenarios. • The sensitivity to key risks/uncertainties of the financial mine model under the different scenarios. • The impacts o f investing in sustainability measures on the overall N P V . M i n e economics for each o f the scenarios were modeled using a scoping level o f modeling and followed standard assumptions (Buchanan et al., 2002; Hustrulid & Kutcha, 1995; Mcintosh, 2003). The underlying assumptions for the economic model for the case study were described earlier in this chapter and the model was formulated in conjunction with a mining engineer 2 4. Assumptions for N P V calculations A n initial spot copper price o f 72.6 cents 2 5 per pound was used in the mine optimization representing the spot copper price in January 2003 when the mine optimization program was run and follows the rule o f thumb that spot copper prices are used in scoping studies (Mcintosh, 2003). The subsequent modeling of capital and operating costs used a long run average copper price of $1.10 per pound 2 6 (Otto et al., 2000) to provide an expected case and the 72.6 cent figure to provide a worst case scenario, however, the original mine model was not re-optimized to account for the long run average price. 2 4 Anoush Bozorgebrahimi collaborated with this research. 2 5 A l l economic data throughout the thesis uses $US unless otherwise specified. 2 6 Subsequent discussions with mining professionals suggest that this value is high for a long run average copper price, however it is representative of current price trends. 60 Capital cost assumptions for each mining scenario are included in Appendix V . Data were extrapolated from equivalent mines using standard formulae 2 7 and the need to build infrastructure and to account for Peruvian salary levels were incorporated into the model. The standard costing method employed in mining engineering for summarizing operating costs and revenues along with capital expenses is the discounted cash flow analysis which produces net present values ( N P V s ) o f future revenues and costs (Buchanan et al. , 2002). In order to calculate the N P V o f the mining project, a discount rate is applied to the annual cash flows. Mcintosh (2003) states that mining company executives generally use a discount rate of 15% when making decisions regarding potential mining projects and this was therefore the discount rate applied in the analysis 2 8 . Financing Considerations M i n i n g operations require an initial outlay of capital to fund the construction and development phases which must be provided through a combination o f loans and equity. Financial analysts and the company board are concerned with the project value expressed in terms o f a positive net present value and a fast payback on investments. There may be some requirements for a minimum length o f mine operation in order to comply with smelter contracts and financing with some insititutions (Batista, personal communication A p r i l 2004). Mcintosh (2003) states that economic evaluation is based on four standard economic measures: the net present value, pay back time and internal rate o f return (which are different expressions o f the same discounted cash flow and formulae) and the competitive cost of the mine 2 9 . A n empirical requirement for a mining operation is that payback should occur before half o f the ore reserves have been mined (Mcintosh, 2003) and the competitive cost is considered to be the most reliable tool for evaluation o f mining projects by financial institutions and mining companies under conditions o f depressed commodity prices. Be low average production costs are considered to be a favourable indication and costs in the lowest quartile prove feasibility using this analysis (Mcintosh, 2003). However, 2 7 The power rule was applied to costs extrapolating from yearly tonnages mined. 2 8 Comments received on drafts of this work suggest that this figure is very high, although it may apply to junior exploration companies. Larger companies use rates closer to 10% and may on occasion use rates as low as 6% (Ringwald, personal communication). 2 9 The net present value represents the difference between positive and negative cash flows for the mine discounted to the present time at a predetermined interest rate (Mcintosh, 2003). The pay back time is the number of years taken for cash inflows to balance cash outflows and do not use a discount rate. 61 no measure alone, determines good project economics. In addition project risks are considered by potential financiers. Project Risks Bureaucratic Expropriation . -Environmental-Violent , Upheaval .,;Lijnked to supply, demand,*;,:'; »*replacement geopolitics, ." industry structure, speculation etc Indirect Link Direct Link Figure 3-4. Risk in the Minerals Industry (based on MacDonald 2001). Project risk is a key concern in determining the feasibility o f a mining project. Macdonald (2001) delineates mining risk into three types: intra-firm, intra-industry and extra-firm, al l of which "interact with each other and change over time and space" (p. 19). Intra-firm risks are related to specific corporate strategies and management, for example level o f specialization, and speed o f expansion; intra-industry risks are those related to the time frame and uncertainties involved in the geologic modeling and permitting aspects of the industry, while extra-firm risks are the social, political and commodity market risks which are outside the immediate control o f a company but wield considerable power over the success or otherwise o f a mining project. These risks and their direct and indirect linkages are shown in Figure 3-4. Commodity price and geological uncertainty risks were considered in the economics section o f this thesis. Political and social risks were considered integrally with the social and institutional modeling. The break even copper price and the N P V at 72.6 cent copper were considered 62 measures of commodity price risk. Geological risks were addressed qualitatively. A more detailed discussion of project risk is included in Appendix V . Costs of Sustainability Measures Direct investment in development initiatives by the company and investment in training programs were estimated in the analysis. Details o f these calculations are included in Appendix V . Summary of Economic Model Economic analysis is an essential element o f project analysis in terms of sustainability. The economic model used in the research used scoping level assumptions, extrapolating costs from equivalent mining operations and using rules o f thumb to construct a model appropriate to the case study deposit. K e y assumptions used in the model included static copper prices o f 72.6 cents and $1.10 per pound, open pit mining and simple flotation processing technologies and salary levels at 80% of North American levels to account salary levels and infrastructure construction in a developing country. A discount rate o f 15% was applied to each of the scenarios and an assessment was made o f the financial viability o f each scenario. This section has presented a standard financial assessment of the case study project. However, the research also considers the distribution o f economic benefits and an analysis of these are included in section below. Socio-economic Distribution The decision to model the distribution o f socio-economic benefits was based on the potential of providing data to describe the differences in the distribution o f economic benefits for the different mine scenarios and thus enable strategies to mitigate perceived inequities. The model sought to address: • How the different scenarios affect the patterns o f socio-economic distribution o f revenues between stakeholders. • How the contributions o f different revenue sources are balanced for the stakeholders under the different scenarios and the implications of this. • The sensitivity o f revenue distribution to certain key assumptions and uncertainties. The distribution o f benefits was modeled using socio-economic analysis, which is often viewed as a sub-methodology o f social impact assessment because it typically occurs in association with the social impact assessment o f a project (Taylor et al., 2004). 63 The methodology makes use of a variety of approaches for the projection of socio-economic impacts. Commonly used approaches include: • The comparative method: Projections are based on past research and experiences in similar cases. • Straight-line trend projections: The future is projected by assuming that current trends will repeat in the future. • Population multiplier methods: Relate changes in infrastructure needs and housing to population size projections. • Statistical significance: This method assesses the probabilistic differences between projections involving the project and projections without the project. • Scenarios: These are logical imaginations of hypothetical futures created by mentally modeling the variables selected. • Expert judgment: Uses people who are familiar with the area to present scenarios and discuss the significant implications of projects and • "Calculation of futures foregone" which is a method for valuing options which are given up by opting for a particular project or plan (option values) (Burdge and Colleagues, 2004). In reality most socio-economic analyses rely on a combination of methods and this is encouraged by Freundenburg (1998) who suggests that such triangulation of data is essential for identification of bias in modeling and collecting data. An important consideration for the comparative method is that communities compared must be "at least roughly similar in terms of basic characteristics (e.g. size, socioeconomic profiles) to ensure comparability" (Taylor et al., 2004 p. 94). The comparative method was used extensively in the socio-economic distribution model in this research as well as later in the community wellbeing model. Assumptions in the Model Economic benefits of mining operations to local and regional communities, and local, regional and national governments are derived through seven main mechanisms: 1. Companies hire local, regional and national workers directly. 2. Companies hire contractors who in turn hire local, regional and national workers. 64 3. Companies contract directly with local, regional and national companies for products and services. 4. There are indirect and multiplier effects o f wage payments to local, regional and national employees reflecting their ability to hire local people (employment multipliers) or to buy goods in local markets (income and employment multipliers) 5. Companies pay taxes to local, regional and national governments and in some cases taxation revenues from central governments are redistributed to the local level (e.g. the Peruvian canon law). 6. Companies carry out community development initiatives (infrastructure and human/social capital development) either directly or through Foundations and partnerships with International organizations, Non-governmental organizations and government bodies (International Institute for Environment and Development, 2002). 7. Companies purchase land and mining rights and may also pay additional compensation to local people (Pasco Font et al . , 2001). A l l o f these revenue sources were modeled except for item 4: indirect and multiplier effects. This item was omitted from the majority o f the analysis because the main aim was to produce data that were comparable between different stakeholder groups. Multipliers would have reduced the equivalency o f the data sets. However, multiplier effects are especially important in the regional context, where indirect economic development is a significant source of economic benefit. For this reason a regional multiplier o f 1.65 was applied to the direct and contracted salaries from the mine to provide a measure o f regional economic development. This multiplier was used by Pasco Font et al . (2001) and was calculated based on specific observations of regional spending by mine employees in Peru. Details o f the calculations for socio-economic modeling are provided in Appendix V I . The revenues were calculated for four different stakeholder groups. Local stakeholders who were defined as those l iving in the project district 3 0, regional stakeholders, who were those l iving in the same "department" as the mine, national stakeholders who included all Peruvian beneficiaries o f the mine and international stakeholders who consisted o f all mine beneficiaries who live outside Peru. 3 0 District is the term for the smallest jurisdictional area in Peru - it would be expected to include one large settlement and all the surrounding communities. 65 Data sources included project data from the case study mine model, a model o f the Peruvian taxation system taken from Otto et al. (2000) and comparative data from other mines. Data from the Antamina and Yanacoccha mines (Antamina mining company, 1998; Damonte et al., 2002; Pasco Font et al . , 2001; Yanacoccha M i n i n g Company, 2003) provided much useful recent data from large Peruvian Andean mines, although ambiguous definition of terms provided some challenges for extrapolation. For some revenue sources Peruvian data were either absent or, as in the case of scenarios which prioritized the training of local people, the model had not yet been applied in Peru. In these cases data from other jurisdictions was extrapolated including data from the Escondida mine in Chile (Castillo et al . , 2001); Inti Raymi mine in B o l i v i a (Loayza et al. , 2001); mines in Northern Canada (Forster, Padget et al. 1994; Parsons and Barsi 2001; Sosa and Keenan 2001); the Red Dog mine in Alaska (TeckCominco, 2001; International Institute for Environment and Development, 2002); and mines in Australia and Namibia (IIED, 2002; Stanton-Hicks & Newman, 2002). The use of data from areas remote from the Andes, was not ideal, but represented a compromise strategy for modeling in situations where Peruvian data were absent and was recognized to increase the uncertainty in the data. Once each individual revenue source had been modeled, the revenues to each stakeholder group under each mining scenario were calculated using discounted cash flow analysis with discount rates o f 15% and 5% representing the mining company discount rate and a social discount rate 3 1. In addition the revenues from different sources to each stakeholder group were differentiated for payment to individuals and families versus community (government) payments and direct development spending by the company. The revenues for each scenario were compared for copper prices o f 72.6 cents and $1.10 to give an assessment o f the sensitivity to changing copper prices of the socio-economic distributions. N o mine scenario The no-mine scenario was also subjected to a socio-economic analysis to enable comparison with the mining scenarios. The approach chosen for modeling the no-mine scenario was to construct a plausible pre-mine subsistence pastoral model, estimate the replacement values o f the subsistence production generated and to extrapolate this through the mine life. Data concerning high Andean 3 1 Cost benefit analysis by governments to determine socio-economic benefits to local communities typically applies either undiscounted cash flows or a social discount rate, which is lower than project discount rates reflecting the greater value attached to long term impacts and benefits by communities (Popp, 2001). 66 carrying capacity for sheep, sheep varieties, expected meat production, the prices of meat and wool and expected potato yields at high altitude were available from the Peruvian Ministry o f Agriculture (2003) and are included in the details o f calculations in Appendix V I . Because o f an inability to locate pricing data for trout, this element of production was omitted from the model. Best and worst case scenarios were constructed for subsistence agriculture to provide a range of values representative o f uncertainty. The best case model incorporated successful agricultural outreach programs. For scenarios A and B , post mine production for the mine area was calculated using the no-mine model as a basis and accounting for pasture degradation. A n absence o f quantitative data describing the outcomes o f mine related agricultural development activities in the local and regional areas led to the exclusion o f this potentially valuable contribution to local communities from the model 3 2 . Annual revenues from the no-mine and post mine scenarios were subjected to discounted cash flow analysis at discount rates o f 15% and 5%> to allow comparison with the mining scenario data. Summary o f Socio-Economic Mode l Eight socio-economic scenarios were described for the case study. Scenarios A - D corresponded to worst case scenarios, where no investment was made in capacity building, training or technical assistance beyond that necessary for permitting. Scenarios A l to D l in contrast represented scenarios where effort was taken to train local people to work in the mine and to train local people in improved agricultural practices. For each scenario the distribution o f revenues to four groups of stakeholders: local, regional, national and foreign were calculated. In addition the revenues were disaggregated by source in order to gain insight into the balance o f contributions flowing to each stakeholder group. 3.4.2 Social Criteria The modeling o f community wellbeing for the scenarios utilized a systematic approach based on the social impact assessment methodology (Burdge, 2004; Burdge and Colleagues, 2004; Freundenburg, 1998; Taylor et al. 2004). The methodology aimed to develop an understanding of: • K e y changes associated with mining in the case study scenarios. These were subdivided into: 3 2 Because of the focus of the sustainable mining community model on converting the capital of the mine into economic development, human and social capital for the local community this omission is important and is discussed in chapters 4 and 5. 67 o Population impacts, o Community/Institutional Impacts 3 3, o Community transitions/changes o Individual and family level impacts o Community infrastructure needs (Burdge, 2004) • How these impacts were expected to change and develop through the life of the mine. • How these impacts were expected to differ for different communities of interest both spatially (communities local to the mine, local to the port, hosts to construction camps etc) and within the same community (politicians, business people, indigenous people, subsistence pastoralists, women etc). SIA Methodology Social Impact Assessment (SIA) provides a "systematic analysis in advance o f the likely impacts a development event (project, policy or plan) w i l l have on the day to day life (environment) o f persons and communities" (Burdge, 2004 p. 2). Vanclay (2003) provides a more encompassing and purpose driven definition: " SIA includes the processes o f analyzing, monitoring and managing the intended and unintended social consequences, both positive and negative, o f planned interventions (policies, programs, plans, projects) and any social change processes invoked by those interventions. Its primary purpose is to bring about a more sustainable and equitable biophysical and human environment". S IA plays an important role in forewarning people so that proactive creative approaches may be taken to the changes and should be applied at the nodes o f change: construction, construction to operations, and closure o f operations and any other major changes introduced en route e.g. major technology changes (Taylor et al., 2004). S IA is underlain by a specific value orientation which promotes equity and sustainable development; it is also a participatory process (Taylor et al., 2004), although this element was not possible for the thesis research. The subjective nature of the SIA methodology was managed through careful application o f variable calculation guidelines in the research. Freundenberg (1998) further suggests the use of researcher sensitivity analysis, interdisciplinary double checks and public involvement techniques to identify gaps and blinders in the research. Institutional impacts were deferred to institutional modeling. 68 Taylor (2004) advocates the use of the analytic inductive approach in SIA. In this approach available (usually secondary) data are analyzed and a conceptual model is formulated for patterns observed. A second period o f data gathering follows and the conceptual model is tested for consistency with new data uncovered. Discrepancies require re-conceptualization until a model has been formulated that is consistent with observations. Effectively carried out inductive analysis is argued as a powerful tool for validation o f conclusions. Another key approach underpinning social impact assessment is the comparative model. This model was introduced in section and involves an understanding o f change which has occurred in similar communities, which have encountered similar projects. The similarity between communities and projects is confirmed through profiling the communities and project and the comparative material is used to project baseline conditions into the future. Most future predictions used in the research are linear trend projections, use multipliers based on projected population changes or use scenarios (Taylor et al. 2004). The social impact assessment (SIA) process has been delineated into 8 steps by Burdge (2004): 1. Describe the proposed action. 2. Describe the history and baseline conditions 3. Begin to identify the full range o f social impacts to be addressed (Scoping) by: a. Identification o f the project stage and time line for the project. b. Establishing the region or zone of influence for the SIA. c. Identification of key stakeholders. d. Identification o f the full range of Social Impacts. e. Thinking about alternatives. f. Locating information and decide on methods. g. Making a work plan. 4. Investigate/Understand project effects. 5. Determine significance o f effects. 6. Propose alternatives. 7. Mitigation/Enhancement. 8. Monitoring. 69 The social impact assessment applied in this thesis included steps 1-4 o f the overall process outline and focused on the scoping steps 3.a-d to reflect the scoping stage analysis applied elsewhere in the research. The proposed history and baseline conditions form the first section (3.1) of the current chapter. The scoping stage of the research led to the recognition o f three different social impact events which required study for the case study mine: construction; the transition to operational stability; and mine closure. From the delineation o f the area of influence o f the mine included in section 3.1, six different communities were identified for consideration in the SIA: the 'Campesino community', the nearest local town, the nearest city, the port city, construction camp towns and the communities along transportation corridors. In addition it was recognized that communities would be heterogeneous and that the interests especially o f vulnerable and marginalized groups including women, indigenous people, elders, children and young people, the poor and the unemployed should be considered (Freundenburg, 1998). The identification o f social impacts was carried out using a list o f social impact assessment variables from Burdge (2004) (Table 3-2) 3 4 and also using the chaining and webbing approach advocated by Taylor et al. (2004). In the chaining and webbing approach, direct social and environmental effects o f the mine were used as starting points for the development o f chains of linked impacts which were then connected together in webs. A n example chain and webbing diagram is included in Appendix X . The social impact variable list was compared with the chain and webbing diagrams and used as a screening tool for each community identified at each social impact node to identify impacts for further investigation in step 4. 