As Of                Filer                Filing    For·On·As Docs:Size              Issuer               Agent

 3/09/09  Iamgold Corp                      6-K         3/09/09    2:2.4M                                   Merrill Corp-MD/FA

Document/Exhibit                   Description                      Pages   Size 

 1: 6-K         Current Report by a Foreign Private Issuer          HTML     17K 
 2: EX-99.1     Technical Report on Tarkwa Gold Mine Dated July 1,  HTML   1.12M 
                          2004                                                   

Exhibit 99.1

 

An Independent Technical Report on the Tarkwa Gold Mine, Ghana

 

 

Report Prepared for

 

Gold Fields Limited and IAMGold Corporation

 

Effective Date:		July 1, 2004
Issue Date:		October 25, 2004

 

 

Prepared under the Guidelines of National Instrument 43-101 and
accompanying documents 43-101.F1 and 43-101.CP

 

 

Report Prepared by

SRK Consulting

3rd floor Windsor Court

1-3 Windsor Place

Cardiff, UK

CF10 3BX

 

Tel : +44 29 20 34 81 50

Fax : +44 29 20 34 81 99

cardiff@srk.co.uk

www.srk.co.uk

 

ISSUED OCTOBER 2004

 

 

	14.3 Environmental Management						55
	14.4 Existing Environment						56
	14.5 Socio-Economic						56
	14.6 Discharge Water Quality						56
			14.6.1			General	56
			14.6.2			Acid mine drainage	57
	14.7 Noise						57
	14.8 CIL Plant						57
	14.9 Tailings Storage Facility						57
	14.10				Rehabilitation and Decommissioning		58
		14.10.1				Rehabilitation	58
		14.10.2				Decommissioning	58
		14.10.3				Reclamation Bond	59
15	HUMAN RESOURCES AND SAFETY						61
16	TECHNICAL ECONOMIC INPUT PARAMETERS						62
	16.1 Introduction						62
	16.2 Basis of the Technical-Economic Input Parameters						62
	16.3 Technical-Economic Parameters						63
	16.4 Special Factors and Operational Risks						63
	16.5 General Risks and Opportunities						63
17	ASSET VALUATION						65
	17.1 Introduction						65
	17.2 Basis of Valuation						65
	17.3 Limitations and Reliance on Information						65
	17.4 Valuation Methodology						66
	17.5 Post-Tax — Pre-Finance Cash Flows						66
	17.6 NPV Sensitivities						68
18	CONCLUDING REMARKS						69

 

List of Figures
Figure 3.1: Locality Map	10
Figure 4.1: Mine Infrastructure	13
Figure 6.1: Regional Geology of Southwest Ghana and Location of Tarkwa	18
Figure 6.2: Schematic cross section from west to east through the principal orebodies	21
Figure 10.1: Monthly Tonnage Reconciliations July 2002 — May 2004	31
Figure 12.1: Schematic of the Heap Leach Process	48
Figure 12.2: Schematic of the CIL Process	49

 

 

	SRK Consulting
	Windsor Court
	1-3 Windsor Place
	Cardiff
	United Kingdom
	CF10 3BX
	e-mail: cardiff@srk.co.uk
	URL: www.srk.co.uk
	Tel: + 44 (0)29 20 34 81 50
	Fax: + 44 (0)29 20 34 81 99

 

Issued: October 25, 2004

 

An Independent Technical Report on the Tarkwa Gold Mine, Ghana

 

1              EXECUTIVE SUMMARY

 

The Tarkwa gold mine (“Tarkwa”) is operated by Gold Fields Limited (“Gold Fields”) the majority share holder, with 71.1% of the issued shares of the owner Gold Fields Ghana Limited (“GFGL”). The proposed transaction means the acquisition by IAMGold Corporation (“IAMGold”) of the Acquired Interests from Gold Fields and its affiliates and the issue by IAMGold of the Consideration Shares to Gold Fields and its affiliates as consideration thereof, resulting in the acquisition by IAMGold of all of the interests of Gold Fields in certain of its subsidiaries which collectively hold all of the mining assets of Gold Fields located outside of Southern African Development Community (“SADC”).

 

Tarkwa is located in south western Ghana approximately 300 km by road west of Accra, the capital, at a latitude 5°15’N and longitude 2°00’W. Tarkwa is located 4 km east of the town of Tarkwa with good access roads and an established infrastructure. The mine is served by a main road connecting to the port of Takoradi, some 140 km to the southeast on the Atlantic coast.

 

Tarkwa operates under mining leases with a total area of 20,700 hectares (“ha”). Five leases relate to the Tarkwa property and are dated April 18, 1997 (expiring April 17, 2027) and two to the Teberebie property dated February 2, 1988 and June 18, 1992 (expiring 2018).

 

Since 1999 all mining has been from open pit extraction, following the closure of uneconomic underground operations. Operations were expanded after August 2000 with the acquisition of Teberebie, increasing the total mining rate from 15.3 million tonnes per annum (“Mtpa”) to 36 Mtpa and the heap leach production capacity from 7.2 Mtpa to 12.6 Mtpa. Until 2004 mining was carried out by a contractor. Following feasibility studies, mining is now on an owner operated basis and processing utilises a conventional Carbon-in-Leach (“CIL”) (4.2 Mtpa capacity) as well as a heap leach (14.4 Mtpa capacity) operation.

 

Steffen, Robertson and Kirsten (UK) Ltd.	Registered Address:	Offices in:
	21 Gold Tops,	Australia
Registered in England and Wales	Newport,	North America
Reg. No. 1575403	Gwent.	Southern Africa
	NP9 4PG	South America
		United Kingdom

 

 

2              INTRODUCTION AND TERMS OF REFERENCE

 

2.1           Terms of Reference

 

Steffen, Robertson and Kirsten (UK) Limited (“SRK”) is a subsidiary of the International group holding company, SRK Global Limited (the “SRK Group”). SRK has been commissioned by the directors of Gold Fields prepare an Independent Technical Report (“ITR”) on Tarkwa in support of the proposed transaction between Gold Fields and IAMGold.

