TN295 No. 9070 2nd Set WmMm ■'■>; i -V ; ■ ■ ■■ ' ■" : ' ■"'■'- ■ !■■.•■'■.■■••:••■■•■• ■■ ■■>■••<: .- SiHH-'Vi ill 11111$ ■hJII L • 1* # % ' *° t ,. °* *"'• ^ ... v. c ^ C^ g a °* °/> a <^ • - , <* \^r\^ -oJfB- / \-Wl\j V^ - ^ \^^\/ v ''*' ^** -A'' ^/ •$&: V«* ••» \-/ -1M« %,** A- *%* ^^^ ^ ^Vk ^ ^*]Mk°* J^S^kSr A.^fe-% -^ o.,*' A A v c d>* ■f: ^v^ ,;#/ *♦ 'fflfs /\ ## ** v % vK* «/\ : -w- ; ^ v ■yX>w^^ Bureau of Mines Information Circular/1986 GOLD AVAILABILITY- WORLD A Minerals Availability Appraisal By Paul R. Thomas and Edward H. Boyle, Jr. UNITED STATES DEPARTMENT OF THE INTERIOR Information Circular 9070 GOLD AVAILABILITY- WORLD A Minerals Availability Appraisal By Paul R. Thomas and Edward H. Boyle, Jr. UNITED STATES DEPARTMENT OF THE INTERIOR Donald Paul Hodel, Secretary BUREAU OF MINES Robert C. Horton, Director T /V^5 As the Nation's principal conservation agency, the Department of the Interior has /\ O responsibility for most of our nationally owned public lands and natural resources. This includes fostering the wisest use of our land and water resources, protecting our fish I and wildlife, preserving the environment and cultural values of our national parks and historical places, and providing for the enjoyment of life through outdoor recreation. (J The Department assesses our energy and mineral resources and works to assure that their development is in the best interests of all our people. The Department also has a major responsibility for American Indian reservation communities and for people who live in island territories under U.S. administration. I Library of Congress Cataloging-in-Publication Data Thomas, Paul R. Gold availability— world. (Information circular) Bibliography: p. 85 Supt. of Docs, no.: I 28.27: 1. Gold. I. Boyle, Edward H. II. Title. III. Series: Information circular iLJnited States. Bureau of Mines); IN295AJ4 ETN420] 622 s [553.41] 85-600217 For sale by the Superintendent of Documents, U.S. Government Printing Office Washington, DC 20402 Ill PREFACE The Bureau of Mines is assessing the worldwide availability of selected minerals of economic significance, most of which are also critical minerals. The Bureau iden- tifies, collects, compiles, and evaluates information on producing, developing, and ex- plored deposits, and mineral processing plants worldwide. Objectives are to classify both domestic and foreign resources, to identify by cost evaluation those demonstrated resources that are reserves, and to prepare analyses of mineral availability. This report is one of a continuing series of reports that analyze the availability of minerals from domestic and foreign sources. Questions about the Minerals Availabil- ity Program should be addressed to Chief, Division of Minerals Availability, Bureau of Mines, 2401 E Street NW., Washington, DC 20241. CONTENTS Page Page Abstract 1 Introduction 2 Acknowledgments 2 Commodity overview 2 Historical perspective on production through 1983 6 Resource overview, 1984 7 Total primary gold availability and production cost evaluation 8 Comparative production costs in major producing countries 8 Potential annual production through 2000 .... 11 Republic of South Africa 12 Historical perspective 12 Resource overview, 1984 15 Total available gold from producing South African mines 17 Economic effect of byproduct production ... 18 Capital and operating costs 19 Effect of exchange rate variation upon industry profitability 20 Annual production potential 22 Ability of other market economy countries to compensate for the expected decline in South African production 24 Gold mining in South Africa 24 Underground mining 24 Surface mining 27 Gold milling in South Africa 27 Low-grade milling 29 Refining and transportation 30 United States 30 Historical perspective 30 Resource overview, 1984 33 Total gold availability and production cost evaluation: sixteen major producers 35 Comparative long-run total production costs 37 Annual production potential through 2000 . 38 Primary gold mining in the continental United States 40 Surface mining 40 Underground mining 41 Placer mining 42 Gold milling in the continental United States . 42 Heap leaching (solution mining) 42 Comparative economics of heap leaching and conventional milling 45 Vat leaching of gold ores 46 Extraction of gold with activated carbon 47 Carbonaceous gold ores 48 Refining and transporation 49 Canada 50 Historical perspective 50 Recent perspective: 1968-83 50 Production by deposit type 52 Lode-type deposits 52 Base metal deposits 53 Placer deposits 53 Resource overview, 1984 54 Production cost and availability: current producers 56 Comparative economics of underground production 56 Total potential gold availability 58 Annual gold production potential 58 Hemlo gold district 58 Background 58 Impact of the Hemlo district upon future annual Canadian gold output 60 Total annual production potential to 1990, all sources 60 Primary gold mining in Canada 62 Surface mining 62 Underground mining 63 Gold milling in Canada 63 Operations 63 Refining and transportation 64 Australia 65 Historical perspective 65 Composition of mine production of gold, 1970-83 65 Minor primary gold operations 67 Major byproduct gold operations 67 Resource overview, 1984 67 Gold availability and production cost evaluation: seven major primary producers . . 67 Major primary and byproduct gold deposits awaiting or under development 70 Mining methods and operating costs 71 Metallurgical methods and operating costs. ... 71 Refining and transportation 72 Brazil 72 Historical perspective 72 Sources of production and production costs .... 73 Possible new sources of production beyond 1984 74 Reserves and resources 75 U.S.S.R 76 Historical perspective, 1754-1983 76 Soviet gold production by deposit type 77 Placer production 77 Lode production 78 Byproduct gold production 78 Reserves and resources 78 Issues concerning Soviet gold production 79 Other important gold producing countries 80 The Philippines 80 Papua New Guinea 80 Chile 81 Zimbabwe 81 Colombia 82 Dominican Republic 83 Ghana 83 Peru 84 Mexico 84 Possible gold projects in countries not evaluated 84 References 85 Appendix.— Methodology of analysis 87 VI ILLUSTRATIONS Page 1. Index of exchange rates: United States, Canada, Australia, and South Africa 5 2. Evaluated gold producing countries and 1983 production levels 6 3. Percentage distribution by country of total contained and total recoverable gold in 111 selected properties 7 4. Potential gold available at break-even costs of production from 111 operations in market economy countries as of January 1984 9 5. Comparative long-term total production costs in selected countries 10 6. Potential annual gold production to the year 2000 from 111 primary operations 11 7. Potential annual gold production to the year 2025 from 111 primary operations 11 8. Goldfields of the Witwatersrand Basin, South Africa 12 9. Gold mining operations of the Orange Free State and Klerksdorp Goldfields 13 10. Gold mining operations of the Far West Rand, West Rand, Central Rand, East Rand, and Evander Goldfields 14 11. Gold ore and refined gold production in South Africa, 1970-83 14 12. Potential total South African gold available at break-even costs of production as of January 1984 17 13. Rate of increase in labor and equipment costs in rand and dollar terms, 1970-83 20 14. Total gold revenues in rands and dollars, 1970-83 21 15. Potential annual South African gold available at various break-even production cost levels, as of January 1984 22 16. U.S. gold production by ore source, 1905-83 31 17. U.S. gold production by ore source, 1968-83 32 18. Location of producing and prospective gold mining operations in the continental United States 33 19. Location of producing and prospective gold mining operations in California and Nevada 34 20. Areas of primary gold deposits in the continental United States 34 21. Potential total U.S. gold available at various production cost levels from 16 major producers as of January 1984 36 22. Potential annual U.S. gold production available at various break-even production cost levels from 16 major producers as of January 1984 38 23. Provincial distribution of Canadian gold production in selected years 51 24. Total mine production of gold in Canada, 1968-83 51 25. Mine production of gold in Canada by type of deposit for selected years 52 26. Location of producing and prospective primary gold mining operations, areas of gold deposits, and historical lode gold mining districts in Canada 55 27. Location of producing and prospective primary gold mining operations in Quebec, Ontario, and Manitoba . 55 28. Contribution of capital and operating cost to total production cost for 15 selected Canadian operations .... 59 29. Potential total Canadian primary gold available at various production cost levels from 15 selected operations as of January 1984 59 30. Potential annual Canadian primary gold available at various break-even production cost levels from 15 selected operations, 1984-2000 59 31. Potential annual primary gold production from three developing mines in the Hemlo District, Ontario, Canada, 1984-2010 60 32. Potential total Canadian primary gold available at various production cost levels from 18 selected operations as of January 1984 60 33. Contribution of three Hemlo District operations to potential annual Canadian primary gold production, 1984-2000 61 34. Australian gold production, 1970-83 66 35. Location of producing and prospective gold mining operations in Australia and areas of primary gold deposits 68 36. Producing and prospective gold mining operations in Western Australia 68 37. Selected major nonalluvial gold mining operations and areas of primary gold deposits in Brazil 73 Vll TABLES Page 1. Primary gold mining operations included in the availability study 3 2. Gold production, 1971, 1981, and 1983, for the 10 largest producing countries 4 3. Relationship between gold price, gold production, and exchange rate variations 5 4. Estimates of total world gold production, by time period 6 5. Estimated total production for five major historic gold producing countries 6 6. World gold production in 1983 6 7. Total demonstrated primary gold resources in 111 selected properties in market economy countries as of January 1984 7 8. Supporting data for figure 4: distribution of total recoverable gold by cost level and country 9 9. Comparative long-run total gold production costs in selected major producing countries 10 10. South African gold ore and gold production, 1885-1983 15 11. 1984 demonstrated gold resources in South Africa and comparison to past production 16 12. Distribution of demonstrated gold resources in South Africa, by reef or reef group and by gold mining field 17 13. Total gold potentially available at increasing cost or price levels from 38 underground operations in South Africa 18 14. Ten largest South African operations in terms of total recoverable gold 18 15. Ten largest South African operations in terms of average annual life-of-mine production 18 16. Ten lowest cost producers in terms of long-run total break-even production cost 18 17. Economic summary data for 38 underground gold mines in South Africa 19 18. Total South African gold potentially available at increasing cost or price levels 20 19. Estimates of potential annual South African production capabilities 23 20. Mine operating cost estimates for underground gold mines on the Witwatersrand, South Africa 23 21. Technical and operational data, within rankings as to mining cost levels, South African underground gold mines 24 22. Comparison of mining productivity changes, 1960 and 1981, based on actual results at a South African mine 26 23. South African milling methods and capacities 27 24. Summary of sand and slimes material available for reprocessing in South Africa 29 25. Mill characteristics of six low-grade surface waste reprocessing operations in South Africa 29 26. Distribution of 16 major primary gold producers in the continental United States, by State and type of mining 33 27. Gold resource data for selected primary gold mines and deposits in the continental United States as of January 1984 35 28. Supporting data for figure 21: total gold potentially available from 16 major producing operations in the continental United States 36 29. Long-run DCFROR for selected U.S. gold properties 36 30. Comparative summary results of 1984 long-run cost determination analyses for producing and nonproducing surface operations in the continental United States 37 31. Potential annual production estimates for 16 major primary gold producers in the United States 38 32. Operations considered possible or probable primary gold producers during 1984-90 39 33. Potential annual U.S. gold production circa 1990, by type of operation 39 34. Combined operational data for major gold producing surface mines in the continental United States 40 35. Comparative operational characteristics of continental U.S. surface mines, low and high mining cost levels 40 36. Classification of producing surface operations by level of mine operating cost and grade of demonstrated resource 41 37. Effect of lower mine operating costs on the ability to mine and process low-grade material at selected producing heap leach operations in the Western United States 41 38. Operational characteristics of producing gold mills in the continental United States, by type of milling method, circa 1983 42 39. Number of U.S. heap leach operations utilizing various mining and processing methods 43 40. Labor productivity and operational characteristics for low-cost and high-cost heap leaching operations in the continental United States 44 41. Comparative economics of producing heap leach and conventional milling operations in the continental United States 45 42. Comparative characteristics of major vat leaching mills in the continental United States, circa 1983 46 43. Variations of major circuit practices at vat leaching gold mills in the continental United States, early 1980's 44. Historical summary of the Canadian gold mining industry 50 45. Gold production from Canadian lode-type deposits, 1973 and 1981, by Province 52 46. Byproduct gold production from Canadian base metal operations, 1973 and 1981, by Province or Territory, number and type of operation 53 TABLES— Continued Vlll Page 47. Demonstrated resource data for selected major Canadian primary gold mines as of January 1984 54 48. Comparative summary results of 1984 long-run cost determination analyses for underground Canadian mines 57 49. Supporting data for figure 29: potentially available primary gold in 15 selected operations in Canada .... 57 50. Summary results of 1984 long-run cost determination analyses for three developing mines in the Hemlo District of Ontario, Canada 57 51. Supporting data for figure 32: potentially available primary gold in 18 major operations in Canada 57 52. Canadian operations considered possible or probable primary gold producers during 1984-90 61 53. Potential annual Canadian gold production by 1990, by type of operation 62 54. Mining types, operating costs, capacities, and productivity for major Canadian gold operations 62 55. Summary of milling methods in use at major primary gold milling operations in Canada, early 1980's .... 64 56. 1981 capacities of Canadian dore bullion, scrap and residue gold refineries 64 57. Historical summary of mine production of gold in Australia 1851-1983 65 58. Operational categorization of Australian mine production of gold in selected years 66 59. Aggregate gold resource data for selected Australian mines and deposits, 1984 67 60. Summary results of 1984 long-run cost determination analyses for major Australian primary gold mines producing in 1983 69 61. Australian deposits considered possible or probable gold producers during 1984-90 70 62. Estimated Brazilian gold production and world ranking, 1975-84 72 63. Expected 1984 Brazilian gold production, by operation and type of mining 74 64. Varying estimates of actual and potential Brazilian gold production in selected years 75 65. Gold production in Russia, 1754-1895, by geographic area 76 66. Varying estimates of gold production in the Soviet Union for selected years 77 67. Summary of 1970 operational data for Soviet gold operations from Dowie and Kaser's 1974 study 77 68. Placer grades at various evaluated operations in market economy countries 78 69. Important lode gold mining developments in the Soviet Union 78 70. Estimates of annual Soviet gold sales, 1966-83 79 71. Annual gold production in 10 important gold producing countries, 1975-83 81 72. Breakdown of 1976 and 1982 gold production in Chile, by category of mining operation 81 73. Estimated distribution of 1981 Colombian gold production, by Province and size of operations 82 74. Gold production in Ghana, 1964, 1980-81, and 1983, by major producing entities 83 75. Possible new gold mining developments of significance in countries not covered in the report 84 UNIT OF MEASURE ABBREVIATIONS USED IN THIS REPORT d/yr day per year mt metric ton 2 °F degree Fahrenheit mt/d metric ton per day ft foot mt/yr metric ton per year g gram pet percent g/mt gram per metric ton 1 ppm part per million ha hectare St short ton kg kilogram tr oz troy ounce 3 km kilometer tr oz/yr troy ounce per year m meter wt pet weight percent m 3 cubic meter yr year GOLD AVAILABILITY— WORLD A Minerals Availability Appraisal By Paul R. Thomas 1 and Edward H. Boyle, Jr. 2 ABSTRACT The Bureau of Mines evaluated the long-term cost and availability of primary gold production from 135 mines and deposits worldwide. Collectively, the evaluated coun- tries represent 93 pet of world gold production. Total recoverable gold available (as of January 1984) from a subset of 111 significant producing mines and developing deposits in 13 market economy countries (MEC's) is estimated at 810 million tr oz. The Republic of South Africa is estimated to account for 87 pet of total recoverable gold. The United States and Canada account for 4 and 4.5 pet of the total, respectively. Eighty-three per- cent of total recoverable gold is available at a constant 1984 break-even price of $400/tr oz, and 70 pet is available at $300/tr oz. South Africa accounts for 90 pet of the gold available at $400/tr oz or less. Major conclusions are that (1) South Africa should remain the largest world pro- ducer through the year 2000, (2) annual MEC output in 2000 should not be significantly different than current output given constant 1984 gold prices of more than $300/tr oz, and (3) there may be a significant decline in production after 2010 in the absence of major new discoveries and continued development of new mines to offset the depletion of the South African mines. 'Economist. 'Geologist. Minerals Availability Field Office, Bureau of Mines. Denver, CO. INTRODUCTION Since the late 1970's the world gold mining industry has been characterized by dynamic growth in production and volatility in price. World production in 1983 was almost 15 pet greater than in 1980, and the short-term trend con- tinues upward. Annual dollar-based price changes since the early 1970's have been unprecedented in both magnitude and volatility. This commodity, which sold for $35/tr oz 3 for over a third of a century, reached an annual average $612/tr oz in 1980. Until 1980, the future price of gold seemed set on an upward trend. Since that year, however, world gold prices have fallen by over 50 pet to slightly over $300/tr oz at the end of 1984. Against this background, this study was undertaken to determine (1) the level of world demonstrated gold re- sources, (2) total and annual availability of gold mine pro- duction, (3) long-run production costs per operation and by national aggregates, (4) the economic and technical factors that impact upon cost and availability, and (5) probable future mine production and cost trends. One statement concerning the results of this study must be made at this point: the unprecedented rise in the price of gold, especially beginning in the mid-1970's, caused ex- ploration and new mine development to expand worldwide at a rapid pace. This pace has continued through 1984, even given the decline in U.S. dollar gold prices since 1980. One reason is the poor economic outlook for base metals, which by comparison renders gold an attractive investment alter- native for the mining industry in general. During the last few years, gold exploration results, mine expansions, and new mine developments have been reported on an almost daily basis. This increase in activity has been most evident in major producing nations such as the United States, Canada, Australia, and Brazil, although seemingly every nation is experiencing some activity. Even South Africa, the industry leader which as recently as 1981 was being written off as a rapidly dwindling producer, has increased its level of production, resources, and new mine develop- ment since 1980. The outlook for future gold price and pro- duction response, however, remains one of uncertainty. Table 1 lists the evaluated primary gold operations by nation and productive status. This list consists of 135 primary gold mines and deposits which were subjected to complete evaluation to determine demonstrated resources and long-run total production costs. These 135 properties represent the great majority of known gold resources and annual production in the market economy countries (MEC's). The study also presents recent information on ap- proximately 65 other (mostly small) properties in various stages of development or initial production. These other properties are all new developments that have been reported in the literature within the last 3 yr. They are perhaps the best indicator of the intense level of dynamic activity that has characterized the world gold industry in recent years. In addition to the above primary mines and deposits, over 200 base metal mines and deposits were evaluated to deter- mine byproduct gold content. No other commodity throughout history has maintained the economic and political importance of gold, and with the possible exception of oil, no other commodity has undergone so much rigorous analysis by so many individuals, com- panies and nations. This Bureau of Mines appraisal makes a unique contribution to this body of literature as a result of its extensive mine coverage and the detailed engineer- ing and economic cost analyses that it has produced. ACKNOWLEDGMENTS Domestic and foreign data collection were performed under contract by Morrison-Knudsen Co., Inc., Boise, ID, and Davy-McKee Corp., San Ramon, CA. Personnel of the Bureau's Minerals Availability Field Office, Denver, CO, evaluated the data and performed the engineering and economic evaluation analyses. COMMODITY OVERVIEW Gold is a unique mineral commodity in that it is, above all, an alternate store of wealth to fiat currencies, most notably the U.S. dollar. The U.S. dollar serves as the primary medium of exchange in international transactions and is the currency in which world gold prices are denominated. For that reason, the following interactive fac- tors all impact directly upon the demand for, and price of, gold: 1. The level of U.S. interest rates; 2. General inflationary trends and expectations; 3. The rate of growth in U.S. money supply; 4. The exchange rates between the U.S. dollar and other major currencies; The troy ounce = 31.1035 g. 5. The balance of payments position in U.S. trade accounts; 6. Government budget deficits; 7. Perceived or real economic, political, or natural crises, etc. The great majority of gold is held either as an invest- ment medium or as a form of insurance to hedge against uncertainty about the value of fiat currency or potential disaster. Only a small percentage of the world's total gold production has been consumed in industrial applications, and much of that metal will eventually be recycled. Gold is a unique mineral commodity also in that long- term production cost increases do not in and of themselves imply rising market prices as is the general case with base metals. This is because typical world annual production of Table 1.— Primary gold mining operations Included In the availability study Table 1.— Primary gold mining operations Included In the availability study— Continued Country and operation name Type 1 Status as of January 1984* Australia: Central Norseman Fimiston Leases Hill 50-Morning Star Mt. Charlotte Mt. Morgan Tailings Project. . . . North Kalgoorlie Telfer Bolivia: Teoponti Brazil: Morro Velho Serra Pelada Canada: Agnico-Eagle (Gold Div.) Camflo-Malarctic Hygrade Campbell Red Lake Con-Rycon Detour Lake Dickenson Dome Giant Yellowknife Golden Giant (Hemlo) Kerr Addison Lac Minerals (Hemlo) Ladner Creek Lupin Project Macassa (Willroy) Pamour Porcupine Sigma Specogna Teck-Corona (Hemlo) Chile: El Indio Colombia: El Bagre La Salada Dominican Republic: Pueblo Viejo . Ghana: Ashanti Prestea Tarkwa Mexico: Pinzan Morado Philippines: Benguet Gold Operations .... Masbate Paracale Republic of South Africa: Beatrix Blyvooruitzicht Bracken Buffelsfontein Consolidated Modderfontein . . . Deelkraal Doornfontein Dreinfontein Consolidated Durban Roodepoort Deep East Rand Proprietary Mine . . . Egoli Consolidated (East) Egoli Consolidated (West) Elandsrand Ergo (East Rand Au & U Co.) . . E. T. Consolidated Fairview (Barberton Mine) Free State Geduld Grootvlei Harmony Hartebeestfontein Joint Metallurgical Scheme Kinross Kloof Leslie Libanon Loraine (Allanridge) Marievale President Brand President Steyn-Video Randfontein Estates RMMM Slimes Project Simmer and Jack St. Helena Stilfontein Unisel Vaal Reefs Venterspost Village Main Reef West Rand Consolidated Western Areas Western Deep Levels u p u p u p u p SD p U p S p SD p U p S p u p u p u p u p u p u p u p u p u N u p u N u p u p u p u p u p S N u N u p SD p u p s p u p u p u p u N u p s p u p u p u p u p u p u p u p u p u p u p u p s p s p u p s p u p u p u p u p u p u p s p u p u p u p u p u p u p u p u p u p s p u p u p u p u p u p u p s p u p u p u p Country and operation name Type 1 j?nUwy^984* Western Holdings Complex U P Winkelhaak U P Witwatersrand Nigel U P Taiwain: Chinkuashih U P United States: Alaska: Apollo U N Bear Creek S P Big Hurrah S P Chicken Unit S P Circle Unit S P Fairbanks Unit S P Golden Zone U N Independence U N Kougarok District S P Livengood Placers S P Mikado U P Nome Beaches S P Peters Creek S P Solomon Unit S P Tuluksak Dredges S P Wiseman Unit S P California: McLaughlin S N Royal-Mountain King S N San Juan Ridge S N Yuba Placer Operations SD P Colorado: Victor Project S P Idaho: Homestake- Yellow Pine S N West End-Garnet Creek S P Michigan: Ropes U N Montana: Golden Sunlight S P Zortman-Landusky S P Nevada: Alligator Ridge S P Battle Mountain S P Borealis S P Carlin Operations S P Goldfield Project S N Jerritt Canyon S P Pinson-Preble-Ogee S P Round Mountain S P Windfall S P New Mexico: Ortiz S P South Dakota: Homestake U P Utah: Mercur S P Washington: Knob Hill U N Zimbabwe: Arcturus U P Athens U P Blanket U P Dalny U P How U P Mazoe U P Muriel U P Old West-Redwing U P Patchway-Brompton U P Renco U P Shamva U P Venice U P 1 S — surface, SD — surface (dredging), U— underground. Any operation utiliz- ing underground mining for a majority of mill feed is classified as underground. Those operations utilizing entirely surface material for mill feed are classified as surface. 2 N— nonproducer, P— producer. newly mined gold is approximately 1,000 to 1,300 mt 4 , whereas total world gold stocks stand at a minimum level of 100,000 mt. With new supply representing only about 1 pet of world stocks, production costs of newly mined gold have little influence upon market price relative to movements of the large volume of aboveground stocks. Also, since most new gold goes into jewelry, bullion, and coins, or is held for security reasons, it is not really "consumed" but rather becomes part of the existing stock. As world stocks increase, the effect of stock movements upon price movements increases as well. The metric ton = 32,150 tr oz. At the level of individual producing mines, gold produc- tion, in general, responds inversely to the market price of gold. However, the production response in the overall in- dustry is composed of two parts. There is the response of current producers and the response of exploration and new mine development. Current producers with the ability to vary the average grade of their mill feed (pay limit) respond to rising gold prices by lowering this cutoff grade or pay limit. Gold output thus falls or remains more or less con- stant even though mill throughput increases. In this case, higher gold prices enable mines to produce less gold without lowering revenues and extend the life of the mine. Converse- ly, falling gold prices for these operations will generally result in the cutoff grade being raised (where possible) which raises production, compensates for the lower gold price, and maintains sufficient revenues but could result in a shorten- ing of mine life. In the case of new mine development, the reaction is different. Rising gold prices will elicit renewed effort at ex- ploration and new mine development. This has been especially true in gold since the dramatic price increases of the mid to late 1970's. The development of a gold mine from deposit discovery to production can take roughly 2 to 10 yr, depending upon a host of technical, financial, and legal factors. The new gold supply that results from rising gold prices is thus temporal in its impact. For example, the high gold prices of 1979-80 resulted in many new mine developments which began producing between 1981 and 1984. To the extent that the current, relatively low, U.S. dollar price of gold causes exploration, new mine develop- ment, and expansion plans to be reduced or delayed, the impact upon production will not be evident for a few years because of this time lag. An examination of recent gold production and price data demonstrates the net effect on world gold production from the interaction of these two responses outlined above. Table 2 contains production data for 1971, 1981, and 1983. From 1971 to 1981, annual gold production from the eight largest MEC producers collectively declined by 9.0 million tr oz. During the same period, the price of gold increased from $41/tr oz to $460/tr oz. South African output declined by 10.2 million tr oz, while the United States, Canada, and Australia posted declines ranging from 81,000 to 570,000 tr oz. Output in other countries (primarily the Soviet Union, China, and Brazil) increased enough to offset around 5.4 million tr oz of this overall decline. The net effect was that total world gold production fell by 5.2 million tr oz between 1971 and 1981. Most of the South African production decline was due to a lowering of the average mill feed grade (pay limit) and the opening up of newer, lower grade areas as a result of rising gold prices. Since 1981, annual South African pro- duction has increased by approximately 762,000 tr oz while the U.S. dollar price of gold has fallen. The price of gold during 1983 averaged around $35/tr oz lower than the average 1981 price and more than $187/tr oz lower if com- pared to the average 1980 price. South African and other MEC private sector producers have, where possible, clearly raised their average mill feed grades to compensate for this declining price in order to maintain sufficent operating revenues. In addition, new mines came into production during the 1979-83 period. In other countries the production responses have been quite different owing to a number of interrelated aspects unique to each country's gold industry; these are dealt with in detail in the individual country sections. Table 2. — Gold production, 1971, 1981, and 1983 for the 10 largest producing countries, thousand troy ounces Change 1971-81 Change 1981-83 -10,268 + 762 + 1,725 + 175 -570 + 601 -116 + 578 + 1,650 + 200 + 1 ,043 + 400 -81 + 444 + 118 + 59 + 529 + 29 + 337 + 170 -5,640 + 3,389 -9,015 + 3,014 -5,244 + 3,283 Country 1971 1981 1983 South Africa 31,389 21,121 21,847 Soviet Union e 6,700 8,425 8,600 Canada 2,243 1 ,673 2,274 United States 1,495 1,379 1,957 China "50 1,700 1,900 Brazil 157 1,200 1,600 Australia 672 591 1,035 Philippines 640 758 817 Papua New Guinea . . 24 553 582 Chile 64 401 571 Total 43,434 37,794 41,183 8 largest market economy countries 1 36,684 27,669 30,683 Total world 46,494 41,250 44,533 "Estimated. 'Less the Soviet Union and China. Sources: Lucas (1), BuMines (20). Table 3 explains the general relationship evident in some major free world producers between the gold price of a nation's currency, the currency price of gold, and the ex- change rate relationship between the U.S. dollar and these other currencies. The relative movements of these variables .have a very significant impact upon the grade and quantity of gold ore that is mined and the amount of refined gold that is produced in these countries. The data presented are for the three largest MEC gold producers. An example us- ing Canada will make the paramount importance of these relationships apparent. In 1978, the U.S. dollar price of gold was $194/tr oz. This meant that Canadian mines needed to produce 4.51 milliounces of gold (0.00451 tr oz) to earn $1 in revenue. This 4.51 milliounces was the "Canadian gold price of the U.S. dollar" in 1978 and dictated the grade and quantity of ore that was mined in Canada. The exchange rate be- tween U.S. and Canadian dollars in 1978 was US $0.88 = Can $1. Thus, the Canadian gold price in 1978 was Can$221/tr oz. Since Canadian primary lode gold produc- tion in 1978 was 1.185 million tr oz, Canadian dollar revenues totaled Can$262 million. At the average 1978 ex- change rate, this was the equivalent of US$230 million. In 1980, the average price of gold had risen by 69 pet to an historic high of $612/tr oz. This was the Canadian equivalent of Can$716/tr oz. The impact of the price rise on production was to allow material of much lower grade to be profitably mined. Thus, the "Canadian gold price of the U.S. dollar" also fell 69 pet to 1.39 milliounces for 1980. Canadian lode gold production in 1980 declined to 1.074 million tr oz as a result of lower grade ore being mined. But revenues in terms of both currencies were more than dou- ble their 1978 level despite this drop in production. In 1983, the opposite result was evident. The price of gold fell 31 pet to $425/tr oz. The "Canadian gold price of the U.S. dollar," by definition, increased 37 pet. Production, responding inversely to price, increased 69 pet owing to pro- ducing mines increasing the grade of ore mined (where possible) and new mines coming into production as a result of the rising prices of the late 1970's. The overall result was that revenues in U.S. dollar terms increased 18 pet despite the decrease in the gold price. Similarly, Canadian dollar gold revenues rose, but by a larger percent (24) owing to the decline in the value of the Canadian dollar relative to the U.S. dollar. Table 3. — Relationship between gold price, gold production, and exchange rate variations Country and currency Year Gold per local currency unit, 10 3 tr oz Change from prior period pet Exchange rate, U.S. dollars per local currency unit Primary lode gold mine production, 10 3 tr oz Gold market price, 1 local currency per troy ounce Value or production 10 6 local currency unit 106 U.S. dollars United States - dollar ($) 1978 1980 1983 Canada - dollar (Can$) 1978 1980 1983 South Africa - rand (R) 1978 1980 1983 5.15 1.63 2.35 4.51 1.39 1.90 5.92 2.09 2.11 NAp -68 + 44 NAp -69 + 37 NAp -64 + 1 1.0 1.0 1.0 580 636 1,553 $194 $612 $425 $112.5 $389.2 $660.0 .8766 .8552 .8114 1,185 1,074 1,824 Can$221 Can$716 Can$524 Can$261.8 Can$769.0 Can$955.7 1.1500 1 .2854 .8991 22,700 21,705 21,847 R169 R476 R473 R3.836.0 R1 0,331.0 R10.333.0 112.5 387.2 660.0 229.5 657.6 775.5 4,411 13,279 9,290 NAP Not applicable. 