3 4 It is recognized that Burdge uses negative rather than neutral terminology for some variables, representing the commonly observed impacts of the changes described. The terminology is used in this thesis to prevent confusion, however, it may be possible to use more neutral language for some variables, which could potentially impact results, especially i f used in a multi-stakeholder environment. 70 Social Impact Variables: The Current List of Twenty Eight. (Burdge. 2004) Population Impacts 1. Population change 2. Influx or out flux of temporary workers. 3. Presence of seasonal (leisure) residents. . 4. Relocation of individuals and families 5. Dissimilarity in age, gender, racial or ethnic composition Community/Institutional Arrangements 6. Formation of attitudes towards the project 7. Interest group activity 8. Alteration in size and structure of local government 9. Presence of planning and zoning activity 10. Industrial diversification 11. Living/Family wage 12. Enhanced economic inequities 13. Change in employment equity of minority groups 14. Change in occupational opportunities Communities in Transition 15. Presence of an outside agency 16. Inter-organizational cooperation 17. Introduction of new social classes 18. Change in the commercial/industrial focus of the area 19. Presence of weekend residents (recreational) Individual and Family Level Impacts 20. Disruption in daily living and movement patterns 21. Dissimilarity in religious and cultural practices 22. Alteration in family structure 23. Disruption in social networks 24. Perceptions of public health and safety 25. Change in leisure opportunities Community Infrastructure Needs 26. Change in community infrastructure 27. Land acquisition and disposal 28. Effects on known cultural, historic and sacred and archaeological resources. Table 3-4 Social Impact Assessment Variables from Burdge (2004). Modeling of the individual impact variables followed the guidelines provided in the Community Guide to Social Impact Assessment - 3rd Edition (Burdge, 2004). Details of the specific assumptions and calculations applied to assess the variables are provided in Appendix VI I . Data sources included the project data for the case study, data concerning population trends, employment statistics, and gender roles from the Peruvian National Institute o f Statistics (INEI, 2004), data concerning projected employment numbers at different stages o f the mine life from the socio-economic model, and comparative data from other mine sites. Burdge's (2004) directions also included threshold limits for significance o f the different S IA variables. When the data were compiled into a matrix, quantitative values were included where these data were meaningful. A frequent notation used was to categorize the impacts as significant (over the threshold value suggested by Burdge), very significant (more than twice the significance threshold) and extremely significant (over three times the threshold level). Summary of Community Well-being Model In summary, the SIA model focused on three key transitions in the development of the case study mine; construction, the transition to operational stability and closure. It employed the SIA methodology to the four mining scenarios and aimed to establish significant differences between 71 the scenarios for each o f the three transitions. Institutional variables identified in the social impact assessment model were deferred to section 5.3.2 below, which focuses on the institutional situation in Peru and its impacts on potential outcomes at the case study. 3.4.3 Institutional Criteria In light o f the growing consensus that institutions play a critical role in facilitating sustainable development, this section o f the research aimed to assess the implications o f the institutional situation in Peru on the social, environmental and economic impacts o f the case study scenarios. Although company institutional factors were recognized as being key to determining the outcomes o f mining projects on local communities and environments, their analysis was not attempted in the research because no specific mining company was identified as the proponent. The understanding o f the role of governance in development and the processes and causes of change in governance is a rapidly expanding field (World Bank, 2003). Many indicators o f governance are qualitative rather than quantitative and predictive tools for institutional modeling are in their infancy (Alcazar & Wachtenheim, 2001). Institutional analysis is a methodology developed by the World Bank which "describes the range o f formal and informal institutions in the public, private and not-for profit sectors and assesses which institutions and institutional linkages are critical to project success (World Bank, 2003). Ross (2001) suggests that government accountability and responsiveness can be measured through gauging the level o f corruption, through measuring how democratic a country is or through assessing how effectively the government addresses health and education concerns. Limited predictive tools for institutional modeling are outlined in some detail in the SIA methodology o f Burdge (2004). Important indicators o f institutional performance in understanding the political settings for projects are outlined by Freundenburg (1999), although detailed guidance on qualifying or quantifying these variables is not provided. Bethelier et al. (2004) have recently derived a framework using four institutional indicators: governance, security of transactions, innovation and regulations, with detailed guidelines for their measurement, which has potential to f i l l this gap, however, the framework, was not available at the time o f the research. In the absence o f more systematic methodologies for predicting the impacts o f the institutional setting, a two part approach was applied to the case study data. The first part involved an assessment o f institutional S IA variables (Burdge, 2004; Freundenburg, 1999), to the scenarios for the case study. The second stage applied quantitative data from Alcazar and Wachtenheim (2001) concerning the leakage o f funds in social programs in Peru to the socio-economic models 72 developed earlier in this research. The aim of this second step was to produce institutionally realistic socio-economic distribution data. SIA Variables The SIA variables assessed were: • Presence of an outside agency, • Inter-organizational cooperation, • Alteration in size and structure o f local government and • The presence o f planning and zoning activity. (Burdge, 2004) • Fruendenberg's variables: o Stability o Changes in the distribution of power and authority o Level o f leadership capacity present o Trust in politicians and social institutions (Freundenburg, 1999) Details o f the calculations of each institutional variable are contained in Appendix VII I . Integration of Institutional Aspects into Socio-Economic Model The second phase o f the institutional modeling applied data concerning leakage levels in public funds in Peru, obtained from Alcazar & Wachtenheim (2001) to the socio-economic data produced in the socio-economic modeling. Two leakage scenarios were applied to the data: an optimistic scenario, in which 30% of revenues were lost through leakage and a pessimistic scenario in which 70% of the revenues were lost. The leakage data were applied only to the revenues predicted to accrue to local and regional governments through the canon redistribution mechanism. Salaries, contract incomes and direct spending by the company on development initiatives were assumed to have no impacts from leakage. The details of these calculations are also included in Appendix VIII . 3.4.4 Environmental Criteria The M i n i n g , Minerals, and Sustainable Development ( M M S D ) project concluded that there are seven principal environmental issues associated with large scale mining (International Institute for Environment and Development, 2002). These are: 73 1. The production o f large volumes o f waste, 2. The challenges o f environmental management, 3. High energy use in the minerals sector, 4. Predicting and managing metals in the environment, 5. Managing threats to biological diversity, 6. Mine closure planning and 7. Min ing legacies such as abandoned mines (IIED, 2002). The modeling o f environmental impacts in the research concentrated on the modeling o f energy use through the surrogate of greenhouse gas emissions (and through this an indirect measure o f the production o f local air pollution in the form o f particulates, SOx and N O x ) and on the production of waste. These two elements were selected to represent long term and short term environmental impacts as suggested in the A P E G B C guidelines (Long & Fail ing, 2001). In addition, the secondary criteria selection process identified a number o f other environmental variables, which were considered qualitatively in the analysis. These are dicussed qualitatively in the results in Chapter 4 and included: • Geochemical considerations, • Ecosystem resilience, • Water management considerations, • Intergenerational equity, • Life cycle Impacts, • Need for materials produced • Materials efficiency. Greenhouse Gas Production Greenhouse gas modeling was based on the GHG Protocol Corporate Standard (World Business Council for Sustainable Development & World Resources Institute, 2004). In addition the GHG Protocol Project Quantification Standard: Road Test Draft (World Business Counci l for Sustainable Development & World Resources Institute, 2003) provided some useful guidance. The modeling began with a scoping step to determine the extent o f the copper life-cycle which 74 would be included in the analysis. The activities included in the analysis through scoping are outlined in Appendix I X . A series of assumptions were then generated in order to provide a structure for the modeling. The assumptions were based on assumptions used by Norgate and Ranking (2000) for modeling greenhouse gas production by base metal mines in Australia and were adapted to provide assumptions which were compatible with the expected conditions at the case study. These are also outlined in Appendix I X . From these assumptions each individual greenhouse gas emissions source was modeled. Details o f the individual modeling and data sources are included in Appendix I X . Once data from each source had been modeled the data were combined to provide data concerning the annual emissions profile of each scenario, as well as the overall lifetime cumulative emission level. Data calculated were compared with published literature quantifying greenhouse gas emissions from copper mining operations. The potential cost o f greenhouse gas emission for each scenario was calculated by applying a figure of $15 Canadian per tonne of carbon dioxide emitted, which corresponds to a guaranteed maximum figure declared by the Canadian government effective for a period of 15 years (DeMarco, 2002). Because Greenhouse gas emissions are related to energy consumption, the cost o f reducing greenhouse gases production can be partially mitigated by savings in energy costs i f energy is used inefficiently or there is a viable alternative energy source available 3 5 . Waste Generation Impacts Waste production impacts were modeled in two ways. The first approach considered the additional ore processing and tailings production resulting from the increased selectivity and smaller block size o f the smaller equipment used in scenarios B and C . In addition the step accounted for increased revenues generated by additional copper produced. The second approach considered the extra waste rock production caused by the increased haul ramp width in scenarios A and B . 3 5 In the Chilean copper industry for example, fuel costs account for 15% of the mining and processing costs and therefore substantial overall savings can result from measures to economize on energy usage (Maldonado, 1998). 75 Tailings Generation Difference in tailings generation between the scenarios related to the fact that smaller machinery is able to be more selective than larger equipment 3 6. This is reflected in the block model for the scenarios in which the smaller shovels for scenarios B and C require a smaller block size and hence are able to mine more o f the deposit than the larger trucks in scenario A . This in turn leads to the production o f greater volumes o f tailings but also to the production o f greater quantities o f saleable copper. The impacts o f extra ore processing are therefore both environmental and economic. Scenarios B and C generate more tailings than scenario A , which creates environmental impact and economic cost, however they also generate more revenue, which depending on the price of copper, may offset the cost o f the increased waste generation. Extra ore processing was modeled by extrapolating data produced by Bozorgebrahimi for an equivalent deposit using block sizes o f 5m and 20m. A t the case study scenarios B and C both used a block size of 15m while scenario A used a block size o f 20m. In addition, the sensitivity o f Bozorgebrahimi's model to copper price was investigated. Details of these calculations are included in Appendix I X . Waste Rock Impacts Differences in predicted waste rock generation between the scenarios were caused by the fact that larger equipment requires wider haul roads in order to operate safely. The research applied a geometric model designed by Bozorgebrahimi (2003) to quantify the additional waste produced by wider haul roads, which is detailed in Appendix I X . The data were based on the assumption that the mine was 820 metres deep at its deepest point. Addit ional waste produced was compared to the overall waste volumes expected for the mine and also was considered in terms o f cost in order to determine significance (detailed in Chapter 4). Summary of Environmental Modeling Environmental modeling comprised modeling o f the greenhouse gas emissions associated with each scenario as well as modeling the variations in waste production and ore processing resulting from different block sizes and equipment selection decisions. The results o f these models are presented in Chapter 4. The environmental modeling was augmented in Chapter 4 by qualitative 3 6 The increased mining cost of smaller equipment may also lead to a reduction in the reserve and thus to the mining of less ore. 76 discussions o f some o f the environmental measures identified during the second criteria selection exercise outlined in section and detailed in Appendix II. 3.4.5 Construction of M C A Matrix Once data had been collected to describe the impacts o f the scenarios in terms o f the different criteria selected, the data were assembled into a multi-criteria analysis matrix. The rationale underlying the initial matrix construction was to provide a matrix which represented the complexity and detail of data which had been produced in the modeling. The matrix was then simplified and processed using quantitative and qualitative approaches as described in section and above. The results o f the criteria modeling and the multi-criteria analysis are presented in Chapter 4. 77 CHAPTER 4. RESULTS This chapter presents the results of the indicator modeling for each scenario. The results are then assembled into the multi-criteria performance matrix and the results of the quantitative and qualitative analysis o f the matrix are presented. The chapter is divided into five sections. The first four sections present the results of the economic, social, institutional and environmental portions o f the modeling described in Chapter 3. In addition, the supplementary criteria discussed in Chapter 2 and Appendix II are included in the appropriate sections o f this chapter. The final section of this chapter integrates the results from the earlier sections o f the chapter into a multi criteria analysis and discusses the comparison of the scenarios. 4.1 Economic Impacts: Results of the Economic Models 4.1.1 Project Economics Economic Performance Scenario A B C Mine Life (Years) 15 35 55 Capital Expenditures 903,088,516 504,099,536 372,271,156 Operating Expenditures 377,561,199 214,147,541 160,494,465 Yearly Income 1,180,920,543 506,108,804 322,069,239 NPV ($1.10) 2,454,656,117 904,905,925 349,194,292 NPV ($0,726) 758,495,011 45,525,065 -202,834,540 Indicators Payback Time ($1.10) 3 4 6 Payback Time ($0,726) 4 6 15 % mined at payback ($1.10) 20 11 11 % mined at payback ($0,726) 27 17 17 IRR ($1.10) 64% 43% 27% IRR ($0,726) 33% 17% -6% Break Even Copper Price $0.52 $0.65 $0.79 Competitiveness (Production cost/lb) $0.35 $0.47 $0.55 Table 4-1 Financial Outcomes for the Mining Scenarios. The key economic outcomes for the mine are shown in Table 4-1. The data use the long run average copper price of $1.10 and a discount rate of 15% as a predictor of probable mine performance and a copper price of $0,726 as a measure o f project risk. A s expected, the profits from a shorter life mine are greater. This reflects both economies o f scale and the relatively high discount rate applied. 78 Risks andfinancing 3.E+09 2.E+09 2.E+09 CO ZD 1.E+09 5.E+08 O.E+00 -5.E+08 -1.E+09 Scenario A Scenario B Scenario C % f | 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 Year Figure 4-1. Payback Time for Scenarios A-C ($1.10 Copper, Discounted at 15%). The capital cost is significantly greater for a shorter life mine (a 79% increase from scenario B to scenario A), which may represent a greater initial risk to the mining company. Figure 4-1 shows the 15%o discounted cash flows for the project with $1.10 copper and indicates payback times of 3, 4 and 6 years respectively for scenarios A through C. All of these payback times pass the hurdle test (described by Mcintosh, 2003) that payback should be achieved before half of the ore is mined. The internal rate of return achieves the minimum standard of 20%> for all scenarios using the long run copper price. For scenario A the break even price (52c/lb Cu) is significantly below the lowest prices observed for copper, as shown in Figure 4-2, which shows trends in copper prices between 1955 and 2002 and demonstrates the cyclic pattern of copper prices. The scenario has some vulnerability due to the large volume of copper produced annually. The case study production would approximate half a million tonnes of copper per year, which compares to global production in 2001 of 16 million tonnes (Ayres et al., 2001) and is significant. In addition the scenario is vulnerable due to the untested technology employed. Scenario B also demonstrates a break even copper price lower than average recorded annual prices, however it uses tested technologies. For scenario C the 79 79 cent break even price is higher than levels that were experienced during the recent down turn in copper price giving the scenario significant risk for early or temporary closures related to copper prices. If, as suggested by the data relating to future copper need 3 7 , demand and supply, copper prices begin to rise, this risk w i l l diminish. Figure 4-2 Annual Average Copper Prices From 1955 to present (in 2000 US$), data from Perez (2004). Costs of Sustainability Measures. Development expenses were approximated as an initial payment o f 0.02% o f capital expenses with an additional expenditure o f 1.2% of net profits once the mine is operational. The investment required in training is shown in Table 4-2 as a percentage o f capital expenditure for initial investments and as a percentage of annual income for annual training costs. 3 7 A more detailed discussion of need for copper is presented in Appendix V . 80 Scenario A B C Capital Expenditure 903,088,516 504,099,536 372,271,156 Yearly Income 1,180,920,543 506,108,804 322,069,239 Training Capex 5,000,000 5,000,000 5,000,000 No. of Employees 3,242 1,741 1,262 Annual Training Cost 6,484,327 3,482,610 2,523,274 Training Capex / Capex 0.55% 0.99% 1.34% Training Annual $ / Income 0.55% 0.69% 0.78% There may also be financial benefits available mine spending to the from on Table 4-2 Costs of Training Programs. sustainability. For example, spending on local training initiatives may result in reduced attrition of the workforce and subsequently lead to lower costs for recruitment and orientation of new employees. In addition productivity drops when new employees are present. A study in Australia produced conservative estimates o f the cost of average employee turnover at a 300 person fly-in fly-out mine at $2.8 mil l ion per year (Beach et al., 2003). It is not known how these costs would compare with Peruvian costs, however i f it is assumed that Peruvian turnover rates are half o f Australian rates and that the costs per employee are also half o f these costs then average turnover costs, for scenario A would be $7 mi l l ion per year, corresponding to a 10% attrition rate. Lowering turnover by 2% would therefore provide the mine with $1.4 mil l ion per year, which is over 20% o f the cost o f the training implemented. Weaknesses in the Economic Modeling Data The key weakness in the project economic data is the failure to carry out a more comprehensive assessment o f uncertainty, although uncertainties are expected to be consistent for each o f the mining scenarios thus permitting comparison. The uncertainty is also coherent with the scoping level of the analysis which expects a level o f accuracy or ± 3 0 % . The mine model was also optimized at 72.6 cent copper and was not re-optimized to account for the higher 1.10 copper price in the later modeling. The project economic model underpins the socio-economic modeling and thus these weaknesses also impact the socio-economic modeling, the results of which are presented below. 