 

This ITR has been prepared in accordance with the rules and companion policies of the Ontario Securities Commission (“OSC”) enacted by Section 143 of the Securities Act, specifically:

 

·        National Instrument 43-101 Standards of Disclosure for Mineral Projects (“NI 43-101”);

·        Form 43-101F1 (the “Form”); and

·        Companion Policy 43-101 CP (the “Companion Policy”).

 

In accordance with the applicable Rules and Companion Policies, this ITR has been prepared under the direction of the Qualified Persons (“QPs”) who assume overall professional responsibility for this ITR. The ITR however, is published by SRK, the commissioned entity, and accordingly SRK assumes responsibility for the views expressed herein. Consequently with respect to all references to QPs and SRK: ‘all references to SRK mean the QP and vice-versa’.

 

The ITR principally comprises a technical-economic appraisal of Tarkwa and has been prepared in accordance with the rules and companion policies stated above, which cover the standards of disclosure for mineral projects under the OSC guidelines which came into effect on 1 February, 2001. It is the recognised format for reporting on the Toronto Stock Exchange.

 

2.2           Purpose of the ITR

 

Tarkwa is operated by GFGL and Gold Fields is the majority share holder, with 71.1% of the issued shares of GFGL. The proposed transaction means the acquisition by IAMGold of the Acquired Interests from Gold Fields and its affiliates and the issue by IAMGold of the Consideration Shares to Gold Fields and its affiliates as consideration thereof, resulting in the acquisition by IAMGold of all of the interests of Gold Fields in certain of its subsidiaries which collectively hold all of the mining assets of Gold Fields located outside of Southern African Development Community (“SADC”).

 

In this regard Gold Fields will transfer to IAMGold ownership of its interests in:

 

·                  Orogen Holding (BVI) Limited (“Orogen”): Orogen holds Gold Fields indirect interests of: 100% of St. Ives Gold Mine in Western Australia (“St. Ives”); 100% of Agnew Gold Mine in Western Australia (Agnew); 80.7% of Cerro Corona development property in Peru (currently under option); 100% of Arctic Platinum Project in Finland; and a

 

3

 

portfolio of other exploration properties and investments;

·                  Gold Fields Ghana Holdings Limited (“GF Ghana Holdings”): GF Ghana Holdings holds Gold Fields indirect interests of a 71.1% interest in and claim on loan account against GFGL, which owns and operates Tarkwa and a 71.1% interest in Abosso Goldfields Limited (“AGL”), which owns and operates the Damang Gold Mine (“Damang”) in Ghana; and

·                  Gold Fields Guernsey Limited, which include a portfolio of exploration properties and investments.

 

For the transfer of ownership Gold Fields will be issued with 351,690,218 IAMGold Shares in addition to a number of IAMGold Shares equivalent to the net cash contributions made by Gold Fields into the Acquired Companies between 24 June 2004 and the Completion Date.

 

SRK has been informed that Gold Fields and IAMGold have signed definitive agreements to implement the Transaction on 29 September 2004 subject to satisfaction of certain conditions precedent.

 

2.3           Scope of Work

 

The scope of work undertaken by SRK comprised a technical and economic review of Tarkwa’s Life of Mine (“LoM”) Plan dated June 30, 2004 and covering the period 2005 — 2028. The review covered the following technical disciplines:

 

·                  Geology;

·                  Mineral Resource Estimation;

·                  Mineral Reserve Estimation;

·                  Planning and Production Scheduling;

·                  Mining Engineering and Open-Pit Design;

·                  Geotechnical Engineering;

·                  Mineral Processing and Metallurgy;

·                  Tailings Engineering and Waste Disposal;

·                  Environmental and Water Management;

·                  Capital Expenditure and Operating Costs; and

·                  Technical Economic Modelling.

 

2.4           Sources of Information and Site Inspection

 

The ITR has been prepared based on information provided to and taken in good faith by SRK by the management of Gold Fields and, specifically, Tarkwa. SRK has not independently verified by means of re-calculation the underlying data, however SRK has:

 

·                  undertaken inspection visits to surface, processing facilities, surface structures and associated infrastructure at Tarkwa during August and September 2004;

·                  held discussion and enquiry following access to key personnel based at the individual site operations and at head office;

·                  reviewed and, where considered appropriate by SRK, modified the Tarkwa Mineral Resources and Mineral Reserve estimates dated June 30, 2004;

 

4

 

·                  reviewed the Tarkwa LoM plan, supporting documentation and the associated technical-economic projections (“TePs”), including assumptions regarding future operating costs, capital expenditures and gold production; and where considered appropriate by SRK modified these; and

·                  examined historical information and results provided by GFGL in support of, in particular, the forecasts contained in the LoM plan and two year budgets.

 

Key documentation provided to SRK in support of this ITR and as generated by Gold Fields is:

 

·                  the Competent Persons Report for Mineral Resources and Mineral Reserves, June 2004; and

·                  a Technical Report on Tarkwa Gold Mine, Ghana, May 2003.

 

SRK has satisfied itself that such information is both appropriate and valid for valuation as reported herein. SRK considers that with respect to all material technical-economic matters it has undertaken all necessary investigations to ensure compliance, both in terms of level of investigation and level of disclosure. In doing so, SRK has not reproduced the information provided to it by Gold Fields without due consideration or appropriate modification.

 

Where fundamental base data has been provided (LoM plans, capital expenditures, operating budgets etc) for the purposes of review, SRK recognises the requirements of 43-101F1 Item 25 and accordingly states that SRK has performed all necessary validation and verification procedures deemed appropriate in order to place an appropriate level of reliance on such information.

 

Dr John Arthur, a QP and Principal Geologist with SRK, and Mr. Rick Skelton, a QP and Principal Mining Engineer with SRK, visited Tarkwa during August 2004. In addition site visits have been undertaken by other SRK team members, qualified in disciplines other than geology and mining, during the period of September 2004. SRK representatives have held discussions with Gold Fields personnel and examined the geology, mineral resources, laboratory facilities, exploration methodology and potential, processing facilities, mining operations, slope stability, groundwater issues, infrastructure, terrain, tailings facilities and environment.