'Based upon the U.S. dollar average annual price of gold converted at the average annual exchange rate. 1.20 1. 10 1.00 - x / o o in S> .90 x UJ Q .80 .70 .SO r~ -v' Australia ^ 'United States » \ /Canada \\ \ A \ \.--^ /\ South Atrica " ><*:*-*^r\\ \ \ \ 1970 1972 1974 1976 1978 1980 1982 1984 Figure 1. — Index of exchange rates: United States, Canada, Australia, and South Africa. Variations in exchange rates have significantly im- pacted industry profitability in recent years. This is because production costs are usually denominated in local curren- cies, but gold revenues, ultimately, are based upon the in- ternational price of gold in terms of the U.S. dollar. If a coun- try's currency declines in value more than the dollar price of gold, the resultant foreign exchange gain can offset both rising production costs and declining prices, albeit not in- definitely. This very important relationship is discussed in detail for South African mines in that country's individual chapter. For now, figure 1 will suffice to underscore the re- cent history of exchange rate variations for four of the largest MEC gold producers. The figure indexes the current value of each currency relative to its 1975 value. As shown, the currencies of Australia, South Africa, and Canada have declined in value, relative to the U.S. dollar, since 1975. For example, the 1983 South African rand had only 65 pet of the purchasing power of the 1975 rand. The following summary presents the general results of this study. It provides comparative results of resource and production availabilities, production costs, and future pro- duction potential as determined for each nation included in this study. It is intended as a general overview. The gold mining industries of each country vary a great deal owing to many country-specific factors. Indeed, each gold mining industry and each gold mine is absolutely unique. For this reason each major producing country is covered in detail in a subsequent chapter. HISTORICAL PERSPECTIVE ON PRODUCTION THROUGH 1983 The point of reference of this study is January 1984. This section places the 1984 resource availability and produc- tion potential in an historical perspective. The intention is to give a general idea of how much gold had been produced prior to January 1984, and the sources of that production, as well as how much gold is currently being produced from present sources. Table 4 provides data on the historical production of gold by time period. It is estimated that as of 1984 between 3.8 and 4.0 billion tr oz of gold had been produced throughout history. This is roughly equivalent in volume to a cube 55 ft on a side. Approximately 60 pet of this historical total has been produced in just the last 53 yr with 45 pet of the production during the last 53 yr coming from just one na- Table 4. — Estimates of total world gold production, by time period Total world . , . , Time period gold production, Areas of ma J°' Production, 10 6 tr oz ranKea Dy size 3900 B.C. - A.D. 1492. 400-500 Africa, Europe, Asia. 1493-1600 23 South America, Africa, Europe. 1601-1700 29 Do. 1701-1800 61 South America, Europe, Africa, Mexico. 1801-1900 374 United States, Australia, Soviet Union, South Africa, Asia. 1901-30 585 South Africa, United States, Soviet Union Australia, Canada, Asia. 1931-83 2,286 South Africa, Soviet Union, Canada, United States, Australia. Total (rounded) 3,800-4,000 NAp. NAp Not applicable. Sources: Mohide (4) and the authors' own estimates based upon numerous other sources. Table 5. — Estimated total production for five major historic gold-producing countries 1 Country First major discoveries Total production 10 6 tr oz Relative reliability of estimate South Africa .... 1872 1,247 Good. Soviet Union . . . 1775 350-450 Poor. United States . . . 1799 335 Good. Canada 1866 215 Do. Australia 1851 190 Do. 'Compiled from numerous sources and authors' own estimates. Table 6. — World gold production in 1983 1 Country Major countries (discussed in detail): South Africa Soviet Union Canada United States Brazil Australia Subtotal (84 pet of total) Important countries (general discussion): China 1 Philippines Papua New Guinea Chile Zimbabwe Colombia Dominican Republic Ghana Peru Mexico Subtotal (13 pet of total) 40 other countries (not discussed) (3 pet of total) Total (100 pet) 44,533 NAp Not applicable. 1 Not discussed owing to a lack of information on gold resources. Sources: Lucas (1), BuMines (20). 1983 gold production, 103 tr oz 1983 ranking 21,847 1 8,600 2 2,274 3 1,957 4 1,600 6 1,035 7 37,313 NAp 1,900 5 817 8 582 9 571 10 430 11 429 12 348 13 303 14 166 15 223 16 5,339 NAp 1,881 NAp NAp f0 .. *M • "A CANADA 2.3 1 UNITEO STATES y V 20 / MEXICO 0.3 DOMINICAN REPUBLIC »-0.3 COLOMBUj BRAZIL PERu\ 0.2 > It emm 0.6 , JTH AFRICA 21.9 PHILIPPINES . 0.8 SXaa. » _ . PAPUA NEW GUINEA S^\ ♦ f AUSTRALIA \ Figure 2.— Evaluated gold producing countries and 1983 production levels (10 s tr oz). tion, the Republic of South Africa. This is why the issue of world gold availability (especially MEC gold availability) centers on the ability of South Africa to maintain its enor- mous productive capacity. The Soviet Union has not released any production data since 1934, when it began a policy of treating all such data as state secrets. Estimates, not surprisingly, have shown wide variation over the years. A reasonable estimate of Soviet Union gold production since 1930 would be between 250 and 350 million tr oz. This means that the Soviet Union and South Africa not only account for 75 pet of current world production but have also accounted for 55 to 60 pet of total production since 1930. World gold production should remain dominated by these two nations for the remainder of the 20th century. Table 5 summarizes historical production for the five countries that have accounted for a majority of world gold production since 1930 and places a reliability estimate on the total production figures. Table 6 provides a breakdown of 1983 world production. Figure 2 shows the 15 countries discussed in this study and their 1983 production amounts. They represent 15 of the top 16 world gold producing coun- tries; China is omitted owing to lack of information. As shown in table 6, the six nations discussed in detail in this study accounted for 84 pet of 1983 world gold pro- duction. The People's Republic of China (ranked fifth) is not discussed at all owing to a lack of information on gold resources. The five MEC's evaluated in detail similarly ac- count for 84 pet of MEC production. RESOURCE OVERVIEW, 1984 Demonstrated resource tonnages and weighted average grades were determined for all 135 primary gold proper- ties (table 1) in the 14 market economy countries that were included in the cost evaluations. This overview concerns itself primarily with a subset of 111 operations that were either known producers as of 1984 or major operations in the late stages of development that appeared certain to come Canada 4.6 pcV United States 4.4 pet Others 4 pet into production during 1984-86. This latter group consists of only four properties: the McLaughlin operation in Califor- nia, United States, and the Golden Giant, Lac Minerals, and Teck-Corona operations in the Hemlo District of On- tario, Canada. The 107 major operations producing in 1984 easily account for over 90 pet of primary gold production and over 80 pet of total primary plus byproduct production in the MEC's. The excluded properties are 16 Alaskan operations of intermittent or questionable production status, 6 nonproducing properties in the continental United States, 1 nonproducing property in Canada, and 1 nonproducing property in Mexico. Cost and availability estimates for these 24 nonproducing properties are discussed in the individual country chapters. Table 7 and figure 3 provide summary data, by coun- try, of total 1984 demonstrated gold resources in the 111 selected properties. The estimates on the left side pie of figure 3 are presented on a contained basis; that is, the total amount of gold contained in the mill feed. This measure accounts for mining recovery and dilution. As is clearly evi- dent, the vast majority of gold on a contained basis is ac- Table 7. — Total demonstrated primary gold resources in 1 1 1 selected properties in market economy countries as of January 1984 Country Total contained gold, 10 s tr oz South Africa United States 1 Canada 1 Brazil Ghana Australia Philippines Zimbabwe Chile Bolivia and Colombia. Taiwan 758.0 40.4 38.4 10.5 7.3 6.7 4.1 3.7 2.4 1.0 .5 Total recoverable qold 10 6 Pet Of tr oz total 716.0 87 32.2 4 36.6 4.5 9.8 1 6.0 <1 6.4 <1 3.5 <1 3.0 <1 2.0 <1 .8 <1 .4 <1 Total 873 819 100 'Estimates differ from totals given in country sections owing to the exclu- sion of certain nonproducing properties. NOTE.— Data may not add to totals shown because of independent rounding. Canada 4.5 pet United States 4 pet Others 4.5 pet Total contained gold Total recoverable gold 873xl0 6 troz 8I9 x lO 6 troz Figure 3.— Percentage distribution by country of total contained and total recoverable gold In 1 1 1 selected properties. counted for by just three countries, the Republic of South Africa, the United States, and Canada. South Africa alone accounts for 87 pet of the total contained gold. The domina- tion of these three countries is due to two factors. First, the Witwatersrand Basin of South Africa is the single largest source of gold the world has ever known. Second, Canada and the United States have benefitted from ongoing explora- tion and development activity due to their having large established mining industries with abundant technical ex- pertise and sufficient financial capital. These countries also have histories of political and economic stability. Australia accounts for just under 1 pet of total contained gold and has the same historically strong mining industrial base as Canada, the United States, and South Africa. This nation has added significantly to its demonstrated resources in re- cent years and has good potential for further additions as exploration and development activity continues. There are two caveats to the above resource analysis which must be made to place world gold resources in the proper perspective. They concern the type of resource data that are reported, and the type of resource occurrence that accounts for production in an individual country. First, the information reported often varies owing to Government policies and also to a reluctance on the part of many mining companies to divulge data on their gold operations. This is due to a number of factors. Among them is the desire to minimize or avoid taxation in countries that tax the value of unmined gold reserves. Also, most public- ly held mining companies are required to report their reserves (as opposed to resources) annually to various in- stitutions. For valid reasons, these annual reported reserves are usually defined as that ore which has been developed on at least three sides and assayed as thoroughly as re- quired. These "proven reserves" are redefined each year based on production, new development, new assays, changes in prices and costs, and, possibly, changes in technology. These proven reserves are justifiably conservative for the above reasons and because many vein-type gold mines are geologically erratic, which makes reasonable inferences of resources difficult or perhaps impossible. Also, exploration work to delineate resources is a costly and time consuming endeavor and will not be done by small operators or poor countries, or perhaps even by large operators that have been in production for a long time. For example, some mines in Canada have annually reported proven reserves sufficient for only 1 to 5 yr of production and have been doing so for 20 yr or longer. Most gold mining operations will not estimate beyond the proven reserve level without a very good reason, such as plans for major capital investments. The second caveat deals with the type of resource oc- currence in an area or country. Some countries, such as Brazil, Colombia, Bolivia, and the U.S. State of Alaska, derive the great majority of their gold production from placer deposits. These operations are generally small scale, intermittent, and nearly impossible in most cases to estimate for contained resources, much less costs of produc- tion. In the case of Brazil, especially, most production emanates from tens of thousands of individuals or small groups mining placer deposits in the Amazon Basin. The Government of Brazil does not really know how much gold is produced or smuggled out of the country. Brazil un- doubtedly has large gold resources, but attempting to measure them is virtually impossible. Thus, there are two forces at work in terms of demonstrated resource estimation. First, the largest and best established areas generally have the best estimation and reporting. Second, those countries or areas with geological occurrences that lend themselves relatively easily to estimation, such as the Witwatersrand Basin, have the best available data in terms of quality and quantity. In the case of the South African mining industry, the quality and quantity of reported data are unsurpassed. Basically, the key to ascertaining world demonstrated gold resources is to know where gold has not been measured or reported in addition to knowing where it has been measured and reported. This study, by necessity, deals only with countries and areas where enough basic information is collected and reported so that it is possible to estimate demonstrated resources with some reasonable degree of confidence. In that sense, the demonstrated resource estimates of this study must be considered conservative. TOTAL PRIMARY GOLD AVAILABILITY AND PRODUCTION COST EVALUATION The same subset of 111 operations were evaluated to determine relative long-term production costs and total gold availability. Figure 4 and supporting data in table 8 sum- marize the overall results of this evaluation. This figure and table relate cumulative gold availability to increasing production cost levels. In addition, table 7 relates total gold availability by nation, and the right side of figure 3 relates total gold availability by national percentage contributions. The major conclusions of the total availability evalua- tion follow: 1. South Africa is by far the single largest source of economic gold production. It represents 716 million tr oz (87 pet) of the 810 million tr oz of total recoverable gold available from these 111 operations. 2. 70 pet of total recoverable gold is economic at a cost level of $300/tr oz or less. This gold is available from just 46 operations, 23 of which are in South Africa. South African mines account for 90 pet of the gold in this cost range. 3. 83 pet of total gold is available at a cost level of $400/tr oz or less. This additional 13 pet is contained within another 25 operations. South African mines also account for 90 pet of total gold available at $400/tr oz or less. 4. The great majority (94 pet) of recoverable gold available from the evaluated demonstrated resources is con- tained within ore bodies mined by underground methods. The individual country percentages South Africa, 98 pet of gold available from underground resources; Canada, 91 pet; Australia, 77 pet; United States, 17 pet. COMPARATIVE PRODUCTION COSTS IN MAJOR PRODUCING COUNTRIES Table 9 and figure 5 contain comparative long-run cost data for South Africa, Canada, the United States, and Australia. A general comparision of production costs reveals that capital costs per ounce of recovered gold are similar among the four countries. For underground resources, the weighted average estimates for capital costs range from $52/tr oz to $73/tr oz recovered gold. Surface mining capital costs in the United States and Australia are effectively the same at $47/tr oz and $48/tr oz, recovered gold, respective- ly. Total capital costs are obviously much greater for underground operations because of the need for such high- cost items as shaft systems and mine plant facilities plus more expensive exploration requirements. However, owing Table 8. — Supporting data for figure 4: distribution of total recoverable gold by cost level and country Break-even cost or price level 1 $200 or less $300 or less $400 or less $500 or less $600 or less $700 or less Over $700 Number of operations (cumulative) 14 46 71 96 105 107 111 Total recoverable gold, 10 6 tr oz (cumulative) 184.5 576.4 667.6 773.1 811.8 812.5 819.0 Cumulative percent of total availability 22 70 83 94 99 99 100 Countries within indicated range, ranked by amount of recoverable gold $200 or less: South Africa, Canada, Brazil, United States, Dominican Republic. $201 to $300: South Africa, Canada, United States, Australia, Philippines, Zimbabwe, Chile, Bolivia. $301 to $400: South Africa, Canada, United States, Australia, Brazil, Zimbabwe, Colombia. $401 to $500: South Africa, Canada, United States, Philippines, Australia, Zimbabwe. $501 to $600: South Africa, Zimbabwe, Canada, United States. $601 to $700: Canada, Zimbabwe. Over $700: Ghana, Taiwan. 'The break-even cost or price level is that point where long-run total production cost per ounce = required price per ounce to obtain a 0-pct discounted cash flow rate of return (DCFROR). 700 -W- 200 300 400 500 600 700 TOTAL RECOVERABLE GOLD, I0 6 tr oz 900 Figure 4. — Potential gold available at break-even costs of production from 111 operations In market economy coun- tries as of January 1984. 10 400 Surface Underground lift" 11 - I PI I KEY ^^$| Breakeven costs [;-:ffi;-;| Operating costs I Capital costs United States Australia South Africa Canada COUNTRIES Figure 5.— -Comparative long«tarm total production costs In selected countries. Tab!* 0. — Comparative long-run total gold production costs In selected major producing countries Country Resource occurrence Capital cost per troy ounce' Operating cost per troy ounce 2 Total capital plus operating cost per troy ounce Weighted Range average Break-even cost or price level 3 Range Weighted average Range Weighted average Range Weighted average $7 -$95 21-126 22-136 44-132 NAp $52 47 62 73 48 $123 $488 117- 401 93- 526 148- 314 240- 310 $247 271 202 222 291 $145 $559 152- 527 134- 598 192- 444 290 360 $299 318 264 295 339 $148 -$573 173- 538 134- 598 200- 477 325- 365 '$285 United States 9 Canada* Surface Underground 354 292 310 Surface 353 NAp Not applicable 'Unrecovered capital investment in mine and and mill plant and equipment, infrastructure, and development remaining as ot Jan. 1984 and reinvestments through life of operation. 2 Totai mining plus milling cost per troy ounce of recovered gold. *The break-even cost or price level is that point where long-run total production cost per ounce = required price per ounce at 0-pct DCFROR 'The break-even cost estimate is lower than the total capital plus operating cost estimate owing to the influence of byproduct credits tor uranium. 'Underground cost data for the 1 mine in this category were withheld to avoid disclosing possible proprietary data. ■Includes 3 Hemlo District operations currently in late stages of development, which will be among the lowest cost world producers and represent 44 pet of total recoverable gold analyzed in Canada Surface mining is insignificant 11 to higher grades, the capital costs on a per ounce basis at underground mines represent only a small part of total pro- duction costs. Long-run operating costs 5 (mining plus processing) per ounce of recovered gold actually favor underground min- ing in general. This is due again to generally higher grades and higher mill recoveries at underground mines. In the case of South Africa, the weighted-average cost estimates are dominated by a number of very large mines which have excellent scale economies. For Canada, the overall estimate is dominated by the 3 new mines in the Hemlo District, which will rank among the 10 lowest gold mines in the world. It is important to stress that this analysis is comparing long-run total production costs per ounce of recovered gold. If one looks only at direct operating costs or current total production costs, the results of this type of comparison would possibly be different, but from a long-term availability perspective, which is cognizant of current and future capital and operating costs and which expresses those costs on a recovered-gold basis, the results favor underground mining. A final comparison of long-run break-even total produc- tion costs demonstrates that all four of these major gold pro- ducing MEC's have, on average, gold mining industries that are economic at long-run prices similar to those that have prevailed in recent years. However, if prices remain in- definitely below $350/tr oz to $400/tr oz many individual mines and new mine developments will become uneconomic. POTENTIAL ANNUAL PRODUCTION THROUGH 2000 This section is intended to provide a general overview of annual production potential for the 111 evaluated primary operations relative to production cost levels and country distribution. The individual country chapters pro- vide general assessments of the production potential of each nation through the end of this century. These chapters deal with all sources of gold production, including byproduct pro- duction from base metal mines. A number of country -specific issues that will undoubtedly impact upon each nation's future production potential (primary and byproduct) are also dealt with separately in the individual country chapters. Figure 6 plots annual production potential through the year 2000 at increasing cost levels. The major conclusions of the analysis are— 1 . Total gold production from the 111 evaluated opera- tions is not expected to decline significantly during the re- mainder of this century even with no major additions to demonstrated resources. Production in the year 2000 is ex- pected to be only 9 pet below the level for 1984, given con- stant 1984 prices exceeding $300/tr oz, owing primarily to a maintenance of production by South African mines. This assumes that market forces are the only factors at play. Clearly, political, social, or military turmoil could greatly change this scenario. 2. In any given year, it is expected that more than 80 pet of total production will emanate from operations with long-run total production costs of $400/tr oz or less. 3. Because of the relative importance of surface mines in the United States and Australia, both countries must con- tinuously replace a portion of their demonstrated resource base to maintain current primary gold production levels. Current demonstrated resources at most mines are suffi- cient for only 5 to 10 yr of production. Both nations have experienced tremendous exploration and development ac- tivity in recent years, which has served to increase annual output. Over the long term, however, the main issue will be one of production maintenance rather than production enlargement. Exploration, development, and production economics in these two nations have come to favor produc- tion maintenance by surface minable resources. 4. Canadian primary gold production potential through the turn of the century should also remain relatively con- stant due to the development of the Hemlo gold district. This district is expected to increase annual output by at least 700,000 tr oz/yr by the end of the 1980's, which would off- set projected declines from other mines. The three mines have initial demonstrated resources sufficient for at least 20 yr of production at full capacity, beginning in 1986. 35 30 n 25 o fClEX}2 r •«\>*s^ FAR WEST RANO A"v *°< tY*NPtn FAR WEST RAND GOLDFIELD 14 Dreifontein Consolidated (East-West -North Dteitorrtein) 15 Western Deep Levels 16 Blyvooruiulcht 17 Doointontein IB Elandsrand 19 Deelkrool WEST RAND GOLDFIELD 20 Kloof 21 Libanon 22 Venterspost 23 Western Areat 240,24b Randfontein Estate* • Egoli Consolidated (taking • limes from Randfontein Estates) 25 West Rand Consolidated LEGEND CENTRA L RAND GOLDFIELD 26 Durban Roodepoort Deep 27 RMMM Slimes Project (Old Ciown Mines Property) 26 Simmer and Jock 29 Village Main Reef EAST RAND GO! DFIELD 30 Witwatertrond Nigel 31 Marievale 32 Grootvlel 33 East Rand Proprietory 34 Consolidated Modderforrtein 35 Ergo 36 Egoli Consolidated EVANDER GOLDFIELD 37 Winkelhaok 38 Bracken 39 Leslie 40 Kinross Areas ore planned extensions for development beyond 1984 (Plans circa 1981-84) -I — t— f- Roilroads irt o o _i < o 500 - i i i - — r 1 i 1 400 - y - 300 - ^/ 200 J~^ - -■. 100 - iiii i . . 100 700 200 300 400 500 600 TOTAL RECOVERABLE GOLD, IC^troz Figure 12.— Potential total South African gold available at break-oven costs of production as of January 1984. 800 Ventersdorp System. Feed grades range from 2.5 g/mt milled tol8.6 g/mt milled. The Leader Reef in the Orange Free State Goldfield has the lowest grade range, 2.5 to 3.5 g/mt; thus, it is milled along with the relatively high-grade Basal Reef materials grading 6.2 to 11.4 g/mt. The Carbon Leader Reef in the Far West Rand Goldfield, at 9.4 to 18.6 g/mt, and the Vaal Reef in the Klerksdorp Goldfield, at 9.0 to 11.0 g/mt, are the highest grade reefs. In terms of distribution among the various goldfields, the Main and South Reef groups, the Kimberly Reef, and the VCR- Elsburg Reef group are the most common, all being present in four of the seven goldfields. TOTAL AVAILABLE GOLD FROM PRODUCING SOUTH AFRICAN MINES The 44 largest primary gold producing operations, ac- counting for 97 pet of total 1983 production, were evaluated. These operations are listed in table 1. Total refined gold potentially recoverable from all opera- tions is estimated at 716 million tr oz: 702 million tr oz from underground mines and 14 million tr oz from surface opera- tions. This represents approximately 16 yr of total world production or 32 yr of South African production at 1984 levels. Figure 12 shows the total amount of refined gold cumulatively available at increasing cost levels from all 44 evaluated operations. In January 1984 dollars, long-term break-even produc- tion costs (as defined in the appendix) range from $148/tr oz to $573/tr oz. Seventy-three percent of potentially available refined gold is recoverable at a cost level of $300/tr oz or less, 85 pet is recoverable at $400/tr oz or less, and more than 94 pet is recoverable at $500/tr oz or less. The price of gold averaged $425/tr oz in 1983, which indicates that the great majority of South African gold resources are economic. This average 1983 price is lower than the $459 average price recorded in 1981 and markedly lower than the $612 average recorded for 1980, but it still represents a 334-pct increase over average 1973 prices. These high prices are responsible for the ambitious development and expansion plans that were announced in 1980-82 in South Africa and which have been incorporated into this analysis. Table 13 shows the total amount of refined gold cumulatively available at increasing cost levels from the 38 underground mines. The weighted average long-term cost of refined gold production for all 38 underground mines is estimated at $285/tr oz. The lowest cost underground mine is estimated at $148/tr oz. This weighted average reflects the influence of a number of exceptionally large and profitable mines. Seven of these 38 underground mines, for example, are estimated to each contain in excess of 40 million tr oz recoverable gold, and an additional 5 underground mines are estimated to each contain in excess of 20 million tr oz recoverable gold. For the six low-grade 18 Table 1 3 — Total gold potentially available at increasing coat or price levels from 38 underground operations In South Africa Gold cost or Break-even DCFROR 10-pct DCFROR price level. Available gold, Number Available gold, Number per troy 10 6 tr oz of mines 10 6 tr oz of mines ounce (cumulative) (cumulative) (cumulative) (cumulative) $200 143.2 3 51.7 2 $300 515,8 21 450.0 17 $400 600.2 28 600.2 28 $500 664.1 36 661.4 34 Over $500 . . 701.7 38 701.7 38 sand and slimes dump reprocessing operations, the weighted average long-term break-even cost is estimated at $335/tr oz refined gold. The overriding importance of a small number of large underground South African mines is demonstrated by the data of table 14. The 10 largest mines in terms of total recoverable gold account for 481 million tr oz. This is 67 pet of total South African gold and 59 pet of evaluated total MEC recoverable gold. The long-term break-even unit cost of these 10 mines ranges from $148/tr oz to $509/tr oz with a weighted average of $273. Table 1 4. — Ten largest South African operations in terms of total recoverable gold 1 Operation name Ownership Dreifontein Consolidated Gold Fields. Vaal Reefs Anglo American. Western Deep Levels Do. Free State Geduld Do. Kloof Gold Fields. Western Holdings Complex Consortium of 4 owners. Randfontein Estates Johannesburg Consolidated. East Rand Proprietary Barlow Rand. President Steyn-Video Anglo American. Harmony Barlow Rand. 'Total recoverable gold = 481 million tr oz: Percent of South African gold = 67. Percent of market economy country gold = 59. Long-run total unit cost: Range = $148 to $509. Weighted average = $273. As shown in table 15, 9 of these underground operations are also listed among the 10 largest average annual pro- ducers on a life-of-mine basis. The 10 operations in table 15 account for approximately 55 pet of expected average an- nual South African production during the 1984-90 period and for approximately 45 pet of expected average annual MEC production from the 111 evaluated operations during this same period. The ownership of this large annual pro- ductive capacity and total recoverable gold is concentrated in five of the large mining houses in South Africa. Given that these mining houses own other mines both inside and outside South Africa, they represent a large percentage of world gold ownership and supply as well. In table 16, the 10 lowest cost underground producers are listed according to this study's estimate of long-run total break-even unit cost. Four of the 10 operations contained in this listing are also among the 10 largest annual life-of- mine producers. All 10 of these operations have long-term break-even unit costs below $250/tr oz. Given South Africa's current dominant production posi- tion and its very large quantity of total and annual gold available at relatively low break-even cost levels, it would appear that its industry currently establishes the minimum floor price of newly mined gold. This means only that free market gold prices could not remain below this average South African cost level for long without causing either ma- Table 1 5. — Ten largest South African operations in terms of average annual life-of-mlne production 1 Operation Name Ownership Dreifontein Consolidated Gold Fields. Vaal Reefs Anglo American. Western Deep Levels Do. Randfontein Estates Johannesburg. Free State Geduld Anglo American. Western Holdings Complex Consortium of 4 owners. Harmony Barlow Rand. Kloof Gold Fields. Buffelsfontein 2 Gencor. President Steyn-Video Anglo American. 'Total average annual output = 12.8 million tr oz. Percent of average annual South African production (1984-90) = 55. Percent of average annual production from 111 operations (1984-90) = 45. 2 Only operation not on 10 largest recoverable gold list. Table 1 6. — Ten lowest cost producers in terms of long-run total break-even production cost Cost level Operation name per troy ounce Kloof 1 Less than $210. Kinross Do. Dreifontein Consolidated 1 Do. Winkelhaak Do. Blyvooruitzicht Do. Buffelsfontein 1 Less than $250 Unisel Do. E. T. Consolidated Do. Western Deep Levels 1 Do. Doornfontein Do. 1 Among the 10 largest average annual life-of-mine producers. jor contractions in world supply of newly mined gold or large South African Government assistance payments to keep that nation's gold mines operating. The two mines with long-run unit costs estimated to be above $500/tr oz (table 13) have long received state assistance. It is important to stress, however, that the gold mines of South Africa are not state owned and are profit motivated. In this regard, the analyses at the 10-pct profitability level determined long- run total costs ranging from $157/tr oz to $618/tr oz, with a weighted average for all underground mines of $302/tr oz. Economic Effect of Byproduct Production Many of the operations produce other mineral products in association with gold. The most notable is uranium; silver, pyrite concentrates, and sulfuric acid are also pro- duced. Silver production for all operations is a byproduct of primary gold production. Revenue generated from silver production is insignificant, generally less than 1 or 2 pet of the total. The total amount of silver recoverable along with the gold from 42 operations is estimated at approx- imately 67 million tr oz. Eighteen operations contain less than 500,000 tr oz silver, and 18 contain over 1 million. Only eight operations are estimated to have average annual pro- duction levels exceeding 100,000 tr oz silver contained in total annual dore bullion production. Most South African uranium production is a byproduct of gold production. In contrast to the situation for gold, data relating to contract prices and production are difficult to obtain. Of the underground operations that produce uranium, six produce enough (or contain enough) that the overall economics of the operation are affected by potential uranium revenues. In only two cases, however, is it felt that uranium revenues are essential to the economic com- petitiveness of the operation. During 1984, four operations announced their intention to cease byproduct uranium pro- 19 duction. One of these operations is converting its uranium plant to a gold treatment plant, and at least one other opera- tion is considering doing the same. A portion of total uranium production is derived from primary uranium mines. One of these has decided to close down indefinitely, and two have converted to become primary gold producers (S, pp. 296-297). The current uranium market situation is quite poor, as is reflected in the conversions occurring in 1983 and 1984. It can be ex- pected that a certain level of uranium production will be maintained owing to political consideration of the South African Government. Capital and Operating Costs Underground gold mining in South Africa is somewhat anomalous in that it is both highly labor intensive and re- quires high levels of capital investments and reinvestments. Data on capital and operating costs as estimated in this study are included in table 17. Total capital investments and reinvestments over the estimated remaining mine lives range from $15 million to $3.8 billion per operation. Table 17. — Economic summary data for 38 underground gold mines in South Africa _ Total or Hange weighted average Operational data: Average annual output 10 3 troz 30-2,100 21,000 Total recoverable gold 1 10 6 tr oz 670-91,450 702 Producing years from January 1984 6-63 NAp Capital and operating cost data: Total capital investment 2 .. .10 6 dollars . . . $15-$3,835 $33,150 Annual capital reinvestment do $1-$98 NAp Capital cost per troy ounce $7-$95 $52 Operating cost per troy ounce 3 $123-$488 $247 Total operating plus capital cost per troy ounce $145-$599 $299 Long-run total cost per troy ounce: Break-even (0-pct DCFROR) $148-$573 "$285 10-pct DCFROR $157-$618 $302 NAp Not applicable. 