4.1.2 Distribution of Socio-Economic Benefits The socio-economic modeling exercise illuminated trends in the annual and overall distribution of revenues between the stakeholders. Further, the exercise demonstrated the relative contributions of revenues from different sources under each scenario. The inclusion o f the no-mine scenario in 81 the model provided a useful contrast to the mine scenarios and demonstrated the economic incentives for several stakeholders of the mining scenarios. Comparison of Total Revenues Accruing to Each Stakeholder Group NPV $1.5 $US (billion) $1 $0.5 $0 A1 B1 C1 D1 B Q Local a Regional s National • Foreign • Total Scenario Figure 4-3. Distributions Discounted at 15% Non Training A - D and Training A1-1) 1. A s shown in Figure 4-3, the 15% discounted revenues to all groups decreased from scenario A to scenario D . Thus, at the initial stage of the analysis, Figure 4-3 suggested that the shorter mine lives produce more favorable socio-economic outcomes for all stakeholder groups than the longer mines. The data also indicated that socio-economic performance at the local level is significantly improved under scenarios where training is prioritized. This trend is demonstrated in Figure 4-4 which shows an increase o f 34% for scenario B and 44% for scenario C when training programs for local people are implemented. Revenues in the region increased by 4% and 8% for scenarios B and C with training respectively. When the regional multiplier o f 1.65% was applied, this improvement increased significantly. National and international levels remained fairly constant between training and non-training scenarios and scenario A showed no improvement for 82 scenarios where training was implemented because there was insufficient time for training to produce results. Despite the use of generous assumptions for the value o f agricultural production (scenario D l ) , local incomes for the no-mine scenario produced less than 27% of the minimum local incomes generated from the mine (this even included the use o f successful training programs, which are not l ikely in a non-mining situation). Expected maximum local incomes for the no-mine scenario are less than 5% of minimum mining revenues. Regional, national and foreign stakeholders also gain no socio-economic benefits from the no-mine scenario. Figure 4-4 Comparison of Local Revenues between training and non training scenarios, 15% discount rate. Socially Discounted Distributions The results o f applying a 5%> social discount rate to local revenues are shown in Figure 4-5. In this figure it can be seen that the improved performance for local stakeholders increased to 61.5% from scenario B to scenario B (training) and 112%> from scenario C to scenario C (training). A more important finding shown in Figure 4-5 is that revenues to local stakeholders increased between scenarios A and B with A and C showing equal performance when social discount rates were used. Regional performance was better for training than non-training scenarios although 83 overall revenues to regional stakeholders were lower for scenarios B and C than for scenario A . National and international revenues both decreased slightly from non-training to training scenarios when modeled using the social discount rate. Decreases were 6% and 11% for national stakeholders for scenarios B and C and 2% and 6% for international stakeholders. Thus it appears that gains by local and regional stakeholders are largely accounted for by losses in revenue at the national level when training programs are put in place. This is logical because the majority of money which goes abroad is the after tax portion of profits, which is fixed in this analysis. Therefore apart from any gains which can be created by employing fewer ex-patriot employees 3 8 all other gains at the local level must be balanced by losses at the regional or national level. $US 2.50E+08 2.00E+08 1.50E+08 1.00E+08 5.00E+07 0.00E+00 E3 No Training B Training Training No Training Figure 4-5. Comparison between local revenues from training and non-training scenarios (Social Discount Rate). Temporal Distribution of Revenues among Stakeholder Groups The temporal distributions in revenues between stakeholders were depicted most clearly in cash flow diagrams. The diagram in Figure 4-6 shows that national and foreign stakeholders begin with large shares of revenues (43% and 44% respectively) while local and regional stakeholders 38 T h e a b i l i t y to e m p l o y n a t i o n a l e m p l o y e e s depends o n there b e i n g su i t ab le l o c a l c a p a c i t y . 84 begin with a small share (3.5% and 9.5% respectively). A s the revenues to local and regional stakeholders climb (at approximately 7%> and 1.2% per year respectively) the revenues to national and foreign stakeholders fall (at approximately 0.9% and 1.0% per year). The cash flow diagram (Figure 4-6) explains the observed variations in revenue distributions to stakeholders when different discount rates are used 3 9 . When a higher discount rate is used, the long term gains to local and regional stakeholders are given less weight, while the short term gains to national and foreign stakeholders are emphasized. In contrast, when lower discount rates are used, the long term benefits to local and regional stakeholders are given more weight and the longer term mines perform better in terms o f the distribution of benefits to all stakeholder groups. Income (US$) 20E+08i 1.6E+08 1.2E+08 8.0E+07 4.0E+07 Local Revenues — Regional Revenues .Nat ional Revenues __Foreign Revenues . • V -4 0 4 8 Note income before mine operation is salaries and compensation etc. 12 16 20 24 Year of Operations 28 32 Figure 4-6 Cash Flows for Scenario B (35 years). It is possible that different discount rates are appropriately applied to revenues accruing to different stakeholder groups, as they describe different time value o f money preferences. Social discount rates may be more appropriate to the local and regional contexts, while project rates may 39 See d i s c u s s i o n o f d i s c o u n t rates i n A p p e n d i x V . 85 be more appropriate to companies and national governments. Notwithstanding this, one mining company executive in Peru commented that communities living near to mine sites generally seek short term benefits (Cantuarias, 2003). Distribution of Benefits by Revenue Source Another set o f data produced in the socio-economic assessment o f the scenarios (Figure 4-7 Figure 4-8 and Figure 4-9) describes the relative contributions o f different revenue sources (salaries, taxation/profits, canon revenues and direct spending on developments) to the income o f each stakeholder group. 1,000,000,000 900,000,000 800,000,000 700,000,000 600,000,000 500,000,000 400,000,000 300,000,000 200,000,000 100,000,000 -0 Income US$ • Total • Salaries S Contract Salaries E9 Mine Taxation H Income Taxes Figure 4-7. Revenue Sources at the National Level (15% Discounted). Figure 4-7 shows the relative contributions o f different revenue sources at the national level discounted at 15%. Taxation revenues dominate in all o f the scenarios, accounting for 69% o f revenues in scenario A , 60% in scenario B and 58% in scenario C . The taxation revenues received by national stakeholders are comprised of both direct taxation o f the mining operation and taxation o f the salaries o f mine workers and contracted employees. Taxation o f the mining operation decreases from scenario A to scenario C from $545 mi l l ion to $120 mil l ion respectively. The relative contribution of income taxes becomes more important from scenario A 86 to scenario C because a greater number of man-hours are required to mine the ore reserve at the longer term mines. Income taxes provide 17% of taxation in scenario A while in scenario C they provide 35% of taxes. A t the national level, the pressure to service external debts and provide services to the Peruvian people means that the overall size o f the taxation revenues w i l l be one of the most important considerations for government decision-makers, thus the national government would be expected to favor scenario A . Community versus Individual/Family Payments A t the local level the balance o f relative contributions to revenues is a key determinant of the outcome o f the distribution o f benefits, because extended families play a key social role and political systems are immature and struggle with corruption. For example: • Payments distributed as salaries or compensation for land purchases w i l l generally benefit only individuals or their families. • Payments distributed through the mining canon and in development projects have, at least in theory, a higher chance of bringing benefits to the wider community. $300,000,000.00 $250,000,000.00 $200,000,000.00 $150,000,000.00 $100,000,000.00 $50,000,000.00 $0.00 • Total 89 Salaries 0 Contracts • Land Purchased E9 Community Development E3 Canon E3 Agricultural Value Figure 4-8 Revenue Sources at the Local Level Scenarios A-D ( 5 % Discounting) 87 Figure 4-8 shows the relative contributions o f different revenue sources at the local level when training initiatives are implemented. For scenario A , company funded community development initiatives and canon revenues predominate; accounting for 83% of the income to the local community. For scenario B , canon and development incomes account for 56% of the total revenues with approximately 33% of the revenues derived from salaries. In scenario C , canon and development revenues contribute 44% to the total revenues, with salaries accounting for approximately 40%> of total revenues. A similar pattern was seen in the regional data. Figure 4-9 shows how the relative importance o f individual/family oriented payments (salaries/compensation) and community focused payments (direct development investment and canon revenues) at the local level was predicted to develop during the mine life. The figure demonstrates the low contribution o f individual payments to local incomes compared with community payments for scenario A (17%> to 83%). The figure also shows the increasing contribution o f individual payments in scenario B . For scenario C , the figure again illustrates the increasing contribution o f salaries and individual payments to local revenues and the low, though consistent community contributions. Figure 4-9 also provides insight into the problems that may occur at mine closure under any o f the scenarios due to the relatively high incomes accruing to the local community which cease at mine closure. Boom and bust cycles have been criticized for their negative impacts on local communities (International Institute for Environment and Development, 2002). The likelihood for each scenario to create these cycles along with other social impacts w i l l be discussed in section 4.2 concerning the social impacts o f closure. Strengths and Weaknesses in the Socio-Economic Data The main weakness o f the data is again the failure to account for uncertainty. Thus, at this point the model represents a simplified deterministic model. The key uncertainties in the model are copper price, geological uncertainties and social and institutional risks. A cursory sensitivity analysis revealed that revenues to all stakeholders were sensitive to copper price and to institutional risk. The impact on institutional risk is described in more detail in section 4.3 concerning institutional impacts. Other key assumptions underpinning the model, which could be further investigated, are: • levels o f employment for different groups o f stakeholders • spending patterns o f these stakeholders • that canon revenues are spent in the local and regional areas 88 3.0E+07 2.5E+07 2.0E+07 C O 1.5E+07 1.0E+07 5.0E+06 O.OE+00 » • * * • • • • • • • • » .. ™ ™-_— ; ••• ....... t ,.'**"'*** j~— ' ~ —" ~ / / / /* / / • / / / __ , . — /—— T" ' -— ' ~" / / *f m\ J« • • •i » • • » • • /* / — -~ m „ „ > ^ ^ 1 » - ~ — — 1 ; ; • "*~" "* / •** Salaries A1 Development A1 Salaries B1 Development B1 - Salaries C1 Development C1 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 Year of operations Figure 4-9. Relative importance of salary and development income sources to the local community over time. • the impacts of training programs on local/regional workforce numbers • the values of subsistence agriculture • income multipliers were omitted from the model (except in separate regional assessment). Summary of Socio-Economic Results A summary of the socio-economic modeling results is shown in Table 4-3. The table shows that for scenarios where local training opportunities are not prioritized, shorter mine life scenarios perform better for all stakeholder groups. However, for scenarios where training is prioritized, medium mine life scenarios perform better for local stakeholder groups, while shorter mine scenarios perform better for regional, national and international stakeholders (using social discount rates). The improved performance of the longer mine scenarios at the local level is derived from increased revenues to the local community from salaries paid to local people. 89 The relative contributions o f revenues from salaries and from canon payments and direct development spending may be important considerations to local and regional stakeholders due to concerns over the effectiveness o f redistribution o f canon payments and conflict arising from the uneven distribution o f benefits and impacts of development projects. Recipient municipalities frequently complain that they do not receive all or part of canon revenues from higher levels o f government and thus communities believe that salaried employment at a mine is the only way to guarantee economic stability in relation to the project. The complaints and frustration are also based on perception and any viable solution w i l l also involve transparency by all stakeholders. Salaried employment can only be given to a portion o f community residents and the high salaries paid by mining companies to individuals can create conflict. A detailed presentation of the social impacts of the mine is contained in section 4.2 below. Summary of Socio-Economic Distribution Data Scenario Stakeholders A B C D No-Training Social Discount Rate Local $215M $148M $100M $4M Regional $534M $391M $357M National $2,657M $1,880M $1,376M Foreign $3,400M $2,163M $1,450M Training Social Discount Rate Local $217M $239M $212M $4M Regional $524M $431M $331M National $2,659M $1,762M $1,225M Foreign $3,412M $2,110M $1,366M Distribution of Revenues Training) Community Payments Local $200M $129M $90M $4M Individual Payments $17M $110M $122M Community/Individual Ratio 90:10 54:46 38:62 Table 4-3. Summary of Socio-Economic Distribution Data. 4.2 Social Modeling: Results of Community Wellbeing Model This section describes the results o f modeling the social impacts and benefits o f the different mine scenarios. It is divided into three sections relating to the three separate social impact assessment points identified in chapter 3: construction, the transition to operational stability and closure. For each o f the junctures social impact variables are discussed in the context o f key stakeholder groups. Impacts on vulnerable populations are identified in specific variables which consider equity. 90 4.2.1 Construction Impacts Summary for Construction Towns The results o f SIA variable modeling for the construction impacts for construction towns are shown in Table 4-4 below. SCENARIO A S/G* e S/G* C S/G* Variable Variable # and name 40 Influx & Out flux of workers 2. Influx and Out flux of temporary workers, (as % of town population) 82.5% negative 37.50% vv negative 22.50 % negative 5. Dissimilarity in age, gender, racial or ethnic composition. 0% female negative 0% female negative 0% fern negative Attitudes & responses 6. Formation of attitudes toward the project. Need specific local data to draw conclusions 24. Perceptions of public health and safety. Need specific local data to draw conclusions Community Cultural impacts 11. Living/Family wage (% of jobs providing living wage) 100% (750) positive 100% (750) vv positive 100% (750) positive 12. Enhanced economic inequities (decrease in local unemployment from project) -50% positive -50% positive -50% V positive 13. Changes in employment equity. Positive for youth and indigenous men, negative for women and elderly 14. Changes in occupational opportunities. (Increase in mining related job categories) 124% 24% 12% V 18.Change in commercial / industrial focus of the area. Reversible and not considered significant 22. Changed family structure. All projects show large increase in single men & decrease in % married. 26. Change in community infrastructure (Extra professionals required) 23 12.5 8.5 X SIG* The significance column is coded: VVV - extremely significant (>3 x threshold), - highly significant (<3 x threshold), V - significant (exceeds threshold) and X not significant. Where impacts are clearly positive or negative they are coded positive or negative. Table 4-4 Results of SIA for construction impacts at construction towns. A comparison between the scenarios reveals that the benefits to the construction community from living wage jobs is extremely significant and is constant for scenarios A - C . A l l o f these scenarios are expected to provide 750 l iv ing wage (labourer) jobs to local men, which represents the limit Details of the calculation of SIA variables are contained in Appendix VII. 91 of available labour. Secondary economic benefits wi l l be higher for scenario A and then for scenario B over scenario C , however, the social impacts are also expected to be higher for the shorter mine scenario. The threshold levels indicated that a town o f 7,000 people may be able to absorb the negative impacts from temporary employment for scenario B , while at least 13,000 people would be required for scenario A . Scenarios A - C have consistent positive employment equity implications for young men and for indigenous men, however they also have negative equity implications for both women and the elderly. There are significant temporary increases in infrastructure need during the construction phase for scenarios A and B , while existing infrastructure should be able to absorb the impacts for scenario C . There is also significant change in commercial industrial focus for the area for scenarios A - C . Scenario D provides no jobs but poses no impacts. 4.2.2 Social Impacts for the Transition to Operations The majority o f significant impacts for the transition to operational stability, were located in the 'Campesino community' , the local town and the regional city, which became the foci o f the analysis. The results are shown in Table 4-5. The different impact profiles for the scenarios related largely to the size of the initial mine workforce and the ability of local and regional residents to fill positions. These factors in turn determined the size o f the immigrant population on the regional city and the impacts that immigrants would have on infrastructure and on class structure. Attitudes to mining projects and interest group activity have been significant influences on the permitting o f mines in Peru and thus this aspect is considered to be important, although the size of the mining operation is believed to have little relation to the impact of this variable. For the transition to operations the most significant impacts were: • The relocation o f community members from the mine site, along with the associated disruption in social networks and the cultural changes in local communities. Relocation changes are constant for all mining scenarios and are considered to be extremely significant, as the threshold level for significance is surpassed 10 fold by the relocation required at the case study. Disruptions in social networks are also probable near the port facility and along new transportation corridors. Social network disruption has been documented to impact the elderly, the poor, indigenous communities and long term residents most severely (Burdge, 2004) and these groups represent significant proportions of those who w i l l be moved and those who w i l l remain and whose networks w i l l be impacted. 92 Scenario A Significant* 6 S/g* c S/g* Population Impacts 1. Population Change - Regional City +9% V +4.8% X +3.5% X 1. Population Change (Campesino Community) WV Population loss is identical for all scenarios and significant at -55% 5. Dissimilarity in age, gender, racial or ethnic composition. % Significant for gender, age and indigenous people for all scenarios at startup. #s small. 4. Relocation of Individuals and Families WV Same for all: Significant for # of families, % elders, poor people and indiqenous people. Community/Institutional Arranqements J — — 6. Formation of attitudes towards the project Attitudes and interest group activity probably significant challenges for project. r * .- • — 7. Interest group activity WV significant for all scenarios 10. Industrial diversification - Local Impact builds with time. Immediate V 6 year V 8 years V 10. Industrial Diversification - Region Significant change only if development is part of cluster. 11. Living/Family wage - Local - (with traininq) V Significant stable employment after 8-9 years only with training. 11. Living Family Wage - Regional City WV Significant for all scenarios. J 12. Enhanced economic inequities. W Mismatch between j o b requirements and local skills base. Most serious in A where training limited. 13. Changes in employment equity. VV There is a mismatch between job requirements and minority skills. 14. Changes in occupational opportunities. V There will be significant changes in employment profiles for all scenarios. LX — Communities in Transition 17. Introduction of new social classes Local-time to threshold. 3 years 5 years 7 years 17. Introduction of new social classes - Reqional City Significant for all scenarios for mine operators, technical & managerial personnel. 18. Change in commercial/industrial focus of the area. Significant at local level. Regional significance if a cluster develops. Individual & Family Level Impacts 20. Disruption in daily living. Significant for local community, port community & ? transportation corridor. 21. Assessing dissimilarity in religious and cultural practices Significant for local and regional populations. 22. Alteration in Family Structure -City Significant differences between workers and traditional residents. Most serious in A then B then C. Population small wrt regional city 23. Disruption in social networks WV Same for all scenarios. Disruption is very significant with 215 households impacted (threshold =15). Community Infrastructure 26. Change in community infrastructure (additional professionals required) 651.18 WV 199.1 VV 153.21 VV SIG* The significance column is coded: WV - extremely significant (>3 x threshold), VV - highly significant (<3 x threshold), V - significant (exceeds threshold) and X not significant. Where impacts are clearly positive or negative they are coded positive or negative. Table 4-5 Summary of Social Impacts in the Transition to Operational Stability 93 • Employment opportunities are considered to be positive in that all jobs pay living wages, however, the mismatch between regional and local skills and the skills required for mine employment has potential to cause conflict, if local and regional residents are not trained to fill these jobs. This impact is particularly significant for scenario A, where there is little time to train local people and large numbers of positions must be filled. For scenarios B ands C , it is predicted that, after training, younger men and indigenous men in the local community may find employment, however, women and elders will probably not benefit from these opportunities. If this impact is not mitigated then the mine could increase the marginalization of these groups in the community and town nearest the mine site. Because many mine jobs do not require physical strength women and able bodied elderly could fill some positions. Vulnerability and livelihoods analyses could provide insight into mitigation of this issue. • At the community level, the numbers of immigrants are unlikely to be significant either at the local city or at the local town except for scenario A, because outsiders are expected to reside in the city. More important changes are expected to occur in the change in occupational opportunities and the introduction of new social classes, as mine jobs become important in the local town and community. One difference between the scenarios is that a significant change in the local class structure caused by mine employment is likely to take longer under scenario C (7 years), while it will take only 3 years for scenario A 4 1 . In this case a longer time period is considered better as it permits a more gradual adjustment. • In terms of infrastructure, the requirements of scenario A are clearly the highest. For all scenarios, it will be important that local authorities reach agreement with the company regarding construction and operation costs for infrastructure. 