 

2.5           Qualifications of Consultant and Qualified Persons

 

The SRK Group comprises 500 staff, offering expertise in a wide range of resource engineering disciplines. The SRK Group’s independence is ensured by the fact that it holds no equity in any project and that its ownership rests solely with its staff. This permits the SRK Group to provide its clients with conflict-free and objective recommendations on crucial judgment issues. The SRK Group has a demonstrated track record in undertaking independent assessments of resources and reserves, project evaluations and audits, technical reports and independent feasibility evaluations to bankable standards on behalf of exploration and mining companies and financial institutions worldwide. The SRK Group has also worked with a large number of major international mining companies and on their projects, providing mining industry consultancy service inputs.

 

5

 

This report has been prepared based on a technical and economic review by a team of consultants principally sourced from the SRK Group’s office in the United Kingdom. These consultants are specialists in the fields of gold geology, mineral resource and mineral reserve estimation and classification; mine engineering and pit modelling, geotechnical engineering; gold processing, hydrogeology and hydrology, tailings management, infrastructure, environmental management and mineral economics.

 

Neither SRK nor any of its employees and associates employed in the preparation of this report has any beneficial interest in Gold Fields or its assets. SRK will be paid a fee for this work in accordance with normal professional consulting practice.

 

The individuals who have provided input to this ITR, who are listed below, have extensive experience in the mining industry and are members in good standing of appropriate professional institutions.

 

Dr John Arthur:		Ph.D, C.Eng, CGeol, MIMMM;
Mr Rick Skelton:		MSc, C.Eng, MSAIMM, MIMMM;
Dr Neil Holloway:		Ph.D, C.Eng, FIMMM;
Dr Ian Brackley:		Ph.D, C.Eng, FSAIMM, MIMMM;
Mr Keith Philpott:		MSc, BSc, C.Geol, Eur Geol, FGS;
Mr Lee Barnes:		MSc, C.Eng, MIMMM.

 

In compliance 43-101F1 Part 5 subsection 5.1 this ITR has been prepared by following Qualified Persons:

 

·                  the Qualified Person with overall responsibility for this independent technical report and the reported Mineral Reserve is Mr Rick Skelton who is an employee of SRK. Mr Skelton is a mining engineer with 35 years’ experience in the mining industry and has supervised numerous due-diligence reviews and various technical studies on similar gold mining assets; and

·                  the Qualified Person with overall responsibility for reported Mineral Resource is Dr John Arthur who is an employee of SRK. Dr Arthur is a resource geologist with 16 years experience in the mining industry and has been responsible for the reporting of Mineral Resources on various gold mining properties, specifically specialising in resource estimation and classification.

 

SRK has previously carried out a number of consultancy commissions for Gold Fields at Tarkwa and these are listed in Appendix B to this report.

 

2.6           Monetary Rates and Units

 

All monetary units, unless otherwise stated, are quoted in US dollars (“USD”).

 

All units conform to metric usage except where otherwise stated. Where gold weight is not expressed in metric units it is expressed in troy ounces (“oz”) where one troy ounce equals 31.10348 grammes (“g”).

 

6

 

2.7           Technical Reliance

 

SRK places reliance on Tarkwa’s QPs that all technical information provided to SRK as at July 1, 2004, is both valid and accurate for the purpose of compiling this ITR. The Tarkwa Mineral Resource and Mineral Reserve estimates have been prepared under the direction of Mr. G. S. G. Chapman, Mineral Resource Manager at Tarkwa who is a full time employee of Gold Fields and has 28 years experience in the gold mining industry and is registered with the South African Council for Natural Scientific Professions (“SACNASP”).

 

2.8           Legal Reliance

 

SRK has placed reliance on representatives of Gold Fields that the following legal aspects pertaining to Tarkwa are correct as at July 1, 2004:

 

·                  that “a statement by the Directors of any legal proceedings that may have an influence on the rights to explore for minerals, or an appropriate negative statement” has been included in the body of the various circulars relating to the proposed transaction;

·                  that the legal ownership and title of all mineral and surface rights has been verified; and

·                  that no significant legal issue exists which would affect the likely viability of a project and / or the estimation and classification of the Mineral Resources and Mineral Reserves as reported herein.

 

2.9           Accounting and Financial Reliance

 

In consideration of all financial aspects relating to the valuation of Tarkwa, SRK has placed reliance of the Financial Officers of Tarkwa that the following information is accurate as at July 1, 2004:

 

·                  unredeemed capital balances;

·                  assessed losses;

·                  opening balances for debtors, creditors and stores; and

·                  working capital and taxation logic.

 

In generating the valuation of Tarkwa, SRK has relied upon the commodity price and macro economic forecasts as included in Section 17, which have been generated by NM Rothschild & Sons (New York) (“Rothschild”), Financial Advisors to Gold Fields.

 

2.10         Warranties and Limitations

 

SRK’s opinion is effective July 1, 2004 and is based on information provided by Tarkwa and Gold Fields throughout the course of SRK’s investigations, which in turn reflect various technical-economic conditions prevailing at the time of writing. These conditions can change significantly over relatively short periods of time and as such the information and opinions contained in this report may be subject to change.

 

The achievability of LoM plan, budgets and forecasts are neither warranted nor guaranteed by SRK. The forecasts as presented and discussed herein have been proposed by Tarkwa’s management and cannot be assured; they are necessarily based on economic assumptions, some of which are beyond the control of Tarkwa. Future cash flows and profits derived from such forecasts are inherently uncertain and actual results may be significantly more or less

 

7

 

favourable.

 

This report includes technical information, which requires subsequent calculations to derive subtotals, totals and weighted averages. Such calculations may involve a degree of rounding and consequently introduce an error. Where such errors occur, SRK does not consider them to be material.