'Refined gold estimated to be recoverable as of Jan. 1984. 2 Unrecovered capital investment in mine and mill plant and equipment, infrastructure, and development remaining as of Jan. 1984 and reinvestments through life of operation. 3 Mining plus milling cost per troy ounce of refined gold. "Less than total operating plus capital cost owing to influence of byproduct credits for uranium. Thirteen operations will require between $500 million and $1.0 billion in total capital investments, eight will require in excess of $1.0 billion, and four will require in excess of $3.0 billion of investments over their estimated remaining mine lives. On an annual life-of-mine basis, total capital rein- vestments are estimated to range between $1.0 million and $98.0 million with 50 pet of the operations in the $10.0 to $50.0 million range. South African tax law allows for the expensing (1-yr writeoff) of capital expenditures in the year incurred. Capital costs per ounce of recoverable gold are remarkably similar for most operations, which indicates that, to a certain degree, capital costs reflect adjustments by the operations to the grade of ore and mining conditions that are encountered. Thirty of the operations have capital costs per ounce of recovered gold ranging between $30 and $60. Operating costs per ounce of recoverable gold show a wide range over all operations ($123 to $488), but at least 50 pet of all operations have estimated costs between $150/tr oz and $250/tr oz. The single most important component of operating cost is labor, which accounts for an average of approximately 48 pet of basic mining costs and 42 pet of basic milling costs. Another factor input of importance is the capital cost of min- ing and milling equipment and replacement parts. Figure 13 plots index data on the rate of increase since 1970 in both of these two major components of South African gold production costs, i.e., basic wage costs (with no allowance for productivity changes, fringe benefits, etc.) and mining and milling equipment costs (9, pp. F1-F2). Both general cost categories have steadily increased since the early 1970's, with wage costs increasing at a faster rate than equipment costs. This is attributable to both rising gold revenues and rising expectations of black mine workers. Changes in labor relations have recently culminated in the legalization of black mine workers' unions. Labor wage costs in South Africa have increased at an average rate of 17 pet annually since 1978. It was in this year that labor wage costs began to rise faster than the cost of mining and mill- ing equipment, which has risen at an average annual rate of 12.5 pet since 1978. Also shown in figure 13 are the U.S. dollar-based equivalents of these two index series. Since 1980, the gap between the cost of labor and equipment in rands and the cost in dollars has widened significantly owing to the devaluation of the rand. This devaluation has offset both rising rand-based production costs and declining U.S. dollar- based gold prices. The operating cost data shown in table 17 reflect a con- stant 1984 dollar analysis. It is interesting though to look at possible future trends in production costs based on the assumption of continued wage cost increases at South African gold mining operations. Toward that end, an analysis was performed that addressed this issue in the following manner. First, the trends in the cost of major fac- tors of production since 1978 were ascertained. Second, this information was used to escalate mine and mill operating costs for a 10-yr period, 1985-94. These escalated costs were then held constant at their 1994 values for the duration of the mines' productive life and were employed in reestimating long-run total break-even unit costs of produc- tion. Lastly, these determined unit costs under the escala- tion scenario were compared to the constant 1984 base case costs derived earlier. Since 1978, combined rand-based mine and mill operating costs have increased at an average annual rate of 16 pet. This analysis assumed that this rate will continue during 1985-94 and that the rand will continue to devalue at an assumed average rate of 8 pct/yr, thus offsetting only one-half of the rand-based cost increase. This results in an annual 8-pct U.S. dollar-based cost increase for all opera- tions during 1985-94. The results of this analysis are given in table 18. As shown in the escalating cost case, 18 operations (contain- ing 53 pet of available gold) have total production cost estimates below $500/tr oz, as compared to 36 operations (containing 95 pet of available gold) under the base case 20 350 300 250 O O 200 if] x* Id 9 150 100 I I I KEY — Labor index, rands — Labor index, U. S. dol lars — Equipment index, rands — Equipment index, U.S. dollars 1970 1972 1974 1976 1978 I960 1982 1984 Figure 13.— Rate of increase in labor and equipment costs in rand and dollar terms, 1 070-83. Table 1 8. — Total South African gold potentially available at increasing cost or price levels: base case versus 8-pct-escalation case Available gold at Number of Gold cost or break-even DCFROR, operations price level, 10 6 tr oz (cumulative) (cumulative) per troy ounce Bgse Escalated Base Escalated case case case case $200 143.2 3 $300 515.8 54.3 21 3 $400 600.2 70.5 28 7 $500 664.1 368.9 36 18 Over $500 701 .7 701.7 38 38 scenario. Thus, without devaluation of the currency to off- set wage cost increases, the economics of gold production in South Africa could be severely affected. This analysis, however, actually serves to underscore the long-term com- petitiveness of the South African mines for four reasons. First, it is not at all implausible to assume that dollar-based production costs in other major producing countries will also arise at an 8-pct average rate. Second, it is also not implausi- ble to assume that the price of gold during the next 10 yr will increase at an average rate of 8 pct/yr in 1984 U.S. dollar terms. Third, even if labor wage costs more than dou- ble, as assumed in this analysis, and U.S. labor wage costs remain constant, average South African wage costs would still be only about one-third of the U.S. level. Fourth, gold is not expected to become a commodity where producers compete for a tight or dwindling market plagued by over- supply or lack of demand because production costs and mine production do not influence market price. The gold price is demand driven, and it is fully expected that South African production will continue to adjust to market prices and re- main the most significant supplier for the rest of this century. Effect of Exchange Rate Variation Upon Industry Profitability The profitability of South African mines has been significantly affected recently by variations in the value of the rand relative to the U.S. dollar. Since September 1983, South African mines have been paid directly in U.S. dollars for their gold production. This has had the effect of transfer- ring foreign exchange gains to the mines, which has helped to offset rising rand-based production costs. For example, during 1981-83 rand-based production costs increased be- tween 28 and 30 pet. But during this same time period the rand-dollar exchange rate depreciated some 27 pet. The result has been that dollar-based gold production costs in South Africa have increased only 1 to 3 pet from 1981 through 1983. This foreign exchange gain has also helped to offset declining dollar-based gold prices. Where transactions in- volve a dollar-rand exchange rate (the "dollar price of the rand") that has declined more than the dollar price of gold, 21 the rand price of gold has actually increased. This helps to maintain profitability for the majority of South African mines. Figure 14 presents data on total gold revenues ex- pressed in rands and U.S. dollars. Of particular importance is the period 1980-83. During 1980, the average annual price of gold reached an historic high of $612/tr oz. South African gold produc- tion was 21.7 million tr oz., which generated total industry revenues of $13.3 billion. At the average annual exchange rate, this was the equivalent of 10.3 billion rands (R). From 1980 to 1983, the average annual dollar price of gold fell 30 pet, and South African gold production, at 21.8 million oz, was essentially the same as in 1980. Thus, dollar-based revenues for 1983 also fell 30 pet to $9.3 billion. Revenues in rands, however, were unchanged at R10.3 billion because the dollar-rand exchange rate had also declined by an equivalent 30 pet. Thus, the devaluation of the rand offset the falling dollar price of gold and helped maintain industry profitability. South African gold production remained more or less constant from 1980 to 1983 because the price of gold in rands, and hence total rand revenues, remained more or less constant. Not all of the foreign exchange gain observed during 1983 was transferred to the mines themselves, since pay- ment in dollars did not become official until later in the year. The impetus for such a payment change is evident from figure 14, where it can be seen that in 1981 the dollar curve dropped below the rand curve due to the rand falling below parity (Rl = $1) with the dollar. The gap between the dollar curve and the rand curve is the average foreign exchange gain accruing due to the rand devaluation. The driving force for this devaluation was the declining dollar- based gold price itself as well as a positive inflation differen- tial between South Africa and the United States. With rand- based costs rising 14 to 16 pct/yr and dollar-based revenues falling, a change to enhance profitability became an economic imperative. Preliminary data for the first 6 months of 1984 show that this trend has continued. The dollar price of gold had fallen to a 6-month average of $381/tr oz, a decline of ap- proximately 9 pet relative to the average 1983 price. The dollar-rand exchange rate during this same period had fallen by approximately 11 pet, thus causing rand-based gold prices to rise slightly (R4/tr oz to R5/tr oz). Thus, declin- ing dollar revenues were again being offset by rising rand revenues, and profitability was being maintained. The most recent data available on exchange rates and dollar gold prices indicate that the rand devaluation has continued at an increasing rate. The dollar price of gold has also continued to decline. If and when the price of gold begins another strong upward spiral, it is possible that the devaluation of the rand will be checked, and an apprecia- tion would be in order given a large enough increase in the dollar price of gold. By definition, a rising dollar gold price means that the value of the dollar is declining relative to gold, but not necessarily relative to the rand. 14 13 - 12 - O 1 — 1970 IIO (A 100 o Id 90 O 80 70 35 30 | 5 £ 20 ~ 15 C 1972 1974 1976 1978 1980 1982 1984 Figure 14. — Total gold revenue* In rands and dollars, 1970-83. 22 Annual Production Potontial How long the gold resources of South Africa last will be determined by the differential between long-term pro- duction cost and market price, current production rates, ex- pansion plans for existing mines, and the development of new mines in the future. It appears that the South African gold industry, far from stagnating or declining, remains very optimistic about the future of gold mining in that country. For example, General Union Mining Corp. (Gencor) has recently announced the development of a new mine (not included in our analysis) in the Evander Goldfield of Eastern Transvaal. The Poplar Mine is estimated (10, p. 10) to contain approximately 60 million mt recoverable ore with an average recovery grade of 5 g/mt over the mine life. This equates to about 9.6 million tr oz recoverable gold. To put the size of the resource base at this one property into perspective, the Poplar Mine alone is estimated to contain more gold than the demonstrated resources of the top 12 producing gold mines in Zimbabwe, which is ranked 11th in world production. At an initial pro- duction capacity of 175,000 to 275,000 tr oz/yr, Poplar would be considered average to small in size relative to existing South African mines, yet it would be as large or larger than either of the two newest major gold mines in the United States — McLaughlin (under development) and Jerritt Can- yon (producing). Poplar may begin producing by 1988. Another new South African mine development, the Beatrix Mine, began producing in December 1983. This mine was included in our analysis. It is estimated to have approximately 40 million mt recoverable ore with a mill feed grade of 6 g/mt. Annual production is currently estimated at 350,000 to 375,000 tr oz refined gold. Accord- ing to production plans for 1985, this mine alone will pro- duce almost as much gold in that year as total projected production for the entire gold mining industry in Zimbabwe. These two new mine developments are simply exten- sions of producing goldfields (Evander and Orange Free State) that in at least one case have been known about for years. In addition, expansions to capacity or maintenance of production by developing new shaft systems at present- ly producing mines essentially represent new mine develop- ment as well. For example, the consolidation of the East and West Driefontein operations will allow the North Driefontein area to be exploited as an addition to this merged operation rather than as a new mine. The North Driefontein area is estimated to contain approximately 27 million mt ore grading 14.3 g/mt or 12.4 million tr oz con- tained gold (11, p. 135). The merger of West and East Driefontein may allow the boundary pillars separating them to be mined. These pillars alone are estimated to contain approximately 900,000 tr oz gold. Another example of South Africa's potential for expan- sion is Randfontein Estates, one of the oldest mines in the country, which began exploitation of the Cooke section (6 to 8 miles south of the old mine) in 1973 (12). The Cooke section, although part of the old Randfontein Estates opera- tion, still represents a new mine development. In addition, sinking and development of the new No. 3 shaft in the Cooke section is expected to be completed by 1985. This new shaft will increase output significantly and is also essentially a new mine development. The preceding examples show that significant develop- ment activity continues in the gold mining industry of South Africa. They also underscore the difficulty of predicting future production and the life of gold-bearing resources based upon a static analysis. The typical assumptions that are usually employed for predicting future gold production in South Africa follow: 1. Output from each mine will remain at current full- capacity levels (which may or may not include expansions underway). 2. No new discoveries will be made. 3. Some new mines in already proven areas will be developed. The result of this type of static analysis is invariably the prediction of rapidly declining total annual production as the demonstrated resources of existing mines are exhausted and not replenished. This results in South Africa experi- encing a decreasing percentage share of the world new gold market and the implication of nearly complete overall resource exhaustion. These two very recent new mine ex- amples (Beatrix and Poplar), along with numerous expan- sions of existing operations, show the potential for error, especially underestimation, that exists under this type of static scenario. In fact, of the 44 evaluated operations, 6 underground and 6 surface waste reprocessing operations were not in production in 1975. With the above discussion as an important caveat, the reader is referred to figure 15, which presents estimates of potential annual gold production available at three different break-even cost-price levels ($300, $400, and $500) for 5- 1984 i _l_ I '•>lIT.T.t7 14 3 f * i "*■ / 1 ° ( *\ * \ 1 *** i Detroit r \ * J MICHIGAN LEGEND • Cities ?? Costed, producing gold mines K Noncosted, producing gold mines X Costed, prospective gold mines X Noncosted, prospective gold mines 500 Scale, km Figure 1 8. — Location of producing and prospective gold mining operations in the continental United States. 34 _ OtE (BOULDER X CREEK)* JERRITT CANYON So* Fronciico ^CARLIN BATTLE MOUNTAIN X tt HORSE CANYON wRELIEF " rikjvnu ALLIGATOR SUMICH BUCKHORN ROGC * * K * PARADISE PEAK W ' NDFALL K SANTA FE W BOREAL IS ^ ROUND MOUNTAIN LEGEND • Cities K Coated, producing gold mints W Noncosted, producing gold mints X Costed, prospectivt gold mines H Noncoited, prospectivt gold mints 200 I Scale, bin Figure 19.— Location of producing and prospocthro gold mining operations In California and Nevada. LEGEND (§^^ Areos of primary Mesozoic gold deposits (J25 Areas of primary Cenozolc gold deposits Q^) Areas of primary Paleozoic gold deposits Scale, km Figure 20.— Areas of primary gold deposits In the con- tinental United States. 35 TacI* 27— Gold resource data for selected primary gold mines and deposits in the continental United States as of January 1 984 Number of Demonstrated resource' Contained gold' Recoverable gold Classification properties ^r^l Snare, pet 10* tr oz Share, pet 10« tr oz Share, pet Maior producing operations. Surface 15 647 93 29 77 216 73 Underground 1 48 7 8_5 23 8_0 27 Total 16 695 100 375 100 29_6 100 Other mines and deposits: 2 Costed: Surface 5 65 5 96 5 13 93 4 47 93 Underground 1 2_5 4 37 7 33 7 Total 6 68 100 5 5 100 4 8 100 Noncosted 3 18 NM NAp 9-12 NAp NM NAp Total 24 NM NAp 14 5-17 5 NAp NM NAo Grand totals: Demonstrated (costedl . . 22 763 NAp 43 NAp 34 4 NAp Demonstrated plus noncosted- 40 NM NAp 52-55 NAp NM NAp NAp Not applicable. NM Not meaningful. 'Mill feed basis, includes adjustments for mining recovery and dilution 2 For detail on names and production status, see table 32. 3 Not subiected to complete demonstrated resource and long-term total cost evaluation copper producers and to a lesser extent lead-zinc producers, ascertain figure. Byproduct gold resources in all of the and because gold production from these mines is a result United States may be as high as 20 million tr oz, with of base metal mining itself and not a driving cause, no fur- roughly half of this total contained in the Bingham Canyon ther attempt was made to address the potential availability Mine in Utah. Total U.S. gold resources, therefore, from of byproduct gold production from copper or lead-zinc mines. both primary and byproduct sources, are estimated to range All significant domestic copper and lead-zinc producers have from a high-probability (demonstrated) estimate of 43 been evaluated in detail in other MAP appraisals (21-22). million tr oz in 22 major primary properties to a lower prob- The 16 major primary gold producing operations in the ability estimate of 79 million tr oz if all known primary and continental United States as of 1984 represent at least 38 byproduct sources of significance are included, regardless individual deposits. This is because 10 of the 16 operations of production status. An independent estimate by are mining or will mine from 2 to 5 individual deposits over Homestake Mining (23) places primary gold resources in the life of the operation as defined by the current level of the United States (assumed to include Alaska) at 69 million demonstrated resources. tr oz. This estimate includes all major producers and well Table 27 provides aggregate resource data for the 16 over 100 explored prospects. The difference between this major producing operations and for 24 other properties study's potential primary gold resource estimates and the (figures 18 and 19 show locations) that are either nonproduc- Homestake estimate is due entirely to a difference in defini- ing, developing, or in the early stages of production. The tion concerning which properties and how many properties resource data for the 16 major producers were estimated to include in the total estimate. Interestingly, the percen- at the demonstrated level. The 24 other properties that are tage split between gold contained in surface and included are those considered as the most likely candidates underground ore is the same, even though the Homestake for full development during the 1984-90 period. Six of these estimate includes many more surface properties and a 24 properties were subject to complete demonstrated higher estimate of gold resources at the Homestake resource and cost evaluation to determine the near-term underground mine at Lead, SD. economics of new gold production in the United States. For As a final note, given that large areas of Alaska and the 18 noncosted properties, the estimates of contained gold the Western United States have not been thoroughly ex- are presented as a range to indicate the lower level of prob- plored for their gold potential, it is felt that gold resources ability that is attached to them in the absence of a com- could be as high as the 100-million-tr-oz level published by plete demonstrated resource evaluation. Lucas (1). This rounded figure is considered to be an inferred In total, primary recoverable gold resources in the con- upper range. In contrast, the resource data herein reported tinental United States as of January 1984 are estimated on and evaluated are "locateable" by known deposit and/or to range from 43 million tr oz demonstrated resource con- operation, tained in 16 producing and 6 nonproducing operations to as high as 55 million tr oz if the 18 noncosted properties of significance are included. TOTAL GOLD AVAILABILITY AND The great majority (93 pet) of demonstrated resource PRODUCTION COST EVALUATION: (mill feed) in the 16 major producing operations is surface- 1 6 MAJOR PRODUCERS minable ore. Only 7 pet occurs as underground ore. Higher ore grades and recoveries for the underground resource The 16 major producers account for over 90 pet of all result in 27 pet of total recoverable gold being represented primary gold production and a majority of total production by underground ore. Still, the majority of recoverable gold in the United States as of 1983. As shown in table 27, these represents a surface resource. Of the six costed non- 16 operations in total contain an estimated 695 million mt producers, only one is a potential underground mine; it of demonstrated resource (mill feed) which contains 37.5 represents 7 pet of the contained and recoverable gold in million tr oz of gold. As of January 1984 there is an this category. The United States is expected to remain estimated total refined gold availability of 29.6 million tr highly dependent upon surface-minable gold resources. oz contained within all 16 operations. The five largest opera- Primary gold resources in Alaska, contained within 16 tions (Homestake, Carlin, Round Mountain, Battle Moun- known, relatively major (mostly placer) operations of inter- tain, and Jerritt Canyon) contain 83 pet of the total mittent production status, are estimated to be 4 million tr recoverable gold. These same five operations, not surpris- oz, a probability of 50 pet is attached to this difficult-to- ingly, account for at least 50 pet of total annual primary 36 600 TOTAL RECOVERABLE GOLD, lO^troz Figure 21 .—Potential total U.S. gold available at various production coat levels from 16 major producers as of January 1984. Table 28 — Supporting data for figure 21; total gold potentially available from 16 major producing operations in the continental United States Break-even DCFROR 10-pct DCFROR Cost or price level, per troy ounce Available gold, 10 3 tr oz (cumulative) Number of mines (cumulative) Available gold, 10 3 tr oz (cumulative) Number of mines (cumulative) $200 or less 2,311 2 4 11 15 16 162 4,957 20,140 22,313 29,594 1 $300 or less $400 or less $500 or less 4,957 20,140 29,556 4 11 14 Over $500 29,594 16 DCFROR = discounted cash flow rate of return. gold production. The six costed nonproducers are estimated to have potential recoverable gold totaling 4.8 million tr oz. Figure 21 and table 28 present the total gold availability estimates and long-run total cost determinations for the 16 major producers. Long-run production costs were deter- mined at the break-even (0 pet) and 10-pct rate-of-return levels. Fully two-thirds of the total gold is available from 11 operations at a cost-price level of $400 or less. These 11 operations (10 surface and 1 underground) are able to realize a long-run rate of return of at least 10 pet given a gold price of up to $400/tr oz. Only four operations can break even at a cost-price level of $300/tr oz or less; these same four can realize at least a 10-pct long-run rate of return at the $300 cost-price level. Fifteen operations can break even, and 14 can realize at least a 10-pct long-run rate of return, at a long-run cost- price level of up to $500/tr qz. The profitability of the 16 major producers and the 6 costed nonproducers is also demonstrated by the data of table 29, which contains rate-of-return determinations at constant-dollar gold prices of $400/tr oz and $500/tr oz. At a constant-dollar gold price of $400, 11 current producers have profitable rates of return ranging from 13.5 pet to over 120 pet. None of the six costed nonproducers is profitable at a constant-dollar gold price of $400/tr oz given the criteria of this analysis (i.e., more than a 10-pct ROR). At $500/tr oz, 14 of the major producers and 3 of the nonproducers are profitable. One producer (surface mining) and three nonproducers (two surface and one underground) require constant-dollar gold prices in excess of $500/tr oz to be profitable according to the criteria of this analysis. The three nonproducers that are not profitable at $500/tr Table 29. — Long-run DCFROR for selected U.S. gold properties, determined at $400 and $500 per troy ounce gold, $10 per troy ounce Ag, and a 10-pct discount rate Gold price, per troy ounce Number of operations DCFROR range. 1 pet 16 PRODUCERS $400 $500 11 5 14 2 13.5-120.0 .0 11.0-120.0 .0- 9.8 6 NONPRODUCERS $400 $500 6 3 3 0.0- 7.4 12.5- 15.0 4.0- 8.0 'The economic model does not calculate rates of return below or above 120 pet. oz all have long-term operating cost estimates above $350/tr oz recovered gold. The three nonproducers that are pro- fitable at $500/tr oz (all surface operations) have operating costs below $250/tr oz recovered gold. The weighted-average operating cost for all five surface nonproducers that were cost-evaluated is estimated at $254/tr oz recovered gold; not suprisingly, the profitable operations are those with costs below the average, while the nonprofitable operations have costs well above the average. Given current (late 1984) gold prices below $350/tr oz, all six of the evaluated nonproducers have long-run cost levels, over the life of the operations, that would result in real economic losses, especially given that typical mine lives 37 Table 30. — Comparative summary results of 1984 long-run cost determination analyses for producing and nonproducing surface operations in the continental United States Nonproducing Producing (5 operations! (15 operations) Weighted Weighted Range average Range average Operational data: Mill teed grade' g/mt 2.0 - 4 8 4.0 0.8 - 7.9 2.2 Average annual output 10 3 tr oz 10- 125 NAp 6- 268 NAp Total recoverable gold 2 10 3 tr oz 94-2.628 NAp 38-7.243 NAp Producing years from Jan. 1984 for 60 pet of the operations 7- 10 NAp 3- 10 NAp Capital and operating cost data: Total capital investment 10" dollars S6-S165 NAp S5-S356 NAo Capital cost per troy ounce $63 - S96 $75 S21-S126 S47 Operating cost* per troy ounce S175-S370 $254 S117-S401 $271 Total operating plus capital cost per troy ounce $271 -$445 $329 $152 -$527 $318 Long-run total cost per trov ounce Break-even 10-oct DCFROR) $329-$458 $351 $l73-$538 $354 10-pct DCFROR ... $449-$577 $468 S180-S567 S396 NAp Not applicaDle. 'Mill teed grades tor each individual property are weighted-average grades ot the entire demonstrated resource minaOle over the lite ol each operation. These individual property grades are then weight-averaged over each subgroup lor comparison 2 Retineo gold estimated to be recoverable as ot Jan 1984 HJnrecovered capital investment in mine and mill plant and equipment, infrastructure, and development remaining as of Jan 1984 plus reinvestments through life of operation "Mining plus milling cost per troy ounce of refined gold. average only 5 to 8 yr. One possible exception is the McLaughlin operation in California, which has an initial 20-yr life, large planned annual capacity, and hence the potential to alter its mining and development plans to ac- commodate depressed prices, at least during the near term. For current producers, the maintenance of a $350 gold price would place 9 of the 16 operations (including 6 heap leaches) in a position of suffering a real economic loss and possible eventual closure. This could reduce expected average an- nual output by the late 1980's by up to 50 pet. In addition, current Gate 1984) gold prices have at the very least delayed the eventual development of many of the 24 nonproducing properties included in this discussion, including 5 of the 6 nonproducers that were cost-evaluated. The following section continues with an in-depth analysis of the comparative economics of current surface producers versus potential or developing surface mines. Comparative Long-Run Total Production Costs Table 30 contains economic summary data for the 15 surface producing and 5 surface nonproducing operations that were evaluated to determine comparative economics of new gold production in the United States. The five non- producing surface operations are believed to adequately cover the range of most prospective future producers. In ad- dition, the estimated average mill feed grades, productive lives, and capital and operating costs per ounce of recovered gold fall within the ranges established by the 15 major sur- face producers. These new operations are, in effect, a replacement resource similar to that being currently mined. The major difference between current and prospective producers lies in total capital investments. For the non- producing properties, total capital investments, by necessi- ty, include the costs of initial development, construction, and purchase of new plant and equipment and related in- frastructure, as well as the estimated costs of capital reinvestments over the life of the operation. These costs range, in total, from $6 million to $156 million for the development of surface mines capable of producing between 10,000 and 125,000 tr oz/yr gold. The majority of new sur- face mines (three of the five in this sample) will have an- nual outputs below 50,000 tr oz/yr and total capital in- vestments below $50 million. For current producers, the major initial capital in- vestments have mostly been recovered through past pro- duction by 1984. Most of the capital costs remaining are those associated with future reinvestments or expansions to capacity. The most notable example of expansion is the Carlin operation in Nevada, where the Gold Quarry deposit is undergoing full development. The initial cost to fully develop this deposit is estimated to exceed $130 million in January 1984 dollar terms. Eleven of the 15 producing operations have total required capital investments and reinvestments as of January 1984 below $50 million. Of particular interest is the fact that even though the five nonproducing operations range from 10,000 to 125,000 tr oz potential annual output, requiring between $6 and $165 million to fully exploit, capital costs on a per ounce of recoverable gold basis fall within the narrow range of $63/tr oz to $96/tr oz. This range probably represents an economic threshold of expected per-ounce capital costs for the development of new surface gold mines in the United States. Any prospective operation that falls significantly above the upper limit of this range will be less attractive for development (especially given that a majority of total capital investment is made prior to production) unless there is an expectation of a compensatingly lower than average per-ounce operating cost. Operating costs for both groups, however, show wide variation between operations, even those of similar size. This is not surprising given that each mine has unique characteristics regarding its geologic or technical and operational parameters. The difference between total capital plus operating costs per ounce of recoverable gold and the break-even cost-price level is greater for current producers owing primarily to greater tax liabilities. Total break-even production costs for the 15 surface producers range from $173/tr oz to $538/tr oz. This range encompasses that derived for the non- producers; as a result there is no significant difference be- tween the two groups in terms of the weighted-average long- term total production costs at the break-even level. A significant difference in total production costs is evi- dent between the two groups at the 10-pct profitability level. All five of the nonproducers require 1984 constant prices exceeding $449/tr oz to obtain this prespecified rate of return. This large increase in required price is a result of the requirement to recover larger total and per-ounce capital costs in less time, and for some, over smaller an- nual production levels. 38 1.8 1.6 14 g 1.2 s- s— • V V \ -^0-$500 £ ,0 Q" 8 _1 O O .6 .... 1 0-$3O0 .4 '—*-^,^ .2 \ 1 1 1 ..... 