4.2.3 Mine Closure Impacts The predicted impacts of closure are summarized in Table 4-6. Because of the length of time that will elapse before mine closure, prediction of closure impacts was considerably more difficult to assess than the impacts for construction and for the transition to operational stability. 4 1 Details of the assumptions behind these calculations are included in Appendix VII. 94 Scenario A Significance B Significance C Significance Population Impacts 1. Population Change - Local -15.6% A/ negative -42.7% A/ A/ A/ negative -39.7% A/ A/ A/ negative 1. Population Change - Regional -4.40% X -1.1% X -0.9% X Community/Institutional Arrangements 10. Industrial Diversification - Local neqative v v v neqative A/A/A/ neqative 11. Living Family Wage - Loss - Local 809 A/ neqative 2311 V A N negative 2234 A/A/A/ negative 11. Living Family Wage - Loss - Region 4989 A/ negative 1284 X 1082 X 13. Changes in employment equity. Indigenous men lose jobs, women and elders may gain comparatively 14. Changes in occupational opportunities. Loss of mining jobs significant at both local and regional level. Worse for local B & C. Communities in Transition 17. Loss of social classes - Local A/ neqative A/AN neqative A/ A/ A/ neqative 18. Change in commercial/industrial focus of the area. Local A/ negative Local A/A/A/ negative Local A/ A/ A/ negative Individual and Family Level Impacts 22. Alteration in Family Structure X Lose young A/ A/ A/ neqative Lose Young A/A/V neqative 23. Disruption in social networks X A/AN negative A/ A/ A/ negative Community Infrastructure 26. Change in community infrastructure.Local 38 AW negative 241 negative 286 V A/A/ negative SIG* The significance column is coded: A/A/A/ - extremely significant (>3 x threshold significant (exceeds threshold) and X not significant. Where impacts are clearly pc negative. ), AW - highly significant (<3 x threshold), A/ -sitive or negative they are coded positive or Table 4-6. Summary of Social Impacts of Closure. There is a significant difference between the closure impacts predicted for scenario A, and those predicted for scenarios B and C. Scenario A produces impacts in terms of job loss in the local community and region and changes in economic focus of the communities which are slightly over the threshold level of significance identified by Burdge (2004). However, scenarios B and C produce extremely significant changes in communities close to the mine site for closure. These changes result from the implementation of training programs encouraging the participation of local people in the skilled workforce for the mine. At closure, the lack of transferability of mining skills to other areas of work and the lack of diversification in the economy of Andean towns is predicted to result in an exodus of skilled people in search of employment opportunities. Social networks will be disrupted as extended families are split up and the family structure of the 95 community w i l l change, as younger families leave in search o f employment. This in turn w i l l lead to very significant decreases in service need in the community as wel l as potential problems in community leadership. STA identifies these impacts with a view to planning for avoidance and mitigation. 4.2.4 Summary of Community Wellbeing Modeling The social footprints o f the four scenarios display significant differences both in the magnitude and in the temporal distribution of impacts and benefits. Scenario A demonstrates the most significant impacts during the construction phase and imparts benefits mainly to regional and national communities rather than to the local ones. Scenarios B and C show less severe impacts during construction and also impart more benefits in terms o f l iv ing wage jobs to the local communities during operations. A t closure, however, the impacts o f these two scenarios are very significant because o f the degree of dependency on the mine that is likely to develop in local communities. For al l three mining scenarios careful attention is needed to the impacts o f waged employment on women and older people. For the no-mine scenario no significant impacts are predicted but no benefits are received either. 4.2.5 Weaknesses in the SIA modeling For effective SIA modeling it is imperative that local people are involved in the modeling process. This is because perceptions and local values are critical to impact identification and mitigation. In addition, because the case study was a composite study location specific data were not available and data were extrapolated from national census data which were up to ten years out o f date. These deficits would be easily resolved in a real case study. In addition, the results did not include secondary jobs related to mine construction. These w i l l include opportunities for renting accommodation, for restaurant and bar operators, to local mechanics and to some extent for ambulatory salespeople and for general merchants. The SIA methodology in itself is rather generic and could be significantly improved by including more detailed analyses o f social capital, community assets and vulnerabilities among other approaches. Final ly , the approach could also integrate a health impact assessment approach in order to fully compensate for the range o f health impacts and benefits which typically occur mining development. 4.3 Institutional and Governance Modeling The aims of the modeling of institutional and governance issues for the case study was to assess the implications of this understanding on the social, environmental and economic impacts of the scenarios. This section outlines the findings o f institutional S IA variable modeling for the case 96 study. The results of applying quantitative leakage data to the socio-economic model are then presented. Finally, the findings o f the institutional modeling are assessed for each scenario. 4.3.1 Institutional Social Assessment Variables The institutional variable modeling attempted to produce quantitative data to describe the impacts o f institutional variables identified in the SIA literature. The key institutional differences between the scenarios related to • The relative sizes o f the mining project, which is expected to impact the size of revenues which must be managed and the character o f the mining company (involving international agencies, multinational company or national company). • The duration o f the mining project, which determines the number of political transitions during the project and the potential improvements or deterioration in governance which w i l l occur during mining. This latter variable w i l l also determine the changing international regulatory and investment climate. The institutional variables identified by Burdge (2004) and Freundenberg (1999) produced useful foci for qualitative reflection concerning the possible evolution o f the Peruvian political context and its potential impacts on the mining operation. There was, however, a lack o f baseline data for some o f the variables and little more than subjective comparison was possible. Variable # 15 Presence of an outside agency The short term presence o f the mining company in scenario A provides less incentive for the development o f mutually respectful relationships with the local community and it is therefore expected that scenario A has the highest potential for conflict between company and community aims. Variable # 16 Inter-organizational Cooperation This variable is expected to be significant for all scenarios due to the large numbers o f permits required by a mining operation in Peru and the poor record of cooperation between agencies. A l l mining scenarios are expected to perform similarly for this variable, although for scenarios A and B international agencies may also be involved, due to the size o f the operations, which have been demonstrated to result in better sustainability outcomes at Latin American mines (Castillo, 2001). 97 Variable #8 Alteration in Size and Structure of Government Changes are expected to be more significant for scenarios A and B at start up because of the larger scale o f construction and the larger workforces involved. Start up impacts are expected to occur in the regional city, local town and in the Campesino community. Scenarios B and C are expected to perform significantly worse at project closure due to the large expected population decreases, which w i l l be concentrated in the local town. Trust in Political and Social Institutions (Freundenberg, 1999) The low levels o f trust in political institutions, low levels of capacity and the slow rates of change documented in governance suggest that scenario A holds little possibility for improved performance during its short life time. Scenarios B and C could both benefit from improved governance and improved trust by local actors resulting from a decrease in corruption, which would be potentially possible over these longer time frames. 4.3.2 Institutional Impacts on Socio-Economic Projections 3.0E+08 2.5E+08 2.0E+08 ZD CD > or 1.5E+08 1.0E+08 5.0E+07 0.0E+00 H Salaries 0 Development (0% Leak) • Development (30% Leak) • Development (70% Leak) ® Total 0% • Total 30% • Total (70% loss) B Scenario Figure 4-10 Accounting for Corruption in Socio-Economic Revenues to Local Communities. 98 The results o f the incorporation o f leakage data into the socio-economic analysis are shown in Figure 4-10. This figure demonstrates that the revenues to local communities under the status quo conditions of corruption are more susceptible under scenario A , because of the dominance o f canon revenues under this scenario. Revenues are more robust under both scenarios B and C where greater proportions of revenue are derived from salaries, however the overall size of development revenues for scenario C is so small that there is potential for conflict between employed community members and those who are not employed and receive no benefits from the mine is high. The potential for improvement in the leakage/corruption situation of Peru for longer term projects further favours scenarios B and C . 4.3.3 Summary of Institutional Modeling. A summary o f the institutional modeling for the scenarios is contained in Table 4-7. Country instability is a possible failure mode for a mining project and thus all projects in Peru at present must contend with the poor stability rating o f the country. The high foreign debt rating o f the country is a contributor to this instability and mining has the potential to ensure revenues to the government which service this debt, while surplus funds remain that have the potential to be used to improve public services. Improved public services have the potential to improve conditions for Peruvians l iving in poverty and thus in the medium term to improve internal stability. Higher taxation revenues from scenario A are beneficial because o f their potential impact on the external debt. Central government taxation revenues for scenarios A - C are $228 mil l ion per year for 13 years, $110 mi l l ion per year for 33 years or $68 mil l ion per year for 53 years, representing N P V s of $1 bil l ion $0.5 bi l l ion and $0.3 bi l l ion using discount rates o f 10% 4 2 . Scenario B performs the best in terms o f revenues to the local community when social discount rates are applied. The data concerning leakage from and effectiveness o f government spending, however, indicate that under the status quo, the effectiveness of government spending is low. In addition, corruption data indicate that canon revenues are unlikely to reach their intended target beneficiaries 30-70% of the time. Under this level o f corruption the revenues for local communities are more robust for the longer duration projects. However, the dependence o f local communities on salary revenues suggests that scenario B may be preferable. In addition the slow rate of improvement for governance reported by researchers (Kaufmann et al., 2001) favours longer projects. 4 2 The 10% figure recognizes that mining companies use an unusually high discount rate and that 10% is a more realistic rate for the Peruvian government. 99 Institutional Modeling Results Scenario Indicator A B C Government Revenues (NPV-10%) $1 billion $0.5 billion $0.3 billion Socio-Economic Local Revenues projection (30% leakaqe) $70 million $95 million $150 million ° ' Initial impact on local governance WV negative V negative V negative Closure Impact on Local Governance X WV negative WV negative Potential Promotion of Improved Governance Poor Good Good Level of Inter-organizational Cooperation International International National The significance column is coded: WV - extremely significant (>3 x threshold), VV - highly significant (>3 x threshold), V- significant (exceeds threshold) and X not significant. Where impacts are clearly positive or negative they are coded positive or negative. Table 4-7. Summary of Institutional Modeling. Corruption and poor governance are costly for a mining company and should be factored in as extra costs for the company. It is suggested that the willingness to pay figure for companies operating in Peru of 7.5% (Rancanatini, 2003) is a suitable figure to include as a measure of the cost of corruption on doing business in Peru. 4.4 Environmental Modeling The final section of the modeling of the case study mine scenarios focused on environmental performance. Whereas the social impact assessment approach chosen for assessing the social impacts of the proposed project took a comprehensive overview approach, the environmental modeling focused on two specific issues: greenhouse gas production and waste production. The results of these two models are presented in this section as sections 4.4.1 and 4.4.2. In addition the environmental section contains a third section 4.4.3 which considers environmental criteria which were not modeled in detail but which are believed to demonstrate significance to the design of the case study mine. These additional criteria were identified in the secondary scoping exercise described in Chapter 2 and are discussed briefly in overview. 4.4.1 Greenhouse Gas Production The greenhouse gas production model results for the entire mine life are summarized in Figure 4-11. The emissions were lowest for scenario A, and increased for scenarios B and C. Overall the tonnage of copper in concentrate produced by the case study mine is 7.3 million tonnes of copper. The emission level per tonne of copper produced for the case study therefore ranges between 0.74 and 0.93 tonnes of carbon dioxide per tonne of copper produced. These data correlate well with the Chilean data from Maldonado (1998) who calculated emissions from Chilean copper mining 100 to be in the range o f 0.9 tonnes o f carbon dioxide per tonne of copper produced. However, they include a high contribution to mining emissions resulting from transportation (haulage) of waste and ore and relatively little for ore processing. In part, the low emissions from processing can be explained by the clean energy available in Peru. In addition, the energy requirement for the mine was calculated from a standard rule of thumb relating to tonnage mined. When this rule of thumb was compared with the probable energy requirements from a brief assessment o f bond work index and energy requirements for grinding it appeared that the energy requirements for grinding had been underestimated in the model. The case study mine processes an average ore grade o f 0.6% copper. Case study emissions are between 21:1 and 16:1 for haulage to electricity while the Chilean emissions are in the ratio of approximately 3:1 (Maldonado, 1998). 6000000.00 5000000.00 TD CD ,f, 4000000.00 CD "O X o Q tz o - O I CO O 3000000.00 2000000.00 1000000.00 0.00 1 ru i • .r-n 1 1 nd~~1 H1 • B Scenario • Construction Transportation • Concrete • Vegetation Emissions D Operations Transportation S Trucks and Shovels M Electricity Figure 4-11 Greenhouse Gas modeling for mining scenarios (Mine Life) Using the Canadian government estimate that a permit for emitting one tonne o f carbon dioxide should not exceed $15 Canadian dollars 4 3 , the relative costs for carbon dioxide emission for the A figure of 1.3 Canadian dollars to the US dollar was used for conversion. 101 three mines are: $62 m i l l i o n 4 4 for scenario A , $75 mil l ion for scenario B and $78 mil l ion for scenario C . It may be important to consider that many analysts have predicted per tonne prices for carbon dioxide emissions significantly higher than the $15 Canadian figure and that the Canadian government's guarantee is valid only for a limited time period. 350000 300000 250000 i d o Q c= o - O 1— ro O 200000 ° 150000 I 100000 50000 • Transportation S3 Trucks and Shovels • Electricity Emmissions A B C Scenario Figure 4-12 Annual Greenhouse Gas Emissions for the Scenarios. Figure 4-12 shows the greenhouse gas emissions per year for each of the three mining scenarios. In this figure it is evident that, although scenario A produces the least greenhouse gas emissions overall, on an annual basis it produces higher emissions. Because o f this, using the Canadian $15 per tonne o f carbon dioxide emissions equivalent and discounting payment at a rate o f 15% the net present value o f carbon emissions permits is highest for scenario A , which has an N P V of $1.9 mil l ion for emissions permits compared with $1.1 mil l ion for scenario B and $0.8 mil l ion for scenario C . These figures have an insignificant impact on mine economic performance. However, decreasing carbon dioxide emissions is likely to be accompanied by a decrease in fuel Figures are calculated in $ U S . 102 consumption, which, due to the fact that fuel consumption can account for up to 15% of mining costs may produce significant savings. 4.4.2 Waste Production Impacts Tailings Table 4-8 demonstrates that the additional quantity of waste produced as tailings under the worst case scenario for the 15 and 35 year mines due to the larger block size required is 3% of the overall tailings volume. The break even price required to cover this cost (85.4c/lb) is considerably lower than the long run average copper price ($1.10) and thus it is envisaged that the extra waste produced should not have a significant impact on the mine economics 4 5 . A s the tailings w i l l be sub-aqueous and there are no concerns regarding acid rock drainage potential there should be no significant additional costs associated with these additional tailings. Scenario A Scenario B Scenario C Scenario D Shovel Capacity 68 CuYds 47 CuYds 38 CuYds N/A Block Size 20m 15m 15m N/A Extra Cu Produced 0 28,000 T 28,000 T N/A Extra Waste Processed 0 5.3MT 5.3 MT N/A Break Even Cu Price N/A 85.4c/lb 85.4c/lb N/A Extra waste to tailings 0 5.3MT 5.3MT N/A % Extra Tailings Volume 0 3%46 3% N/A Table 4-8. Tailings Production Impacts of the Scenarios. Waste Rock Additional waste rock volumes occasioned by the increase in ramp width to accommodate bigger trucks are shown in Table 4-9 below. A s the bottom rows o f the table show, the maximum extra tonnage o f rock to be moved for scenario B with a three lane haul road is 8.5 mil l ion tonnes and for scenario A the maximum additional tonnage is 15 mil l ion tonnes. These figures compare with an overall tonnage o f waste to be moved o f 3 bil l ion tonnes. The additional tonnage o f waste to be moved therefore represents 0.28-0.5%) o f the overall tonnage o f waste rock and approximately 4 5 From discussions with mining industry professionals, long run prices used in design may actually be around this 85 cent price. 4 6 No adjustment was made to the reserve size due to the probable higher cost of operating smaller equipment. 103 Scenario A B C Truck Capacity 380 320 260 Truck Width (m) 10 9.15 8 Road Width (2 Lane) 40 36.6 32 Road Width (3 Lane) 50 45.75 40 Extra Width (2 lane) 8 4.6 0 Extra Width (3 lane) 10 5.75 0 Add Waste m3 (2 lane) 7.2E+06 4.1E+06 0 Add Waste m3 (3 lane) 9.0E+06 5.2E+06 0 Add Waste Tonnes (2 lane) 1.2E+07 6.8E+06 0 Add Waste Tonnes (3 lane) 1.5E+07 8.5E+06 0 Table 4-9. Extra Waste Production Results - Waste Rock. $5 to $8 mil l ion dollars in costs using standard formulae for costs o f dril l ing, blasting, loading and haulage. 4.4.3 Additional Environmental Criteria Table 4-10 shows the additional criteria identified through the use of sustainability frameworks in order to ensure that no aspects o f the 'Case study' mine that were important to sustainability had been omitted from the analysis. The table lists the criteria under five separate themes, assesses whether there are likely to be differences between the three mining scenarios for each criteria (Trend) and then assigns a value o f low, medium or high to the probable significance o f the differences. Theme Criteria Trend Significance Geochemical Considerations Mobilization of toxics Yes Moderate Ecosystem Resilience Ecosystem resilience preserved Yes Low Fauna, prevent mobility of herds Yes Low Access to Ecosystem No Low Water Management Considerations Lead to permanent drop in groundwater Yes Moderate Substantially change local water courses Yes Moderate Enable locals to manage watershed Yes Moderate increase risk of waterborne disease Yes Low Intergenerational Equity Increase consumption of scarce resources Yes High Reversibility Yes High Prevent future access to resource Yes High Lifecycle Impacts Impacts of Transportation Yes High Impacts of Use Yes Moderate Enhance recyclability Yes High Table 4-10. Additional Criteria Identified through Sustainability Frameworks. 