 

2.11       Disclaimers and Contingency Statements for US Investors

 

The United States Securities and Exchange Commission (the “SEC”) permits mining companies, in their filings with the SEC, to disclose only those mineral deposits that a company can economically and legally extract or produce from. Certain terms are used in this report, such as “resources”, that the SEC guidelines strictly prohibit companies from including in filings.

 

Mineral Reserve estimates are based on many factors, including, in this case, data with respect to drilling and sampling. Mineral Reserves are derived from estimates of future technical factors, future production costs, future capital expenditure and future gold price. The Mineral Reserve estimates contained in this report should not be interpreted as assurances of the economic life of Tarkwa or its future profitability. As Mineral Reserves are only estimates based on the factors and assumptions described herein, future Mineral Reserve estimates may need to be revised. For example, if production costs increase or product prices decrease, a portion of the current Mineral Resources, from which the Mineral Reserves are derived, may become uneconomical to recover and would therefore result in Mineral Reserve estimates with lower tonnages and contained gold.

 

The LoM plans, the TePs and the Financial Model (“FM”) include forward-looking statements in compliance with the requirements of NI 43-101. These forward-looking statements are necessarily estimates and involve a number of risks and uncertainties that could cause actual results to differ materially.

 

8

 

3              PROPERTY DESCRIPTION AND LOCATION

 

3.1           Location and Access

 

Tarkwa is located in south western Ghana approximately 300 km by road west of Accra, the capital, at a latitude 5°15’N and longitude 2°00’W. Tarkwa is located 4 km east of the town of Tarkwa with good access roads and an established infrastructure. The mine is served by a main road connecting to the port of Takoradi, some 140 km to the southeast on the Atlantic coast (Figure 3.1).

 

3.2           Mining Leases

 

Tarkwa operates under mining leases with a total area of 208.28 km2. Five leases relating to the Tarkwa property are dated April 18, 1997, expiring on April 17, 2027. Two mining leases dated February 2, 1988 and June 18, 1992 and expiring in 2018, relate to the Teberebie property. There are no additional reconnaissance or prospecting licences held by Tarkwa.

 

SRK notes that the current LoM Plan envisages extraction from Teberebie until the end of 2022. SRK does not envisage a problem with the extension of the lease to cover continued operations in this area if required, however this cannot be guaranteed.

 

There is an area of overlap between the Tarkwa and Damang licenses, in that over one area Damang has rights to a depth of 30 m and Tarkwa rights below this depth (Figure 4.2). Damang is the adjacent property to the north that is owned and operated by AGL, which Gold Fields has a 71.1% interest in (Section 9).

 

Table 3.1: Mining Leases

 

		Lease		Date				Area
Location		Number		Granted		Expiry Date		(km2)
Tarkwa		WR 637/97		April 18, 1997		April 18, 2027		22.27
Tarkwa		WR 640/97		April 18, 1997		April 18, 2027		34.19
Tarkwa		WR 639/97		April 18, 1997		April 18, 2027		49.51
Tarkwa		WR 638/97		April 18, 1997		April 18, 2027		43.00
Tarkwa		WR 641/97		April 18, 1997		April 18, 2027		43.34
Teberebie		WR 1518/96		February 2, 1988		February 1, 2018		15.97 (includes 1518/96)
Teberebie		WR 1520/96		June 18, 1992		June 17, 2018		See above
Total								208.28

 

3.3           Information on Equity

 

GFGL was incorporated in Ghana in 1993 as the legal entity holding the Tarkwa concession mining rights and opencast operations commenced early in 1997 with the winding down and closure of the original underground operations. Gold Fields is the operator of the mine and majority share holder with 71.1% of the issued shares of GFGL. IAMGold, through its wholly owned subsidiary Repadre Capital Corporation, holds 18.9%, and the Ghanaian Government a 10% free carried interest, as required under the Mining Law of Ghana.

 

9

 

The Government of Ghana has a 10% non-participating interest in all exploration and mining ventures with the right to purchase an additional 20% equity interest in the mining venture at a fair market price.

 

Royalties of 3% - 12% of mineral revenue are paid to the Government of Ghana. The current 3% royalty rate is based on a 30% operating ratio. Where the operating ratio is below 30% the rate is still 3% but the unused margin can be carried forward to reduce the operating ratio in subsequent years. Where the operating ratio is over 30% but less than 70%, the rate of royalty is 3% plus 0.225% of every one percent by which the operating ratio exceeds 30%. Where the operating ratio is 70% or more, the rate is 12%. The formula for calculating the operating ratio is revenue versus operating costs and capital cost allowances (royalty payments, and taxes imposed on income are not allowable operating costs).

 

Companies may export gold to any foreign refiner upon approval of the Government. The Government has a pre-emptive right to purchase the gold production at fair market value, although this right has never been exercised. Offshore foreign currency retention accounts for the receipt of foreign currency, including the proceeds from gold production, are permitted.

 

3.4           Environmental Liabilities

 

Environmental issues and liabilities are discussed in Section 14.

 

11

 

4              ACCESSIBILITY, CLIMATE AND INFRASTRUCTURE

 

4.1           Topography, Climate and Vegetation

 

The topography of the area comprises a series of prominent ridges and valleys (to approximately 200 m height) within a generally low lying plain. A large low lying flood plain lies to the northwest. No major rivers traverse the mining area, the nearest one being the Huni River some 8 km to the northwest which flows along the concession boundary. The area is drained by minor streams towards the west.

 

A tropical climate at Tarkwa is characterized by two wet seasons (March-July and September-November) with a mean annual rainfall of approximately 2000 mm. Evaporation averages approximately 130 mm per month. Mean monthly temperatures range from 21-32°C. Although there may be minor disruptions to operations during the wet season there is no long term constraint on production due to climate at particular times of the year.

 

The vegetation comprises a mixture of tropical rain forest and semi-deciduous forest. Deforestation due to subsistence farming by the local population has degraded the natural vegetation in the environs of the mines to secondary forest, scrub and cleared land. No primary forest is found on the concession.