1 1 1984 1966 1988 1990 1992 1994 1996 1998 2000 Figure 22. — Potential annual U.S. gold production available at various break-even production cost levels from 16 major producers as of January 1984. Annual Production Potential Through 2000 As noted previously, the United States is heavily depen- dent upon surface mining for maintaining gold production. On a relative, worldwide basis, surface primary gold mines in the United States (both current producers and those awaiting or under development) are generally low-grade, short-life operations with annual outputs in the 20,000- to 90,000-tr-oz-range. There are notable exceptions such as the Jerritt Canyon, Round Mountain, Battle Mountain, Carlin, and McLaughlin operations. These five operations each pro- duce or will produce in excess of 100,000 tr oz/yr gold and have remaining productive lives (based on current demonstrated resources) extending at least through the mid-1990's; three of the five operations are expected to pro- duce beyond the year 2000. By contrast, the majority of the remaining producing and developing mines, as well as those considered most likely for development during this decade, generally represent annual production below 60,000 tr oz and have remaining or expected productive lives of 5 to 10 yr. Thus, the United States, unlike Canada or South Africa, will have to replace a large part of its demonstrated resources every 8 to 10 yr to maintain current production levels. This section details expected annual production through 1990 for the 16 major evaluated primary gold producers (as of the reference point January 1984) and incorporates estimated annual output from those operations that are cur- rently under development or in the early stages of produc- tion and those that are considered the most likely can- didates for development. Two caveats are necessary. First, with a large number of small, short-life operations con- tinually coming into and going out of production, it is very difficult to forecast potential output in any given year. Se- cond, estimating output for developing or potential opera- tions by necessity involves margins of error because develop- ment plans may change, be delayed, or be canceled altogether. This has become especially true during the last few years as gold price increases have failed to live up to the expectations of 1979-80. The purpose of this exercise is not to prophesy but rather to indicate the direction of future events given current information. The authors do not feel that the margins of error inherent in this type of analysis are significant enough to violate the basic trends and implications. Table 31. — Potential annual production estimates for 1 6 major primary gold producers in the United States Year Number of producing operations Total potential annual production, 10 3 tr oz 1984 1985 1986 1987 1988 1989 1990 1992 1994 1996 1998 2000 16 16 16 15 15 15 13 11 9 7 4 3 1,378 1,425 1,585 1,555 1,528 1,506 1,398 1,352 1,252 957 830 700 Figure 22 and table 31 present estimates of long-run break-even total production cost and annual gold availabil- ity, covering 1984 through 2000, for the 16 major produc- ing operations that were evaluated. In the case of the Carlin operations in Nevada, the developing Gold Quarry deposit is included as part of the overall ongoing operation and not as a separate developing deposit. During 1984-90, annual primary gold output from these 16 operations is estimated to increase from 1.378 million tr oz in 1984 to 1.585 million tr oz in 1986 and then decline to 1.398 million tr oz by 1990. During this 7-yr period, three operations (two heap leaching and one conventional surface) are expected to have ceased production. Also during this time, the large Gold Quarry deposit of the combined Carlin operations is expected to reach capacity production, which will add around 150,000 tr oz/yr to total output. The addi- tional output from the development of Gold Quarry will compensate for the three smaller operations that are ex- pected to be fully exploited by 1990. During the period 1990-2000, total production is ex- pected to further decrease as current demonstrated resources are depleted. In the absence of additions to demonstrated resources, 10 more operations are expected to close by the year 2000, leaving 3 of the 16 current pro- ducers still active after that time. Total production in 2000 from these three large operations will still represent 51 of 1984 production. This illustration underscores a basic characteristic of the U.S. gold mining industry: Most of the operations are small and their lives are relatively short, 39 Table 32. — Operations considered possible or probable producers during 1984-90 State and operation name Mining method and ore type' Status Estimated annual capacity, 10' tr oz Estimated first lull year ol production NEAR-TERM PRODUCERS AND PROSPECTS California Grey Eagle Jamestown Mesquite Rich Gulch Idaho: Dewey Thunder Mountain Montana: Beal Montana Tunnels Nevada: Buckhorn Dee (Boulder Creek) Horse Canyon Paradise Peak Relief Canyon Ruby Hill Sante Fe Sumlch South Dakota: Annie Creek Washington: Cannon Subtotal (rounded) California: McLaughlin Royal-Mountain King San Juan Ridge Idaho: Homestake- Yellow Pine Michigan: Ropes Nevada: Goldfield Pro)ect Subtotal (rounded) Grand total (rounded) 'OP — open pit, UG— underground. OP milling ore do OP heap leach OP milling ore do. do. do. do. OP heap leach ..do .do OP milling ore OP heap leach UG milling ore OP heap Teach UG milling ore OP heap leach UG milling ore NAp Initial production 1983 Developing 1984 Development decision expected late 1984. Feasibility study completed in early 1984. Initial production 1983, expanding 1984. Feasibility study underway 1984. Feasibility study underway 1984. . do Initial production mkt-1984. Developing 1984 Initial production late 1983. Feasibility study underway 1984. Developing 1984 Feasibility study stage 1984. Feasibility study underway 1984. Developing 1984 Nominal production late 1983. Developing 1 984 NAp 40 120 80 100 36 25 45 55 30 30 35 90 25 36 30 30 20 140 970 1984. 1985. 1986. 1986. 1984. 1986. 1986. 1987. 1985. 1986. 1984. 1986. 1985. Post-1987. 1987. 1986. 1984. 1985. NAp. EVALUATED FOR LONG-JERM PRODUCTION COSTS OP milling ore . .do do do UG milling ore OP heap Teach NAp NAp Developing 1984 Feasibility study stage 1984. Developing 1984 Explored deposit Developing 1984 Feasibility study stage 1984. NAp NAp 130 44 40 90 22 12 340 1,310 1985 Post-1986. 1986 Post-1987. 1987 Post-1988. NAp. NAP. which means that the resource base must be constantly replaced to maintain current output levels. Table 32 lists 24 operations that are either developing or considered probable for development between 1984 and 1990. With only one exception, the list was held to the criteria of (1) there is at least 20,000 tr oz annual produc- tion potential, (2) the property is at least in the feasibility study stage, and (3) there is an initial demonstrated resource sufficient for at least 5 yr of production. A tabulation that employed more liberal production, resource, or development progress criteria, and that included all explored prospects for which published information is available, could easily list more than 100 properties. These 24 operations, in total, represent approximately 1.31 million tr oz of additional annual production capacity that could potentially come on-stream between 1984 and 1990 if all 24 were indeed developed. Sixteen of the 24 opera- tions, representing approximately 1.0 million oz/yr (75 pet of the total estimated annual production potential), are con- sidered highly probable near-term producers. It is impor- tant to stress that 20 of the 24 operations would be surface producers, which implies that the United States will con- tinue to remain dependent upon surface mining for a majority of its gold production. Also, at least one-third of the operations will employ heap leaching. Total production costs for these 20 surface operations, in general, should not be significantly different from those of current surface producers. The information of tables 31 and 32 is combined in table 33 to arrive at estimates of potential total annual gold pro- duction between 1984 and 1990, assuming constant byproduct and placer gold production based upon the 1977-83 period. No attempt was made to estimate poten- tial future byproduct gold production, the majority of which Table 33. — Potential annual U.S. gold production circa 1 990, by type of operation r.,„„ r.i „„=,,.,«„ Total or range, Type of operation 1Q3 tr Q * Primary: Major producers as of January 1984 1,400 Developing, in early stage of production, or considered probable for development as of circa 1984: 1 Possible 1,310 Probable 1 ,000 Placer 2 20- 30 Base metal (byproduct gold) 2 300-400 Totals: Possible 3,140 Probable 2,700 'Restricted to properties with annual production potential of at least 20,000 tr oz that are at least in the feasibility stage, and that have a demonstrated resource sufficient for at least 5 yr of production. 2 Assumed to remain constant at 1977-83 average levels. emanates from copper mines, given the very uncertain outlook of the U.S. copper industry. Similarly, no attempt was made to estimate potential future placer production given that the percentage contribution of placer production is very low, usually around 2 pet. As shown in table 33, output by 1990 is expected to ap- proximate 2.7 million tr oz, assuming that the 1.0 million tr oz of new production considered highly probable is developed and output of current producers is maintained. If all 24 operations are developed, then total output could approximate 3.1 million tr oz. It must be remembered, however, that this would be a "best case" scenario in terms of the price of gold and the results of feasibility studies and future developments. 40 PRIMARY GOLD MINING IN THE CONTINENTAL UNITED STATES The 16 major gold producing operations in the continen- tal United States as of January 1984 can be classified into three basic types: 14 surface (bench-berm), 1 floating plant- dredging operation, and 1 underground mine. Separate discussions of the three major mining types follow. Surface Mining Of the 14 producing surface operations evaluated as of January 1984, only 2 were in production in 1968 and only 5 were in production as late as 1979. Thus, two-thirds of the major surface mines have been brought on-stream in the last 5 yr. Six of these nine new operations, and 8 of the 14 in total, utilize heap leaching of the ore, rather than con- ventional milling. In terms of ore grades, the eight current heap leaching operations are treating material grading from 1 to 5 g/mt with a straight average of 2.5 g/mt, while the six operations milling their ore are treating material rang- ing from 1.7 to 7.9 g/mt with a straight average of 5.3 g/mt. From a long-term availability perspective, only 3 of the 14 operations will be treating 100 pet millable ore over their entire productive lives, since 2 of the 5 that are currently milling their ore have at least some portion of their long- term resource planned for heap leaching. Table 34 presents combined operational data for the 14 major surface mines. Included are estimates, circa 1982-83, of mining labor levels (including prorated administrative employment), ore capacity, waste capacity, ore plus waste capacity, waste to ore ratios, mine productivity (defined as metric tons of ore plus waste per worker-shift), and estimated gold output for 1983. Table 34 demonstrates two major points concerning the surface mining of gold ores in the continental United States as of 1983. First, the evaluated mines are highly efficient, with an average pro- ductivity measure of 238 mt ore plus waste being mined per worker-shift. Second, the weighted-average recoverable grade of the ore mined, estimated to be slightly less than 3 g/mt, is relatively low. Both of these characteristics are due to four major technological and economic developments since the early to mid-1970's, as follows: 1. The availability of larger and more efficient earth- moving equipment. 2. The major increase in the price of gold. 3. The development of and refinements to the heap leaching method of gold extraction. 4. The development of and refinements to the various techniques of extracting gold by the use of carbon. It is difficult to ascertain general characteristics of the surface mining operations analyzed in this study. Stripping ratios vary between operations, and even within an opera- tion, as do the ore and the waste haulage distances. These two items, especially the stripping ratio, depend more upon the gold grade of the defined ore reserve and can vary from year to year depending upon the price of gold. An opera- tion will attempt to adjust to the type of resources that is available. As analyzed, the stripping ratios for the operations range from 1.0/1.0 to 9.8/1.0, the haulage distances for ore range from 0.3 to over 12 km, and the transport distances for waste range from 0.3 to 1.5 km. Over their estimated productive lives, the 14 surface mining operations will mine from as many as 35 different pits (deposits). Table 34. — Combined operational data for major gold producing surface mines in the continental United States Category Value Total estimated employment in mining' 1,100 Annual ore capacity 2 1 6 mt . . 14.3 Annual waste capacity 2 1 6 mt . . 51.2 Total annual ore plus waste capacity 2 10 6 mt . . 65.5 Overall waste-ore ratio, weighted average 2 3.6:1.0 Overall productivity, ore plus waste per worker-shift 3 . . mt . . 238 Estimated 1983 gold output 10 3 tr oz. . 1,350 Estimated recoverable gold grade" g/mt. . 2.94 'Includes prorated administrative labor; circa 1982-83 estimates. 2 Circa 1982-83. includes prorated administrative labor. "Circa 1983 ores. Table 35. — Comparative operational characteristics of continental U.S. surface mines, low and high mining cost levels Operational characteristics Low-cost High-cost mines mines Number of operations Ore plus waste moved 10 6 mt/yr. . 4.5 Overall mining productivity, average per shift' mt. . Operations using contract mining: Number Percent Operations employing heap leaching: Number Percent Typical excavation equipment: Operations using shovels and front-end loaders: Number Percent Operations using front-end loaders only: Number Percent Shovel size m 3 . Front-end loader size m 3 . Typical transport equipment: Average number of trucks per operation Truck size mt . . 45 Operations using crawler tractors with trailers: Number Percent 9 14.0 290 1 11 3 33 0.30 7 78 2 22 3.8-8.0 4.6-10.0 12 -77 3.8 30 5 -2.5 170 3 60 5 100 1 20 4 80 4.0 -5.3 8 -45 1 20 'Metric tons of ore plus waste moved per worker-shift; prorated administrative labor. Mine operating cost estimates, based on 1984 dollars per metric ton of ore, range from about $3.50 to $11.56. These values reflect the total cost of moving waste and ore, which is then burdened solely to the ore tonnage. For com- paring operating costs among operations, it is more valid to look at the costs on the basis of the total material moved (ore plus waste basis). When this is done, the producing sur- face gold mines in the continental United States basically split into two different classes of operations: (1) low-cost operations, which can move a ton of material for $1.06 to $1.54, and (2) high-cost operations with corresponding costs of $2.23 to $4.15. Table 35 summarizes the important characteristics that differ between the two classes. Several generalizations regarding these two classes can be made: 1 . Those operations with high costs on a ton of ore plus waste moved basis are low-tonnage, 100-pct heap leach operations with mostly contract mining (60 pet) using smaller front-end loaders (FEL's) 3.8- to 5.3-m 3 capacities and trucks (30- to 45-mt capacities). For this study, where an operation was contracting out its mining, the estimated operating costs had a 20-pct profit factored into the estimate, representing contractor profit. 2. The low-cost operations are high-tonnage operations, treating mostly milling ore (66 pet of the operations) with 41 Table 36. — Classification of producing surface pet less on a per-ton-of-material-moved basis, or about 57 operations by level of mine operating cost and grade pc t less on a per-ton-of-ore basis. This allows material of demonstrated resource grading 67 pct less to be heap-leached at a total production cost (mining plus leaching) that is only 30 pct ($29/tr oz ■ recoverable gold) higher. w . . Overall weighted-average Class of operation „~~tf™ci 9 rade of demonstrated operations' resource, g/mt Underground Mining Low-cost surface mines, L w^cost surface mines 6 3 ' 45 Underground mining in 1983 represented only about heap leach ore ' 5 1.18 15 to 20 pct of total U.S. gold production from primary ores. High-cost surface mines, As of early 1984, the only significant underground gold — eap eac : mine in the continental United States is the Homestake ipoes not include the 1 dredging operation in California. 1 operation with Mine j g th Dakota which has been in continuous pro- both millable and heap leaching ore is counted twice. , ,. „ „._ ' „ ._ , , r , duction for 108 yr. One other significant underground noncontract mining. They employ a combination of shovel- primary gold producer closed in late 1983. Two mines are FEL excavation with truck haulage using large equipment currently under development: the Ropes Mine in Michigan (4.6- to 10.0-m 3 FEL's and 45- to 77-mt trucks). and the Cannon Mine in Washington. Several other pro- 3. Contract mining appears to be favored when the spective underground primary gold deposits were in various operation plans to heap-leach the ore and the reserve for stages of exploration and study as of 1984; however, it is mining is indicated to be less than 10 yr, or where climatic felt that for the near term the Homestake Mine should con- conditions limit the mining season to 9 months of the year tinue to dominate gold production from underground or less. deposits in the United States. Table 36 summarizes the overall gold grade of the Several aspects of the Homestake mining operation are demonstrated resources for the surface mining operations. of interest. First, as of the early 1980's the lowest level of As shown, the demonstrated resources have been classified workings was about 2,450 m in vertical depth, or slightly into three categories: (1) low-cost surface mines excavating over Vh miles deep. Second, the increase in the price of gold "milling" ore, (2) low-cost surface mines moving ore for heap since the early to mid-1970's has allowed the feed grade to leaching, and (3) high-cost surface mines moving ore for be lowered significantly, which in turn has allowed the mine heap leaching. The weighted-average gold grade for each to increase its use of high- volume, low-cost mining methods of these three categories represents the overall weighted- such as blasthole open stoping and vertical crater retreat average grades for the entire demonstrated resource (VCR) stoping. For example, in 1981, 43 pct of the mine's evaluated for all of the operations in that category. As ex- ore output came from blasthole stoping and VCR stoping, pected, those operations mining ore for milling in a conven- and 57 pct came from higher cost, cut-and-fill mining. Third, tional mill have the highest overall grade at 3.45 g/mt. Of as of the early 1980's, major exploration of the deposit below more interest is a comparison of the overall grades at the the 2,450-m level was being conducted. These three aspects low-cost and high-cost heap leach operations. These grades basically define the characteristics and concerns of the mine are 1.18 and 2.21 g/mt, respectively. This difference of 1.03 as of the 1980's, which are the increasing depth of mining g/mt is significant because it represents additional revenues and the need for flexibility in mining methods appropriate of approximately $9.30/mt, assuming a gold recovery of 70 to changing economic conditions and gold price levels, pct and a gold price of $400/tr oz. Because of the few operations involved, operating costs Table 37 attempts to show how lower mine operating at underground gold mines in the United States are not costs at high-tonnage surface mines compensate for a lower discussed in detail. In general, the costs are or should be grade of material that has to be leached. As shown, the min- similar to the underground mining costs being experienced ing cost at the low-cost heap leach operations is nearly 54 at most of the Canadian mining operations. Table 37. — Effect of lower mine operating costs on the ability to mine and process low-grade material at selected producing heap leach operations in the Western United States Low mine High mine operating cost 1 operating cost 2 Range Average Range Average Mining cost per metric ton of ore plus waste $1.41 -$1.50 $1.46 $2.23-$4.15 $3.21 Mining cost per metric ton of ore leached $3.00-$3.52 s$3.26 $6.48-$9.54 "$7.57 Grade of material leached g/mt.. 1.03- 1.28 1.16 1.82- 5.00 3.52 Recovery factor pct.. NAp 0.70 NAp 0.70 Recoverable gold per metric ton of ore leached g.. NAp 0.81 NAp 2.46 Ore required to produce 1 tr oz recovered gold mt. . NAp 5 38.4 NAp 6 12.6 Mining cost per troy ounce of gold recovered NAp 7 $125 NAp 8 $96 NAp Not applicable. 'Data for 2 100-pct heap leaching operations, costs as of Jan. 1984. 2 Data for 3 100-pct heap leaching operations; costs as of Jan. 1984 include contract mining. determined as follows: $1.46/mt ore plus waste multiplied by the average stripping ratio of 2.23 for the 2 properties included. "Determined as follows: $3.21/mt ore plus waste multiplied by the average stripping ratio of 2.35 for the 3 properties included. 5 Determined as follows: 31.1035 g/tr oz divided by 0.81 g/mt = 38.4 mt required to produce 1 tr oz gold. determined as follows: 31.1035 g/tr oz divided by 2.46 g/mt = 12.6 mt required to produce 1 tr oz gold. determined as follows: 38.4 mt multiplied by $3.26/mt ore leached = $125. determined as follows: 12.6 mt multiDlied bv $7.57/mt ore leached = $96. Difference: low-cost relative to high-cost operations, pct -54 -57 -67 NAp -67 NAp 30 42 Table 38 Operational characteristic* ol producing gold mill* In tho continental United States, by type of milling method, dree 1983 NAo Not applicable Number of operations Annual ore capacity, 10 3 mt Annual gold recovery, 10 s tr oz Mill feed grades, g/mt gold Recoveries of gold, pet Range Weighted average Range Weighted average 9 10,165 7,550 50 6,400 405 924 13 22 0.95-5.5 1.63-7.9 NAp NAp 1.7 48 8.8 .14 60-85 70-94 NAp NAp 71 3 7 1 86.0 920 Dredging (gravity separation) 1 800 Total 18 24,165 1.364 NAp NAp NAp NAp Placer Mining Because of limitations on the minimum size of the gold producing operations analyzed in this study, only two ma- jor placer operations, one producing and one in study as of 1984, were evaluated. The producing operation began production in 1980. It is near Marysville, CA, and required the renovation of a dredge that had last operated in 1967. In renovation, the maximum excavation depth was increased to allow excava- tion from a depth of 42.7 m below the water level of the dredging pond. The operation also involves a large amount of overburden removal using FEL's and conveyor belts to increase the area available for dredging. The operation has been so successful that plans as of 1982 were to renovate a second dredge to operate in a nearby area. The evaluated nonproducing placer operation would have an output capacity equivalent to that of a major pro- ducer and would be somewhat unique in concept. Initially, bulldozers and FEL's will scrape overburden into trap loaders feeding conveyor belts for transport to waste or reclamation areas. With increased mining depth, an ad- vanced overburden removal system will be required to remove overburden on the extreme western edge of the mine. Following the overburden removal stage, low-grade fluvials will be mined by bulldozers and processed in skid- mounted trommels and concentrating plants. The concen- trating plants will be stationed in an interim processing water pond. As mining progresses, these concentrating plants will be positioned on benches. In a third stage of mining, high-grade gravels will be excavated with a backhoe, a hydraulic breaker, a ripper mounted on a backhoe, and an FEL. When the lowest bench is formed it will be approximately 12.2 m above bedrock, and subsequent excavations from this bench will be below the water table. These excavations will fill with water to form a pond capable of floating a barge equipped to process the high-grade gravels. These three mining stages will be initiated at staggered intervals so that each is in advance of the other. In the concentrating process, trommels will wash and size low-grade gravel to less than 9.525 mm; high-grade gravel will be sized to less than 19.05 mm. A desliming tank will separate the heavy metals from clay. Screens and pulsating jigs will produce a gold concentrate, which will flow to the amalgamation circuit. Amalgam (a mixture of mercury, gold, and black sand) will be retorted in an air- tight metal tank; retorted gold will then be melted in a fur- nace and formed into bullion. When both these placer operations have attained their full capacity, as proposed circa 1982-83, they should account for about 100,000 tr oz combined annual production. However, based on the demonstrated resource analyzed in this study, the operations have mine lives extending only to 1992-95. The economics of these large placer operations appear to be favorable at the reported grades of the material being mined or planned for mining. Weighted-average grades range from 0.1 to 0.5 g/mt. Long-run production costs are estimated at $300/tr oz to $350/tr oz recovered gold. It must be remembered that placer operations this large are always high-profile operations in terms of environmental aspects. Thus, even though the economics of these opera- tions appear favorable, a "boom" of similar developments should not be expected. GOLD MILLING IN THE CONTINENTAL UNITED STATES The 16 major gold producing operations in the continen- tal United States as of 1983 included 8 operations that were using heap leaching to produce gold bullion, 6 that were using conventional crush-grind-vat leach methods to recover gold, 1 that was using a combination of heap leaching and conventional vat leaching, and 1 that was using a dredge and gravity methods for gold recovery. An additional opera- tion of interest due to its beneficiation technique was one that was producing a bulk flotation concentrate product and then leaching the tails from flotation to produce dore bullion. Table 38 summarizes the pertinent estimated opera- tional data covering the four basic types of milling opera- tions. The Carlin operation in Nevada, which utilizes a com- bination of heap leach and conventional milling, is ac- counted for in table 38 as one operation in each of two categories. The data in the table represent circa 1983 opera- tional data and do not necessarily reflect expected or pro- posed changes to the capacities and milling methods that have been evaluated in this study when addressing the economics of these operations over their total productive lives. The operation that was floating its ore and leaching the tailings closed down in late 1983 and has only an estimated 3-yr reserve remaining. This operation was not included in the cost analyses but is briefly discussed from a comparative technical perspective. The two predominant methods in use, heap leaching and conventional vat leaching, are discussed in detail. Heap Leaching (Solution Mining) Heap leaching of gold ores is a recent development in the U.S. gold mining industry. A correct term for this method would be "solution mining" of prepared ore dumps with cyanide solutions. The method was developed to economically treat low- grade gold ores (primarily less than 3 g/mt) through much lower capital requirements and lower operating costs per ton of ore treated in the milling stage. It is a particularly 43 good method when treating disseminated oxide ores of the type discovered in the 1960's and 1970's in the Western United States. In most cases, the material to be leached is first delivered to a crusher. The crushed ore is then delivered by truck to specially prepared leach pads. A cyanide solu- tion of appropriate strength and pH is then evenly distributed on the heap through sprinklers and allowed to percolate through. The gold particles dissolve in the cyanide solution, which is collected in sumps and pumped to the gold recovery plant. The gold is extracted from the solution by one of two methods, the carbon-in-pulp (CIP) method or the Merrill-Crowe deaeration-zinc dust precipitation method. In the first case, the gold in solution is adsorbed onto granules of carbon which are then stripped of the "loaded" gold by a hot NaOH solution. This solution is in turn fed to electrowinning cells where the gold deposits onto cathodes of steel wool. In the second method, the product is a zinc-dust precipitate, which is filtered. The cathodes (in CIP) or the precipitate (in Merrill-Crowe) are then melted in crucible furnaces along with fluxing materials such as borax, niter, and silica. The resultant product from the smelting is a dore bullion of precious metals grading anywhere from 35 pet gold and 55 pet silver to 96 pet gold and 3 pet silver. In practice, there are many variations to the "normal" description of heap leaching given above, and subsequent discussion will attempt to describe some of the variations being practiced as of 1983 in the United States. The economics of a heap leaching operation are very dependent upon the gold recovery that is obtainable. Because the ore grades are so low (less than 1 g/mt leached in some cases) it is difficult to exercise effective grade con- trol. Thus, heap leaching operations can experience grade variations from month to month or year to year of as much as 25 to 50 pet, and the operations have to have a built-in flexibility to be able to increase tonnage and reduce operating costs through economies of scale. Location is also important for two reasons. First, one of the major advantages to the heap leaching method is the low capital cost required for the gold recovery operation. Thus, if the proposed mine is in a remote location with at- tendant high cost requirements for infrastructural in- vestments, this capital cost advantage could be negated. Se- cond, heap leaching works best in dry or arid and temperate to hot climates. Cold temperatures negatively affect nor- mal heap leaching operations because the solubility of gold decreases greatly below 50° F and because the leaching solution could freeze. Wet climates, especially those prone to intense rain storms, pose a problem in that sudden deluge would rapidly dilute the leaching solution and could cause the collection ponds to overflow. As to the mineralogy of the ore itself, one principle is of paramount importance: The best recovery of gold will be obtained where the most contact can be made by the leaching solution on the majority of those gold particles amenable to dissolution. In this regard, high clay contents in the ore are undesirable because clay causes uneven per- colation of the leaching solution. High amounts of sulfide minerals in the ore are also undesirable because of preferen- tial leaching of base metals such as copper, lead, and zinc and because some of the gold particles will be "locked-up" in sulfide minerals. Carbonaceous ores (ores containing a high amount of organic carbon) are difficult because the car- bon particles in the ore preferentially absorb the gold before it can be recovered in the leach solution. Oxidized ores are favorable for heap leaching because the solubility of gold is improved by higher oxygen contents during the solution stage. In addition, oxide ores usually contain more gold par- ticles in the "native" form, and base metals are not pres- ent in large amounts. Another favorable type of heap leaching ore is material where the gold mineralization is disseminated throughout the ore rather than clustered in specific zones because percolation will contact more of the gold particles. All of these mineralogic factors have a ma- jor effect on the overall recovery of gold in the heap leaching process. As shown in table 38, nine major producing heap leaching operations were analyzed for this study. Five of the operations are located in Nevada; New Mexico, Col- orado, Montana, and Idaho each have one major heap leaching operation included in the analysis. One of these operations, a small producer in Colorado, is an anomaly in attempting to categorize it as either a heap leach or a vat leach milling operation. This operation incorporates aspects of both methods in that its ore is placed into large vats enclosed in a building because of climatic conditions and then is subjected to an intense cyanide leach. Another of the nine heap leaching operations is also anomalous in com- parison with the others in that this operation simply places run-of-mine ore (no crushing and no agglomeration) on the leach pad, builds berms around the top of the pad, and develops a pond of cyanide leach solution on top of the ore heap rather than using sprinklers. This operation also does not remove the leached ore from the pad at the end of the leaching cycle, choosing instead to construct a new pad for subsequent leaching. It is estimated that the nine heap leach operations ac- counted for 405,000 tr oz of gold production in 1983, which represents nearly 30 pet of total primary gold production from the 16 major producers evaluated in this study. This 1983 production level is an eight-fold increase over the reported 1979 production from solution mining of gold ores, which was slightly more than 50,000 oz of gold. This eight- fold increase is easily explained when it is noted that seven of the nine operations have come on-line since late 1979. The ore grades ranged from 0.95 to about 5.5 g/mt, with a straight average grade of 2.3 g/mt and a weighted-average grade of 1.7 g/mt of leached material. Estimated recoveries of gold at the nine heap leaching operations range from 60 to 85 pet with a straight average of 75 pet and a weighted- average recovery of 71 pet. The range of recoveries from 60 to 85 pet defines fairly well the two extremes to be ex- pected in the heap leaching of predominately oxide ores. It is difficult to summarize the nine major heap leaching operations in production in the United States during 1983. At least one major operational characteristic varies among the nine operations. The most common methods, along with the number of operations utilizing them, are listed in table 39. Somewhat surprisingly, as of 1983, only four of the nine operations were agglomerating their crushed ore prior to placement into heaps. Agglomeration of the ore with water and binders, such as cement, "balls" the ore into particles of similar shape and size, allowing a more even distribu- Table 39. — Number of U.S. heap leach operations utilizing various mining and processing methods Crushing of ore prior to leaching 7 Agglomeration prior to placement into heaps 4 Distribution of solution with sprinklers 8 Extraction of gold using the carbon-in-pulp method 6 pH control with lime 7 Distribution of ore and building of heaps with trucks and bulldozers 7 Rehandling of leached material 8 Asphalt base for leach pads 6 44 tion of the percolating solution within the entire heap. It is especially recommended where the ore contains a fair amount of clay minerals or fines material since, in both cases, segregation due to placement will cause impervious layers to develop within the heap and the cyanide solution will be diverted along these impervious boundaries. Not surprisingly, six of the operations were using the carbon-in-pulp method of extracting the gold from solution. The three operations that were using the conventional Merrill-Crowe system of extraction were all producing dore bullions which contained more silver than gold and have appreciable silver contents in the ore, in which case car- bon methods are not preferred. The majority of the operations control the pH level of the cyanide solution with lime rather than caustic soda (NaOH), and must distribute and build their heaps with trucks and bulldozers. The two anomalies regarding heap building are an operation that uses a traveling gantry for distribution onto the heap and an operation that does its heap leaching in large vats in an enclosed building where a front-end loader distributes the ore. The most variable production aspects among the nine operations occur in four areas: (1) the leach cycle time, (2) the number and sizes of the leaching pads, (3) the configura- tions of the ore heaps, and (4) the number of months in a year that leaching is conducted. Leach cycles (including leaching and the removal and treatment of leached material) range from 4 days to 3 yr; the majority are in the 1- to 3-month range. The storage capacity of permanent leach pads ranged from 20,000 to 1.0 million mt of ore. The configurations of heaps range from "pyramidal" to "furrowed." In terms of the leaching season, six of the nine operations were limited to 7 to 9 months of the year, while three were conducting their leaching virtually year round. Of these three year-round operations, only two could do this because of natural climatic conditions; the third operation was heating its leach solution to allow full leaching for 9 months of the year and part-time leaching for the other 3 months. However, heating the cyanide solution causes ap- preciable loss of cyanide. The estimated operating costs for the heap leaching stage at the nine operations range from $2.39/mt to $10.52/mt ore. These costs represent the costs incurred from the point of ore delivery to the crusher or to the heaps, depending upon the particular situation, and include the costs of rehandling the leached material at the end of the leaching cycle, which in some cases is done by the mining contractor. The two highest estimated operating costs of $8.75 and $10.52 reflect one operation with higher than nor- mal labor costs and one operation with extremely high cyanide consumptions. Thus, the more normal range of heap leach operating costs, on a per ton of ore basis, is $2.39 to $7.37, a range that encompasses seven operations. This nor- mal range, in turn, appears to split into two separate classes: a low-cost class in the range of $2.39/mt to $4.05/mt, and a high-cost class in the range of $4.86/mt to $7.37/mt. A comparison of various characteristics for these two classes is shown in table 40. The items listed are felt to be the ma- jor determinants of whether a particular heap leaching operation will be in the low-cost or high-cost class. As shown, labor productivities average nearly 48 pet higher at the low-cost operations, which is mostly a reflec- tion of the larger pad sizes. The low-cost leaching opera- tions also show an advantage in the high proportion that do not have to crush and/or agglomerate their ore and that can operate year- round. Table 40. — Labor productivity and operational characteristics for low-cost and high-cost heap leaching operations in the continental United States Low-cost High-cost heap leaching heap leaching Number of operations 3 4 Annual ore capacity: Range 10* mt/yr. . 