104 Geochemical Considerations For geochemical concerns the key difference between the scenarios was the number o f years during which waste rock, piles o f marginal ore and rock faces were expected to be exposed to water and air before they were protected from the elements through reclamation activities. A s the Andean climate consists of wet and dry seasons metals which build up in waste rock piles at the beginning o f the dry season are flushed out of the waste piles and into local ground and surface waters with the onset o f precipitation at the beginning o f the wet season. Under scenario A , it is expected that reclamation w i l l proceed more rapidly than under the longer mine life scenarios. This may have a significant impact on the short term mobility o f metals in local watersheds. It may also be significant in the medium and long term depending on the kinetics of neutral leaching o f molybdenum, as wel l as the effectiveness o f reclamation efforts at limiting entry o f water into waste dumps and tailings and hence reducing molybdenum mobility. In their favour the longer mines have a greater possibility that innovative and effective strategies for mitigating neutral leaching and metal mobilization w i l l be developed during the mine life. Ecosystem Resilience The longer mine lives o f scenarios B and C imply a longer period of disturbance o f the natural vegetation in the mine area. They also provide a longer period o f time during which the mine area is controlled by the mining company for regeneration on revegetated areas. Because a key limitation to the recovery o f vegetation is grazing pressure on the land (Riehm & Manticorena, 2002), re-vegetation and control o f the land during the mine life are expected to produce superior results for longer mine lives. However, it is probable that a sizable fraction o f re-vegetation and reclamation w i l l have to wait until the end of the mine life and thus scenario A is seen as probably preferable in terms o f ecosystem resilience. There wi l l also be small variations in the mine footprint related to the extra volumes o f waste produced by scenarios A and B . These differences are not expected to be significant in terms o f potential ecosystem recovery. If greater quantities o f metal leaching occur with the longer life mines then this factor may cause significant damage to ecosystems downstream of the mine site. There is also wetland on the mining concession which w i l l be destroyed by the pit. While Peru does not require that wetlands are replaced, responsible ecosystem management requires this replacement. Such wetlands may serve a secondary objective as a passive water treatment system protecting streams lower in the watershed from contamination. 105 Water Management Considerations. Mining wi l l necessitate dewatering o f the aquifers in the glacial overburden in order to facilitate a dry pit. After the mine closes dewatering w i l l cease and it is expected that the mined out pit wi l l flood. Because the mine is at high elevation and close to the top o f the watershed there is little surface area upstream of the mine to provide large volumes o f water to fill the pit rapidly. The water balance for the mine site is, however, expected to be positive due to the high rainfalls received. The environmental impact assessment o f the Antamina mine, which is in an analogous topographic setting states that the pit, is expected to take up to 75 years to flood following operations (Antamina Min ing Company, 1998). Throughout the mine life and flooding o f the pit, water resources w i l l need to be managed carefully to ensure that stream flows are adequate for downstream aquatic life and water users. Water shortages during this period may impact crop yields for downstream irrigated crops. Water management may therefore be required for up to 90 years for scenario A , 110 years for scenario B and 135 years for scenario C . Under these circumstances a shorter mine life is predicted to create a lower risk for water management problems in the long term. After flooding, the reservoir should become an asset to the local people in controlling downstream water flows during the dry season. Inter generational Equity Min ing is a non-renewable use of a natural resource. Mineral deposits represent concentrations of chemical elements and compounds rare in the earth's crust, which have been sufficiently concentrated to allow their extraction and use. In this sense, mining consumes scarce resources, which w i l l therefore not be available for future generations. For society, metals have the potential to be reused and recycled in a very efficient manner. However, current societal practice includes recycling efficiencies of only 20% (Ayres et al . , 2001), considerable wastage and design for cradle to grave use rather than for cradle to cradle reuse (Lovins et al., 2001). Current known copper reserves w i l l last for approximately 50 years at current levels o f use (Ayres et al . , 2001). Because demand for copper is predicted to increase, the current reserves are predicted to have a shorter life expectancy and copper is expected to demonstrate a shortfall. A s copper prices increase, it is l ikely that substitutions w i l l be found either for copper itself or for the need to transmit electricity over relatively large distances 4 7, which is the main use o f copper for which potential substitution is not currently available. Copper is thus expected to maintain a relatively For example, more local generation would require less copper in distribution infrastructure. 106 stable price and level o f demand at twice current levels of consumption until supply runs out, although there is potential for the 'bottom to fall out of the market" due to innovation in electrical design and technologies. For the communities, which live in the area of a mineral deposit the mineral resource can provide benefit only once. In addition, despite the potential for mineral wealth to be converted into human, social and built capital in the vicinity o f the mine site and for conservation of the natural capital of the environment, the norm is for mineral wealth to bypass local communities and to benefit national and foreign elites (Ross, 2001). Under these conditions a community would be wise to assess the credentials o f any mining proponent carefully and to ascertain the level o f participation in decision-making that w i l l be afforded them and the revenues and other benefits which are likely to accrue to them. The community would also be wise to ensure that project economics allow for adequate returns on investment for the local community. Only where it is clear that long term benefits w i l l remain in the community is it possible to argue that mining can be sustainable in terms o f intergenerational equity, because the mining wealth w i l l be passed along to future generations in the form of improved social and human capitals, solid environmental management practices, no significant environmental legacies and increased community wellbeing. Where commitments and positive track records are not in place, the principle o f intergenerational equity may suggest either not mining the deposit or delaying mining until better conditions exist. In addition the predicted increase in demand for the medium term may favour communities who wait before exploiting their copper resources, however there is some risk in adopting such a strategy, as mining is expected to peak sometime in the 21 s t century with earliest peaks predicted in 2050-2060 and latest peaks in 2080 (Ayres et al., 2001). Life Cycle Considerations A t present copper produced in Peru is generally shipped as concentrate to North America, Europe or Japan and Korea for smelting ( B H P , 2003). Once it has been smelted the majority is converted to copper wire and used in electrical applications (Ayres et al. , 2001). More minor uses include copper piping and telecommunications, although copper use in the latter is being substituted by optical fibre and wireless. Dissipative uses include brake linings in automotive vehicles (Ayres et al. 2001), corrosion from copper facings on buildings, use in copper chemicals as biocides and in ecosystem uses. Copper also accumulates temporarily in long lived uses (industrial equipment, electric utilities, building facings etc) and in short term uses (domestic appliances, automotive uses). Copper recycling is estimated to be in the range of 12 to 42% in the United States, with the percentage of refined copper from scrap in the range of 11-33%. M u c h scrap metal is exported 107 from the United States to China for recycling. Recycl ing efficiency is estimated at 30% for the U S while it is estimated at greater than 80% in Germany (Ayres et al., 2001). Whi le considerable effort has been focused by mining companies on optimizing individual elements o f the mine production system, integrative research into systems approaches is relatively new and is limited to elements that are wholly within the control o f the mining company. Examples o f this integrative approach include mine mi l l integration and the introduction o f solvent extraction-electrowining refining technology, which enables ore concentration and refining to occur at the same location. If this integrative approach is broadened to include transportation, manufacture, use, reuse, recycling and waste disposal elements o f the copper life cycle, then there may be potential for radical improvements in resource use efficiency. In order for the case study to profit from these improvements, the mining operation w i l l either have to incorporate flexibility into the design or else to delay the commencement o f production. 4.5 Performance Matrix and Decision-Analysis The results from the individual modeling exercises were combined into an integrated performance matrix for each of the 4 scenarios A - D for the case study mine. A n attempt was made to represent all significant information identified in the criteria modeling and the range and uncertainty measures in the matrix. The performance matrix is shown in Table 4-11. Two forms o f data processing were applied to the data. The first performed a quantitative analysis o f the data aimed at producing a mathematically optimum decision. This data processing approach was adopted despite the fact that the technical data modeling achieved up to this point were tentative and did not include a satisfactory inclusion of probability. The approach was included as an exploration o f the scaling methods and to gain Insight into the value and limitations o f the approach. Weights w A i to w D n were included in the analysis in recognition o f the fact that different weights would be assigned to each measure. The second data processing approach used the quantitative data produced and applied a qualitative analysis to the data. This second analysis used consequences tables to identify important tradeoffs between scenarios and used the insights gained from the criteria modeling to generate alternative scenarios for the case study. 108 Table 4-11. Multi-Criteria Performance matrix for Scenarios at 'The case study' mine. Scenario Criterion A B C D Project Economic Parameters Capital Expenditures $903 million $504 million $372 million 0 NPV ($1.1 OCu) $2,455 million $905 million $349 million $10 million % mined at payback ($1.10 Cu) 20% 11% 11% N/A Break Even Copper Price $0.52 $0.65 $0.79 Competitiveness (Production cost/lb) $0.35 $0.47 $0.55 Cost of Sustainability Measures (NPV -15%) $37 million $24 million $19 million Socio-Economic Distribution ($1.10 Cu - unless indicated) Total Revenue - Local - Training -15% discount $73 million $52 million $37 million $10 million Total Revenue - Regional - Training -15% discount $184 million $107 million $71 million 0 Total Revenue - National - Training -15% discount $951 million $502 million $323 million 0 Total Revenue - Foreign - Training -15% discount $1,167 million $563 million $332 million 0 Local Revenues 5% Discount $171 million $254 million $212 million $28 million Local Revenues 5% Discount - 30% corruption loss $126 million $216 million $188 million $28 million Local Revenues 5% Discount - 70% corruption loss $67 million $165 million $155 million $28 million Local Ratio Salaries: Development(5% Discount -30% corruption) 1/4.6 1/0.7 1/0.4 1/0.0 Local Ratio Salaries: Development(5% Discount-30% corruption) 72.6c Cu 1/2.4 1/0.7 1/0.09 1/0.0 Social Impacts Construction Families Relocated 100 100 100 0 Networks Disturbed 3.65:1 2.65:1 2.45:1 1:01 Jobs Provided to Local People 750 750 750 0 Negative Impacts to Local Women and Elderly (M/F ratio) Infrastructure Stress Yes 2.4 x threshold Yes - 1.2 x threshold No No Operations Population Change - City 9% 4.40% 3.50% 0% Social Structure Change - City City Infrastructure requirements (# professionals) 651 199 153 0 City: Local Inequities Local all in unskilled Gradual transition to skilled jobs - training Stays same Local Communities -Disruption Significant disruptions for local communities, port and transport. Local Communities -Time to Significant Employment not achieved 8 years 9 years Closure Population Change - Local -15.60% -42.70% -39.70% 0 % 109 Population Change - City -4.40% -1.10% -0.90% 0 % Family Wage Job Loss - Local 809 2312 2235 0 Family Wage Job Loss - City 4990 1284 1083 0 Loss in Local Infrastructure - # professionals -38 -242 -237 0 Institutional Issues Central Government Revenues (NPV-15%) $659 million $318 million $187million 0 Initial impact on local governance Very significant Significant Significant None Closure Impact on Local Governance Not significant Very significant Very Significant None Potential Promotion of Improved Governance Poor Good Good None Interorganization Cooperation International International National None Environmental Issues Greenhouse Gas Production - Total 6.0 M tonnes 6.8 M tonnes 7.0 M tonnes Greenhouse Gas Emissions per year 0.37 M tonnes 0.18 M tonnes 0.12 M tonnes Potential Cost Emissions Permits (NPV-15%) $2.1 million $1.2 million $0.8 million Additional Tailings Production 0 5.3MT 5.3MT N/A % Increase 0 3% 3% N/A Additional Waste Mined (Tonnes) 15 million 8.5 million 0 0 % Increase 0.50% 0.28% 0.00% 0 Cost Increase $8 million $5 million 0 0 Geochemical Considerations May Mobilize less May mobilize more May mobilize more Least mobility Ecosystem Resilience 15 year disturbance 35 year disturbance 55 year disturbance Least Disturbance Water Management <90 year disturbance <110 year disturbance < 130 year disturbance No Disturbance Intergenerational Equity Most risk of no 2nd generation benefits Medium risk of no 2nd generation benefits Most risk of no 2nd generation benefits best situation for future generations Life Cycle Considerations Worst Medium Better ?Best Table 4-11. Multi-Criteria Performance matrix for Scenarios at 'The case study' mine. 4.5.1 Quantitative Data Processing of the Multi-Criteria Performance Matrix. Several of the measures included in the table represented different expressions for the same data. For example, extra tonnage of waste mined was included as well as an estimated cost for mining the extra waste. Other measures included in the table aimed to present an expression of the risk involved in a scenario. For example, including the break even copper price for scenarios provided a measure of the sensitivity of project economics to copper price. In order to facilitate comparison between the scenarios it was decided to simplify the matrix as much as possible. Simplification involved: 110 Overall Scenario Multi-Criteria Performance Matrix Scenario Project Economic Parameters A B C D weight NPV ($1.10 Cu) 10 4 1 0 WA1 Break Even Copper Price 7 4 0 0 WA2 Competitiveness (Production cost/lb) 10 7 6 0 WA3 Average Revenue distribution factor 10 7.1 5.1 1.4 WA4 Average Revenue distribution factor 10 6.5 4.5 0.7 WA5 Technology Risk 0 10 10 0 WA6 Socio-Economic Distribution Average Revenue distribution factor 10 5.8 3.8 0.3 WA7 Local distribution 5.5 10.0 8.3 0.0 WA8 Local corruption risk 5 6 7 5 WA9 Social Impacts Construction Relocation 0 0 0 10 WB1 Living Wage Jobs Provided 10 10 10 0 WB2 Infrastructure stress 0 2 3 4 WB3 Operations City Population Change 0 1 1 2 WB4 Local Community disruptions 0 0 0 10 WB5 Local Community significant jobs 1 10 9 0 WB6 Closure Local Population 6 0 1 10 WB7 Regional Population 1 1 1 2 WB8 Local Infrastructure Loss 2 0 0 10 WB9 Institutional Issues Initial Institution Stress 0 5 5 10 WC1 Closure Institution Stress 8 0 0 10 WC2 Potential Improvement 3 10 10 0 WC3 Interorganizational Cooperation 10 10 5 0 WC4 Environmental Issues Greenhouse Gases 1 0 0 9 WD1 Waste Increases 0.15 0.0011 0.0025 10 WD2 Geochemical Mobility 3 1 1 10 WD3 Water Management 3 1 0 10 WD4 Intergenerational 0 5 0 10 WD5 Life Cycle 0 3 6 0 WD6 Table 4-12 Summary of Multi-Criteria performance matrix. 1. Eradicating measures which were dependent or demonstrated considerable overlap. 2. Identifying measures which might indicate failure modes for the project under a specific scenario 111 3. Scaling measures where possible on a 10 point scale. This was achieved by identifying the measure which represented the best outcome for the project and setting its value at 10. Other values were calculated by scaling based on the best outcome value. (Table 4-32). 4. Summing the scaled values for each scenario assuming that the sum of weights for economic, social, institutional and environmental variables were evenly weighted (ie. Z w A l -A n = I w B l - B n = Z w C l - C n = ZwDl-Dn=0.25) (Table 4-13). Whi le these assumptions are not realistic from a practical perspective. The exercise provides a theoretical "sustainability perspective" and provided some tentative insights. Summary of Mult -Criteria Analysis Table Scenario Criteria A B C D Economic 7.8 6.4 4.4 2.0 Social 3.4 3.8 3.7 4.4 Institutional 5.3 6.3 5.0 5.0 Environmental 1.2 1.7 1.2 8.2 Average 4.4 4.5 3.6 4.9 Table 4-13 Summary of MCA performance matrix using "balanced sustainability weighting" • Relative Performance on Environmental Indicators The widest range in performance between scenarios is shown by environmental measures. While it may be logical to consider the difference between the deposition of 4.5 bi l l ion tonnes of waste rock and no tonnes o f waste rock to be significantly different, the overall measure suggests that the mine w i l l produce significant long term contamination problems. This assumption may be valid at many mine sites, however, the specific geochemistry o f the case study deposit does not seem to justify such an extreme variation in scores. This finding suggests that differential weightings should be applied to the individual environmental measures to provide a meaningful summation o f environmental performance. • Relative Performance on Economic Indicators The next largest difference in performance is shown by economic indicators and this seems to be a valid difference. Again the weighting between the different economic measures could be improved upon as intuitively it appears that the aggregate measures may be under-representing the differences in economic performance. • Relative Performance on Social and Institutional Indicators 112 Institutional and social performance show little difference between the scenarios in the model. In these measures there seems to be some balance between negative impacts and positive benefits for the scenarios. It may be very important to disaggregate these impacts and benefits because they may accrue to different groups with some groups carrying most of the impacts and others carrying most o f the benefits. Such a situation is likely to result in conflict. • Lack of Validity of Average Data Given the weighting problems and the problem of hidden tradeoffs identified above the data were found to have little validity for comparing the scenarios at this juncture. 4.5.2 Qualitative Analysis of the Multi-Criteria Performance Matrix Tables 4-15 to 4-19 show the qualitative assessment o f training and employment opportunities resulting from scenarios A and B . These scenarios showed overall similar performance in the quantitative analysis but had very different social and economic profiles. The qualitative approach explicitly recognized the trade-offs involved between the two scenarios and aims to formulate improvements to the scenarios taking into consideration both costs and benefits. Training/Employment Benefits Scenario A Scenario B Better Option Increased Local Employment Low (unskilled & temporary) High (45% local employ ment with skilled jobs). Revenues to the Local Economy High ($171 million) High ($254 million) Vulnerability of Revenues to Corruption High Moderate Mine profits High $2,455 million Moderate $905 million Costs Cost of Training Programs Low $5 million Moderate ($36 million) Loss of Local People at Mine Closure Low High New Costs Table 4-15. Trade - offs concerning employment in the mining operation. The methodology was based on Gregory (2000) and the best option represented one possible improved scenario. Ideally the trade-offs analysis is a participatory process, however in the research initial insights were identified by the researcher based on the Peruvian Stakeholder interests identified in Appendix I V . Table 4-15 shows the blank trade-off matrix, which encourages the bartering of costs and benefits to find an improved solution. Table 4-16 shows one possible solution to the barter. The improved solution proposed involves delaying mine operations and implementing training programs focusing on both mining skills and 113 other transferable skills in the interim. The proposed solution also contemplates alternative financial management mechanisms for canon revenues to ensure revenues over a longer period for scenario A. The costs of this approach are a delay in production and associated delay in receiving revenues by stakeholders. This is counterbalanced by enabling the mine to operate a more profitable faster mine life and hence providing increased revenues for all. Training/Employment Benefits Scenario A Scenario B Better Option Increased Local Employment Low (unskilled & temporary) High (45% local employ-ment with skilled jobs). Delay mine - Train locals for shorter mine. Revenues to the Local Economy High ($171 million) High ($254 million) Improved local benefits with local hire. Vulnerability of Revenues to Corruption High Moderate Institute independent foundation for portion of canon revenues. Mine profits High $2,455 million Moderate $905 million High $2,000 million - (10 yr delay -5 yr construction) Costs Cost of Training Programs Low $5 million Moderate ($36 million) Moderate $36 million Loss of Local People at Mine Closure Low High Flexible work schedules. Transferable skills. New Costs | Delay in receiving revenues. Table 4-16 Possible solution to trade-offs matrix for employment/training. A similar process is utilized in Table 4-17 in regard to construction impacts on women and families, which were identified in the SIA as a significant concern for all mining scenarios. Construction Impacts on Families and Women Benefits Scenario A Scenario B Better Option Employment High (750) locals) High (750) Build Slower. More Local. Employment. Costs Arrival of Outsiders High 82.5% Moderate 32% Build Slower. More Local. Employment. Loss of Local Workforce. High (750 locals) High (750 locals) Schedule building around local agricultural priorities. Impacts on Local Women High (100% jobs to men) High (100% jobs to men) Jobs to women, slower building, more local hiring, New Costs & Benefits Costs to company for slower process. Payback -can build faster more profitable mine. Table 4-17. Trade - offs concerning construction impacts on women and families. The potential solution in this case would complement the solution shown in Table 4-16. The solution involves a slower construction schedule which is compatible with local agricultural schedules in order to limit migration of employees and the impact on the community. In addition the solution proposed in Table 4-17 suggests focusing on employment opportunities for women. The costs of this approach are a delay in production and associated delay in receiving 114 revenues by stakeholders. This is counterbalanced by enabling the mine to operate a more profitable faster mine life and hence providing increased revenues for a l l . Relocation Benefits Scenario A Scenario B Better Option Revenues to International High Moderate Revenues to Nation High Moderate Revenues to Region High Moderate Revenues to Local High Moderate New Benefits Improves the benefits in profits of the mine. Costs Families Relocated High High Take the time to actually relocate families rather than pay them off. Disruption of Networks High High Pay attention to networks prior to and after relocation. New Costs Delay in mining impacts profits Table 4-18. Trade - offs concerning relocation. A third trade - off matrix is presented in Table 4-18 focusing on relocation impacts o f the mining operations for scenarios A and B . The potential solution mandates a true relocation process, rather than payment o f compensation for land. Finding appropriate land for relocation o f families is time consuming and therefore mining companies frequently favour payment of compensation, however, this strategy has resulted in significant impacts for people who are not used to operating in a cash economy but know how to survive from subsistence agriculture. Environmental Benefits Scenario B Scenario D Better Option Sustainable Agriculture Moderate Low Fund farm capacity. Costs GHG Emissions High Low More fuel efficient mine Waste Increases High None New waste handling Geochemical Mobility High Low Key Issue - No Data Water Management Impact High Low Key Issue - No Data Intergenerational Moderate Low Delay Production until benefits guaranteed. New Costs and Benefits Delay: costs Table 4-19. Trade-offs for environmental issues. A final trade - off matrix is presented in 4-19. In this case, mining scenario B is compared with the no-mine scenario D for environmental impacts. The key finding o f this exercise is that the 115 environmental modeling carried out in the research does not correspond to the main environmental concerns of the stakeholders and that further technical information is required in order to formulate improved alternatives. 