 

4.2           Local Infrastructure and Economy

 

The industries of the Tarkwa region are mining (gold, manganese) and subsistence farming which occurs throughout the area with cocoa, plantain, pineapple, cassava, maize, yam and some oil palm, rubber and coffee being the principal crops.

 

The mine has access to the national electricity grid from the Volta River Authority (“VRA”). The electricity supply to Tarkwa is sourced from the VRA Tarkwa substation via a 34.5 kilovolts (“kV”) line. The Tarkwa substation is part of the 161 kV ring network, located between the Takoradi and Prestea substations. The substation is well situated, having total capacity of 550 megawatts (“MW”) in close proximity and with Prestea being connected to the Cote d’Ivoire power line, which has a capacity of 240 MW. In addition, Tarkwa is connected to the hydro-electric generating stations at Akasombo via the transmission grid on a ring feed arrangement. The installed capacity at Tarkwa is about 91 megavolt ampere (“MVA”), consisting of one 25 MVA at 11 kV and two 33 MVA at 34.5 kV. The current maximum demand is a 42 MVA.

 

A main asphalt road that connects the mine to the port of Takoradi services the Tarkwa area. From Takoradi an easterly coastal road connects with the capital Accra. Domestic communications are provided by way of a microwave link and landlines. International communications are provided by a direct satellite system and land lines.

 

12

 

5              HISTORY

 

5.1           Ownership and Operations

 

Mining activities around the town of Tarkwa date back to the late 19th century when several small mining companies operated an area then known as the Abontiakoon Concession. In 1935, the property was taken over by the Amalgamated Banket Area Limited (“ABA”), which also acquired the old Tarkwa mine and the properties of Mantraim, Pepe, Akontansi, Kotraverchy, Patet-Sommahoo and Teberebie, and Awunaben-Ahoonabe areas.

 

The amalgamation resulted in an expansion programme including the sinking of the Abontiakoon Vertical Shaft (“AVS”), the construction of new residential buildings and workers compounds, the construction of a workshop and store at AVS from 1936 to 1938 and the completion of a new 30,000 tonnes per month (“tpm”) central mill in 1940. The Tarkwa South Mine was dewatered in 1940 and development undertaken to link with Mantraim Vertical Shaft, also being sunk at that time. Prior to the takeover by ABA, eight vertical shafts and numerous pits had been excavated. Operations at Akontansi had resumed in 1937 but ceased at the end of 1943. The Pepe workings were established in 1940 and mined as open cuts.

 

In 1943, as a result of World War II, the mine was closed and only essential plant equipment maintained. Operations were resumed in 1946, with a temporary suspension of activities for three months during 1947 due to a general strike.

 

In 1950, the amalgamation of ABA, Gold Coast Banket Areas Ltd (Fanti Mines) and South Banket Areas (Tamso / Effuanta Mines) became effective with the administrative and mining operations being directed by ABA.

 

Operations were again suspended toward the end of 1955 because of general strike action. After the strike in February 1956, ABA was given a financial grant by the Government of Ghana to assist with the long term planning and operation. In 1956, the working of the Pepe opencast mine was stopped and in 1958 the Mantraim incline shaft was closed. In 1960 the Fanti Main shaft was closed and allowed to flood. In 1960, owing to the refusal of the Ghanaian Government to grant further subsides to ABA, all workings were abandoned and allowed to flood.

 

Realizing the effect this would have on the livelihood of the mine employees in the Tarkwa district, the government purchased the entire site from ABA. In May 1961, the property was commissioned as one of the properties of the State Gold Mining Corporation (“SGMC”). The Tarkwa mines were renamed Tarkwa Goldfields Limited in 1963.

 

The Apinto Shaft was sunk between 1973 and 1976 to access additional sources of ore. In 1986, a rehabilitation programme financed by the World Bank helped sustain the Tarkwa Goldfields Limited operations. In June 1993, the Government of Ghana entered into an agreement with GFGL under which GFGL would operate the mine under a management contract.

 

14

 

In 1996, a pre-feasibility study into an open pit / heap leach operation, undertaken on behalf of GFGL by SRK concluded that such a project was economic. This study was followed up with a feasibility study and the subsequent approval to proceed with the project.

 

The initial development, known as Tarkwa Phase I, was completed in April 1998 at a cost of USD149 million (“M”) and entailed a total mining rate of 14.5 Mtpa comprising 9.8 Mtpa waste and 4.7 Mtpa heap leach feed ore.

 

An expansion, known as Tarkwa Phase II was completed in July 1999 at a cost of USD10M. This increased the mining rate to 20.7 Mtpa and ore feed to the heap leach pads to 7.2 Mtpa.

 

In August 1999, GFGL suspended all underground mining operations at the Apinto shaft and AVS sections for economic reasons. The milling plant continued to process the remaining ore and clean-up material until shutdown in December 1999. At that stage, GFGL withdrew totally from the underground operations resulting in the flooding of the underground workings.

 

In August 2000, following the acquisition by Ghanaian Australian Goldfields Limited (“GAG”) of the Teberebie lease and operations, GFGL acquired the northern part of the Teberebie lease from GAG. The facilities, comprising the Teberebie open pit and heap leach pads and associated equipment were re-commissioned at a cost of USD11 M and placed into production. This expansion, known as Tarkwa Phase III, increased the mining rate by 15.3 Mtpa to 36 Mtpa and the heap leach production capacity from 7.2 Mtpa to 12.6 Mtpa.

 

Until 2004, open pit operations were carried out by a contractor African Mining Services (Ghana) Pty Ltd. (“AMS”) a joint venture between two Australian contractors; Henry Walker Eltin and Ausdrill. Following a bankable feasibility study (“BFS”) of utilizing a conventional CIL plant (completed in December 2002 by Lycopodium Pty Limited “Lycopodium”) and a study into owner mining operations, projected output targets were increased to a 16 Mtpa heap leach operation and a 4.2 Mtpa CIL operation and owner mining operation with effect from mid 2004.