344-3,265 475-780 Average 10 3 mt/yr. . 2,290 550 Pad capacities 10 3 mt. . 250-1,000 20-136 Labor productivity 1 per worker-shift, mt: Range 80-92 42-68 Average 87 59 Percentage of operations using— Crushing 33 100 Agglomeration 75 Waste removal 67 1 00 Year-round operation 67 25 1 1ncludes prorated administration labor In the normal range of operating cost levels for heap leaching, the costs for direct and indirect labor constitute the largest single item, ranging from about 30 to 38 pet of the total for the low-cost class and 28 to 38 pet for the high- cost class, with both classes averaging about 33 pet. The miscellaneous cost category, including maintenance materials, insurance, miscellaneous supplies, and water, represents the second largest portion of the total operating cost, with the low-cost operations averaging 31 pet and the high-cost operations averaging 33 pet. Reagent costs are the third most important cost category, averaging about 18 pet at the low-cost operations and about 21 pet at the high-cost operations. Electricity costs and energy costs (gasoline, heating oil, and propane) constitute the smallest portion of total operating costs in both classes at 18 pet on average for the low-cost class and 13 pet on average for the high- cost class. The following items deal with some generalizations con- cerning the costs of specific operations that are intended to give the reader a rough idea of the cost levels associated with these tasks. First, for operations of smaller capacity and in the high-cost class, crushing costs appear to range from about $1.25/mt to $2.20/mt ore. Rehandling of waste range from about $0.35/mt to $1.72/mt ore. Second, one gold operation reported a cost for constructing an asphalt-based leach pad of slightly more than $15 per square meter in 1982. Third, in all cases, cyanide costs represent the first or second largest item in the costs for reagents, while costs for cement, lime, soda ash, carbon, and nitric acid vary among the top three individual reagent costs from opera- tion to operation. To summarize, heap leaching is a relatively new method of recovering gold and could be considered as still in the development stage because the operational characteristics make it an ideal method of experimentation. With low capital and operating costs, heap leaching has enabled gold ore of very low grade to become competitive with much higher grade gold deposits. The method, however, requires high efficiency in the mining and milling operations, along with a high degree of built-in flexibility to the operation. Also, at present gold prices, the method is limited to cer- tain types of ores and is subject to infrastructural and climatic constraints. In addition, the gold grades of the material leached and recoveries experienced can be ex- tremely variable, and much attention should be paid to in- itial metallurgical tests of the ore before proceeding with a heap leaching operation. A recent Bureau publication (24) discusses heap leaching technology, lists 118 leaching opera- tions, gives details on 26 key operations, covers Federal and 45 Table 41. — Comparative economics of producing heap leach and conventional milling operations in the continental United States Entirely surface Entirely surface mining and mining and heap leaching conventional milling operations operations Number of operations Total capital investment remaining 1 Capital cost per troy ounce: Range Weighted average Total (mining plus processing) operating cost per troy ounce: Range Weighted average Total operating plus capital cost per troy ounce: Range Weighted average Break-even (0-pct DCFROR): Range Weighted average 1 0-pct DCFROR: Range Weighted average Producing years from January 1984 8 7 operations less than $20 million; 5 operations less than $10 million. 3 operations greater than $60 million; 4 operations greater than $40 million. 'Unrecovered capital investment in mine and mill through life of operation. $26 to $126 $21 to $82. $47 $46. $117 to $401 $146 to $332. $288 $247. $152 to $527 $182 to $397 $335 $214. $173 to $538 $187 to $428. $349 $268. $180 to $567 $210 to $489. $360 $304. 4 operations, 3 to 6 yr; 1 operation, 6 yr; 3 operations, 7 to 10 yr; 4 operations, 12 to 15 yr. 1 operation, 51 yr. plant and equipment, infrastructure, and development remaining as of Jan. 1984 plus reinvestments State requirements, and provides an extensive bibliography of over 160 references on gold and silver heap leaching. In terms of annual output, the nine heap leaching opera- tions evaluated in this study range from 6,500 to 83,000 tr oz of gold production, with an average annual output of 45,000 tr oz per operation. This compares with annual gold outputs ranging from 74,000 to nearly 200,000 tr oz/yr at the seven evaluated conventional vat mills with an average output of 132,000 tr oz/yr. In addition, the mine lives of the heap leaching operations tend to be somewhat less than those of conventional milling operations. Thus, probably at least three average heap leaching operations have to be developed in order to replace the annual gold production capacity at one average conventional milling operation in the United States. Comparative Economics of Heap Leaching and Conventional Milling The economics of heap leaching differ from those of con- ventional milling in four significant ways: 1. Total initial capital investments in plant and equip- ment, infrastructure, and development required to bring the operation into production are lower. 2. Total capital reinvestments over the mine life are lower. 3. The preproduction period required to bring the opera- tion into production is shorter. 4. The payback period, due to significantly lower capital investments, is shorter. Table 41 presents comparative data on eight producing surface operations employing 100 pet heap leaching and five producing surface operations employing 100 pet conven- tional milling. The most significant economic difference bet- ween the two is in required total capital investments over the mine life (the sum of items 1 and 2 above). Seven of the eight evaluated heap leach operations have total estimated capital investments, remaining as of January 1984, of less than $20 million, and five of the eight have total capital investments of less than $10 million. By contrast, four of the five conventional milling operations have total required capital investments exceeding $40 million, and three opera- tions have total capital investments exceeding $60 million. The barrier to market entry, therefore, into gold produc- tion from heap leaching of surface material is much lower than for conventional milling operations. The relatively low total capital costs required for the development and opera- tion of heap leaching deposits thus render them viable in- vestment opportunities for small mining companies with low capitalization. This is the primary reason for the in- crease in the number of heap leaching operations that have come into production since 1979. Conventional milling operations are generally larger in terms of annual output, require greater investments in facilities such as mill plant and equipment, and require higher capital reinvestments over the mine life. But since the grade of the ore is higher, total ore resources greater, and recoveries better, these conventional milling operations have essentially the same capital costs per ounce of recoverable gold. Total operating costs per ounce of recoverable gold (mining plus milling or leaching cost) significantly favor conventional milling, on a weighted- average basis. This is not only because of higher grade ore and higher milling recoveries but also because of higher annual production levels. Break-even, long-term total production costs per ounce of recovered gold for the eight evaluated heap leach opera- tions range from $173 to $538, with a weighted average of $349. Seven of the eight operations fall in a range from $173 to $425. The increase in gold prices in recent years and the development of the heap leaching process have served to broaden the market in terms of the sources of gold produc- tion in the Western United States. On average, the five con- ventional milling operations have lower total production costs than the heap leaching operations, thus demonstrating economies from larger scale production and higher grade ore as an offsetting factor to higher total capital investments. 46 As analyzed in this study, the heap leach operations, on average, have shorter mine lives than the conventional milling operations. Seven of the eight heap leaching opera- tions have estimated remaining lives of less than 11 yr, and four operations have remaining lives of only 3 to 6 yr. By contrast, four of the five operations employing conventional milling have remaining productive lives of between 12 and 15 yr. Vat Leaching Of Gold Ores During 1983, eight of the major primary gold produc- ing operations analyzed in this study were using conven- tional milling techniques (vat leaching) to treat at least some of their ore production. Conventional milling, or vat leaching, as defined for this study, is any mill that crushes and grinds its ore and performs cyanide leaching in agita- tion vats. Seven of the eight mills produce precious metal (dore) bullion by crushing, grinding, leaching, and gold ex- traction with various methods plus smelting. The remain- ing mill was producing a bulk flotation concentrate contain- ing most of the gold and silver and then leaching the tail- ings from flotation to produce a small amount of dore bullion; this operation has since closed down. In terms of geographic location, four of the eight mills are located in Nevada with one mill each located in Utah, Montana, South Dakota, and Washington. The oldest of the mills was con- structed in 1937, another basically dates from 1952-53 when a major renovation was made, one dates from the mid-1960's, and one was originally constructed as a copper mill in the late 1960's and then converted to a primary gold mill in 1978-79. The remaining four mills are all relatively new, two being commissioned in 1981 and two brought in- to production in 1983. Table 42. — Comparative characteristics of major vat leaching mills in the continental United States, circa 1983 Mills less than Mills more than Total or 5 yr old 10 yr old average Number of mills 5 Combined annual ore capacity 10 3 mt. 5.17 Combined annual gold production capacity 10 3 tr oz. 518 Weighted-average feed grade g/mt. 3.7 Weighted-average mill recovery pet. 84 2 2.38 419 6.1 93 7 7.55 937 4.4 87 If the copper mill that was converted to primary gold milling is included as a "new" mill, then the five mills brought on-line in 1979-83 represent sufficient capacity as of late 1983 to treat slightly more than 5.2 million mt/yr ore to produce nearly 520,000 tr oz/yr gold. The three older mills, as of late 1983, had the capacity to treat about 2.4 million mt/yr ore to produce nearly 420,000 tr oz gold. Com- bined, the eight major mills were probably producing about 940,000 tr oz gold in 1983, or more than two-thirds of total U.S. primary gold production. Since the mill in Washington State that was producing a bulk flotation concentrate and dore bullion was shut down late in 1983 and has a limited (2- to 3-yr) reserve of ore, the following discussion will concentrate on the seven ma- jor vat leaching mills currently in operation. Table 42 summarizes the operational characteristics of the two groups of conventional gold mills: those less than 5 yr old and those more than 10 yr old. The five newer mills have productive capacities ranging from 445,000 to 1.7 million mt/yr ore to produce between 73,000 and 196,000 tr oz/yr gold. The two older mills range from 700,000 to 1.7 million mt/yr ore milling capacity to produce from 128,000 to 279,000 tr oz/yr gold. Of particular interest in table 42 are the lower feed grades and lower recoveries at the newer mills compared to the older mills. The lower feed grades at the five newer mills reflect the increase in the price of gold since the mid-1970's, which enabled lower grade deposits to be brought into production. Even though the two older mills were treating ore in 1983 that graded 50 pet higher than the newer mills' ore, they have also lowered their feed grades in response to the gold price increase. For example, in 1974 these two mills produced about 530,000 tr oz gold from 2.2 million mt ore, whereas in 1983 the two mills pro- duced about 420,000 tr oz gold from 2.4 million mt ore, a decrease of 29 pet in the weighted-average feed grade and a 21-pct decrease in overall gold production. Yet, using average gold prices of $160/tr oz for 1974 and $425/tr oz for 1983, the indicated 1983 revenue from these two mills' pro- duction is 214 pet higher than 1974's estimated revenue. Even if the estimated revenues for 1974 and 1983 are com- pared on a constant-dollar basis, the estimated revenue for 1983 is still 15 pet higher than 1974's estimated revenue despite a 20- to 30-pct decrease in feed grades and production. As in Canada and Australia, it is difficult to categorize the major vat leaching gold mills in the continental United States because the ores all have slightly different characteristics. Table 43 shows the variations in practices Table 43. — Variations of major circuit practices at vat leaching gold mills in the continental United States, early 1980's Major circuit Ore blending Comminution Coarse free gold recovery . Special treatments Extraction of gold from cyanide solution Carbon reactivation Special byproducts Method or practice Carbonaceous with noncarbonaceous ores Single-stage crushing; grind with SAG mill and ball mill 2- or 3-stage crushing; grind with rod and/or ball mill Launder traps in grinding circuit Jigs and tables with grinding circuit Separation into major sand and slimes fractions Oxidation of pyrrhotite with air Oxidation of carbonaceous ore with chlorine gas Carbon-in-leach (CIL) with electrowinning onto steel wool Carbon-in-pulp (CIP) with electrowinning onto steel wool Carbon-in-pulp (CIP) with zinc dust precipitation Merrill-Crowe deaeration-clarification with zinc dust precipitation In gas-fired kilns In kilns or with nitric acid wash Mercury production by retorting steel wool or filter cake Number of mills using method or practice 47 at the seven major vat leaching gold mills as of the early 1980's. The categorizations shown point out six major aspects of interest. First, only two of the seven mills include some provi- sion for recovery of coarse, free gold in their circuits. One mill recovers slightly more than 20 pet of its total recoverable gold with the use of launder gold traps in its grinding circuit, while the other mill recovers from 10 to 20 pet of its gold by using jigs and tables to treat the cyclone underflow from the ball milling circuit. In total, only about 7 pet of the annual gold recovered at the seven milling operations represents coarse, free gold recoverable by gravity methods. Second, carbon methods for extracting gold from the cyanide leach solution are used in six of the eight separate extraction circuits involved. The two extraction circuits that use the older Merrill-Crowe deaeration-clarification method are mills that are more than 10 yr old; these mills have ob- viously found no reason to change the original extraction method at their older circuits. However, 100 pet of the ex- traction circuits constructed at major vat leaching mills in the continental United States within the last 10 to 11 yr have been carbon method circuits. Third, all six circuits using carbon methods for extract- ing gold from cyanide solution reactivate the stripped car- bon particles in gas-fired or indirectly fired kilns. At one operation, the stripped carbon is sometimes reactivated by using a nitric acid wash. Fourth, mercury is produced at three of the seven mill- ing operations. The mercury is present in the steel wool cathodes from electrowinning or in the filter cake from zinc dust precipitation. At those operations producing mercury for sale, the cathodes or filter cake are smelted and the vaporized mercury is condensed to liquid form for sale in flasks. The residue from this first smelt is then resmelted in the presence of fluxes such as niter, borax, and silica to produce the final dore bullion product containing the precious metals. The amount of mercury produced is not large, and revenues from mercury are miniscule compared to those from gold and silver.* The three operations that produce mercury as a byproduct are the same three that treat carbonaceous ore as a portion of the total ore feed. Fifth, one of the seven vat leaching operations makes a major split of its ground ore into a sand fraction (about 60 pet of the total tonnage treated) and a slimes fraction (about 40 pet of the total tonnage treated). The sand frac- tion is leached in vats with gold extraction by the Merrill- Crowe-zinc dust precipitation method, while the slimes frac- tion is pumped 4.8 km to a vat leaching plant which uses carbon-in-pulp absorption, stripping with a hot caustic solu- tion, and electrowinning of the gold onto steel wool. Sixth, and possibly of most interest, is that three of the seven operations must contend with carbonaceous feed of a refractory nature. Material referred to as "refractory" is not rigorously defined; basically it is any material with a high organic carbon content in proximity to the gold par- ticles and/or a fair amount of the gold "locked up" in pyrite that results in very low gold recoveries when treating the ore with normal cyanide leaching techniques. To obtain reasonable (over 70 to 80 pet) gold recoveries with this type of ore, it is necessary that the ore be oxidized prior to the cyanide leach. *Only two of the three operations have reported their level of mercury pro- duction. The values given approximated 50 and 250 flasks of mercury per year. Of the six major points discussed above, it is appropriate that two of these major points should be expanded upon in separate discussions. These two items, the predominant use of carbon extraction methods and the handling and treat- ment of carbonaceous (refractory) gold ores, are the two most important developments in the vat leaching of gold ores in the United States since the late 1960's. Interestingly, both of these developments were heavily dependent upon joint research efforts of major U.S. gold mining companies and Bureau of Mines metallurgical research facilities. Extraction of Gold With Activated Carbon The first major attempts to utilize the ability of ac- tivated charcoal or carbon to absorb complex metal ions in the processing of gold ores occurred in the 1940's and 1950's. The Golden Cycle mill in Cripple Creek, CO, the Getchell mine near Golconda, NV, and the Idria Mine in the coun- try of Honduras are cited as three of the earliest plants to utilize the carbon-in-pulp process. These earlier plants were recovering the gold-loaded carbon from the slurry by screen- ing or flotation and then recovering the gold from the car- bon by burning the carbon or sending the gold-loaded car- bon to a smelter. Unfortunately, the fixed gold price of the 1950's and 1960's caused these three operations to close, and the CIP process was temporarily forgotten (25, p. 95). Three developments in the late 1960's and early 1970's caused a resurgence of interest in the CIP method. First, the price of gold was allowed to find its own level, raising the possibility of higher prices in the future. Second, the Bureau of Mines Research Center at Reno, NV, developed a hydrometallurgical method to strip the gold from the car- bon particles using a hot, caustic-cyanide washing solution under elevated pressures. The resulting solution contain- ing the stripped gold was then sent to a specially developed electrolytic cell (the Zadra cell) for electrowinning the gold onto cathodes of steel wool. Third, the largest gold producer in the United States, the Homestake operation in South Dakota, was attempting to find a solution to increased labor costs at its slimes leaching plant, where the requirement to filter-press 1,800 mt/d of slimes was becoming burden- some (25, p. 95). As a result, Homestake Mining Co. and the Bureau of Mines Reno Research Center conducted a joint pilot plant operation at Lead, SD, in 1971. The results were favorable, and a full-scale 2,177-mt/d CIP plant was constructed in 1972-73. This plant was the first major CIP plant con- structed in the United States to use the new caustic solu- tion washing-electrowinning procedures. A second CEP plant was constructed by Homestake Mining Co. at Creede, CO, in 1975-76, and a third major plant came on-stream in 1979. This plant was at Duval Corp.'s Battle Mountain operation, where a copper flotation mill was converted to a CIP gold ore processing plant. In 1980-83, four other major vat leaching plants using carbon methods to extract the gold came on-stream. As practiced currently in the CIP method, carefully sized particles of activated carbon are contacted with the pulp (leached ore plus cyanide solution) from the vat leaching stage. This contact takes place in stages in a series of tanks. The barren activated carbon is introduced to the last tank in the series and is advanced countercurrently to the pulp flow. Thus, the most "unused" carbon is contacted with the pulp that should have the least amount of gold available for absorption onto the activated carbon. 48 The gold-loaded carbon particles, suspended in a slurry, are then screened from the slurry and washed in a hot caustic-cyanide solution, usually under elevated pressures, to extract the gold from the carbon particles. The gold in the wash solution can be recovered either by electrowin- ning onto steel wool cathodes or by precipitation with zinc dust. The stripped carbon can be reused in the process after reactivation. This is usually accomplished by heating the spent carbon in a kiln, but washing the spent carbon with nitric acid will also reactivate it. The above is a generalized description of the CIP method. Needless to say, there are technical variations in each plant in operation. A second, more recent, carbon method is called the carbon-in-leach (CIL) method. In this method, the carbon particles are contacted with the pulp in the vat leaching stage itself, rather than after the leaching stage. This method is usually employed when the ore to be fed to the leaching stage contains a fair amount of organic carbon; the idea is to mitigate the deleterious effects on gold recovery caused by organic carbon in the ore. This is ac- complished by introducing the activated carbon into the leaching stage where the deleterious effects occur. As shown in table 43, of the six "gold extraction with carbon" circuits at major vat leaching mills in the continen- tal United States, five use the CIP method and one uses the CIL method. Also, five of the six circuits use electrowin- ning onto steel wool cathodes, while one circuit uses zinc dust precipitation of the gold in the caustic-cyanide solu- tion. Also, as noted previously, all six of the circuits reac- tivate the stripped carbon by heating it in kilns, with one of the circuits sometimes using a nitric acid wash. The advantages and disadvantages of carbon methods of extraction versus the conventional Merrill-Crowe deaeration-clarification and zinc dust precipitation method of extraction are still being debated and involve many technical questions beyond the scope of this analysis. Original claims were that both capital costs for equipment in the carbon extraction circuits and operating costs are lower, mostly owing to lower labor requirements for the ex- traction circuit. In cases where all of the other factors con- tributing to overall capital costs and overall operating costs are equal, these claims will be true. The predominance of the use of carbon methods in the United States appears to be due to three major factors: 1. Labor requirements and maintenance and/or replace- ment costs are lower with carbon methods. This is in keep- ing with the stress placed on efficiency at all of the newer U.S. gold milling operations. 2. New mill circuits (constructed within the last 10 yr) are predominant in the United States. 3. The technology of carbon methods is probably most advanced in the United States. Carbonaceous Gold Ores Carbonaceous ores, those containing a high amount of organic carbon, are fairly common in the Southwestern United States. Because the organic carbon particles in this type of ore will adsorb gold that is present in the cyanide solution ("preg robbing"), recoveries of gold are very poor without some type of treatment to destroy as much of the organic carbon as possible before the cyanide leaching stage. There are no hard-and-fast rules as to the effect that the organic carbon content in a particular ore will have on the overall gold recovery without special treatment. 8 However, the general relationship is probably close to the experiences of the Carlin operation in Nevada, which was recovering only slightly more than 34 pet of the gold when milling its carbonaceous ore without special treatment versus as much as 83 pet of the gold when using an oxidation with chlorine gas treatment on the same type of ore (26, pp. 103-104). Three of the seven vat leaching mills analyzed in this study have had to contend with carbonaceous ores. Two of the three operations have developed the preoxidation- chlorinoxidation method (also referred to as the "double- oxidation" method) of treating carbonaceous ores. The third operation, which is much lower grade than the other two, decided that oxidation of its carbonaceous ore was too cost- ly and chose instead to blend the carbonaceous ore with its regular noncarbonaceous ore in a ratio of about 1:9, thereby attempting to control the effect of the carbonaceous ore on lower gold recovery. As a result of this decision this opera- tion has the lowest overall gold recovery of the seven mills analyzed, with 70 pet being considered as good as can be expected with that particular blend of the two ore types. The chloric oxidation method of treating carbonaceous ores was initially investigated beginning in 1967 with a joint research program between the Carlin Gold Mining Co. and the U.S. Bureau of Mines Metallurgical Research Center at Reno, NV. After extensive bench-scale testing, a pilot plant was constructed in 1969 to test the use of either chlorine or sodium hypochlorite as the oxidizing agent (26). The chlorine gas method was selected because generating sodium hypochlorite by electrolysis of the pulp involved high capital costs which were not justified by the size of the carbonaceous reserve involved. The chlorine oxidation cir- cuit began treating ore on a fairly large scale at Carlin in 1971-72. In 1977, a preoxidation with air stage was added prior to the chlorine oxidation stage to oxidize some of the pyrite and pyrrhotite prior to the main oxidation stage. The second U.S. mill to utilize chlorine oxidation was con- structed during 1979-81 and basically uses the same methodology as at Carlin. At one of the two operations that treat carbonaceous ore by oxidation, 20 pet of the total annual ore milling capaci- ty represents carbonaceous ore; at the other mill, fully 50 pet of the total annual ore capacity is for carbonaceous ore. Recoveries being experienced in the 1980's ranged from 83 to 87 pet of the gold in the carbonaceous ore at the two opera- tions using the preoxidation-chlorine oxidation method. A brief description of the process follows. First, when treating carbonaceous ore, the two basic types of ore, car- bonaceous and noncarbonaceous, are handled separately in the crushing and grinding stage. The ground carbonaceous ore is slurried, and the pulp is heated using steam to 100° F at one operation and to 180 "Fat the other. The heated pulp is then sent to agitation tanks, where air is mixed with the slurry to oxidize as much of the pyrite, pyrrhotite, and other sulfide minerals as possible prior to the chlorine ox- idation stage. The preoxidized pulp is then sent to another series of agitation tanks, where chlorine gas and air are introduced to oxidize the remaining sulfides and the organic carbon particles. The method is expensive. For example, this study estimates that the "double-oxidation" method using "Work by the U.S. Bureau of Mines and the U.S. Geological Survey as of the early 1970's had classified ores containing 0.25 to 0.8 pet organic car- bon as carbonaceous ore and those containing 0.06 to 0.25 pet organic car- bon as noncarbonaceous. However, as noted by Guay and Peterson (26, p. 103) the organic carbon assay was never found to be a highly useful tool in the metallurgical development work at the Carlin, NV, mill. 49 chlorine adds from $5.30/mt to $6.50/mt ore milled to the normal operating costs expected for straightforward cyanidation vat leaching, with the additional costs repre- senting mostly chlorine consumption (13.6 to 22.7 kg/mt of ore milled) and steam generation for heating. Thus, it is not surprising that the two operations using the double- oxidation method with chlorine gas also are the two highest grade operations of the seven vat leaching mills studied, with average feed grades of about 6.5 g/mt to nearly 8.0 g/mt. A rough measure of the effect that this additional operating cost can have is that with a carbonaceous ore grading 6 g/mt, the added operating cost of $6/mt ore would represent an additional cost of $37/tr oz recovered gold, assuming an 85-pct recovery of gold with the process. As summarized in table 42, the total combined annual ore capacity for the two vat leaching operations constructed prior to 1974 is 2.38 million mt/yr ore to produce 420,000 tr oz/yr gold. Individual feed rates as of the early 1980's for the two operations were 700,000 and 1.7 million mt/yr with feed grades of 5.5 and 6.3 g/mt, respectively, or 6.1 g/mt on a weighted-average basis. Total mill recoveries averaged 90 and 94 pet, respectively (93 pet on a weighted-average basis). In comparison, the total combined ore capacity for the five newer mills constructed since 1978 is 5.17 million mt/yr ore to produce about 518,000 tr oz/yr gold. Individual ore capacities at these newer mills range from 445,000 to 1.7 million mt/yr and average 1.04 million mt/yr. Individual outputs of gold at the newer mills range from 74,000 to 196,000 tr oz/yr gold for an average output of 104,000 tr oz/yr gold. Feed grades at the newer mills range from 1.6 to 7.9 g/mt with a weighted-average feed grade of only 3.7 g/mt, nearly 35 pet lower than the comparable value for the two older mills. Similarly, overall gold recoveries at the newer mills range from 70 to 90 pet with a weighted average of 84 pet, nearly 9 pet lower than the comparable value for the two older mills. The lower weighted-average gold recoveries at the newer mills reflect two factors. First, overall recovery of gold decreases as lower grade ores are processed, especially ores below 2.5 g/mt. Second, one of the five newer mills is processing a certain amount of unoxi- dized carbonaceous ore. All seven of the vat leaching operations produce dore bullion as the final product. The dore bullions range in grade from 47 to 96 pet gold and have gold-silver ratios rang- ing from about 1:1 to as high as 48:1. The estimated mill operating costs at the seven vat leaching operations range from $9.44/mt to $21.82/mt ore feed. This cost range includes appropriate weightings for those operations treating certain percentages of noncar- bonaceous and carbonaceous ores. The estimated cost range for vat leaching of noncarbonaceous (normal) ores ranges from $9.44/mt to $17.76/mt ore feed. The requirement to utilize double oxidation with chlorine gas to treat car- bonaceous ores is estimated to add $5.30/mt to $6.50/mt to the normal vat leaching operating cost, reflecting additional costs for chlorine and steam generation. Indicated levels of productivity, including adminis- trative labor at the older mills, range from 15 to 20 mt ore milled per worker-shift with an average of 17.5 mt. Produc- tivity at the newer mills is more than double ranging from 26 to 75 mt ore milled per worker-shift, and averaging 37 mt. Despite the reasonably high efficiencies at the vat leaching operations treating gold ores, the cost of direct and indirect labor, as a percentage, represents the single most important component of the overall mill operating cost, ranging from about 27 pet to over 50 pet with an average of 34 