4.5.4 Summary of the application of M C A to the case study. A structured approach to decision making has been supported based on empirical research which demonstrates human cognitive limitations when faced with complex decisions (Clement & Reilly, 2001). In fact people have been found to systematically employ cognitive short cuts limited and misleading insights about uncertainty and individual preferences to arrive at suboptimal decisions (Clemen & Reilly, 2001; Gregory, 2000). In the case study the decision context demonstrates considerable complexity as well as a wide range of interested stakeholder groups and would therefore seem ideal for the application of this methodology. The MCA structured in the research, along with the technical modeling undertaken provide some useful tentative insights, as well as practical data which can be used to refine later analysis and reorient the modeling towards more relevant criteria in some areas. A more detailed discussion of the insights gained from the results is presented in Chapter 5, which discusses research findings and concludes the thesis. 116 C H A P T E R 5. DISCUSSION A N D C O N C L U S I O N S 5.1 Summary of the Research This thesis has presented an exploration o f M C A and the A P E G B C guidelines as a vehicle for more holistic mine design. The research used a composite case study based in the Peruvian Andes to provide realistic social, institutional, environmental, geological and engineering data which set the context for a comparison study o f four different scenarios. Three o f the scenarios involved mining the ore body but differed in the mine life chosen, the mining equipment selected, the number o f employees hired and in the related infrastructure and institutional required. The fourth scenario was situated in the same geographical and temporal location as the mining scenarios but left the ore in the ground; it represented the no-mine option. Two different processes were used to select holistic assessment criteria for the multi-criteria analysis. A n initial set o f criteria was selected by a small interdisciplinary group using the A P E G B C Sustainability guidelines to structure selection. The criteria incorporated long and short term effects, direct and indirect impacts and cumulative effects. A second criteria selection process applied indicators from a framework designed to assess the sustainability implications of mining projects (7 Questions to Sustainability) and screening tools designed to identify the sustainability issues associated with different engineering project types (FIDIC screening tools: van de Putte, 2001). This second criteria selection step aimed to build a more comprehensive framework for assessment which did not omit any significant issues. Both sets o f criteria were used in the final M C A , although the initial criteria were modeled in more detail. Once the criteria for detailed modeling had been identified, suitable indicators and measures were selected based on available data, methodologies and modeling techniques. The approaches used for modeling were: • Economic Data: Economic Evaluation. • Socio-Economic Data: Socio-Economic Model ing . • Community Wellbeing: Scoping level S IA. • Institutions and Governance: SIA of institutional variables and socio-economic modeling of leakage. 117 • Greenhouse Gas Emissions: WRI/WBCSD greenhouse gas emissions calculation tools. • Extra Waste Production: Borzorgebrahimi (2004)'s methods. Once results had been produced from the criteria modeling, the results were simplified; overlapping categories were removed and scaling was applied. The numerical data produced in this way were combined in a linear additive MCDA process, which weighed social, economic, environmental and institutional considerations equally to provide insight into the process and its outcomes. The multi-criteria matrix was also subjected to a qualitative analysis process using consequence tables to identify trade-offs and improve on alternatives. This second exercise involved consideration of the interests of stakeholders, based on Peruvian research outlining the concerns and demands of different Peruvian mining stakeholders. Overall tentative insights were gained into the issues of weighting for the case study and some preliminary suggestions were derived for improving mining scenarios to produce a more sustainable mine design. The results demonstrated a number of challenges inherent in the modeling and multi-criteria analysis processes, which are the focus of the discussion portion of this chapter. Following the discussion, a conclusions section synthesizes the lessons learned from the discussion and evaluates the success of the research in providing a foundation for the development of more holistic mine design tools. The conclusions identify potential modifications to improve on the prototype mine design tools developed in the research. The chapter ends by identifying gaps in the literature and providing suggestions for future research. 5.2 Discussion The aim of the research was to explore a practical approach for integrating the principles of sustainability into mine design. Overall the focus of this discussion section is the lessons learned from the research. It is structured around 5 main themes: 1. Insights from building and profiling the case study, 2. Insights from applying the prototype mine design tools to the case study, 3. Insights into sustainable mining at the case study, 4. Insights on the effectiveness of prototype mine design package, 5. A discussion of potential future applications for the holistic mine design approach. 118 5.2.1 Insights from building and profiling the case study The use o f a composite case study created challenges for the research because much of the baseline data was necessarily rather generic. This was especially true for the environmental and social sections. For example, the labour model for the mine was extrapolated from similar mining operations rather than employing a project specific human resources model and population demographics, employment profiles and infrastructure requirements were based on averaged census data for Peruvian communities. It is believed that the fact that it was possible to create useful data despite these challenges is a positive sign. In a real project development situation, it should be possible to collect specific data, triangulate conceptual data with field observations and stakeholder participation and generate a significantly more reliable case study profile upon which to base predictions. The researcher's familiarity with the Andean area was found to be particularly useful in building the social and institutional portions o f the case study. The need for familiarity with an area is not limited to gaining an understanding o f the more subtle aspects o f culture. It includes data that may at first sight appear to be clear such as taxation structures. Loca l knowledge is critical in designing a mine both through stakeholder participation in the early phases o f project design and through hiring specialists who are familiar with a particular cultural setting to profile the mining project, the holistic mine design approach appears to increase the importance o f this aspect. The participatory approach could also be used to focus the baseline study on specific areas o f interest. Without this focus it is l ikely that resources w i l l be wasted building lengthy inventories which w i l l have no particular utility to impact prediction, project development or monitoring. 5.2.2 Insights from applying the prototype mine design tools to the case study Decision and criteria structuring. The main methodological influence for the study was the A P E G B C Sustainability Primer. This guided the selection o f the criteria for modeling and also contributed to the concentration on M C A for comparing the scenarios. The main reference used to structure the M C A was the Brit ish Department for Transport, Local Government and the Regions Multi Criteria Analysis Manual (Dodgson et al., 2001), which is the key reference suggested in the A P E G B C Sustainability Primer (Long & Failing, 2001b). The M C A methodology described in Dodgson provided guidelines for comparing quantitative and qualitative variables and enabled the incorporation o f criteria which could not be expressed in economic terms. The reference also provided detailed guidance for the quantitative manipulation of M C A matrices to compare alternatives. 119 Unfortunately, the reference failed to address the importance of specifying the decision context and structuring the decision analysis (Clemen & R e i l l y , 2001). In addition stakeholder elicitation in selecting appropriate criteria, explicit definition of criteria attributes and the use o f the values focus to improve upon alternatives rather than simply distinguish between alternatives were not addressed. These are all critical elements in an effective decision analysis process (Gregory, 2000; Hammond et al., 2001). In hindsight, the results o f the multi-criteria analysis and variable modeling could have been significantly more relevant to the decision context, i f greater attention had been given to structuring and decision objectives initially. Both A P E G B C and FIDIC guidelines begin with prescriptive lists of the types o f criteria to be investigated. These lists are intended to ensure that important elements of sustainability are not omitted from the analysis, which is a laudable aim. However, it may be more valuable to begin from stakeholder concerns and apply lists as a check for comprehensiveness after criteria have been selected. In addition, both A P E G B C and FIDIC guidelines were deficient in addressing social issues, political issues and higher level environmental concerns when compared with the 7 Questions framework. Similarly the 7 questions may give inadequate weighting to economic and environmental criteria, because it implicit ly gives equal weighting to each o f the 7 question areas considered. Insights from the criteria modeling. The modeling applied in the research included tools at various stages o f definition and development. They included established techniques defined in documented guidelines (economic evaluation), innovative approaches which have been designed with the rigour o f the international standards organization to provide for cross sector comparability (greenhouse gas modeling), approaches which are being standardized, but are still characterized by diversity in measures and sampling and frustrate extrapolation between sources (socio-economic and social impact assessments) and tools which are in the initial stages o f development (governance modeling). In addition one modeling approach designed by a research colleague was used (waste generation impacts: Bozorgebrahimi, 2003). Economic and environmental variables were significantly easier to model than the social variables. In part this can be explained by the relative newness o f the social modeling approaches and the lack o f standardization. However, it is probably a truism that social variables are intrinsically more difficult to model. When applied to the social context, quantifying outcomes lends both credence and magnitude estimates to what can otherwise be rather vague assertions. 120 However, "the things that have . . . . been counted are not necessarily the only things that count" (Freundenburg, 1999 p. 101) and there may be a tendency to overlook equally important uncounted issues because a subset o f issues have been measured. A n example o f this tendency shown in the research is the relative ease o f modeling the distribution o f tax revenues and wage benefits to a local community compared with modeling the numbers o f living wage jobs expected to accrue to local community members and the likely consequences (positive and negative) o f the use o f this income. The weight given to quantified social data mandates transparent communication o f the uncertainties incorporated into social impact data because otherwise rough estimates can easily be mistaken for precise details. Uncertainty modeling would be a logical refinement for the models presented in this thesis. In addition, it is important that indicators be used in the specific context and conceptual framework of their design, which w i l l frequently be negotiated between stakeholders. It may also be important to ensure that the analysts employed in the social, socio-economic and institutional assessment o f projects base their measurements upon a comprehensive application of social science principles and theory. In the same way that engineers base their designs on an inherent body o f theoretical knowledge, sociologists base their models on a different but equivalent conceptual framework. Thus just as an engineer knows where to locate design parameters, rules o f thumb and understands how data from equivalent and non-equivalent sources can be extrapolated based on an understanding o f the theoretical constructs o f anthropology, sociology, psychology and economics guide social assessors. This enables the determination o f which variables should be measured, the scope of the study, the uncertainty underpinning the analysis and appropriate formats for communicating the information to different audiences. The absence o f these conceptual underpinnings in determining how to extrapolate data, how to predict behaviours based on known patterns and concepts and whether important sociological constructs had been omitted was a significant weakness o f the research. Questions such as, " H o w could case study data from Northern Canada be applied to the Andean situation?" and "How applicable are North American impact assessment criteria?" were troubling as there were clear parallels and obvious differences between the two contexts. These weaknesses on the part o f the researcher, were mitigated through the use o f clearly structured guidelines for impact assessment. However, the researcher had a limited capacity to interrogate the tools. Despite a lack o f in depth sociological knowledge, the SIA framework, appeared overly generic, in its assumption that threshold levels o f impact would be the same across different community types. This aspect would be somewhat mitigated by participatory application, however, other methodologies may be required to supplement SIA in order to mitigate this deficiency. Approaches derived from 121 vulnerability theory, sustainable livelihoods and social capital analysis seem to be candidates for this. Finally, the health impacts and benefits o f mining operations were omitted from the analysis. This is an important omission and represents an opportunity to improve on the model presented. Health Impact Assessment would appear to be an obvious initial choice o f methodology for exploration. In terms o f the need to understand institutions and power structures, institutional analysis is a dynamic and developing field o f research. The majority of the data that exist concerning governance is qualitative and perceptions based. This data was insufficient to determine the impacts o f poor governance on project outcomes under different scenarios. Some recent research has focused on quantitative measures and characterizing governance regimes according to specific measurable criteria and modeling is expected to develop significantly improved predictive capacity in the near future. For the more developed assessment tools, standard guidelines do not include some issues which are important in terms o f sustainability. It is important to develop standard methodologies which account for the benefits and costs of sustainability measures. A tool which shows potential in this respect is the financial modeling o f environmental, social and political risk. This could be a powerful mechanism for driving the industry towards more sustainable practice because it would quantify risks in the economic language o f business. It should be noted though that quantifying subjective risks in financial terms presents challenges. Environmental modeling in the research focused on a subset of environmental variables, which were not the optimal variables for designing the case study mine. In fact, a useful lesson from the research was that environmental modeling at the project level should probably focus on variables which have a direct impact on local people's lifestyles and livelihoods and on government concerns. Waste production and greenhouse gas emissions are both important environmental concerns, however focusing on these two variables over a relatively narrow portion o f the copper lifecycle proved frustrating and unproductive. A much more useful exercise would be to consider the entire copper lifecycle and to focus on system optimization, although perhaps the focus for such modeling needs to be at the company level rather than the individual project level. The greenhouse gas modeling also suffered from an inadequate data base and modeling difficulties which require attention i f the predicted data is to be reliable. Overall the criteria modeling produced data o f mixed utility for the decision context and objectives. The individual tools also demonstrated varying levels o f reliability, standardization and comprehensiveness. However, as an initial exploratory package o f tools applied for a scoping 122 stage project, the results produced by the modeling are considered to be good enough to provide insights into the relative merits o f the mining scenarios. Insights from the multi-criteria matrix and data processing. Initial attempts to draw valid inferences from the M C A matrix were frustrating. There was no one scenario which clearly outperformed all of the others, so it was necessary to use more complex data processing techniques. There were insufficient data available to determine criteria threshold levels and weighting data were not available. The quantitative data processing, used estimates where necessary. The original matrix was designed to portray as much detail concerning each criteria as was possible in a concise format. Reducing the matrix to scaled numbers and then to a series o f four numbers and eventually to one index for each scenario produced what at first appeared to be meaningless numbers. In part this was due to the mathematical orientation of the reference materials used (Dodgson et al., 1997). Once more practical decision analysis literature were employed (Clemen & Rei l ly , 2001: Gregory, 2000; Hammond et al., 1999) the primary goal o f decision analysis tools, to improve thinking and to sharpen communication about critical concerns and tradeoffs, became evident and applicable. The results o f the quantitative and qualitative assessments produced, while far from rigorous, provided valuable insights and guidance for structuring better alternatives for sustainable development o f the mineral resource. Reliability of Results Uncertainties in the data arose from a number o f sources. Generally speaking, uncertainty can be categorized as scientific uncertainty or estimation uncertainty (WRI & W B C S D , 2001c). The former is the consequence o f imperfect knowledge about the performance o f a scenario and the latter includes uncertainties implicit in the modeling method or the parameters input into the model. In terms o f SIA modeling, Freundenberg (1999) cites studies showing that standardized methodologies for predicting population multipliers were found to have average errors of 50%. A l l social variables calculated using multiplier techniques would therefore be expected to have an uncertainty o f at least 50%>. Uncertainties may be even higher for variables which have been extrapolated from statistical databases where parameters have not been clearly determined and methodologies for data collection are unknown. Both o f these difficulties apply to the projected contract worker data and thus these estimates are considered very uncertain (>50%> uncertainty). Where social variables combined multiplier data and employment data (municipal service provision) uncertainties would be expected to be in excess o f 70%. 