 

The existing surface operation exploits narrow auriferous conglomerates from various pits: namely (from west to east) Kotraverchy, Akontansi Ridge, Akontansi East, Pepe, Mantraim, West Hill and Teberebie.

 

5.2         Historical Production

 

Production for the past eight years is as shown in Table 5.1 below:

 

15

 

Table 5.1: Historical Production from Tarkwa

 

Financial Year (July to June)		Ore Processed (Mt)		Metal Production (oz Au)
2004		16.0		550,000
2003		15.2		540,000
2002		14.9		544,000
2001		11.7		440,000
2000		8.0		296,000
1999		5.0		206,200
1998		0.2		55,235
1997		0.2		48,845

 

5.3           Historical Exploration

 

Historical exploration is outlined in Section 7.

 

5.4           Historical Mineral Resource and Mineral Reserve Statements

 

As an operating mine, the Mineral Resource and Mineral Reserve estimates at Tarkwa are updated annually as at June 30. Reconciliations are carried out systematically at Tarkwa between actual production and the Mineral Resource and Mineral Reserve estimates on a monthly and annual basis. These are described further in Section 10.

 

16

 

6              GEOLOGY

 

6.1           Regional Geology

 

The Tarkwa orebodies are located within the Tarkwaian System which forms a significant portion of the stratigraphy of the Ashanti Belt in southwest Ghana. The Ashanti Belt is a north-easterly striking, broadly synclinal structure made up of Lower Proterozoic sediments and volcanics underlain by the metavolcanics and metasediments of the Birimian System. The contact between the Birimian and the Tarkwaian is commonly marked by zones of intense shearing and is host to a number of significant shear hosted gold deposits including Prestea, Bogoso and Obuasi.

 

The Tarkwaian unconformably overlies the Birimian and is characterized by lower intensity metamorphism and the predominance of coarse grained, immature sedimentary units which from oldest to youngest are:

 

·      Kawere Series (250 — 700 m) — poorly sorted, polymictic conglomerates and quartzites with no significant mineralization;

·      Banket Series — well sorted conglomerates and quartzites with clasts generally considered to be Birimian in origin and containing significant gold mineralization, hosting the Tarkwa orebody. In the Pepe area the Banket Series is approximately 32 m thick and at Kotraverchy up to 270 m thick;

·      Tarkwa Phyllite Series (120 — 140 m) — fine grained chloritic siltstones, mudstones and schists with no significant mineralization; and

·      Huni Series (1370 m) — fine grained massive quartzites with no significant mineralization.

 

6.2           Local Geology

 

The local geology is dominated by the Banket Series which can be further sub-divided into a footwall and hangingwall barren quartzite separated by a sequence of mineralized conglomerates and pebbly quartzites. The stratigraphy of the individual quartzite units is well established with auriferous reefs interbedded with barren immature quartzites. The units thicken to the west and current flow parameters indicate a flow from the east and north-east. Tarkwa’s geological location is shown in Figure 6.1.

 

Sedimentological studies of the detailed stratigraphy within individual reef units have led to the recognition of both lateral and vertical facies variations. The modelling of these has resulted in the recognition of a cycle of events from initial channel formation and rapid downcutting of the central channel (basin downwarp time units T1 and T2) through a period of uplift and reworking (T3) and finally a period of meandering channel bars and flow reduction leading to the development of low grade silty interbeds. This sequence has been recognised in each of the main reef units with the T3 sequence recognised as the principal episode of gold deposition and concentration.

 

17

 

6.3           Structure

 

Structurally, the Tarkwaian belt has been subject to moderate folding and at least five episodes of deformation are thought to be represented at Tarkwa. Thrust faults occur between Pepe and Akontansi and between Akontansi Ridge and Kotraverchy. The original deposition occurred in a listric basin environment with associated low to steep angle normal faulting. Subsequent compression and folding led to development of thrust faults and reversing of previous normal faults. The final stages involved further thrusting in a southwest direction.

 

This has led to the individual deposits at Tarkwa having significantly different structural controls and differing degrees of structural complexity. At Pepe, the orebody forms on a gentle anticline with limb dips from horizontal to a maximum of 35°. The area is cut by north-east trending normal faults but no significant fault losses are reported. At Akontansi East, the strata dip north-west at 18° and are cut by north-east trending normal and reverse faults with low angle thrusts occurring at the east and west margins. The area is also intruded by numerous dykes and sills and is considered to be structurally complex. Throughout Akontansi, numerous small scale structures occur which are only identified either during grade control or mining with reverse faulting leading to duplication of the reef adjacent to fault planes. At Kotraverchy, the north-east plunging anticline is cut obliquely by a number of shallow angle thrusts and cross cut by numerous strike slip faults. The Kotraverchy area is recognised as being structurally complex and as a result the grade control drill spacing will be reduced to 12.5 x 25 m in the next phases of grade control drilling.

 

The geological model at Tarkwa has been refined considerably since mining commenced and the large amount of grade control and mining data has allowed much more detail to be introduced to the model. This has, in turn, led to the development of the domaining models described in a following section. There is still a considerable amount of small scale faulting which is only identified through grade control drilling and it is likely that this is having an influence on the reconciliation between the diamond drilling and grade control resource models, although the primary cause is probably un-modelled strike changes.

 

6.4           Deposit Types

 

The orebody at Tarkwa consists of a series of sedimentary Banket quartz reef units similar to those mined in the Witwatersrand Basin of South Africa. The operation is currently mining multiple reef horizons from seven open pits and there is further potential for mining underground in the future.

 

6.5           Mineralization

 

Mineralogically, the gold occurs as free particles with an average size of between 50-150 µm and is mostly entrained within the silicified interstitial matrix between conglomerate clasts. Gold is generally found native with only minor electrum and with only 3-7% silver.Accessory oxides present in the ore include magnetite, goethite, ilmenite and rutile.

 

The principal orebodies are shown in Figure 6.2.