pet for all seven milling operations. At the newer mills, the second largest cost item, as estimated in this study, is the miscellaneous costs (all costs except direct and indirect labor cost, reagent costs, and energy costs), which have a wide range as a percentage of the total cost, generally from 16 to 46 pet with an average of 33.5 pet for the five newer mills. At the older mills, the miscellaneous cost items repre- sent the third most important cost element, averaging 17.5 pet of the total cost. The estimated costs for reagents, ex- pressed as a percentage of the total milling cost, are fairly consistent with a range from 15 to 26 pet of the total cost and an average of 22 pet for all seven operations. The costs for electrical energy and energy in the form of steam, pro- pane, and fuel oil constitute the smallest individual major cost component at the vat leaching operations. As a percen- tage of the total operating cost, energy costs range from 5 to 27 pet with an average of 15 pet. Refining and Transportation It is estimated that as of 1981, total gold refining capaci- ty in the continental United States was about 6.0 million tr oz/yr. This total represents capacity for refining dore bullion, anode slimes from copper electrowinning, copper- lead-zinc concentrates, and scrap and other residues. In 1981, copper-lead-zinc concentrates, anode slimes from cop- per refining, and other residues were being processed at eight facilities with a total production capacity of about 2.0 million tr oz/yr refined gold. Scrap and residue were being refined at 22 small facilities having a total production capacity in 1981 of about 800,000 tr oz/yr. Only three facilities— Englehard Minerals' refinery at Newark, NJ, Handy and Harman's refinery at Attleboro, MA, and Homestake Mining's refinery at Lead, SD— were accepting the main product of primary gold mining operations (dore bullion) along with scrap and residue. The three refineries accepting dore had a combined estimated production capaci- ty of slightly more than 3.6 million tr oz/yr refined gold, with most of the capacity being at Engelhard's refinery. In 1981, the primary producers of gold in the Western United States were probably shipping about 900,000 to 1.0 million tr oz/yr dore bullion by air to these Eastern U.S. refineries. Since 1981, two important developments have occurred in the geographical location of the refineries accepting dore bullion. First, Johnson Matthey Investments, Inc. of Lon- don, England, began construction of a major precious metals refinery in a suburb of Salt Lake City, UT. The $10 million facility was planned to have an initial production capacity of 1.5 million tr oz/yr refined gold and 5.0 million tr oz/yr refined silver with the design enabling the doubling of pro- duction capacity with minor changes. The refinery was dedicated on April 22, 1983. The second development oc- curred 1 yr later on April 5, 1984, when Englehard Minerals announced that it would close its refinery in Newark, NJ (27). The relationship between these two developments is not entirely clear, although the shift makes geographic sense with the increase of primary gold production in the Western United States. The initial production capacity of 1.5 million tr oz/yr refined gold at the new Johnson-Mathey refinery is more than sufficient to handle the output from the primary gold operations evaluated in this study. For example, if the out- put of gold from the Homestake, SD, operation and the gold production from the small operation that closed in late 1983 50 are not considered, the 15 remaining major surface gold pro- ducers would account for about 1.05 million tr oz gold pro- duction in 1983. Thus, with the capability of easily doubl- ing production at the Salt Lake City refinery, there will be sufficient refinery capacity to easily handle an additional 1.0 million tr oz annual primary gold production, should such a development occur between now and 1990. However, a large gold price increase could result in a large increase in recycled scrap material, which could swamp the refin- eries, as happened in 1980. CANADA HISTORICAL PERSPECTIVE It is estimated that approximately 215 million tr oz gold were produced in Canada during the period 1858 through 1983. Canada has been the second-ranked market economy producer (behind South Africa) and the third-ranked world producer (behind South Africa and the Soviet Union) since overtaking the United States in 1930. Provided that the gold price remains about $400/tr oz, Canada should maintain this ranking into the 21st century, although some projec- tions assert that Brazil may take over the third-ranked posi- tion by 1990. Table 44 summarizes the major historical production developments of the Canadian gold mining industry. Ex- cept during 1974-81, Canada has produced over 2 million tr oz/yr gold since 1930. Fully 88 pet of total gold produc- tion from Canada has occurred since 1930. Major, sus- tainable levels of gold production were only attained with the development of the lode gold deposits in Ontario dur- ing 1909-25. Figure 23 summarizes the Provincial distribution of gold production for selected years through 1983. As shown, gold production in the early years of this century was dominated by placer operations in British Columbia and Yukon Ter- ritory, whereas production from Quebec, Ontario, and the Northwest Territories was negligible. By 1933, these short- lived placer operations were replaced in importance by pro- duction from underground operations in Ontario and Quebec, which accounted for 85 pet of total production. Gold production since the 1960's has been dominated by Quebec, Ontario, the Northwest Territories, and British Columbia, which collectively accounted for 93 pet of total production in 1983. RECENT PERSPECTIVE: 1968-83 Figure 24 depicts total mine production of gold in Canada for 1968-83, a period when the average annual gold price increased tenfold from $40/tr oz to $425/tr oz. The period just prior to 1968 had recorded a 34-pct production decrease from 4.629 million tr oz in 1960 to 3.062 million tr oz in 1967 (20). This decline is directly attributable to steadily increasing costs of production in the face of a fixed gold price. By 1970-71, when the United States lifted the gold backing for the dollar, production had further declin- ed to slightly above 2.0 million tr oz/yr (20), the lowest level of production since 1930. The freeing of the gold price in 1970-71 did not halt the decline in Canadian production, which fell into a narrow range of 1.628 to 1.735 million tr oz/yr for a 9-yr period between 1973 and 1981 (20). Because the majority of Canadian gold production was from vein- type lode deposits during this period, it is most likely that the additional 20-pct decrease from the 1970-71 period to the 1973-81 period was caused more by a lowering of the mill feed grade at the surviving operations than by the ac- tual closure of gold operations. A major turnaround in gold production has occurred since 1981, with production in 1983 being 36 pet higher than in 1981. However, putting this increase into perspective, the 1983 level of production at 2.27 million tr oz (20) is basically the same as 1970 production and still about 400,000 tr oz less than production in 1968. It is estimated that the currently developing mines of the Hemlo District will add around 676,000 tr oz of annual output by 1989. Yet even with this addition, the projected output for 1989 will only be at levels prevalent in the 1960's and still far below the historical highs. Table 44. — Historical summary of the Canadian gold mining industry Year or Period Occurrence or development 1823 Placer gold discovered on the Chaudiere River in Quebec Province. 1846 Silver veins reported in the vicinity of Thunder Bay, Lake Superior Region. 1852 Free gold discovered in quartz at Mitchell Harbour, Queen Charlotte Island. Causes the first auriferous quartz "rush" in British Columbia. 1857-65 Formal introduction of Canadian decimal currency occurs in 1858. First official annual gold production values are released for 1858, which totals 34,104 tr oz, all from British Columbia. Many placer gold discoveries occur in several Provinces during 1858-66. 1866 First discovery of gold in Canadian Precambrian shield near Madoc, Ontario. 1869 Gold discovered in the Yukon River. 1894-1900. . . . Annual gold production increases substantially from 54,600 tr oz in 1894 to 1.3 million tr oz in 1900. Of the increase, 1 .2 million tr oz (96 pet) comes from British Columbia and the Yukon Territory as a result of the discovery of placer gold at Klondike, Yukon Territory, in 1896. 1906 Annual gold production reaches a low of 406,000 tr oz as a result of a drastic decline in production from the Yukon placers. 1909 Dome, Mclntyre-Porcupine, and Hollinger Claims are staked in the Porcupine District of Ontario. 1911-17 Teck-Hughes, Wright-Hargreaves, and Lake Shore Claims are staked in the Kirkland Lake District of Ontario. 1922 Annual production reaches 1.2 million tr oz, passing the 1 .0-million-tr-oz mark for the first time since 1902. Ontario accounts for 1.0 million tr oz, or 83 pet, of the total. 1 925 First discovery of lode gold in the Red Lake District of Ontario. 1930 Canada's annual production of gold surpasses U.S. production for the first time and passes the 2.0- million-tr-oz/yr level for the first time. 1935 Annual production passes the 3.0-million-tr-oz/yr level for the first time; Ontario accounts for 2/3 of the total. 1936 First cyanide mill in Canada is constructed in Nova Scotia. 1937-41 Annual production reaches 4.1 million tr oz in 1937 and soars to 5.3 million tr oz in 1941. 1942-45 Gold production drops 53 pet from the 1941 level to about 2:5 million tr oz in 1945. 1951-62 Annual gold production ranges from 4.1 million tr oz/yr to 4.6 million tr oz/yr, demonstrating fairly good stability. 1963-67 In 1963, Canadian gold production falls below the 4.0-million-tr-oz/yr level for the first time since 1937. 1968-83 Annual production continues to show a decline from 2.7 million tr/oz in 1968 to a low of 1.6 million tr/oz in 1980. First results from drilling in the Hemlo District, Ontario, announced in 1981. Annual production rebounds from 1980 low to 2.3 million tr oz in 1983. Sources: References 20 and 28. 51 Atlantic Provinces 3 per 1902, I.C3xiO°Toz Others „_■ • «_ | P CT \ Proine °n3vinces \ 1933, 2.S8x I0 6 T0I Other? 3pctv Yukon \ 4pcr^ 1973, l.9SxiO"troz 1983, i27xl0 6 troz Figure 23.— Provincial distribution of Canadian gold pro- duction in selected years (pet and 1 6 tr oz). 1968 1969 1970 1971 1972 1973 1974 19/5 1976 1977 1978 1979 1980 1901 1982 1983 Figure 24. — Total mine production of gold In Canada, 1068-83. 52 1965- 3.587 xiO s *roz 1980- f.627x l0 6 noz 1983 -2.280 xl0 6 troz Figure 28. — Mine production of gold In Canada by typo of deposit for selected years. PRODUCTION BY DEPOSIT TYPE Figure 25 presents Canadian gold mine production for selected years broken down by deposit type (29, p. 57; 30, p. 62). As shown, 82 pet of total mine production in 1965 was accounted for by "auriferous quartz" or lode-type deposits. Base metal mines, producing gold as a byproduct of copper, lead, or zinc production, accounted for 17 pet, and placer deposits represented the remaining 1 pet. By 1980, production from lode-type deposits had declined by almost 2.0 million tr oz and represented only 60 pet of total pro- duction. Production from base metal mines had also fallen but by only approximately 50,000 tr oz and represented 34 pet of total production. Production from placer deposits had risen by around 50,000 tr oz and represented 6 pet of total production in 1980. By 1983, production from lode-type deposits had increased by 730,000 tr oz and accounted for 75 pet of total production. Byproduct gold production from base metal mines had fallen by 1.0 million tr oz, while placer production had increased slightly. A comparison of total Canadian production with produc- tion from lode-type deposits demonstrates that the level of total production is extremely dependent upon the output from lode-type deposits. For example, 94 pet of the 1.3-million-tr-oz decline in total production from 1965 to 1980 was due to declines in output from the lode-type deposits. Likewise, the 647,000-tr-oz increase in total pro- duction from 1980 to 1983 was due to increased output from lode-type deposits, which increased production enough to both offset the declines in output from base metal mines and add to overall annual production. The following sec- Table 45. — Gold production from Canadian lode-type deposits, 1973 and 1981, by Province Province Number of operations 1973 1981 Percent of total lode production 1973 1981 Ontario Quebec Northwest Territories . . . British Columbia 12 6 4 11 10 6 3 59.5 23.0 17.5 .0 46.0 37.5 13.5 3.0 Total 22 30 100.0 100.0 Sources: References 31 and 32. tions detail the geographical distribution of the three deposit types and discuss the number of mines producing from each deposit type by region. Lode-Type Deposits Table 45 summarizes the provincial and operational aspects of lode deposit production of gold for 1973 and 1981, which are believed to exemplify the changes that have taken place in the Canadian gold mining industry. In 1973 there were 22 primary lode gold mines (owned or operated by 17 companies) producing 1.4 million tr oz/yr gold. By 1981, there were 30 primary lode gold mines (owned or operated by 25 companies) producing 1.1 million tr oz/yr. In both years, over 80 pet of lode gold production originated in Ontario and Quebec. Three of the mines producing in 1973 had closed by 1981, while 11 new mines had come on-stream— 6 in Quebec, 2 in the Northwest Territories, and 3 in British Columbia. 53 Table 46. — Byproduct gold production from Canadian base metal operations, 1973 and 1981, by Province or Territory, number and type of operation 1973 1981 Province or territory British Columbia Quebec Ontario Manitoba and Saskatchewan Yukon Territory Newfoundland New Brunswick Total operations Total production 10 3 tr oz NAp Not applicable. Sources: References 31 and 32. Gold Number and Gold Number and production, type of production type of pet operation pet operation 35 1 Fe-Cu, 48.0 1 Ag, 1 Ag, 1 Fe-Cu, 1 Pb-Zn-Cu, 1 Ag-Cu, 10 Cu. 2 Pb-Zn-Cu, 9Cu. 29 6 Pb-Zn-Cu, 22.5 4 Pb-Zn-Cu, 6Cu. 4 Cu. 15 3 Pb-Zn-Cu, 7.0 1 Pb-Zn-Cu, 2 Ni-Cu, 2 Ni-Cu. 2Cu. 14 4 Pb-Zn-Cu, 13.0 4 Pb-Zn-Cu, 1 Cu. 1 Cu. 3 1 Pb-Zn, 6.0 1 Pb-Zn, 1 Cu. 1 Cu. 3 1 Pb-Zn-Cu, 2.0 1 Pb-Zn-Cu, 1 Cu. 1 Cu. 1 1 Pb-Zn-Cu 1.5 1 Pb-Zn-Cu 100 42 NAp 100 35 NAp 528 NAp NAp 462 NAp NAp Production from the lode deposits, however, had fallen for two basic reasons. First, 9 of the 11 new operations only accounted for approximately 84,000 tr oz of total annual production by 1981, or an average of less than 10,000 tr oz/yr per operation. Second, the high gold prices of 1980-81 caus- ed the larger operating lode mines to decrease their mill feed grade for these years. Production began turning around in late 1981. In the period from November 1981 through the first quarter of 1982, nine new lode gold mines came into production with capacities ranging from 10,000 to 70,000 tr oz/yr; the ma- jority were in the 20,000- to 50,000-tr-oz/yr range. These new operations accounted for at least 300,000 tr oz of addi- tional lode gold production in 1982 and, combined with the raising of mill feed grades at the larger operations due to lower gold prices, led to an increase of slightly more than 400,000 tr oz in total lode gold deposit production from 1981 to 1982. An additional 150,000 tr oz of lode production added between 1982 and 1983 came mostly from the final phases of expansion plans at producers instituted in 1981-82, since the net gain in the number of producers during this time was only two. During 1983 5 mines opened and 3 closed, leaving a net of 41 primary lode producers as of early 1984 (.32). Base Metal Deposits Byproduct gold production from primary base metal mines has fallen steadily from 602,000 tr oz in 1965 to 455,000 tr oz in 1983. Table 46 gives a breakdown of gold production from base metal deposits by Province and type for 1973 and 1981. As shown, the decrease of approximate- ly 66,000 tr oz from 1973 to 1981 coincided with a decrease in the total number of operations in production from 42 to 35; 13 mines closed (5 Pb-Zn-Cu, 6 Cu, 1 Ag, and 1 Cu-Fe) while 6 mines were opened (2 Pb-Zn-Cu, 1 Cu, 1 Fe-Cu, 1 Ag, and 1 Cu-Ag). The majority of the decrease occurred in Quebec and Ontario; two Pb-Zn-Cu and two Cu mines closed in each Province with no new operations put into production. In 1981, approximately 73 pet of byproduct gold produc- tion from base metal mines was produced by only 12 of the 36 operations. These 12 all produced in excess of 15,000 tr oz/yr and were comprised of 3 Pb-Zn-Cu, 1 Ni-Cu, 1 Cu-Ag, and 7 Cu operations. As of early 1984, 11 of the 12 opera- tions were still producing. Two major points concerning the economics of byproduct gold production from base metal mines can be made: 1. The economic viability of byproduct gold producers depends upon the primary and coproduct base metals be- ing produced rather than the economics of gold. 2. Control of gold grades and recoveries is very limited. Assuming that prices for copper, lead, zinc, and nickel reached their lowest levels during 1982-83, then byproduct gold production in Canada may have bottomed out at about the 450,000-tr-oz level. Combining this with the assump- tion that the 12 major producers represent a "core" of pro- duction during the worst of economic conditions, then an- nual byproduct production would have a low-level range of about 350,000 to 450,000 tr oz and a high-level range of 450,000 to 550,000 tr oz. Placer Deposits In the early to mid-1960's placer gold production in Canada was fairly steady at a level of 40,000 to 50,000 tr oz/yr, representing around 1 pet of total output. In 1967, a sharp drop occurred and placer production fell to below 10,000 tr oz/yr, remaining there until 1975. Since 1975, placer production has steadily increased. Production in 1983 totaled approximately 114,000 tr oz and represented 5 pet of total Canadian production. In 1973, when placer produc- tion for the year was only 9,804 tr oz, one sluicing opera- tion in British Columbia and 32 sluicing operations in the Yukon Territory provided the bulk of production. The opera- tions in the Yukon Territory accounted for 66 pet of the out- put, the one operation in British Columbia accounted for 32 pet, and a number of very small operations in Alberta, Manitoba, and Saskatchewan accounted for the remaining 2 pet (31). By 1981, nearly all the creeks in British Colum- bia and the Yukon Territory with a history of gold produc- 54 Table 47. — Demonstrated resource data for selected major Canadian primary gold operations as of January 1984 NM Not meaningful. 'Mill feed basis, includes adjustments for mining recovery and dilution. 2 As of early 1984, includes 1 property in the financing stage. Classification Number of operations Recoverable 1 resource, 10 6 mt Contained 10 6 tr oz gold' Pet Recoverable gold 10 6 troz Pet Major producing Major developing 2 14 4 111 132 22.0 20.0 NM NM 20.6 NM 18.9 NM Total Underground ore Surface ore 18 17 5 243 NM NM 42.0 37.9 4.1 NM 90 10 39.5 NM 36.1 91 3.4 9 tion had been staked. Because of the recent boom in placer production, it is difficult to estimate the average size of the more important operations. Indicated production capacities of the larger operations are only about 2,000 to 3,000 tr oz/yr. The major points to be made concerning placer pro- duction in Canada are— 1. Because of hoarding and other nonreporting, it is dif- ficult to estimate placer production. 2. The majority of placer production comes from two districts in British Columbia (Atlin and Cassiar) and three districts in the Yukon Territory (Dawson City, Mayo, and Kluane Lake). 3. With gold prices above $400/tr oz, a steady annual placer production of 50,000 to 150,000 tr oz can probably be expected. 4. Many of the placer operations in Canada are rework- ing deposits for the third or fourth time, and many are on- ly economical when treating a large volume of feed material. RESOURCE OVERVIEW, 1984 Estimates of reserves and resources in Canada have in- creased greatly since the late 1970's. The Canadian Govern- ment estimate of 1984 reserves was more than three times higher than the 1978 estimate and 35 pet greater than the 1983 estimate. These 1983 and 1984 estimates of 27 and 36 million tr oz, respectively, are conservative and repre- sent a minimum level of availability. The estimates are based upon information that is circa 1982-83 and by defini- tion cover only "operating mines and deposits committed for production as of January 1 of each year" (33). The estimates probably cover mostly material that is considered to be economically minable by the companies submitting the information. It is believed that the estimates include both primary and byproduct sources of gold. This study (table 47) analyzed a demonstrated gold resource, as of January 1984 (mill feed basis), of 22.0 million tr oz, contained within the 14 largest and most significant primary producing mines. These operations, with one ex- ception, each contain in excess of 450,000 tr oz gold in the demonstrated resource. In addition, a 1984 demonstrated gold resource contained within three developing primary gold mines in the Hemlo District of Ontario and one poten- tial large-scale surface mine is currently estimated at 20.0 million tr oz. Total Canadian primary demonstrated gold resources contained in the 18 major operations and deposits included in this study are thus estimated at 42.0 million tr oz. Based on the analysis, this contained gold is appor- tioned 90 pet to underground resources and 10 pet to sur- face minable resources. On a recoverable gold basis, the ap- portionment is basically the same, 91 pet underground and 9 pet surface, with the slight change due to different overall recoveries. Only 1 evaluated operation (a nonproducer) would be a 100-pct surface mine, whereas 14 operations would be 100-pct underground producers and 3 operations would produce from both surface and underground resources. The surface resource present at three underground producers is insignificant in total quantity, and it is expected that these operations will exhaust their surface material in a few years. This resource situation stands in sharp contrast to that in the United States, where a large majority of the resource is surface minable material. The potential amount of demonstrated byproduct gold contained within 31 producing and nonproducing copper, lead zinc, and silver deposits totals approximately 15 million tr oz. However, a potential of only about 6 million tr oz is contained within currently producing operations, some of which are under severe economic pressure and are ques- tionable future producers. This reduces the probable availability of this byproduct gold. In addition, byproduct gold grades are generally less than 1 g/mt, recoveries are less than for primary operations, the time frame of availability is considerably longer, and the relative share of byproduct to total production has fallen from 34 pet in 1980 to 20 pet in 1983 as base metal production has de- clined. The very poor performance of base metal prices has impacted severely upon a determination of economic reserves of gold associated with base metal production, as in the case of the aforementioned Canadian Government estimates. Taking this fact into consideration, together with the more recent information supporting this study's higher estimate of Hemlo gold resources, places these two sets of estimates in close relation to each other. An independent estimate by Homestake Mining (23) based upon 1983-84 data places primary gold reserves in Canada at 51 million tr oz of contained gold. This estimate includes current producers, the major Hemlo developments, and a large number of smaller producers and potential pro- ducers. Reserve and resource estimated for Canada's large number of smaller producers and explored deposits (poten- tial producers) are very difficult to quantify with any degree of accuracy. This 51-million-tr-oz estimate is considered to be a reliastic inference of the upper range of known poten- tial primary gold resources in Canada. Estimates of total contained gold resources in Canada thus range, circa 1983-84, from a minimum of 36 million tr oz of "total reserves" (official Canadian estimate for operations in production or committed for production as of January 1, 1984) through this study's estimates of 42 million tr oz of demonstrated primary gold resources, or 48 million tr oz of total demonstrated resources for both primary and byproduct operations, to potentially as high as 51 million tr oz of primary gold resources estimated by Homestake Mining. All three sets of estimates vary between 15 and 30 pet of each other, and differences are due mostly to definitions concerning the number and sizes of operations included in the estimate. 55 OCC* N 9 ® ® o o LEGEND Costed, producing gold mines Costed, prospective gold mines Nancosred, prospective gold m C.t, Major historical lode gold mining camps (districts) Areas of gold deposits and occurrences Mesozoic eiposures Areas of gold deposits and occurrences Precambrian exposures Figure 26. — Location of producing and prospective primary gold mining operations, areas of gold deposits, and historical lode gold mining districts In Canada. LEGEND Costed, producing gold mines Noncoetex), producing gold mines Costed , prospective gold mines Noncosted, prospective gold mines Lot* A*> Nipigoni "» Agnico-Eogle ,_(GoM Division) Hemlo Operotions V^ tuuam (LocMmjral.Prooert,, R„,^KirV K /Sopn (i * l « ,l 2i. l ., !««"»?7 « . 400 V) O O 300- P 200 KEY Y///% 10-pct DCFROR I J Operating costs I Capital costs m rZ Hi! w 2 n i 1 R i I z - Figure 28.— Contribution of capital and operating cost to total production cost for 15 selected Canadian operations. 700 600- 500 S 400- 8 soor- o _l £* 200 KXl- 10-pct DCFROR \-J' - ,--"<. / Break-even DCFROR 25 TOTAL RECOVERABLE GOLD, lO^troz Figure 29. — Potential total Canadian primary gold available at various production cost levels from 15 selected operations as of January 1984. This study currently estimates that the three proper- ties in total contain approximately 72.5 million mt of recoverable ore with weighted average diluted grades from 5.8 to 10.6 g/mt. Total contained gold is approximately 16.4 million tr oz. Table 50 contains summary data of the economic analyses. The results underscore the highly favorable economics of these operations. Break-even cost determinations range from $153/tr oz to $191/tr oz gold with a weighted average of $170; cost determinations at the 10-pct DCFROR level range from $218/tr oz to $294/tr oz with a weighted average of $255. Operating costs per ounce are quite low ($93 to $125), and given the large amount of available gold, capital costs per ounce are also low ($45 to $54). These cost estimates are significantly lower than those determined for all three of the newest gold producers in Canada and in- deed are lower than those for all but the largest and highest grade older producers that were also evaluated. The major reason for the favorable economics at the Hemlo properties is the capability to mine high-grade material over very large mining widths, which allows the use of high-volume min- ing methods that are inexpensive by comparison to the methods that must be employed at most of the other underground gold mines in Canada. Average annual production levels at the three Hemlo operations will range between 90,000 and 253,000 tr oz, and initial estimates of productive lives range from 26 to 34 yr. Total available gold from each property ranges from 3.0 to 6.6 million tr oz. As of early 1984, the three operations are estimated to have a total potential of producing 16.0 million tr oz of refined gold over their lives. 2 1964 1990 1992 1994 r996 1998 2000 Figure 30. — Potential annual Canadian primary gold available at various break-even production cost levels from 15 selected operations, 1984-2000. 60 Impact of the Hemlo District Upon Future Annual Canadian Gold Output The impact of these three operations upon annual Cana- dian gold output will be dramatic. Figure 31 shows com- bined annual output for these three mines for 1984-2010 at a break-even price-cost level of $200. According to this scenario, the Golden Giant Mine will begin producing in late 1984, the Teck/Corona property will have its initial pro- duction in 1985, and Lac Minerals (the largest of the three operations) will start producing in 1986. Full-capacity pro- duction, as initially designed, will be achieved by the Teck/Corona and Golden Giant Mines in 1987, with the Lac Minerals property expected to reach its full-capacity level in 1989. From 1989 through 2006 a combined annual out- put of 676,000 tr oz is anticipated. Production would decline thereafter if no further exploitable resources are added, which is considered unlikely. In any event, just the con- tained gold already demonstrated in these three properties will sustain a high level of production for about 20 yr. In table 51 and figure 32, total Canadian gold availabili- ty is depicted for the original 15 evaluated properties com- bined with the available gold of the three Hemlo district mines. With 100 pet of the 16.042 million tr oz of recoverable gold at the three Hemlo operations available at a break- even cost-price level of $200/tr oz or less and below $300/tr oz at the 10-pct DCFROR level, these mines should come to represent the low-cost "core" of the Canadian gold min- ing industry. For example, when combined with the other 15 operations discussed earlier, the Hemlo operations in- crease the total available gold in Canada at the $400 cost- price level by 136 pet. At a 10-pct rate of return, the total available gold at the $400 cost-price level is almost three times higher when the Hemlo mines are included. The effect of the Hemlo mines on future annual output will be to significantly raise available production from major Canadian gold mines at all cost-price levels. Figure 33 pro- vides estimates of future production potential through the year 2000 for all 18 primary gold operations. Production from the three developing Hemlo mines will ensure that production from the major primary operations surpasses 2 million tr oz /yr during 1988-90 and remains above 1 million tr oz through the year 2000, even in the absence of further increases to demonstrated resources and annual capacity at the other properties. Cumulative Canadian output during the 1989-2000 period, for all major mines producing at $410 tr oz or less, rises to 15 million tr oz with production from Hemlo, an increase of 85 pet. At $510 tr oz or less, cumulative output during the same period rises to 19.6 million tr oz, an in- crease of 69 pet. Production from the Hemlo gold district should ensure that Canada remains the third or fourth largest world gold producer for the foreseeable future. TOTAL ANNUAL PRODUCTION POTENTIAL TO 1990, ALL SOURCES This study concentrate upon 18 major primary opera- tions which form the core of both current and future pro- duction potential. But the gold mining industry of Canada is complex and under dynamic growth with smaller opera- tions beginning production and closing each year. An analysis of all primary producers in operation during 1981-82 determined that 19 additional lode gold producers were not cost-evaluated in this study. It was further deter- !=! .4 - 1 1 1 1 ■/ \o-$200 ' 1. | I 1984 1990 1995 2000 2005 20K) Figure 31. — Potential annual primary gold production from throa developing mines In the Hemlo District, Ontario, Canada, 1984-2010. 500 O 300 Breok^even DCFROR IO 15 20 25 30 35 TOTAL RECOVERABLE GOLD, I0 6 troz 45 Figure 32.— Potential total Canadian primary gold available at various production cost levels from 18 selected operations as of January 1984. mined that these 19 nonevaluated mines accounted for about 375,000 to 400,000 tr oz annual gold production and that only 5 of the 19 had annual output clearly exceeding 25,000 tr oz. Thus, the majority of the operations that were not evaluated are very small compared to those that were evaluated. In fact, a few of these mines were closed, either permanently or temporarily, by the low prices of the 1983-84 period and most of the remainder have reserves equivalent to only 3 to 5 yr of production. A survey of the 1983-84 period has determined some 15 other smaller properties (out of a much larger total) that appear to have had enough encouraging exploration work to warrant either feasibility studies, that are undergoing mine development at this time, or that are in the early stages of production. Table 52 lists these 15 properties that are considered the most likely to be in production as an ad- ditional or replacement resource between 1984 and 1990. An approximate total production level of 470,000 tr oz/yr is possible by 1990 if all 15 are indeed brought into produc- tion. The individual estimates are based upon the most re- cent published information and are subject to revision as the studies and developments progress. The following are some of the major economic aspects of the 15 properties shown in table 52: 1. All of these properties will begin production (if they are actually developed) with at least 5 to 8 yr worth of 61 25 KEY 3 Hemlo Other producers 1984 1986 1988 1990 1992 1994 1996 1998 2000 Figure 33. — Contribution of three Hemlo District opera- tions to potential annual Canadian primary gold produc- tion, 1984-2000. Table 52. — Canadian operations considered possible or probable primary gold producers during 1984-90 Province and property British Columbia: Bralome-Pioneer Takla Lake Manitoba: Agassiz San Antonio Ontario: McBean Renabie (Cullaton Lake) . Shoal Lake (Duport) Stock Township Quebec: Aquarius Bousquet-Cadillac Croinor-Abigold D'Or Val mines Horn Mine Russian Kid Saskatchewan: Bootleg Total NA Not available. NAp Not applicable. Status as of 1983-84 Estimated additional annual production capacity, 10 3 tr oz Estimated first year of production Under study In exploration Under study ... do In production, April 1984 Expanding In exploration Under study Developing ... do In exploration Developing ... do ... do In production, early 1984 NAp 24 30 26 24 23 60 40 35 25 23 40 23 21 28 10 470 NA NA NA NA 1984 1984 NA 1987-88 1984 1985 NA 1985 1985 1985 1984 NAp reserves available, and a fair proportion will begin produc- tion with 10 yr or more worth of reserves. 