123 The greenhouse gas prediction methodology includes a separate module for calculating the uncertainty (World resources Institute and World Business Council for Sustainable Development, 2001c). The methodology involves calculating the uncertainty for each individual parameter measured and combining these uncertainties to calculate the overall uncertainty. Because the uncertainty at scoping level is considered to approximate 15-30%, this is the uncertainty assumed for the parameters. Modeling uncertainty is stated in the greenhouse gas literature to be up to 20% (WRI &WBCSD, 2001) and thus this figure was applied. For this reason the overall uncertainty is expected to be approximately 35% (assuming normal distribution48). The high level of uncertainty in the data points to the need for closer scrutiny of uncertainty in a more refined mine model. These uncertainties should be presented integrally with the quantitative data in order to present a clear picture of the level of understanding and/or the state of knowledge in a particular area. The data presented in the research should be considered as initial indications to be tested in later refinements, probably involving consultation with experts and stakeholders. Overall the trends in the data are seen as reliable based on the consistent use of modeling drawn from leading authorities in the various fields considered. Assumptions used have been clearly set out in the thesis. A combination of providing clear assumptions and adapting recognized methodologies throughout the thesis, provides a strong foundation for further extrapolation of the model. The results of the modeling, although incorporating considerable uncertainty, are rigorous enough to be considered of good quality for a scoping stage of analysis. Weighting also presented a significant uncertainty in the research. Without access to a group of stakeholders with whom to discuss the importance of different variables it was impossible to achieve an adequate weighting scheme for the research. The conceptual "balanced sustainability weighting" applied illuminated some of the difficulties in using the weighting methodology quantitatively even when this uncertainty is no longer present in the data. These difficulties included the fact that impacts and benefits may accrue to different sectors of a stakeholder group and linear additive weighting may therefore be problematic. 5.2.3 Insights into sustainable mining at the case study The assessment of the capacity of the approach to promote more sustainable mining of the case study deposit is preliminary. The research was more useful in suggesting issues which should be prioritized for further investigation in pre-feasibility and feasibility mine design processes. For 48 Calculated from uncertainty of a x uncertainty of b = V(a2+b2). 124 example, relocation must be carefully examined along with the impacts o f different training strategies on mine employment and on mine closure. The role o f institutions in determining the distribution of economic benefits among stakeholders appears to be very important as wel l as a consideration o f the distribution o f benefits among different demographic groups at the local level. The research also suggested impact mitigation approaches for some variables. For example, there may be some flexibility in selecting communities to house construction camps. The size of the community is suggested by Burdge (2004) to have important bearing on the impacts o f construction on a host community. The research suggested minimum town populations required to mitigate impacts. In addition specific attention to training local people and the particular types o f skills developed may be key to transferability o f skills at mine closure. Finally, by focusing on the gender impacts o f mine employment it may be possible to generate innovative mitigation strategies for the typically severe impacts o f mining projects on some sectors of local women. The qualitative analysis showed that many o f the trade-offs in mine development for the case study were between the expected revenues o f the mine and the duration o f development in the local area which was related to mine life. Scenario A performed exceptionally well for traditional economic variables. In addition, it produced significantly better revenues for all stakeholder groups, except at the local level when social discount rates were applied. Its poor local economic performance was related to low levels o f local employment. Most o f the negative impacts o f the scenario appeared to be related to the insufficient length of time at the beginning of the project for training local people, accomplishing an effective relocation strategy and managing construction without interfering with traditional activities or requiring large migrant workforce populations. Although most mining companies are concerned with completing mine construction as soon as possible in order to reduce interest payments on borrowed capital, for the case study, it may be advantageous to delay construction 4 9. This might enable the permitting o f a much more profitable mine, while avoiding the significant social risks which would be associated with a rushed relocation process, low levels of local hiring, potentially poorer environmental performance and large revenues accruing to local governments before capacity for responsible use can be developed. Clearly, stakeholder involvement in design could be useful in deliberating this type o f strategy or suggesting other more satisfactory approaches. The World Bank might be a potential funding source for "delayed mining" loans. 125 5.2.4 Assessment of the prototype holistic mine design package Multi-criteria analysis ( M C A ) offers some promise as a tool for structuring the complex decisions involved in holistic mine design and criteria modeling is easily adopted into an engineering framework, as it applies tools analogous to the engineering process. The approach taken to modeling risk in the research was to vary key variables within expected ranges to determine sensitivities. Copper price was the most common variable utilized and this analysis was largely restricted to the economic variables. Despite significant challenges, by following published guidelines carefully, it was possible to produce useful data and gain insight into potential impacts and benefits of different mine scenarios and to suggest improvements. The lessons learned from the current research project become more poignant in their ability to guide effective spending o f these resources. Constraints on the Process One concern with the approach selected was the intensive and time consuming nature o f the M C A process. In part the cumbersomeness o f the process was a result of the use o f a large number o f criteria. This might be mitigated by structuring the M C A around a series o f core objectives identified by stakeholders. In part, however, it was a result o f learning curve effects, which are likely to affect any technical professional who attempts to broaden their ski l l base to encompass sustainability concerns. In a commercial engineering practice, the large amount o f time required for capacity building in this respect may be a significant obstacle to wider implementation. Establishing interdisciplinary teams which work together on project design is one potential strategy to catalyze the learning process. In this way learning can be a process o f sharing data with qualified colleagues in complementary fields, who can provide guidance concerning theoretical concepts while learning about the more technical aspects o f mine design in exchange. Most mining and mining consulting companies do not yet have this in house social (and sometimes even environmental) capacity. Therefore a significant constraint to the integration of sustainability issues into mine design appears to be a reticence to invest in capacity building and in the hiring o f qualified individuals to perform social analysis. It is still the norm to find social impact assessment in the hands of technical professionals at many project sites (Taylor, 2004; Burdge & Opreysek, 1994), who must presumably struggle through the process in a similar fashion to this researcher. The "The Min ing Team" ( T M T ) approach currently being pioneered by researcher and practicing professionals in the mining industry, which is a grassroots capacity building initiative in tools and approaches for collaborative mine development also shows promise in overcoming this obstacle (Kent, et al., 2004). 126 Effectiveness in representing sustainability concerns In addition to providing useful information to decision-makers, from a sustainability perspective it is important that key elements o f sustainability have not been overlooked in the analysis. In this sense, the environmental variables, which were o f less use to the decision-making process, may have important implications for the mining company and also for the planet, even i f no stakeholders identify them as priority concerns. B y relying on issues that are prioritized by stakeholders, there is a risk o f ignoring more deeply entrenched issues that fall into paradigmatic blind spots. These issues may be more appropriately categorized as elements o f the strategic sustainability agenda for the mining industry and mining companies rather than part o f the tactical sustainability agenda represented in the mine design context. There is however an important relationship between strategic and tactical sustainability. The strategic sustainability vision must guide the tactical agenda, i f the company is not to risk the discovery that a series o f apparently logical tactical steps have led them in a direction which does not coincide with the long term strategy. The practical experience of the research suggests a tension between the pragmatic, values based, decision-analysis process suggested by Gregory (2000) and a purist and more technical principles driven sustainability agenda. Underlying this tension is a lack of consensus over what mining looks like in a sustainable society. A consensus vision would serve to resolve the higher level criteria debates. It would create internal alignment within the company and thus would permit the development o f a culture capable o f applying the systems engineering approach. It would also produce a structure and a culture for both prioritizing criteria to be modeled and assessing the relative merits o f different scenarios to be compared. The tension may therefore serve a useful purpose. It may be used to create a deliberated high level vision at either the company or the industry level. I f this vision is based on a rigorous sustainability framework and is applied proactively, it may serve to prevent a recurrence of new, unintended consequences in the future. There appear to be different roles for the pragmatic and rigorous sustainability approaches, with a values based approach (structured through careful technical data) showing promise to guide a project improvement / approvals agenda and a higher level application o f sustainability principles approach being better suited to company long term strategic planning. 127 5.2.5 Future Applications A possible future mine design process is shown in Figure 5-1. The diagram demonstrates a mine design process where social and environmental impacts are considered integrally for each iteration of the mine design (represented by the black arrows). The social and environmental models are fully integrated into the mine design process, and the process is informed by appropriate stakeholder input (diagonal arrow). A number o f innovative and developing tools are utilized in order to determine the impacts o f the mine on social, institutional, environmental and economic spheres (small boxes: T C A -Total Costs Accounting, M C A , R A - Risk Assessment, L C A Life Cycle Assessment are shown). A number o f strategies have promise for generating the immediate "next steps" towards integrating sustainability into a future mine design process, such as the one pictured above: 1. Participatory Involvement The research indicates the important role that participatory processes may play in holistic mine design, in the specific context of the decision studied, this is problematic because o f the difficulties o f involving external actors in confidential, conjectural and ongoing design processes, the time involved in achieving consensus decisions (Cormick et al. , 1996), the problems for communities in truly comprehending the impacts on their community of a future mining project (International Institute for Environment and Development, 2002), as well as concerns raised by observers as to whether this type o f process produces results that are any more sustainable (van der Putte, 2001). Critics suggest that sustainable outcomes can be derailed by self interested parties with a short term focus (Thomas, 1995). However, the most progressive mine design processes in Northern Canada and Australia show trends to more collaborative process and Geology Simula tion Engineering Cost Stakeholders Figure 5-1. Potential Future Mine Design 128 clearer delineation o f community specific impact and benefit criteria (International Institute for Environment and Development, 2002). The T M T initiative proposes collaborative management of mining projects by a representative stakeholder board (Kent et al . , 2004) and similar to collaborative management pilot programs which have been introduced in other natural resource industries (Gregory, personal communication). I f participatory mine design processes are to be advocated, it is essential that proponents consider very carefully how the input from participation w i l l be incorporated into the mine design and which decisions wi l l be included. 2. Interdisciplinary Teams Involving interdisciplinary teams in the design o f mining operations is another logical step forward. The cross disciplinary fertilization which could occur in such teams has the potential to catalyze the development o f social science, environmental and process capacity by mine engineers, as well as educating members o f other disciplines about the technical criteria and constraints employed in designing a mine. It is very likely that such exchanges would lead to innovations in mine design, as new insights are integrated. A more formal interdisciplinary brainstorming process could also be applied to the entire mine design facilitated by a sustainability specialist. This type o f process is increasingly common in consulting companies, especially those specializing in environmental issues (personal communication with consultants). In addition, Placer Dome reports the use o f such processes at the height o f the first wave o f their sustainability initiative (Ian Pond, personal communication). In situations where elicitation o f objectives directly from stakeholders is not possible, expert brainstorming sessions may be able to provide insight into the expected concerns o f stakeholders, especially where experienced social scientists and trained social impact assessors are involved. 3. Encourage and coordinate with Regional Development Plans Another approach is for companies to work with international and national agencies to encourage the development o f comprehensive regional development plans or collaborate with existing plans. The impact and benefit criteria for the mine can be specified and coordinated with the strengths and needs of the region to produce the most effective mine for the area. This could lead, for example to the selection o f more labor intensive mining techniques in areas where unemployment was a problem. Alternatively, mining revenues could be effectively diverted to fund sustainable employment initiatives derived within the regional development plans. A n example o f coordination with regional development plans is in the Kalimantan region of Indonesia where U N C T A D are involved in a comprehensive regional development plan with several mining 129 companies (Kalimantan Go ld Corporation, 2003). The importance o f integrating mine development into plans for community development has also been raised in the Canadian North, where it is believed that the mineral potential of the region must be developed in a coordinated fashion based on sustaining revenues to local communities over generations (National Round Table on the Environment and the Economy ( N R T E E ) , 2001). 4. Use Strategic Sustainability to Guide Process Finally, from the company perspective, some o f the most promising work integrating sustainability into project design comes from Non-Governmental Organizations, who are at the forefront o f sustainability consulting. These organizations include The Natural Step, SustainAbility and the Rocky Mountain Institute, as wel l as several University Business Schools who have developed a focus on sustainability such as the Schulich Business School in York Ontario. A l l o f these organizations apply a similar strategic approach to sustainability when working with corporations. • Bu i ld awareness and educate concerning the principles o f sustainability and the requisites for a sustainable society. Envision a long term sustainable vision for the business. • Make an inventory o f the impacts and benefits of current business practices. • Design a strategy for improving business practice towards the vision in incremental steps. • Monitor and evaluate progress. It is suggested that the adoption o f this type o f integrated sustainability strategy holds considerable promise for mining at the company level. It provides the necessary vision and tools to guide a rational and cost effective transition to sustainability. The pioneers o f this approach in mining are B H P - B i l l i t o n who are working with the Natural Step in Australia (Herbertson, personal communication). Their work includes backcasting to create a vision for the company, Li fe Cyc le Analysis ( L C A ) o f different ways o f providing energy and steel-making services, as well as ongoing research, benchmarking and evaluation (Norton et al., 2001). 130 5.5 Conclusions The research investigated the hypothesis that a holistic mine design process, which integrates social, environmental and institutional criteria on an equal footing with the geological, engineering and economic criteria, can improve the sustainability outcomes o f the mine. Specifically the research aimed to: 1. B u i l d a holistic baseline model o f a case study mineral deposit. 2. Create a prototype holistic design assessment process. 3. A p p l y the prototype process to the case study in a pilot test. 4. Reflect on the insights about the case study gained through the design process. 5. Assess the effectiveness o f the holistic assessment with respect to facilitating sustainable mining development. This conclusion provides a brief assessment o f the success o f the research in investigating the initial hypothesis and in achieving each specific aim, drawing on the insights considered in the discussion. The thesis ends with suggestions for research and concluding remarks. The research has demonstrated that including broader criteria in the assessment o f scenarios for mine design process, can provide a more complete picture of the impacts and benefits o f each alternative. In addition, the research demonstrates the strong potential for the multi-criteria analysis and decision analysis methodologies for structuring and analyzing information to enable the transparent deliberation o f the sustainability impacts o f different mine scenarios. In consideration of each specific research objective addressed above: 1. Build a holistic baseline model of a case study mineral deposit. The research demonstrates that a holistic analysis must begin from a holistic baseline. A t present it is not industry practice to construct a holistic profile o f the project until much o f the mine design process has been completed. The research demonstrated that tools are available to assist in profiling all relevant aspects o f a mining project. The research also suggested that familiarity with the project area and participatory processes would be helpful both in producing a meaningful baseline study and in providing foci for the producing baseline studies which would be relevant to later modeling. 2. Create a prototype design process of tools capable of assessing the sustainable development outcomes of different mine design scenarios for the case study. The prototype package created in the research applied the multi-criteria analysis to criteria selected principally by applying the 131 processes suggested in the APEGBC sustainability primer. A secondary criteria selection process employed more comprehensive sustainability indicators and screens to ensure completeness. The methodologies selected for criteria modeling were standard financial appraisal, socio-economic analysis, social impact assessment, social impact assessment of institutional variables and manipulation of the socio-economic analysis to reflect the institutional situation, greenhouse gas modeling and modeling of extra waste produced. The results of this modeling along with qualitative consideration of the secondary criteria were incorporated into a multi-criteria performance matrix, as suggested by the APEGBC methodology. The matrix was processed through a quantitative linear additive process and through a more subjective consequence tables approach. 3. Apply the prototype process to the case study in a pilot test. This objective was achieved. 4. Reflect on the insights about the case study gained through the design process. The research produced initial insights which were outlined in the discussion above and appear to be useful in: a. Prioritizing areas for investigation during pre-feasibility and feasibility studies. b. Suggesting potential impact mitigation strategies. c. Identifying trade-offs. It is expected that these initial insights into sustainability at the case study will be further developed through discussions, workshops and further refinements of the model. The use of consequence tables as a communication tool between stakeholders shows particular potential. 5. Assess the effectiveness of the process methods and tools with respect to facilitating sustainable mining development. The combination of tools used in the prototype holistic mine design package demonstrated promise as the basis for designing more sustainable mines. Specific modifications to the package of methodologies used in this thesis which might improve performance include: a. Use the approach in an interdisciplinary team. This would reduce the potential for errors caused by misunderstanding of concepts and assumptions across disciplines. b. Improve the assumptions and baseline data used in the model. c. Use a participatory approach (preferably with stakeholders and a real project) to select the initial design criteria, rather than a technocratic approach. It may be 132 helpful to use comprehensive indicator frameworks to ensure completeness o f the criteria selected. d. Structure each criterion carefully to determine specific measurable attributes that w i l l inform decision-making. e. Ensure that criteria-modeling is specific to the information needs of the project. f. Include a more detailed estimation o f the investments required for sustainable development o f the mine and the projected revenues and risk reductions from the investments. g. Ensure participation and the inclusion of local knowledge in the social impact assessment. Consider complementing SIA with livelihoods and vulnerability theory approaches and social capital analysis to ensure that important cultural elements are included in the analysis. h. Incorporate comprehensive health data into the approach. Health Impact Assessment appears to be a promising methodology for this work. i . Seek out leading edge institutional modeling and prediction strategies so that this important element can be incorporated into mitigation efforts. j . Mode l more directly applicable environmental criteria. Assess the greenhouse gas modeling and baseline data to identify the cause o f the inconsistency with measured data from other copper mines. k. Perform the quantitative and qualitative analyses o f the multi-criteria matrix in a participatory setting (preferably with stakeholders) in order to improve on project alternatives and address trade-off in a transparent fashion. 