 

19

 

The mineralized reefs are identified below from the base upwards (younging):

 

·      AFc — 0.2-3.0 m thick, only occurs in the west and is scoured by the A1 in the east. Well sorted with rounded clasts of quartzite and visible gold;

·      A1 — between 2-7 m thick, moderately to poorly sorted conglomerate and thin quartzites with occasional visible gold;

·      A3 — occurs up to 7 m thick, moderately sorted, the conglomerate forms as thin discontinuous lenses within a package of cross bedded quartzites, visible gold is rare;

·      B2 (or B) — up to 3 m thick, very coarse quartzites with thin lags of sub-rounded pebble conglomerate;

·      CDE — up to 8 m thick and can be subdivided into the lower C reef and upper E reef both of which are conglomeratic and are separated by the D reef quartzite; and

·      G — varies from 2-6 m thick poorly sorted conglomerate with clasts of quartzite and phyllite.

 

20

 

7              EXPLORATION AND DRILLING

 

7.1           Exploration History

 

Initial exploration commenced in June 1993 at Pepe. To date, the vast majority of drilling has targeted the open pit potential although a number of deeper holes have investigated the underground potential at Akontansi and Pepe/Mantraim. Prior to GFGL’s involvement, SGMC carried out selected drilling mostly along the Pepe Anticline targeting shallow underground potential. However, these holes are not used owing to data quality issues. Data from GAG at Kotraverchy and from Pioneer Goldfields at Teberebie is included in the GFGL database.

 

7.2           Exploration Methodology

 

Exploration is initially carried out using diamond drilling to produce continuous core sampling through the sequence of mineralized reefs. Core is logged and halved with one half retained for quality control and validation purposes. The remaining core is sent to Transworld laboratories in Tarkwa for assay. Check assaying is carried out at SGS laboratory which is also based in Tarkwa. Core drilling is initially carried out on a wide spaced grid of 200 m along strike and 100 m in the dip direction (400 x 200 m in some cases). This grid is then in filled to a final spacing of 100 x 100 m. Core logging and sampling is carried out based on the recognition of geological boundaries and marker horizons.

 

Grade control is carried out by close spaced infill drilling of the exploration grid using reserve circulation (“RC”) drilling on 25 x 25 m grid spacing. In some areas of known structural complexity this spacing is reduced to either 25 x 12.5 m or 12.5 x 25 m. The Kotraverchy pit area is planned to be drilled using this closer spaced grade control.

 

7.3           Drilling

 

Currently, a total of 1,770 exploration holes have been drilled of which 1,358 have been drilled by GFGL. The GFGL drilling has primarily been carried out using diamond coring methods; however, the total includes 211 RC drillholes. Total exploration drilling to date amounts to some 230,000 m.

 

The following list summarizes the data available for completion of the 2004 model:

 

·	Diamond Drilling:		1,274 holes		(123,960 m)
·	PQ Holes:		8 holes		(547 m)
·	RC on a 100 m grid:		227 holes		(11,780m)
·	RC on a 50 m grid:		56 holes		(4,483 m)
·	RC on a 25 x 50 m grid:		175 holes		(2,074 m)
·	Grade control holes (RC):		13,873 holes		(454,543 m)

 

All grade control drill holes have also been captured in the geological database. The primary database captures the following:

 

22

 

(1)               The collar positions of all RC and diamond cored holes.

(2)               Down-the-hole survey data.

(3)               Lithological data.

(4)               Assay data.

(5)               The final stratigraphic zoning of all boreholes.

 

Mining software geological databases are used for final data storage and data manipulation. During import of raw data into the Surpac database, validation routines are carried out.

 

23

 

8              SAMPLING

 

8.1           Sampling Method and Approach

 

Initial exploration drilling is carried out using diamond coring with sampling at 25 to 50 cm intervals from the hangingwall contact to the footwall. Samples are also taken in the footwall and hangingwall to test for low grade dilution.

 

Sampling during RC drilling and grade control is carried out on the basis of vertical 1 m composites with sampling starting at the hangingwall of the uppermost reef horizon and continuing below the footwall of the lowermost package. An initial split is carried out at the drill rig using a 3-tier riffle splitter and a 3.5 kg sample is bagged for analysis. Representative rock chips are collected during the drilling and logged on site by a geologist to establish the stratigraphic context of the sampling and to provide a geological description of each sample.

 

Belt sampling is carried out at both processing facilities, referred to as the North and South plants, with a sample taken every five minutes and composited to four hours to produce a 4 kg sample. This procedure has been in place since March 2002 and prior to this belt samples were undertaken at 15 minute intervals. This sample is then pulverized and assayed in triplicate. The average of the three assays is reported as the final belt assay.

 

8.2           Sample Preparation, Analysis, Security and Data Verification

 

All original assaying for grade control samples is carried out at the SGS laboratory in Tarkwa, while for core samples Transworld is the primary laboratory and SGS the secondary. Both GFGL and SGS have comprehensive quality control procedures and protocols in place which involve duplicate and repeat assays, blanks, standards and external laboratory checks. The following checks are carried out:

 

·                  GFGL produce a duplicate sample for 5% of all samples taken (1 in 20) which involves the production of two samples from the same sample interval during the initial sample collection at the drill site, and provides information on sample repeatability and sampling error;

 

·                  the laboratory produces a repeat sample for 5% of all samples (1 in 20) from the initial coarse reject to check for laboratory preparation errors;

·                  10% of all sample pulps are re-assayed as a check on analytical error;

·                  all samples which assay >3.0 g/t are re-assayed;

·                  each batch of 50 fire assays includes one standard and one blank as well as, on average, two duplicates and two repeats; and

·                  5% (1 in 20) sample pulps are re-assayed by an external laboratory to check analytical precision. Currently the external laboratory is Transworld laboratory in Tarkwa.

 

The SGS laboratory is ISO 9001 and 9002 compliant and is aiming for ISO 17025 compliance by the end of 2005. Transworld has not yet achieved ISO compliance but is reported to be aiming to achieve this in the near future.