2. All but one of the planned mines will be underground with expected feed grades ranging from 4.5 to 9.0 g/mt. 3. The planned surface mine is anticipating mill feed grades of 2.8 to 3.0 g/mt. 4. The range of production capacities is between 10,000 and 40,000 tr oz/yr gold with an average planned size of slightly over 25,000 tr oz/yr. 5. Because they mostly involve "new looks" at past pro- ducing operations, the "exploration" programs for the underground operations necessarily involve a high amount of underground development with extensive rehabilitation and extensions to previous workings to gain access to areas for exploration. Thus, most exploration and development programs for these deposits take 4 to 5 yr from the initial exploration stage through the feasibility study stage and are high cost, ranging from $3 million to over $10 million, depending, of course, upon the "target" size and amount of development required. However, because of the high ex- ploration requirements, once a decision is made to go ahead with a project, the only additional capital requirements will be for mine equipment, milling facilities, and associated in- frastructure items, and an average-sized operation will only take between 1.5 to 2 yr to get into full production. For the following complete scenario of prospective future gold production, it has been assumed that byproduct gold production in 1990 will average between 350,000 and 62 Table 53.— Potential annual Canadian gold production PRIMARY GOLD MILLING MAJOR by 1990, by type of operation OPERATIONS IN CANADA Type of operation ^'tom/d" 98 ' 'm ?' ^ discussion of estimated mine operating costs at the Prim 18 primary gold operations can be summarized by examin- Major producers (includes ing 20 separate mine operating cost estimates (17 Hemlo District . 2,100 60-68 underground and 3 surface mining estimates). Although a Developing, in early stage of . .. c , , , , . ° . — . ° production, or considered majority of primary gold production in Canada comes from probable for development underground mines, some of the operations utilize both sur- Base metal operations 2 ^ ace anc * underground material in their overall production. (byproduct) 350-550 1 3-17 Two examples of this would be the Giant Yellowknife opera- Placer operations 2 100-150 3-4 tion> which mines sur f ace and underground material oa : — Z-L simultaneously, and the Detour Lake operation, which is cos??vI!uaK d ,0 C ° mple,e demons,ra,ed resource and long - ,erm total currently mining only surface material but is expected to 2 Assumed to remain constant at 1977-83 average levels. progress to an entirely underground mine by the late 1980's. 555,000 tr oz, which is similar to output levels of the Table 54 presents ranges of the 20 estimated mine 1981-83 period. Again, it is difficult to estimate potential gating cost f for *J? " ' P™nary gold operations in Canada byproduct gold output since the near-term economic outlook J at wer ? evaluated in J 18 study Included in the table are e i j j • • _x • c- -i i the number of estimated costs in the range shown, the mm- for copper, lead, and zinc remains uncertain, bimilarly, ., , , ... , ? . .i . ? ! , .. , . ., . ■ ,1 . . , , , / ing methods and capacities represented in that range of placer production (a minor contributor to the total and the ° .. , , e r „ . . i most variable) is expected to range no higher than 100,000 fP^ atl ? g c ° sts ' ar \ d a meas , ure of ° vera11 m ™ g . pr ,° d " C " *- i cn nnn + ■ i aan tivity for the underground operations which includes to loUjUUU tr oz in iyyu. , iii i ^ i *• ii--jj- m _ • .v e j. xuico underground labor and prorated surface and administrative To summarize the foregoing discussions, table 53 , , & ^ presents this study's estimates of the expected range of total gold production by 1990 in Canada from both primary and Surface Mining byproduct sources. The estimated range anticipates the development of new mines, the expansion of some current Surface mining of primary gold is not significant in mines, and the permanent closure of others which for full Canada. As of late 1984, there were only two major mines potential would probably require a gold price of at least in Canada producing gold ore entirely with surfacing min- $450/tr oz to $500/tr oz. As shown, total Canadian gold pro- ing (one of which was not evaluated for costs in this study duction in 1990 could range between 3.0 and 3.4 million and is not included in table 54). In addition, there was one tr oz, which would represent the highest level since 1965 major mine at which surface ore represents a portion of the and an increase of approximately 1.0 million tr oz over 1984 mill feed. There is also one major proposed surface opera- production levels. Around two-thirds of this increase is ex- tion (shown in table 54) which, as of 1984, was still in the pected to come from the development of the three major financing stage. Estimated waste-to-ore ratios of the sur- Hemlo properties. The remainder is expected to come from face mines evaluated for costs ranged from 0.56 to 3.5 mt a number of smaller producers. The single most important waste per ton of ore. The estimated surface mine operating point to stress is that the major primary operations that costs in Canada for the mines with capacities greater than were cost-evaluated represent (at 67 pet) the largest source 2,800 mt/d ranged from $1.25/mt to $2.70/mt material of assured long-term gold production and basically deter- moved, depending almost entirely upon the mining capaci- mine the overall economic competitiveness and growth pro- ty of the operation. The higher cost surface mine shown in spects of the Canadian gold mining industry. table 54 represents a sporadic producer which has Table 54. — Mining types, operating costs, capacities, and productivity for major Canadian gold operations Range of mine Number of Capacities Estimates of overall under- operating costs, dollars per operating costs Description of mining methods within range within range, mt/d ground mt mining production, 1 per worker-shift ton of ore in range Underground: $10.00 to $20.00 4 Long-hole open stoping, some shrinkage. 1,670-6,500 12.5-31.2 20.01 to $30.00 5 Cut-and-fill, shrinkage, room-and-pillar, long- hole and blasthole open stoping. 1,400-5,600 5.9-13.8 40.00 to $50.00 3 Cut-and-fill, long-hole and blasthole open stoping, shrinkage. 1,600-1,900 2.8-10.0 50.01 to $60.00 2 Open stoping, cut-and- fill, square-set stoping. 600-1 ,000 3.0- 3.5 Greater than $60.00 3 Cut-and-fill, narrow- vein shrinkage. 400-800 1.7- 3.0 Total 17 NAp NAp NAp Surface: Less than $15.00 2 0.56-3.5:1.0 waste- to-ore ratio. Bench (berm). 2,800-15,700 NAp Greater than $15.00 1 3:1 waste-to-ore ratio. Bench (berm). 300-350 NAp Total 3 NAp NAp NAp NAp Not applicable. 'Includes prorated and administrative labor. 63 anomalously high surface mining costs due to the very low capacity and erratic nature of production from year to year. Surface mining costs in Canadian primary gold operations appear to be about 15 to 25 pet higher than surface mining costs at comparable U.S. operations, mostly reflecting a higher proportion of costs attributable to labor. It is not ex- pected that surface mining will assume a major role in primary gold production in Canada in the near future. Underground Mining The vast majority of primary gold production in Canada has been and will continue to be produced by underground mines. It is very difficult to generalize or simplify a discus- sion of the underground mines producing primary gold in Canada, especially in regards to operating costs. As shown in table 54, the underground operating costs estimated in this study range from slightly more than $10/mt to over $60/mt ore feed to the mill. Table 54 shows the three interrelated factors — the predominant mining method in use, overall underground productivities, and daily capacity — that have the greatest influence on the overall cost of mining. These categories are, in turn, directly related to the economic geology of the gold occurrence (grade of gold over a defined width of ore body) and conscious deci- sions on the part of the operation's management as to how to mine that defined ore body. Thus, the predominant min- ing method employed at various times may differ depend- ing upon the amount of ore and the gold grade that manage- ment wishes to be supplied to the mill. One actual (admittedly extreme) example of this in- terplay between desired ore feed grade, management deci- sions, and the resultant costs should suffice to give the reader a better understanding of the complex of factors in- volved in determining the overall cost of gold extraction. In the last 1970's, a small (450-mt/d) underground mine was utilizing a relatively high-cost cut-and-fill mining method to produce ore grading a very high 12 to 13 g/mt. After the spectacular runup in gold prices during 1978, 1979, and early 1980, a decision was made by management to increase the operation's milling capacity by nearly three times, which meant that a comparable increase to the underground mining capacity was required. To achieve this increase in milling capacity, management decided to con- vert the mining method from the high-cost cut-and-fill method to a low-cost, high-tonnage long-hole open stoping method. (See table 54 for a comparison of the relative operating cost of these methods.) The change was instituted quickly, but possibly too quickly because the recoverable yield of gold at the mill fell from 12.92 g/mt of ore in 1979 to 7.8 g/mt in 1980 and to an extremely low 4.5 g/mt in 1981. This was a decrease of 65 pet in the recoverable grade over just a 2-yr period and a decrease of 44 pet in overall gold production from 1979 through 1981, despite a 60-pct in- crease to the mill's ore feed capacity. In addition, increased capital expenditure and declining output coincided with fall- ing gold prices during 1980 and 1981. In 1982, to save the operation, the company decided to revert to its high-cost cut- and-fill method in the majority of the stopes, thus raising the feed grade back up to its normally high range (40). It should be noted that most of the major primary gold producers in Canada that have come into production within the last few years, or that are planned for production in the near future, either utilize or plan to utilize low-cost underground mining methods such as blasthole or long-hole open stoping or relatively large-scale open pit methods. For example, of the nine underground mining costs in the two lower ranges shown in table 54, seven are operations that have been brought into production since 1968 or will be in production in the near future. This is partly a reflection of the increase in the price of gold leading to the economic ex- traction of lower grade ore bodies and partly a reflection of the discovery of a new type of ore body, the Hemlo type, which has amazingly large thicknesses (3 to 30 m) that can be mined utilizing low-cost, high-tonnage underground methods. The depth at which major underground gold mines in Canada are operating does not, as of this date, appear to have a major influence on the overall mining cost. Hoisting distances as of the early 1980's range from about 200 m to around 2,200 m depths. The only major effect on mining costs caused by the depth of mining at the evaluated Cana- dian operations is due to losses of productivity at those operations which have to make more than one transfer of rock or more than one transfer of workers from the surface to the working area and vice versa Direct plus indirect labor constitutes the major cost item in these Canadian underground operating cost estimates, averaging 48 pet for the seven long-hole or blasthole open stoping operating cost estimates and 63 pet for the 10 opera- tions using either a mix of low- and high-cost mining methods or solely high-cost methods such as cut-and-fill, shrinkage and square-set stoping. GOLD MILLING IN CANADA OPERATIONS The 18 Canadian primary gold operations evaluated in this study represent 19 individual mills that are or will be treating ore from 25 individual mines and at least 26 dif- ferent ore types. All but 4 of the 19 mills were in produc- tion as of 1983, with all 4 nonproducers in the financing or development stage as of 1983. Of the 15 mills in operation in 1983, 7 were originally constructed during the 1930's, 3 were initially built in the 1940's, 2 were commissioned in the 1960's, and 3 were brought into production during 1980-82. As of 1983, only 1 of the 15 mills was accepting ore for custom milling from mines in the immediate vicinity, while another large mill was contemplating such a policy owing to projected deple- tion of its own ore reserves. Design capacities as of the early 1980's for the 15 mills in operation ranged from 430 to 2,700 mt/d with 5 mills treating less than 1,000 mt/d and 10 mills treating more than 1,000 mt/d. In some cases, utilization as of the early 1980's was as low as 75 pet of design capacity. In general, this reflected hoisting constraints or conscious managment decisions rather than mill inefficiencies. Estimated mill feed grades employed in this analysis for the 15 operating mills range from 3.4 to slightly more than 21 g/mt. Of the 15 mills evaluated, 12 were estimated to have overall gold recoveries of 90 pet or greater (straight average of 94 pet), while 3 either have or are expected to have overall gold recoveries of 83 to 85 pet. These lower recovery operations seem to reflect both that overall gold recovery drops off as the feed grade falls below 4 g/mt and that ores where the gold is intimately associated with pyrite appear to have lower recoveries. Estimated mill operating costs range from $10/mt to close to $35/mt ore feed. Four of the mills have operating costs in the $10/mt to $12/mt milled range, seven have operating costs in the $12.01 to $17 range, and four have 64 operating costs in the $17.01 to $35 range. The mills with operating costs below $12/mt have large capacities (1,300 to 1,700 mt/d) and simple flowsheets (crush, grind, vat leach, Merrill-Crowe-zinc dust precipitation, smelting to dore bullion). Those mills in the $12.01 to $17 range have more widely varying capacities (570 to 2,700 mt/d) and slightly more complex flowsheets with flotation and/or roasting of sulfides involved. Of the four mills with operating costs above $17/mt mill feed, two have relatively complicated flowsheets involving gravity concentration, flotation, and roasting; one operation has an extremely remote location; and two operations have relatively small capacities. Of note is the fact that all four of the mills with operating costs above $17/mt also have very good mill feed grades of greater than 10 g/mt ore milled. Table 55 summarizes the methods of gold recovery in use as of the early 1980's at the major Canadian gold mills analyzed. As of 1981, only four operating mills had jigs in circuit with their grinding units to recover coarse, free gold. A high percentage of Canadian mills were utilizing a flota- tion and/or roasting stage for treatment prior to the cyanide leaching stage. These two points reflect the basic nature of most Canadian primary gold ores, in which the gold is highly associated with sulfide mineralization (especially arsenopyrite, pyrite, and pyrrhotite) and is usually present in a finely divided, disseminated state rather than as coarse grains. Also of interest is that 14 of the 15 mills utilize the Merrill-Crowe-zinc dust precipitation method of extracting gold from the cyanide solution, while only one mill utilizes the relatively new carbon-in-pulp method as the main 9 ex- traction method. The predominance of the Merrill-Crowe- zinc dust precipitation method reflects the fact that 12 of the 15 mills were constructed prior to 1969, before the period of the first major developments in carbon extraction of gold technology. Still, of the three mills added during 1980-82, two have opted for the more conventional Merrill-Crowe method over the newer carbon extraction methods. In Canada, there are no large-scale heap leaching opera- tions such as are found in the Southwestern United States, and none are foreseen in the immediate future. The long period of winter weather is a major negative factor in large- scale heap leaching. Also, all of the large-scale, economic heap-leaching operations in the United States are surface- mining mostly oxidized, primary gold deposits, a type of deposit that Canada does not appear to have in abundance. Refining and Transportation As of 1981, five gold refineries in Canada were accept- ing dore bullion (primary mine production) for refining along with gold scrap and residue (secondary sources). The five refineries and their estimated 1981 production 9 As of 1981, 2 of these 14 mills were using minor C/P circuits to recover gold from their precipitated roaster gases. Table 55. — Summary of milling methods in use at major primary gold milling operations in Canada, early 1980's Milling method Number of mills (by major milling circuit) using the circuit Comminution circuit: Crushing-grinding 15 Jig recovery of free gold 4 Treatment prior to extraction of gold by cyanide solution: Pre-aeration or other pretreatment prior to cyanide leach 4 Float for copper concentrate (for sale) 2 Float-regrind-roast concentrate 4 Float-regrind concentrate 1 Float-regrind pretreat concentrate 1 Float-roast concentrate 2 Float-roast concentrate-regrind calcine 1 None 4 Extraction of gold: Vat leach with cyanide solution 15 Precipitation of gold, production of dore bullion: Merrill-Crowe-zinc dust precipitation, smelt to dore bullion 14 Carbon-in-pulp (extract and strip gold from solution)-electrorowinning onto steel wool-smelt to dore bullion 1 Other processes: Electrostatic precipitation of roaster gases, recover gold by leaching and CIP treatment of precipitates 2 Leach tails from flotation 1 Production of As 2 3 from roaster gases 3 capacities for refined gold production are shown in table 56. In addition to the refineries listed in table 56, Inco Metals Co. has a precious metals refinery at Sudbury, On- tario, capable of producing about 225,000 tr oz/yr refined gold as a byproduct from treating the anode slimes resulting from the refining of copper and nickel. Noranda Mines Ltd. also has a precious metals refinery at Montreal East, Quebec, capable of producing about 720,000 tr oz/yr byproduct gold from anode slimes resulting from the elec- trolytic refining of copper. The majority of the refining capacity shown in table 56 for treating dore bullion from the gold mines is conveniently located in relation to the Quebec and Ontario lode gold mines. The Royal Canadian Mint, located in Ottawa, is believed to handle most of the newly mined dore bullion in Canada. For analysis, it was assumed that all of the dore bullion produced at operations in Ontario, Quebec, and the Northwest Territories was sent to the Royal Canadian Mint at Ottawa for refining, with the dore bullion produced in British Columbia being sent to the refinery in British Col- umbia. The two operations producing gold-containing cop- per concentrates as well as dore bullion were assumed to be sending the copper concentrates to Noranda's smelting and refining facilities in Quebec. It is not clear where the proposed large-scale surface operation analyzed in this study will send its bullion if and when it is in production. However, indications are that its production level will be too high for the refinery in British Columbia, and its bullion will probably go to the Royal Canadian Mint refinery. As shown in table 56, total Canadian refinery capacity Table 56. — 1981 capacities of Canadian dore bullion, scrap and residue gold refineries Company Province City Estimated production capacity, 1 3 tr oz/yr Delta Smelting and Refining Co. Ltd. Englehard Industries Johnson Matthey Ltd. Royal Canadian Mint Trimount Refining and Smelting British Col Ontario ...do ...do ...do NAp umbia . . Richmond 250 Aurora Toronto Ottawa Richmond Hill 630 2,000 7,000 60 Inc. Total NAp 9,940 NAp Not applicable. 65 in 1981 was nearly 10 million tr oz/yr refined gold from new dore bullion, scrap, and residues. The 18 properties evaluated in this study would account for no more than about 1.3 to 2.5 million tr oz/yr dore bullion production. Thus, it is believed that even with the Hemlo deposits com- ing on-stream in the late 1980's, there is sufficient refin- ing capacity to handle the increases in gold production fore- seen in Canada by 1990. AUSTRALIA HISTORICAL PERSPECTIVE Total gold production between 1851 and 1980 from all Australian deposits is estimated to be approximately 190 million tr oz, or an average of 1.46 million tr oz/yr for the 130-yr period. Approximately 33 pet (63 million tr oz) of this total has come from alluvial mining of tertiary and recent river gravels. The remaining 127 million tr oz was produced from lode deposits (41, p. 141). The first official recordings of gold discoveries in Australia were in 1851 in the Provinces of Victoria and New South Wales, although the first actual discovery had oc- curred as early as 1839 in Victoria Province. Of the discoveries in the two Provinces, the Victorian discoveries were the most important. Gold production from alluvial deposits in Victoria peaked in 1856 at 3 million tr oz/yr, and by 1860 attention had turned from alluvial production to hard rock mining in the Province. By 1891, production in the Victoria goldfields had fallen to about 500,000 tr oz/yr, and the search for gold had been expanded to other provinces, most notably Western Australia (42). By 1893, both the Kalgoorlie and the Norseman Fields Gode deposits) in Western Australia had been discovered and were pro- ducing. Table 57 provides a brief synopsis of the historical production picture in Australia. The table concentrates on reported historical production of Victoria Province and the Kalgoorlie Goldfield of Western Australia since these two areas have accounted for an estimated 118 million of the 190 million tr oz of total production. COMPOSITION OF MINE PRODUCTION OF GOLD, 1970-83 Figure 34 plots reported mine production of gold in Australia for 1970-83 (2-3). As shown, the initial reaction to free gold markets was an increase in production to 775,000 tr oz in 1972. However, the combination of a reces- sion and declining gold prices caused production to decrease to 502,000 tr oz for 1976. The sustained increase in the price of gold from September 1976 into 1980 led to a slight re- bound, yet by the end of 1981 production was only up 25 pet over the low level of 1976. It was only in 1982 and 1983 that the effects of higher gold prices, which caused a surge in new mine development, began to show up in substan- tially higher production levels, exceeding the 1-million-tr- oz/yr level in 1983. This was the first time that Australian mine production of gold had exceeded the 1-miilion-tr-oz mark since 1963 and represented more than double the 1976 production level. The aggregate production values shown in figure 34 must be related to an operation-by -operation analysis to bet- ter understand the dynamics of the gold mining industry in Australia. With that in mind, table 58 provides "snap- shots" of certain critical years during the period 1970-83. Byproduct and coproduct gold production steadily de- clined on a percentage basis over this period, from 46 pet Table 57. — Historical summary of mine production of gold in Australia, 1851-1983 1 Year or period Occurrence or development 1851 First official discoveries in Victoria and New South Wales Provinces. 1859 Production reaches 3 million tr oz/yr from Victoria Province. 1860's . . Victoria Province produces an average of 1.6 million tr oz/yr, representing essentially 100 pet of Australian production and 40 pet of world production. 1890's . . . Victoria Province production falls to an average of 500,000 tr oz/yr. Western Australian Goldfields (Kalgoorlie and Norsemen) discovered and in production. 1903 Total Australian production reaches what proves to be a peak of 3.8 million tr oz. This coincides with the historical peak production at the Golden Mile (Kalgoorlie) of 1.0 million tr oz in 1903. Production in Victoria Province has increased to the 800,000-tr-oz/yr range. 1920's . . . Victoria Province production falls to an average of only 65,000 tr oz/yr for the decade. Kalgoorlie's production declines to 325,000 tr oz/yr, on average. The decline from production levels of the early 1900's was attributed mostly to rapid inflation of costs following World War I, which caused lower grade operations to close. 1950-67. . Victoria Province's production further declines to an average of 44,000 tr oz/yr throughout the period. Total Western Australian production stabilizes in a range of 650,000 to 850,000 tr oz/yr. 1968-76. . Increasing costs and low gold prices cause total Australian production to decrease from 787,000 tr oz in 1968 to an all-time low of 502,000 tr oz in 1976. Victoria Province's production reaches an almost negligible level of 2,050 and 6,600 tr oz in 1971 and 1972, respectively. Western Australian production decreases to a low of about 250,000 tr oz in 1976. 1977-83 . Total Australian production increases dramatically from 502,000 tr oz in 1976 to 1.035 million tr oz in 1983. Western Australian production increases from 250,000 tr oz in 1976 to 750,000 tr oz in 1983. Increase coincides with large rise in the price of gold. 'Compiled from numerous sources and the authors' own estimates. in 1972 to 41 pet in 1976 to 35 pet in 1979 and only 21 pet in 1983. The number of operations in this category, however, remained nearly constant. Throughout the 1970-83 period, the number of major gold producing operations (primary plus byproduct and coproduct producers) ranged from 9 to 13, standing at 12 as of 1983. From 1979 through 1983, the primary producers that could be considered major producers (the core of the industry) had only increased their annual gold output by 40,000 tr oz even though total Australian production had increased by 444,000 tr oz. Ninety percent of the remaining 404,000-tr-oz increase from 1979 through 1983 had come from as many as 21 minor primary gold operations (those with 5,000 to 30,000 tr oz of annual out- put), many of them representing new operations brought into production since 1981. Overall for 1983, 36 pet of total Australian gold production came from 7 major primary gold producers, 35 pet came from 21 minor primary producers, 21 pet came from byproduct gold producers, and 8 pet came from very small, primary gold producers (those producing less than 5,000 tr oz/yr). 66 1970 1971 I 1979 I960 1961 1982 1963 Figure 34. — Australian gold production, 1 970-33, Table 58. — Operational categorization of Australian mine production of gold in selected years 1 Year Total production, 10 3 tr oz Description of operations producing gold 1972 1976 1979 1983 755 95 pet (715,000 tr oz) of total production derives from 13 operations, with 5 of the 13 representing byproduct producers accounting for about 46 pet. The remaining 5 pet comes from 20 small, primary gold operations (produc- ing less than 2,000 tr oz/yr, on average). Western Australia comprises 43 pet and the Northern Territory 33 pet of total gold production. Geographic distribution of major gold producers follows: Western Australia — 5 Au mines; Northern Territory — 1 Au-Cu-Bi mine, 1 Au mine; Tasmania — 1 Pb-Zn-Au mine, 1 Cu-Au mine; Queensland — 1 Cu-Au mine; New South Wales — 1 Pb-Zn-Au mine; Victoria — 2 Au mines. 503 9 of the 13 major operations of 1972 are still producing. It is estimated that these major producers account for virtually 100 pet of total Australian mine production. Only 4 of the 9 major operations are primary gold producers. The 5 byproduct operations account for 41 pet of production. Western Australia comprises 42 pet and the Northern Territory 35 pet of total gold production. Geographic distribution of major gold producers follows: Western Australia — 2 Au mines; Northern Territory — 1 Au-Cu-Bi mine, 1 Au mine; Tasmania — 1 Pb-Zn-Au mine, 1 Cu-Au mine; Queensland — 1 Cu-Au mine; New South Wales — 1 Pb-Zn-Au mine; Victoria — 1 Au mine. 597 8 of the 9 major operations of 1976 are still producing. In addition, 1 new major producer, Telfer in Western Australia, has been brought into production since 1976. The 9 major operations of 1979 are estimated to account for 93 pet of total production (555,000 tr oz) with 6 pet coming from small primary gold operations, mostly in Western Australia. The same 5 byproduct producers of 1972 are still producing in 1979. Western Australia comprises 59 pet and the Northern Territory 25 pet of total gold production. Geographic distribution of major gold producers follows: Western Australia — 3 Au mines; Northern Territory — 1 Au-Cu-Bi mine, 1 Au mine; Tasmania — 1 Pb-Zn-Au mine, 1 Cu-Au mine; Queensland — 1 Cu-Au mine; New South Wales — 1 Pb-Zn-Au mine. 1,035 An estimate of the distribution follows in troy ounces: 7 major, primary gold operations (over 30,000 tr oz/yr). Includes 6 operations in Western Australia — 305,000 tr oz; 1 operation in Queensland — 70,000 tr oz. 21 primary gold operations (5,000 to 30,000 tr oz/yr). Includes 18 operations in Western Australia — 330,000 tr oz; 1 operation in Queensland — 10,000 tr oz; 2 other operations — 25,000 tr oz. 5 byproduct operations (no production criteria). Includes 1 operation in Northern Territories, 1 operation in New South Wales, 2 operations in Tasmania, 1 operation in Queensland. Very small, primary gold operations (<5,000 tr oz/yr.). Includes unknown number of operations, probably at least 35 in total, majority in Western Australia. 'Compiled from numerous sources and the authors' own estimates. 67 MINOR PRIMARY GOLD OPERATIONS None of the 21 minor operations producing in 1983 were in production prior to early 1981. Most of these operations came into production with only 5 yr or less of reserves and an average capacity of about 17,000 tr oz/yr. These opera- tions, which required little time to be brought into produc- tion, are mostly open pit operations, and many of them use heap leach processing methods. They are characterized by low capital and operating costs and are very flexible in terms of operational characteristics. Many of the operations consist of several small deposits feeding a central or custom mill; thus, it is virtually impossible to keep accurate account of production costs, reserves, and resources at these minor producers. To put the size of these minor primary producers in Australia into perspective, it is expected that by 1988, over 20 new operations of similar size will be required just to replace the 1983 production level represented by these minor, primary producers. Likewise, to add 350,000 to 400,000 tr oz of additional annual production from this category of operation would require that as many as 40 new operations be brought into production. It is believed that a gold price of $350/tr oz to $450/tr oz is a sufficient economic incentive for these minor, primary gold producers, but the generally small amount of resources available for treatment could be a major concern. It is possible, depending upon the price of gold, that these operations could account for 500,000 to 700,000 tr oz of production in any given year, but it would probably be difficult to sustain this level very long. MAJOR BYPRODUCT GOLD OPERATIONS The five major byproduct gold producers in Australia that have produced throughout the 1972-83 period are the Tennant Creek, Mount Chalmers, Broken Hill, Rosebery/Hurcules, and Mt. Lyell operations. At the Tennant Creek operations of Peko-Wallsend in the Northern Territories, overall production of gold declined from 208,000 tr oz in 1972 to 115,000 tr oz in 1983. However, the company felt, as of 1980-81, that it had a good chance of finding at least four or five ore bodies similar to the Warrego ore body (41, p. 141). It is believed that, if found, these would be utilized more as replacement mines to pro- long the life of the operations rather than to increase pro- duction dramatically. At present, the mine has 11 yr worth of proven reserves. The Mount Chalmers operation is a copper mine that produces 30,000 to 40,000 tr oz/yr gold. Estimated reserves at this long-time producer represent only about 2 to 3 yr worth of production. The other three major byproduct gold producers in Australia all have 10 to 30 yr worth of proven reserves. Based on the proven reserve situation described above, it is expected that the 1979-83 level of byproduct gold pro- duction of 200,000 to 220,000 tr oz/yr should be easily main- tained for 10 to 20 yr; however, unless new discoveries are made, there is very little room for major increases in byproduct production above that level. RESOURCE OVERVIEW, 1984 The level of demonstrated gold resources in Australia has been changing rapidly over the last few years as ex- ploration and development activity have intensified. As a consequence of this high level of activity, any estimate of gold reserves or resources will likely be outdated in a very short time. The Australian Government estimated "demonstrated economic resources" of gold as of December 31, 1979, at ap- proximately 9.0 million tr oz. As of December 31, 1980, this estimate was revised upward to 10.7 million tr oz. Total identified resources at that time were estimated at 14.0 million tr oz. An additional 4.8 million tr oz was classified as "contained in potential ore presently being developed." (43, p. 171). A summary of the aggregate resource data estimated for this study is listed in table 59. For this study, a 1984 demonstrated resource of 6.7 million tr oz of contained gold was estimated for the seven largest primary producers. In addition, seven primary lead-zinc and copper properties were estimated to contain approximately 2.5 million tr oz of byproduct gold. Also reported in this study are preliminary estimates of approximately 11.0 million tr oz of gold contained in 17 other new properties in various stages of exploration and development, or in producing mines that were too small to meet the established criteria for complete economic evaluation. These latter resource estimates are reported separately and are not included as "demonstrated" resources owing to the absence of sufficient data to enable a more complete geologic verification of the resource estimates. The Olympic Dam/Roxby Downs proj- ect and the Deep Leads project are listed separately owing to their very large estimates of around 60.0 million tr oz of potentially contained gold. Figures 35 and 36 show the location of major primary and byproduct gold properties in Australia, designated by production status and inclusion in the cost analyses. GOLD AVAILABILITY AND PRODUCTION COST EVALUATION: SEVEN MAJOR PRIMARY PRODUCERS The seven evaluated primary gold operations in total were estimated to contain approximately 64 million mt of demonstrated resources as of January 1984. Mill feed grades Table 59. — Aggregate gold resource data for selected Australian mines and deposits, 1 984 Classification Nurngff Major demonstrated primary producers 7 Other primary properties 1 17 Byproduct gold producers 7 Total 31 Possible major gold producers: Olympic Dam and Deep Leads 2 NAp Not applicable. ^or details on name, current development status, and estimated annual output, see table 61. Contained gold, 103 tr oz Recoverable gold, 10 3 tr oz 6,700 1 1 ,000 2,500 20,200 60,000 6,400 NAp NAp NAp NAp 68 INDIAN OCEAN INDIAN OCEAN Tasmon Sea Hobart TASMANIA LEGEND • Cities |yl Costed, producing gold mines ® Byproduct gold producers ® Possible gold producers, 1984-90 \%%\ Areas of Paleozoic exposures \\J Areas of Proterozoic environments fc-:?l| Archean environments of Western Australia I u 500 Seal*, km Figure 35. — Location of producing and prospective gold mining operations in Australia and areas of primary gold deposits. WILUNA TAILINGS "> / ^HORSESHOE LIGHTS ^SONS OF GWALIA BIG BEir MT. MAGNET X V'ULL 50-MORNING STAR) PAODINGTON BLACK HILLS NEVORIA PonniYRr- PERSERVERANCE PARINGA FIMIS10N LEASES MT CHARLOTTE Kalgoorlie LEGEND • Cities >? Costed, producing gold mines H Noncosted, producing gold mines X Possible gold producers 1984-90 500 I Figure 30. — Producing and prospective gold mining operations in Western Australia. 