5.4 Future Directions for Research A s exploratory work, this thesis is situated at the forefront of what w i l l hopefully become a series of studies which develop the prototype model presented in this thesis into a fully functional integrated design process, which balances the range o f sustainability considerations to produce better mines. This section synthesizes the insights gained from the present research into suggestions for future research. The basic model used in this thesis would benefit from a thorough evaluation, preferably from a number o f different disciplinary perspectives to assess for weaknesses and strengths. The 133 financial model could be refined to account more realistically for infrastructure, and sustainability investments in environmental, institutional and social areas. More detailed modeling o f relevant environmental variables could be performed and refinements made to the socio-economic, institutional and social impact modeling. A change which would significantly enhance the reliability o f the results would be to provide a clear quantification of uncertainty in terms of the expected ranges of values for each criterion and their distribution. Many o f the calculations required for this have already been carried out or require minor modifications to the spreadsheets. Experimentation with participatory processes for identifying mine design criteria and for participatory analysis of trade-offs and improving on scenarios would be valuable. This would serve both as a vehicle for developing capacity in the industry, would produce insights concerning stakeholder preferences and might generate innovative collaborative solutions to challenges which result from the inequitable distribution o f impacts and benefits from mining operations. This is already underway in the work o f Kent et al. (2004). Once it has been refined, it would be interesting to apply the methodology at a pipeline mining project or at a mining operation which is planning for a critical change in its life cycle. Mine closure projects might provide particularly useful case studies, because existing data could be used to calibrate models. A n y practical application to a real situation, would have the advantage of providing an opportunity to evaluate the success of the model in predicting outcomes at a later date. The research also identified a number o f gaps in the literature concerning the impacts o f different mine variables on community impacts. A n absence o f recorded information concerning the relative performance of international and domestic mining companies, larger and shorter mining operations, or different mitigation strategies frustrates the potential to predict which strategies w i l l result in better or worse performance for the mine with any confidence. Systematic data demonstrating the social impacts o f mines over time are a particularly important gap in the literature. Gender analysis o f impacts, and research describing the role o f social capital, institutional variables and the health impacts of mining operations were particularly scarce and would aid in designing mines which maximized benefits in these areas. 5.5 Concluding Remarks The research in this thesis was based on the conviction that mines have the capacity to contribute positively to sustainable development and could play a particularly important role in catalyzing 134 regional and rural development in the developing world. The track record of mining companies in achieving this potential has been disappointing and one of the reasons identified for this situation was hypothesized as being the concentration on limited engineering and economic analyses during mine design. This technical approach has led to the treatment o f social and environmental concerns as inflexible impacts to be mitigated once a completed mine design has been created. The approach advocated in this thesis incorporated environmental, social and institutional criteria directly into the mine design process on an equal footing with the traditional technical design criteria. The criteria were used in a prototype holistic mine design process to assess the relative sustainability performance of four different mining scenarios for a case study. The results of the research show that multi-criteria analysis, a tool which enables the comparison o f criteria which cannot all be reduced to economic measures has considerable potential as the basis for a package of holistic mine design tools. For holistic mine design tools to achieve widespread use they w i l l need to be refined in order to produce consistent, credible results. They w i l l need to adequately characterize all relevant impacts and benefits of a mining scenario, provide guidance for impact mitigation and be applicable within reasonable time frames and budgets. It is believed that this thesis has provided a foundation for such mine design tools and provides important insights for their future development. 135 THIS PAGE LEFT INTENTIONALLY B L A N K . R E F E R E N C E S Agricola , G . (1556). De Re Metallica, H . C . Hoover & L . H . Hoover (translators), New York , N e w York : Dover. Alcazar, L . and Wachtenheim, E . (2001). A Public Expenditure Tracking Survey in central government transfers to municipalities in Peru: A Detailed Look at the Vaso de Leche Program, Washington DC: Inter-American Development Bank. Al lenby, B . (2001) Earth Systems Engineering Editorial in The Bridge V31 #1 p . l , Washington D C : National Academy o f Engineering. Al t ie r i . M . (2001). Indigenous knowledge re-valued in Andean Agriculture in Newsletter of the Centre for Research and Information on Low External Input and Sustainable Agriculture (ILEIA) 12(1) pp.7-11. Alvarez Campillay, V . (2001). Towards sustainable development indicators for the minerals industry. Santiago, Chile: Chilean Copper Commission (Cochilco) (In Spanish). Anderson, . M . and Woodrow, P, (1989). Rising from the Ashes, Development Strategies in Times of Disaster, London, U K : Intermediate Technology Publications. Antamina (1998). Environmental impact assessment, available at www.minem.gob.peru (In Spanish), accessed November 12 t h, 2002. Arabsheibani, G .R. (1990). Tourism Multiplier Effects on Peru, Brasi l ia D .F . , Brazi l : Catholic University o f Braz i l , available through h t t p : / / w w . u c b . b r / e c o n o m i a / r e v i s t a / T O U W S M % 2 0 M U L T I P L I E R % 2 0 E F F E C T S % 2 0 O N % 2 0 P E R U . P D F accessed November 10 t h, 2003. Aramayo, A . , (1998). Changes in Agriculture and influence on peasant strategies, (Cambios en la Agricultura e Influencia sobre las Estrategias Campesinas), in A . Zoomers (Ed.) Peasant strategies in the South East of Bolivia: Interventions and Rural Development in Chuquisaca and Potosi (Estrategias Campesinas en el Surandino de Bol iv ia . Intervenciones y Desarrollo Rural en el Norte de Chuquisaca y Potosi), (In Spanish), L a Paz, Bol iv ia : K I T - C E D L A -Centro de Information para el Desarrollo (CID). 137 A S O C A M (Sustainable Agriculture for Mountain Peasants) (2002). Lessons learned in Andean Forestry (Lecciones Aprendidas en Foresteria Andina), Quito, Ecuador: Sustainable Agriculture for Mountain Peasants ( A S O C A M ) (In Spanish). Auty, R. M . (1993). Sustaining development in mineral economies: The resource curse thesis, London, Routledge, Avery, D . (1974). Not on Queen Victoria's Birthday, London, U K : Coll ins. Ayres R., Ayres, L . and Rade, I (2002). The Life Cycle of Copper, its Co-Products and By-Product,. International Institute for Environment and Development: London, England. Beach, R., Brereton, D. , and Cliff , D . (2003). Workforce turnover in fly-in-fly-out mining operations in Australia: An exploratory study, Brisbane, Australia: Centre for Social Responsibility in Min ing . Berthelier, P., Desdoigts, A . , and Ould Aoudia J. (2004). 'Institutional Profiles': Presentation and Analysis of an Original Database of the Institutional Characteristics of Developing, in Transition and Developed Countries, Dijon, France: Laboratoire d'Economie et de Gestion ( L E G ) , Burgundy University available through http://ideas.repec.org/p/lat/legeco/2004-01.html accessed July 17th, 2004. B H P Bi l l i ton , (2003). Energy Smart Program Exceeds Target at EKATI diamond mine, available through http://www.bhpbilliton.com/hsecReport/2003/caseStudies/cs_environmentl8.html accessed March 15th 2004. Blaikie , P., Cannon, T., Davis, I., and Wisner, B . (1994). At Risk: Natural hazards, people's vulnerability, and disaster. London, U K : Routledge. Botts, S., (2003). The challenge o f sustainable Development in Peru, in Proceedings of the conference of the American society of mining engineers. June, 2003. Cincinnati, U S A . Bozorgebrahimi, E . (2003). The evaluation of equipment size effects on open pit mining. Unpublished Doctoral Dissertation. Vancouver, Canada: The Faculty o f Graduate Studies - Department O f M i n i n g Engineering - University O f British Columbia. Brown, L . R., Lenssen, N . , and Kane. H . (1995). Vital Signs: The Environmental Trends That Are Shaping Our Future. N e w York , N Y : W . W . Norton & C o . Bruntland, G . (ed.), (1987). Our common future: World Commission on Environment and Development, Oxford, Oxford University Press. 138 Buchanan, D . , Chalmers, N . , Hevhoe, M . , Jones, M . , Shaw, T. (2003). An Introduction to Modeling Project Finance an Edumine Course, Vancouver, Canada: Edumine. Bunn, D . (1984). Applied Decision Analysis. N e w York: M c G r a w - H i l l . Burdge, R.J. (2004). A community guide to social impact assessment, 3 r d edition. Middleton W I : Social Ecology Press. Burdge, R.J . (2004b). A n Ex-Post Facto Analysis of the Social and Economic Impacts o f Reservoir Development in R. Burdge and Colleagues (2004) The Concepts, Process and Methods of Social Impact Assessment. Middleton, WI: Social Ecology Press. Burdge, R J . and Colleagues, (2004). The Concepts, Process and Methods of Social Impact Assessment 3rd edition. Middleton Wisconsin, U S A : Social Ecology Press. Burdge, R., Fie ld , D . and Wells, S. (2004). Ut i l iz ing Social History to Identify Impacts o f Resource Development on Isolated Communities: The Case o f Skagway, Alaska in R. Burdge and Colleagues (2004) The Concepts, Process and Methods of Social Impact Assessment. Middleton, WI : Social Ecology Press. Burdge R.J . and Opreyszek, P. (1994). On M i x i n g Apples and Oranges: the Sociologist does Impact Assessment with Biologists and Economists" in Burdge R. J . (Ed) A Conceptual Approach to Social Impact Assessment. Middleton, Wisconsin: Social Ecology Press. Cantuarias, F . (2003). Multi-stakeholder spaces in Peru: The Case of Antamina (Espacios Multiactores en el Peru: El Caso Antamina. Presented at an international seminar on round table processes and development in mineral zones: L i m a . International Development Research Centre, Ottawa Canada Castillo, J . , Sanchez, J . , Kunze, V . and Araya, R . (2001). Chi le : Size does matter in G . M c M a h o n and F. Remy (Eds.) Large Mines and the Community. Ottawa, O N : International Centre for Research and Development. Caterpillar (2003) Caterpillar Performance handbook 33rd Edition, Peoria, 111: Caterpillar. Carton, W. and Dunlap, R. (1978). Environmental Sociology - a new paradigm. The American Sociologist, 12: 41-49. Cement Association o f Canada, (2003) About Cement and Concrete available through http://www.cement.ca/cement.nsf/internetyA17D88E3DF75A10C85256917006C6BF17opend ocument accessed A p r i l , 27 t h , 2003. 139 Chang Barker, K . (2003). Return on investment for international education. Burnaby, B C : British Columbia Centre for International Education. Chevron Phillips Chemicals (2001/2003). Introduction to Mineral Processing: Froth Flotation. The Woodlands: Texas: Chevron Phillips Specialty Chemicals. C11FAD (Cornell International Institute for Food Agriculture and Development) (1999) International Indigenous Agricultural Systems: Andean Region in CIIFAD 1998-1999 annual report pp. 65-70. Ithaca, N Y : Cornell International Institute for Food Agriculture and Development. Clayton, C , Pakalnis, R., and Meech, J . (2002) A knowledge based system for selecting a mining method available through http://www.mining.ubc.ca/ipmm/Papers/Clayton.pdf accessed March 16th, 2004 Tool also available through Edumine at http://www.edumine.com/xedumine/miningmethod.htm accessed March 16th 2004. Clement, R. and Rei l ly , T. (2000). Making hard decisions 2nd Edition. Pacific Grove, C A : Duxbury. Cobo, Martinez J. (1986/7). Study of the Problem of Discrimination against Indigenous Peoples, E/CN.4/Sub.2/1986/7, United Nations. Cochi lco (Chilean Copper Commission) (2002) Research on Mine Closure Policy. London, England: IIED available through http://www.iied.org/mmsd/mmsd_pdfs/044_cochilco.pdf acessed M a y 4th, 2004. CoDevelopment Canada, (2002). Walking Together: Guide to Facilitate a Transformative Process (Caminando Juntos: Guia para facilatar un proceso transformador) Vancouver, Canada: CoDevelopment Canada (In Spanish). Cohen Duncan, L . (1999). Roman Deep Vein Mining. University o f North Carolina web pages available through http://www.unc.edu/~duncan/personal/roman_mining/deep-vein_mining.htm accessed July 16th, 2004. Connor D . (undated). From Decide Announce Defend, to Profile Educate Participate. Victor ia , Canada: Connor Development Services Limited. Available at http://www.islandnet.com/~connor/dadtopep.html accessed March 13th, 2004. Cooney, J . (2001) NGOs - Your New Partner in Mining, Friend or Foe, presented at the 12 t h Symposium Minera l Economic Society o f the Canadian Institute o f Min ing , January, 2001. Cormick, G . , Dale, N . , Emond, P., Sigurdson, S., and Stuart, B . (1996). Building Consensus for a Sustainable Future: Putting Principles into Practice. Ottawa, O N : National Round Table on the Environment and the Economy. 140 Costa, S. and Scoble, M . (2003). A multidisciplinary approach to integrating sustainable development into mining engineering. Journal of Cleaner Production. Oxford: Elsevier, accepted October 2003. Coulas, M (2001). Copper in Natural Resource Canada Yearbook, Ottawa, Canada available through http://www.nrcan-rncan.gc.ca/mms/cmy/content/2001/25.pdf accessed February 21 st, 2004. Crawford, J. and Hustrulid, W . (1979). Open Pit Mine Planning and Design, N e w York , N e w Y o r k : American Institute o f M i n i n g , Metallurgical and Petroleum Engineers. Culverwell , M . , Lee, B . and K o z i e l l , I. (2003). Towards and Improved Governance Agenda for the Extractive Sector. Royal Institute for International Affairs: London. Daly, H . (1990). Towards Some Operational Principles o f Sustainable Development. Ecological Economics 2 pp. 1-6. Damonte, G . , Leon, C . & Davi la , B . (2002). Peruvian National Study in Community Development Activities of Mining and Natural Resource Companies in Latin America and the Caribbean. International Development Research Centre: Ottawa. O N . Canada. Danielson, L . , (2002). The role o f the minerals sector in the transition to sustainable development. In World summit on sustainable development: opinions. London, England: International Institute for Environment and Development. De Ferranti, D . , Perry, G . , Lederman, D . and Maloney, W . (2002) From Natural Resources to the Knowledge Economy: Trade and Job Quality, Washington D . C . : The Wor ld Bank. DeMarco, E . (2002) Canada ratifies Kyoto and takes steps to implement climate change plan in Climate Change and Sustainable Development Committee Newsletter 6(2), Chicago, Illinios, U S A : American Bar Association. D F I D (Department For International Development) (2000) Sustainable Livelihoods Guidance Sheets. London: D F I D . D F I D (Department For International Development) (2003) Peru: Voices of the Poor (Peru: VOces de los Pobres) London, England: D F I D available through http://www.fco.gov.uk/Files/kfile/Voices%20of%20the%20Poor.pdf accessed November 3 r d , 2003(In Spanish). Dies, R. (2002). Sustainability in Professional Engineering And Geoscience: A Primer- Part 3: Practice Specific Module - Min ing . Sustainability in Professional Engineering and Geoscience Sustainability Committee o f the Association for Professional Engineers and 141 Geoscientists of British Columbia. Burnaby: Association of Professional Engineers A n d Geoscientists of British Columbia. Dodgson, J . , Spackman, M . , Pearman, A . and Phillips, L . (2000) Department for Transport, Local Government and the Regions Multi Criteria Analysis: A Manual. London, England: Department for Transport, Loca l Government and the Regions Dohm, G . (1979) Circular Analysis - Open Pit Optimization in Crawford, J. and Hustrulid, W . (Eds) Open Pit Mine Planning and Design. Society o f M i n i n g Engineers of the American Institute o f Min ing , Metallurgy and Petroleum Engineers: N e w York , New York . Elert, F. (1999) Density o f Concrete in Physics Fact book available through http://hypertextbook.com/facts/1999/KatrinaJones.shtml accessed A p r i l , 17th 2004. El-Swaify S .A. and Yakowi tz D .S . (1998). Introduction to Mult iple Objective Decision Making For Land, Water A n d Environmental Management. In Mult iple Objective Decision Mak ing For Land, Water A n d Environmental Management, Proceedings Of The First International Conference On Multiple Objective Decision Support Systems (Modss) For Land, Water And Environmental Management: Concepts, Approaches, And Applications Chelsea, M I : Lewis Publishers pp. 3-8. Federal Reserve, (2004) Federal Reserve Statistical Release, N e w York , N e w York : Federal Reserve available through http://www.federalreserve.gov/releases/hl5/data.htm accessed July, 10th, 2004. Fitzgerald, G . , Taylor, N . and Mclintock, W . (2002) The impacts of resource sector restructuring on occupational and community identity paper presented at the Annual meeting o f the International Association for Impact Assessment (2002) Den Haag, Netherlands. Available through http://www.tba.co.nz/frst_projects/IAIA_2002_paper.pdf accessed March 17th, 2004. Francis, G . (1999). cited in Veiga, M . , Scoble, M . , and Macalistair M . L . (2001). M i n i n g With Communities. Natural Resources Forum 25, pp. 189-202. Freudenburg, W . R . (1999). Tools For Understanding The Socioeconomic A n d Poli t ical Settings For Environmental Decision M a k i n g In Dale, V . A n d English, M . (Eds.) Tools To Aid Environmental Decision Making. N e w York , N e w Y o r k : Springer Verlag. Fussier, C , Cramer, A . and van der Vegt, S. (2004) Raising the Bar: Creating Value with the UN Global Compact. Sheffield, England: Greenleaf publishing. 142 Garner, A . , Keoleian, G . A . , Industrial Ecology: An Introduction, A n n Arbor, Michigan: National Pollution Prevention Center for Higher Education, University o f Michigan.: Gibson, G . (2001). Building Partnerships: Key Elements of capacity building London, England: International Institute for Environment and Development. Gizikoff. , K . , Jones, C . and Bloodgood, M . (1997). Monitoring indicators of Sustainability at Highland Val ley Copper, Logan Lake, B . C . in Proceedings of the 21st Annual British Columbia Mine Reclamation Symposium in Cranbrook, B C , 1997. Glave, M . and Kuramoto, J. (2002). Peru in M i n i n g Minerals and Sustainable Development Latin America team. Ottawa, O N : Mining Minerals and Sustainable Development in South America. International Development Research Centre. (In Spanish). Gluschke, W . (2004) Peru sets a new royalty rate available at http//www.mineweb.net/sections/mineralfinance/327394.htm accessed July 12 t h , 2004. Georges, C . (2004): The 20 l h century glacier fluctuations in the Cordillera Blanca (Peru). Arctic, Antarctic, and Alpine Research in press available at http://geowww.uibk.ac.at/glacio/LITERATUR/georges03.pdf accessed June 14th 2003 accessed M a y 29 t h , 2004. Graedel, T. and Allenby, B . (1994) Industrial Ecology, Upper Saddle River, N e w Jersey: Prentice Ha l l . Graf Lambsdorff, J. (2002). 2002 Corruption Perceptions Index in Transparency International, Global Corruption Report. Washington D C : Transparency International. Gregory, R. (2000). Us ing Stakeholder Values to make Smarter Environmental Decisions in Environment 43(5) pp. 34-44. Golden, J., Levy, T., and Hauptman, A . (2001) Recent discoveries concerning Chalcolithic metallurgy at Shiqmim, Israel in Journal of Archaeological Science (2001) (28) pp. 951-963 Guerrero J.P and Hoffbauer, H . (2002). Budget transparency in Latin America, in Transparency International, Global Corruption Report Washington D C : . Transparency International. Hammond, J., Keeney, R. and Raiffa, H . (1999) Smart Choices. Cambridge, Mass: Harvard University Press. Handelsman, S.D. and Fahrni, K . C . Use of Computer at Granisle Copper, paper presented at Symposium on Decision-Making, Vancouver, British Columbia, 1971 143 Hawken, P., Lovins, A . , and Lovins L . H . (1999). Natural Capitalism: Creating the Next Industrial Revolution. Snowmass, Co: Rocky Mountain Institute Heijungs, R., Guinee, J. , Huppes, G . , Lankreijer, R. , Udo de Haes, H . , Wegener, D . , Sleeswijk, A . , Ansems, A . , Eggels, P., Duin, R., and van Goede, H . (1992) Environmental life cycle assessment of products. Guide and backgrounds. Leiden, The Netherlands: Leiden University Press. Hodges, A . (2003) APEG Sustainability Policy (Draft), Burnaby, B C , Canada: Association o f Professional Engineers and Geoscientists of British Columbia. Hoffman A . and Ventresca, M . (Eds.) (2002). Organizations, Policy and the Natural Environment: Institutional and Strategic Perspectives. Stanford, CA: Stanford University Press. Horswi l l , D. (2000). Sullivan Mine: Sustainable Development available through http://www.teckcominco.com/articles/operations/su-sustainable.htm accessed July, 16th, 2004. Hustrulid, W . and Kutcha M . (1995). Fundamentals of Open Pit Mine Planning and Design. Burlington, Vermont: Ashgate Publishers. INEI (Peruvian National Institute for Statistics) (1995). Women in Peru ( L a Mujer en el Peru), L i m a , Peru: Peruvian national Institute of Statistics available through http://www.inei.gob.pe/biblioineipub/bancopub/Est/Lib0039/nOO.htm accessed November 1st, 2003. INEI (Peruvian National Institute for Statistics) (2004). District Information Data Bank (Banco de Informacion Distrital), L ima , Peru: Peruvian National Institute for Statistics available through http://www.inei.gob.pe/ accessed June, 2003 (In Spanish). Institution o f C i v i l Engineers, (2003). Engineering Sustainability Journal. London, England: Institution of C i v i l Engineers. Intergovernmental Panel on Climate Change (2000). Land Use, Land-Use Change, and Forestry, Cambridge, U K : Cambridge University Press. Intergovernmental Panel on Climate Change (2001). Climate Change 2001: Synthesis Report Watson, R .T . and the Core Writ ing Team (Eds.). Geneva, Switzerland: Intergovernmental Panel on Climate Change. 144 International Cyanide Management Institute, (2002). International Cyanide Managem