69 Tabic 60.— Summary rasults of 1084 long-run coat determination analyses for major Australian primary gold mines producing In 1083 Grand totals Entirely surface mining Primarily underground mining Ranoe To,al of Ranoe To,al or _ weighted average ™ weighted average Operational data: Mill feed grade g/mt... 1.0-5.0 3.6 4.0-12.0 7.0 Annual output lOHroz... 45-65 110 25-120 304 Total availability' 10>lroz... 400-1,100 1,491.2 225-1,600 4,876.3 Producing years from Jan. 1984 9-17 NAp 6-23 NAp Capital and operating cost data: Total capital investment! 10* dollars. . . $21-$51 $72.5 $30-$130 $357.3 Capital cost per troy ounce NAp $48 $45-$135 $73 Operating cost per troy ounce* $240-$310 $291 $150-$315 $222 Total operating plus capital cost per troy ounce $290-$360 $339 $192-$344 $295 Long-run total cost per troy ounce: Break-even (0-pct DCFROR) $325-5365 $353 $200-$477 $310 10pd DCFROR $340-5380 $368 $242-$S13 $345 NAp Not applicable 'Refined gold estimated to be recoverable as of Jan. 1984 •Unrecovered capital Investment In mine and mill plant and equipment. Infrastructure, and development remaining as ol Jan. 1984 and reinvestments through lite ol operation. •Mining plus milling cost per troy ounce of refined gold. NAp 414 6,367.5 NAp $4298 NAp NAp NAp NAp NAp were estimated to range from 1.1 g/mt (for a tailings reprocessing operation) to 11.6 g/mt. Total gold contained in the 64-million-mt resource is estimated at 6.7 million tr oz, with total recoverable refined gold estimated at approximately 6.4 million tr oz. Two of the seven primary producers are old (primarily underground) operations which closed in the 1970's due to low market prices and rising production costs. With the rapid increase in gold price in the late 1970's, these mines were reopened with the expectation that prices would re- main high or continue to increase. Recent decreases in the price of gold have again put the long-term economics of these two operations into question; both require long-term prices significantly above the $400 base price in order to break-even. Another two of the primary producers are also old (most- ly underground) operations that have remained in con- tinuous production throughout their lives and are highly profitable at $400 gold, and two of the evaluated operations represent new producers that have been in production 5 yr or less. Of these two "new" operations, one is a surface mine that remains profitable at current gold prices and one is an underground producer that is considered subeconomic at the current base price. Also included in this analysis is a tailings project that has been producing profitably for several years. Average annual output for the seven evaluated major primary gold operations ranges from 25,000 to 119,000 tr oz/yr. The seven operations should represent a combined output of 458,000 tr oz of gold pro- duction as of 1984. Table 60 presents summary data on the cost determina- tion analyses for the seven primary gold producers. This table details the relationship between the major economic cost-determining factors such as mill feed grade (a measure of resource quality), available annual production (a measure of output quantity or resource flow), total refined gold poten- tially recoverable (a measure of total product quantity or resource stock), the number of future producing years as of January 1984 (the time element), total capital investment to be recovered, and capital and operating costs per ounce of recoverable gold. In the case of the evaluated Australian surface mines, total capital investments to be recovered over the life of the operation range from $21 to $51 million, whereas in the case of the underground mines the range is from $30 to $130 million. However, owing to higher grade material at the underground mines, capital costs per ounce of recoverable gold are not significantly different. For the seven major primary gold producers herein evaluated, long-run total break-even production costs average $353/tr oz for surface production and $310/tr oz for underground production. Long- run production costs, which include a 10-pct rate of return on invested capital, average $368/tr oz for surface produc- tion and $345/tr oz for underground production. Clearly, an operator interested in quick payback and low total initial capital investments will favor surface prop- erties, especially small ones. This is why, as shown in table 58, a significant percentage of 1983 production is accounted for by some 21 small, primary gold operations, most of which are surface mines with typical outputs ranging from 5,000 to 30,000 tr oz/yr. On an individual mine basis, this analysis indicates that the three new or reopened major underground operations in Australia represented gambles in that gold prices of around $500 are required in order for all costs to be recovered while attaining a minimum 10-pct rate of return. In the late 1970's, when the development or redevelopment decisions for these operations were made, the long-term price outlook was much more optimistic than that of 1984. As was shown in the Canadian analysis, older produc- ing mines that have remained in continuous operation possess a number of economic advantages. They have recouped most or all of their initial capital investments and generally have higher grade ore and larger annual output capacities. For these older producing operations, the pro- duction decision is largely dependent upon operating costs, with capital investment recovery representing only a small part of total production costs. For example, at one new pro- ducing operation, approximately $132/tr oz gold produced is required to recover $29.5 million in capital investments over 9 yr, whereas one of the established producers requires only $71/tr oz gold produced to recover $127.3 million in capital investments over 15 yr. Separate analyses were performed in which long-run rates of return were determined for each of these seven mines given gold prices of $400, $500, and $600 per ounce, respectively, and a 10-pct discount rate. In all cases the price of byproduct silver production was held constant at $10/tr oz. The results indicate that at $400 gold, three of the seven operations have zero or negative long-run rates of return, i.e., these three operations require gold prices in excess of $400/tr oz for long-run profitability. Four of the seven mines (containing 76 pet of the 6.4 million oz of recoverable gold) are highly profitable at the base price of $400/tr oz. At $500 gold, six of the seven mines are profitable by the criteria of this analysis with long-run rates of return 70 Tatol* fl 1.— Australian deposits considered possible or probable gold producers during 1084-00 Province or Territory and operation Northern Territory: Pine Creek . New South Wales: GoonuaWa-Parkes Queensland: KkJston Mt Rawdon Red Dome (Mungana) . . South Australia: Olympic Dam-Roxby Downs Western Australia: Bamboo Creek Big Bell Black Hills Golden Crown Harbor Lights Horseshoe Lights Nevoria Paddinglon Paringa Porphyry-Perseverance Sons of Gwalia Wiluna Tailings Total NAp not applicable. 'Byproduct. Status as ol 1983-84 Feasibility study stage Explored deposit In development, Initial production 1985 Feasibility study stage do Feasibility study stage, development approval received 1984 In development Explored deposit Feasibility study stage In development do In production In development do In production Feasibility study stage In development In production NAp Estimated annual production lOMrgj 50 180 190 125 70 '105 30 45 40 35 100 30 25 80 35 30 20 30 Estimated first year of full production Post 1985. Post 1985. 1988-87. Post 1985. Post 1985. Post 1989. 1986. Post 1985. 1987. 1986. 1987. 1984. 1985. 1966. 1984. 1986. Post 1985. 1985. 1,120 NAp. exceeding the 10-pct discount factor. Long-run prices in ex- cess of $513/tr oz are required for all seven mines to operate profitably. These analyses indicate that the three new or redeveloped mines may not remain in production for the entire life of their demonstrated resource unless long-run gold prices remain above $400/tr oz. The other four opera- tions clearly demonstrate economic production at current 1984 prices. The seven evaluated producers each have sufficient cur- rent demonstrated resources to allow for a minimum of 8 yr of production assuming sufficient gold prices. Two of the seven are expected to produce until at least the turn of the century but will account for only around 100,000 tr oz of combined output in 2000. These latter two mines are low- cost producers and are economic at current prices. Clearly, for Australia to maintain production at current levels will require the continued discovery and development of new mines. The potential for new mine development is exam- ined in the following section. MAJOR PRIMARY AND BYPRODUCT GOLD DEPOSITS AWAITING OR UNDER DEVELOPMENT During the last few years, exploration and development activity in the Australian gold industry has been very dynamic. Many new exploration projects and development decisions have been reported in the literature in just the past 2 yr. Table 61 lists 18 major primary and byproduct gold deposits that have basically been fully explored and were in various stages of study or development as of 1983-84. Several of the properties have already produced small amounts of gold. The criterion for inclusion in this list was a minimum of 100,000 tr oz contained gold in the reported reserves. These deposits represent the largest con- tained gold reserves being reported for the entire country as of 1983-84. Of the 18 deposits, 16 are primary gold deposits and 2 are byproduct gold products. Of the 16 primary gold deposits, 12 are located in Western Australia, 3 are located in Queensland and 1 is located in the Northern Territories. Ten of the 12 primary gold deposits in Western Australia will likely produce at a level of at least 30,000 tr oz/yr with an average output of 50,000 tr oz/yr. It is estimated that the 12 deposits in Western Australia could account for 500,000 tr oz/yr of additional gold production if they all come into production during the same time period. It is more like- ly that by late 1986 or 1987, 9 of the 12 will be in produc- tion and could account for around 350,000 tr oz of additional Western Australian production. The four primary deposits located in Queensland and the Northern Territory are more questionable as to their possible development because they are all lower grade deposits in the range of 2.3 to 3.4 g/mt. As of late 1984, it appeared that one of the deposits would definitely be developed with another likely to be developed and the re- maining two definitely "on hold." The three deposits in Queensland would represent an estimated 385,000 tr oz of additional annual gold production if all were brought into production in the same time period. It must be emphasized that some of the individual prop- erty data included in this table must be considered speculative at this time. Within this list of 18 properties are two noteworthy mine developments, both proposed as surface mines, that appear to have a high probability of eventual production. One of these is the Harbour Lights pro- ject in Western Australia, which is estimated to contain up to 1 million tr oz of gold and could begin initial production by late 1985 (44, p. 230). The other project of note is the Kidston project in Queensland, estimated to contain around 2.2 million tr oz of gold, which could be in full production by late 1985; preliminary estimates for the first 5 yr of pro- duction average around 190,000 tr oz/yr (45, p. 199), which would make it Australia's largest producing mine. One potential byproduct operation that warrants atten- tion is the Olympic Dam/Roxby Downs primary copper pros- pect in South Australia. Olympic Dam is estimated to con- tain 2 billion mt of minable material grading 0.6 g/mt gold, or approximately 38.6 million tr oz of contained gold (46, p. 62). This deposit, if developed as presently planned, would have a very long life, probably in excess of 50 yr, and thus would not represent a major influence on annual Australian gold production since its annual gold output would probably be only slightly greater than 100,000 oz/yr. The develop- ment of Olympic Dam/Roxby Downs remains very much in doubt at this time, even though the project has received development approval. The project has some major technical and economic obstacles which must be overcome before any definitive development decision can be anticipated. 71 MINING METHODS AND OPERATING COSTS A unique characteristic of the major Australian gold pro- ducers evaluated in this study, when compared to opera- tions in other major gold producing countries, is the tenden- cy for the operations to be combinations of surface and underground mines. For example, of the seven primary gold producing operations evaluated for costs in this study, two are 100-pct surface mining operations, two are 100-pct underground operations, and three are combinations of underground and surface mining operations. The two opera- tions with 100 pet of their production by underground methods are large operations with annual ore capacities of 400,000 and 850,000 mt/yr; however, their ore grades are low (in the 4- to 5-g/mt range). The two operations with 100 pet of their production from surface mining methods con- sist of a conventional bench mining operation with annual ore capacity close to 500,000 mt/yr at a gold grade of over 7 g/mt, and a dredging operation that treats tailings from a prior milling operation at a rate of about 3.5 million mt/yr for a grade of slightly over 1 g/mt. The three operations utilizing a combination of surface and underground methods are fairly small, with total ore milling capacities ranging from 100,000 to 230,000 mt/yr and overall weighted-average mill feed grades of 4.7 to 10.8 g/mt. 10 Of the five operations with at least some underground mining, two are using sublevel stoping methods, one uses a combination of cut-and-fill and shrinkage stoping, one uses an overhead stoping method, and one utilizes cut-and-fill with mill tailings. As of 1982-83, hoisting depths at the Australian gold mines were relatively shallow, ranging from 120 m to more than 1,300 m. Daily ore capacities ranged from 120 to 2,800 mt/d with 5 to 15 pet waste rock also being hoisted. Stoping heights for the nonsublevel stoping operations ranged from 1.8 to 3.0 m, while the sublevel stoping operations were dealing with stope heights of 3 to 10 m on average. Estimated underground mining productivities (including supervision) range from 4 to 15 mt per worker-shift. Estimated underground mining costs range from $15/mt - $20/mt to $50/mt - $60/mt. Labor costs are the single most important item in the estimated mining costs, representing 47 to 65 pet of the total. It is difficult to categorize the major surface mining operations producing gold in Australia. As of 1981-82, an- nual ore capacities of the evaluated surface mines showed a wide range from 18,000 to 480,000 mt/yr, with waste-to- ore ratios ranging from zero to 40 mt waste per ton of ore. The larger pits are generally planned to bottom at a depth of 60 to 120 m. These depths presently appear to be the limit of any of the producing or planned surface gold mines in Australia. Most of the mining of overburden and ore in Australia is done by contractors since most of the surface mining operations begin with short projected mine lives and the reserves necessary to recoup major investments in min- ing equipment are not present. This situation, combined with relatively small capacities, fairly lengthy ore haulage distances to mills, and the somewhat remote locations of the operations, results in surface mining costs that are significantly higher ($1.50/mt to $1.90/mt material moved) then those at major U.S. surface gold mines. Direct labor costs are estimated to account for 25 to 35 pet of the total operating cost for the Australian surface mines evaluated in detail in this study. 10 Underground ores grade 5.0 to 13.6 g/mt, while surface ores grade from 1.5 to 6.0 g/mt. The proportions of surface and underground material range from 90 pet underground and 10 pet surface to 62 pet underground and 38 pet surface. METALLURGICAL METHODS AND OPERATING COSTS Of the seven major primary gold operations evaluated, six are treating primary ore and one treats reclaimed tail- ings from a prior sulfide flotation operation. The nine types of ore at these seven operations are treated in six mills. Two of these mills are using the carbon- in-pulp method of extracting gold from the solutions pro- duced by cyanide leaching, while four use the conventional Merrill-Crowe method of extracting the gold from leach solu- tions. All of the mills treating primary ore contain a gravi- ty separation-amalgamation circuit in closed circuit with their grinding stage to recover the coarse free gold in their ores, which is estimated to range from at least 10 pet to as much as 50 pet of total gold at individual operations. Of the nine basic ore types involved, two are refractory ores and require roasting of flotation concentrates followed by cyanide leaching of the calcine resulting from roasting. Two other ore types are primary sulfide ores requiring pro- duction of a flotation concentrate which is then leached with a cyanide solution. The remaining ore types include two completely oxidized ores, two that are classified as "primary ores," and one consisting of sulfidic tailings. The operations that treat refractory ores are interesting in that they con- duct a cyanide leach on the flotation concentrates prior to roasting and then use a second cyanide leach of the calcine after roasting of the concentrates. One of these operations then goes on to include a third cyanide leach, this time leaching the tailings from flotation. Mill recoveries (total gold recovery) when treating nonrefractory, primary ores are high, ranging from 90 to 97 pet. The best recovery is at operations treating oxidized ore with high free gold content. Recoveries when treating refractory ores are lower, in the range of 83 to 90 pet. The recovery for the tailings retreatment operation is very low at 40 to 45 pet, basically reflecting the sulfidic nature of the original ore and the low grade of the tailings. A low recovery of only 83 pet is also indicated at one of the opera- tions which treats a clayey, sulfide ore along with two dif- ferent primary gold ores. The product of all of these mills is in the form of dore bullion, which is produced by smelting the various gold-containing precipitates (steel wool with electrowinning, "sponge gold" from retorting of amalgams, and zinc dust precipitate from the Merrill-Crowe process) with fluxes. These dore bullions are cast into bars or but- tons in the 800- to 1,200-tr-oz range. The dore bullion from the operations analyzed ranges from 55 pet gold and 35 pet silver to 92 pet gold and 4.5 pet silver, the remainder is com- posed of base metals such as copper, lead, zinc, and iron. All of the dore bullion from the major operations was as- sumed to be transported by airfreight to one of the two ma- jor gold refineries in Australia. These refineries are located at Perth in Western Australia and at Melbourne in Victoria Province. For the mills treating primary, nonrefractory ores, estimated operating costs (including smelting to dore bullion) range from about $9/mt to $19/mt ore treated. The lower cost operations (below $15/mt) reflect higher tonnage operations (above 200,000 mt/yr). Generally, those opera- tions treating refractory ores incur additional costs in the range of $3/mt to $5/mt which represent mainly the addi- tional costs of flotation and roasting. It must be remembered, however, that this is simply a general trend noticed from comparison of the operations involved in this study. It is difficult to generalize about costs in the 72 Australian gold mining industry, especially when including costs in the administrative and general categories, since these categories reflect a variety of circumstances from relative remoteness, to physical layout of the operations, to company policy. A few final points should be made regarding gold metallurgy in Australia. First, there were no large-scale heap leaching operations in Australia as of the 1983-84 period, although there were several small heap leaching operations with maximum capacity in the 20,000- to 30,000-tr-oz/yr range. Second, the operating costs developed in this study apply only to the larger scale milling opera- tions (above 150,000 mt/yr ore feed) which are using vat leaching techniques. Third, several of the major gold mills in Australia evaluated in this study set aside some portion of their capacity for custom milling of ores from small pro- ducers in the vicinity of the mill. REFINING AND TRANSPORTATION As of 1981, four primary gold refineries were accepting gold bullion in Australia: one in Victoria Province, one in Western Australia, and two in New South Wales. The refinery in Victoria is located near Melbourne and has an estimated capacity to produce 160,000 tr oz refined gold and 320,000 tr oz refined silver per year. The refinery in Western Australia is operated by the Perth Mint and is located in the city of Perth. It has an annual capacity to refine about 600,000 tr oz gold and 65,000 tr oz silver. In New South Wales, Matthey Garrett Pty. Ltd. has a refinery at Kogarah capable of producing 320,000 tr oz/yr gold and 640,000 tr oz/yr silver. The fourth gold refinery accepting bullion was Broken Hill Associated Smelting Pty. Ltd.'s Port Pirie complex in South Australia, which has the capability to produce 6,500 tr oz refined gold and 900,000 tr oz refined silver per year, mostly from its own lead con- centrates. A fifth (nonprimary) gold refining facility is located at Port Kembla in New South Wales, where Elec- trolytic Refining and Smelting Co., Ltd., produces about 65,000 tr oz gold and 450,000 tr oz silver per year from treat- ment of anode slimes resulting from copper refining. The above capacities are circa 1981-82 and represent a total capacity to produce around 1.2 million tr oz refined gold and 2.4 million tr oz refined silver per year. The ma- jority of the major, primary gold producers evaluated in this study send their bullion to the Perth Mint's refinery in Western Australia. A Government organization, the Gold Producer's Association (GPA), handles the insurance, refin- ing, and marketing of gold bullion being sent to the Perth Mint. Although the GPA takes responsibility for shipment of the bullion from the mine site to the Perth Mint, the transport charge is paid directly by the mines. It appears that as of 1984, Australia's total capacity for gold refining was being pressured by increasing mine production levels. BRAZIL HISTORICAL PERSPECTIVE Beginning in the late 17th century and continuing in- to the late 18th century, Brazil was the largest producer of gold in the world. Mohide (4, p. 129) reports an estimated production from 1691 through 1780 of about 24 million tr oz with 9.4 million tr oz being produced during 1741-60. Most of the production during the 18th century came from placer deposits in the southern part of the country (fig. 37). With placer production beginning to decline in the late 18th century, the Morro Velho Mine in Minas Gerais was developed in 1834. Since that time, this mining operation has represented the basic core of Brazil's reported (registered) gold production up to the late 1970's. Also dur- ing this same time period, Brazilian gold production has been consistently eclipsed by discoveries and developments in the United States, Australia, South Africa and Canada, so that as of the mid-to-late 1970's, Brazil was the 11th to 13th ranked largest producer in the world, as shown in table 62. Several developments since the late 1970's have re- quired a reexamination of Brazil's position in the world gold mining industry. As shown in table 62, estimates of Brazil's gold production show a tremendous fourfold increase be- tween 1979 and 1980 with the higher levels of production continuing through 1984. This increase in production has elevated Brazil into the ranks of the top six producers in the world, following South Africa, the Soviet Union, and Canada, and close behind or essentially equal to the Peo- ple's Republic of China and the United States. The inter- related factors that have caused or reflect this increase in Brazilian gold production are believed to be — 1. The increased U.S. dollar price of gold. 2. Increased access to the Amazon Basin region through improved infrastructure and modes of transport (e.g., helicopters). 3. The discovery of the Serra Pelada gold deposit. 4. Improved cruzeiro payment terms for Government purchases of gold output by small miners in the Amazon Basin. 5. Poor economic conditions leading to dislocation from normal industries. These five factors have had two major effects on estimates of gold production in Brazil. First, all five of these factors have been responsible for what is essentially a gold rush into the Amazon Basin. This gold rush is similar in many ways to those in past world history, but it involves a geographic area of much greater size. Second, the fourth factor, improved cruzeiro payment terms by the Brazilian Government, has succeeded in attracting more of the out- Table 62. — Estimated Brazilian gold production and world ranking, 1975-84 Y Total production, 1 World ^[ 10 3 tr oz ranking 1975 172 13 1976 240 12 1977 280 11 1978 301 11 1979 320 11 1980 ' 1,300 4 1981 1,200 6 1982 1,500 8 1983 1,600 6 1 984 2 2,120 4_ 'Compiled from U.S. Bureau of Mines Minerals Yearbooks (20): 1976, 1980, 1982. All figures differ substantially from those appearing in latest available official Brazilian sources due to the inclusion of estimates for unreported pro- duction by small-scale operations (garimpeiros). Officially reported figures for major mines in 10 3 tr oz follow: 1978-129; 1979-107; 1980-131 (revised); 1981-415. Estimated (47, p. 2). 73 1 1 1 . 1 1 1 ii i - 1 ■■.■yy//.XaribDeany^^;.£.$ea:yYs.-y PfiNAM^X... VENEZUELA GUYANA l%^£surinam% FRENCH kGUIANA-" : :;_ : 'x : : v ;: .-:-: :: . : : ; &&■■■'■■■■■ ■ywjk^-y >. : . : . : : : : :::::: :':-:': : xv: : x ; :: 53 LEGEND City or town © Morro Velho Operations Areos of Preeambrian shield © Jacobina exposures ® Serra Pelodo Areos of port production L i ■ i 500 1 Scale, Km Figure 37. — Selected major nonalluvial gold mining operations and areas of primary gold deposits in Brazil. put from workings in the Amazon Basin into the category of "registered" gold production, rather than being smug- gled out of the country or not reported, as was prevalent in the past. This has led to higher reported production figures. According to the Finance Ministry of Brazil, so- called clandestine gold mining operations have dropped from 83 pet of total Brazilian production in 1979 to 30 pet in 1983 (48 p. 281). Some idea of the size and intensity of this Amazon Basin gold rush can be obtained by noting that the Serra Pelada deposit was discovered in February 1980. By September 1981, it was estimated that 50,000 garimpeiros (individual miners) were working Serra Pelada at its highest level of activity (49, p. 24). In a recent article in the Wall Street Journal, it was stated that 250,000 garimpeiros were operating in the entire Amazon Basin area, up from 97,000 3 yr earlier (50, p. 33). SOURCES OF PRODUCTION AND PRODUCTION COSTS Brazilian gold production for 1983 has been cited by several sources as 1.51 million tr oz, an increase of about 84 pet over 1982 production of about 803,000 tr oz. Expec- tations for 1984 were that 2.12 million tr oz would be pro- duced (48, p. 281); however, the latest available figures in- dicate that gold production for 1984 will be slightly less than expectations at about 2.025 million tr oz (50, p. 33). Fully 85 pet of 1984 production will represent production from garimpeiros, with the remaining 15 pet being produced by organized companies. By comparison, in 1982 around 66 pet came from garimpeiro operations, while 34 pet came from organized mining companies. Thus, basically all of the in- crease in Brazilian gold production since 1982 has come from garimpeiro operations. Table 63 categorizes the major sources of 1984 Brazilian gold production according to which operations are organized and which are not and also as to which are underground, hardrock operations and which are alluvial-eluvial-fluvial operations. As shown, only two primary gold operations, Morro Velho and Jacobina, are organized, company-run underground, hard rock operations and they account for on- ly 209,000 tr oz (10 pet) of expected 1984 production. Less than 1 pet (16,000 tr oz) is produced from surface and underground copper operations conducted by Caraiba Metais. The remaining 1.895 million tr oz, or 89 pet, is pro- duced from alluvial-eluvial-fluvial operations, 95 pet of which is produced by garimpeiro operations. Because so much of Brazil's gold production represents small, nonmechanized production by tens of thousands of garimpeiros mining alluvial-eluvial-fluvial gold deposits, 74 Table 63. — Expected 1984 Brazilian gold production, by operation and type of mining 1 Expected 1984 production mt 10 3 troz Type of mining Company operations: Morro Velho Jacobina Paranapanema Dragagem Fluvial Caraiba Metais Other companies Subtotal Garimpeiro operations: Tapajos River Cumaru River Mato Grosso State Rio Madeira Gajas State Belero area Serra Pelada Amazonas State Maranhao State Roraine River Amapa State Others Subtotal Grand total 'Compiled from various sources and the authors' own estimates. 5.0 1.5 1.0 .6 .5 1.4 161 48 32 19 16 45 10.0 321 15.0 482 10.0 322 10.0 322 5.0 161 4.5 145 3.5 113 3.0 97 1.4 45 1.0 32 .6 19 .5 16 1.5 48 56.0 1,802 66.0 2,120 Hard rock (underground). Do. Alluvial-eluvial-fluvial surface). Do. Byproduct-hardrock (surface-underground). Unknown. NAp. Alluvial-eluvial-fluvial (surface). Do. Do. Do. Do. Do. Do. Do. Do. Do. Do. Unknown. NAp. NAp. it is impossible to analyze the majority of Brazil's gold pro- duction as to development schedules and production costs. Production levels vary greatly at these types of operations, and production costs in the normal sense of the term are meaningless. Therefore, an availability analysis that at- tempts to relate an expected total or annual level of gold production in Brazil to a long-term total production cost can address only about 10 pet of Brazil's 1983-84 production. It is not possible to predict annual production, much less total potential availability, from the garimpeiros' opera- tions. The long-term production cost of gold at these opera- tions is essentially zero because the gold is produced in a more or less subsistence economy (i.e., no alternative form of employment) by semimechanized or manual methods in- volving very low-cost capital equipment such as motors to pump water and gravel. Some operations do use heavier equipment such as front-end loaders and trucks; this "mechanized" mining is technically illegal, according to the Government, but will probably increase in the future. The only cost that the Government incurs for that portion of the garimpeiros' production that is purchased is the marginal cost of printing new cruzeiros, which is also essentially zero. From a long-term availability perspective, the chief concern is that this high-grade, small-scale mining by the garimpeiros may render many deposits uneconomic to later mechanized mining, thus rendering some portion of the ultimate gold potential of the Amazon unavailable. The Morro Velho underground mine and the Serra Pelada surface operation are estimated to contain approx- imately 10 million tr oz total gold. Morro Velho is an economic gold producer at a gold price of $400/tr oz. Based upon this study's estimate of demonstrated resources and current output levels, Morro Velho is expected to produce past the turn of the century. The Serra Pelada operation, currently being mined by tens of thousands of garimpeiros, was evaluated as if it were operated as a large-scale, mechanized surface mine. The analysis under this scenario determined a long-term total production cost of less than $100/tr oz. The Government of Brazil has expressed a desire to convert Serra Pelada to a large-scale, mechanized operation run as a company endeavor, but has not pursued this option owing to the prob- lem of having to deal with tens of thousands of displaced garimpeiros. POSSIBLE NEW SOURCES OF PRODUCTION BEYOND 1984 With the spectacular increases in gold production that have occurred so far during the 1980's in Brazil, it is natural to speculate on how much higher Brazil's annual produc- tion level could go. As can be seen from the widely varying estimates of table 64, this is a difficult question to answer. This table presents different estimates of future Brazilian gold production to 1990 and the "best" estimates of actual gold production for 1980-84. The 1980 Brazilian Government estimate of expected production in 1984 of 5.048 million tr oz has proven to be much too optimistic; estimated actual production was 2.12 million tr oz. It is difficult to say how much of the shortfall from the expectations of 1980 was due to the nonimplemen- tation of the plans that the estimate was based upon, and how much was due to operational or economic factors. A comparison of the 1980 estimates with actual results should make one cautious about the two circa 1983-84 estimates of possible production levels in 1990. As shown, the Brazilian Government has mentioned a goal of producing nearly 13 million tr oz/yr by 1990 which would probably make Brazil the second largest gold producing country in the world, depending upon which estimate for the Soviet Union is used. The estimate for the same period by Mining Journal Research Services is much more conservative at an expected production of slightly more than 4 million tr oz/yr by 1990 and is probably more realistic since it only reflects a doubling in gold production between 1984 and 1990. Even this lower estimate of gold production would still make Brazil the third-ranked gold producing country in the world, behind only South Africa and the Soviet Union. For the near term, the Gold Institute predicts total Brazilian production of 2.581 million tr oz in 1985, a 22-pct increase over estimated 1984 production. 75 Table 64. — Varying estimates of actual and potential Brazilian gold production in selected years, thousand troy ounces Best estimates of Bra ~i a ^f V ?a«n ent Gold lnsti,u * .** .^Mfc ^ : |«* \/ ^ V* •»& \/ • ,4> ,o"e. <*> ,v . * v . ■A . . . «Z»_ A v ... A • 4* ** o v fe. °* i0 Y • !.•• * o * r « » * ' <^, ./ \ W^*\"W/\w.A '-Sr.- /\ wj$s ^% y.^%X .^.JiUkrS .y.-^.V /..^fc.% c u * ^... V^V \^-\/ v^*/ V^\^ • ^sfe-. V>* •^S&. ^ ^ •4fe'- ^ / ♦ y ». J .H^; ^^ ■ liKHli JHtWlfBiWlnnMlSraBli lUiHIf mSmsmsst UBRARVOFCONGRES" ™"I HRf! Iflfl I •';. JHj • $#ll* | 8 j i| l 8 |j j