 

24

 

9              ADJACENT PROPERTIES

 

The mining leases held by GFGL are adjacent to the mining lease held by AGL approximately 35 km to the north. Ownership of AGL mirrors that of GFGL, and AGL conducts operations at the Damang gold mine to the north of Tarkwa.

 

Bogoso / Prestea Gold Mine (“Bogoso”) is an open pit mining operation owned and operated by Bogoso Gold Limited which is 90% owned by Golden Star Resources Limited (“Golden Star”) and 10% by the Government of Ghana, and lies south of Bogoso Township in the west of Ghana approximately 120 km from the port of Takoradi.

 

Iduapriem Gold Mine (“Iduapriem”) lies 6 km south of Tarkwa and is an open pit mining operation owned and operated by AngloGold Ashanti Limited (“AngloGold”), with 10% held by the Government of Ghana.

 

Technical information on Bogoso and Iduapriem is available in public domain documents published by Golden Star and AngloGold respectively that are available on these companies’ websites.

 

Akoon Gold Mine is situated to the north of Tarkwa and was an underground operation from 1912 to 1999. Gold Coast Resources Inc. currently owns an option on this closed facility.

 

26

 

10            MINERAL RESOURCES

 

10.1         Introduction

 

The process of estimating the Mineral Resource at Tarkwa involves the use of a number of factors based on historical results and planned dilution. Both exploration and grade control drill results are used for geological modelling and grade interpolation. Currently, Mineral Resources are classified according to the guidelines set out in the South African Code for Reporting Mineral Resources and Mineral Reserves (the “SAMREC Code”). SRK considers that the derivation of the Tarkwa Mineral Resource and Mineral Reserve also conform to the definitions and guidelines prepared by the CIM Standing Committee on Mineral Resources and Mineral Reserves and approved by the CIM Council of the Canadian Institute of Mining, Metallurgy and Petroleum in August 2000 (the “CIM Standards”).

 

Historically, the Mineral Resource model derived from exploration drilling, has consistently underestimated the tonnage when compared to the production results. This has led GFGL to carry out a comprehensive study of the historical results comparing the original wide spaced exploration drilling (diamond cored), the grade control (RC) drilling and the production data from individual pits.

 

The current 2004 model replaces the previous model (2001) and has been produced using the data and assumptions which have resulted from a two year study. The methods used for geological modelling, calculation of dilution, and grade interpolation have not been changed since the 2001 model. The inclusion of sedimentologically based domains is, however, new. In addition, a tonnage factor has been introduced to the Mineral Resource estimates in an attempt to better reconcile the actual production tonnage with that estimated from the exploration model.

 

10.2         Geological Modelling

 

10.2.1      Modelling Procedures

 

The process for creating the geological model used as the basis for the Mineral Resource has changed recently. Previously, the orebodies were modeled as a series of three dimensional (“3-D”) footwall and hangingwall digital terrain model (“DTM”) (triangulation) surfaces using the results from the available drilling. GFGL has now switched to producing a 3-D wireframe solid model for each reef based on block modelling of the reef thickness which produces a more realistic estimate of volume.

 

The first stage of modelling the exploration areas is to create a footwall triangulation surface using the intersection point in each borehole. The vertical thickness (“VT”) of the reef is then calculated for each borehole and this variable is interpolated into a 25 x 25 m block model using an inverse distance squared model. The interpolated VT value is added to the footwall triangulation surface to create a series of 25 x 25 m spaced points which are triangulated to create a physical hangingwall surface.

 

27

 

Currently, the Kotraverchy, Akontansi and Teberebie orebodies have not been re-modelled in this way (86% of Mineral Resource). The Pepe and Mantraim North areas are currently using the new 3-D surface method for geological modelling. Plans are in place to convert the remaining areas over to the new method although the timeframe for this is not in place.

 

Wireframe modelling is carried out using a set of guidelines which have been drawn up on site. The purpose of this is to standardize the modelling procedure and provide GFGL geologists with a standard protocol for the following:

 

·      risk area identification;

·      database checks;

·      wireframe checks; and

·      statistical checks of validity.

 

The geological modelling is initially based on the exploration diamond core drilling. However, where closer spaced (25 m) RC drilling has been carried out for grade control, the assay results and drill logs from the RC data are combined with the results from the diamond drilling. The inclusion of grade control drilling in the modelling procedure has led to significant changes in the position of the orebodies in some areas of the operation, when compared to the initial model based on wide spaced (100-200 m) exploration drilling. In some cases these changes are manifested as changes in the elevation of the individual reefs by up to 20 m in places due to revised interpretations. This will influence the stripping ratio in the immediate area and therefore will affect the pit optimization.

 

The inclusion of the closer spaced drilling also allows better definition of variations in orebody thickness. However, there does not appear to be a significant difference between the original modelled thickness and that produced using the grade control data.

 

The orebodies at Tarkwa are cut by numerous faults which generally run perpendicular to the strike and are predominantly reverse faults. Movement on the faults is in both dip and strike directions, although vertical displacements are generally less than 20 m. The faults have been modelled as 3-D planes and the orebody reef models have been clipped against these faults to prevent duplication of reef volumes.

 

A number of igneous intrusions also cross cut the reef orebodies and can cause areas to be sterilized. However, the major intrusions have been identified during the exploration drilling, modelled as 3-D wireframes and taken into account in the resource estimation process. Minor intrusions do occur which are not picked up by the initial exploration drilling at the 100 m spacing. However, these are usually picked up by the grade control drilling.

 

SRK has reviewed the geological modelling procedure and protocols and is in general agreement with the methodology employed. It is obvious that the exploration drilling is not accurately defining the location and shape of the orebody on a local scale and the updated

 

28

 

model is still underestimating tonnage. However, the grade control drilling appears to be adequate and is currently running approximately 18-24 months ahead of mining which allows the changes to the geological model between exploration and grade control to be incorporated in the annual mine planning.

 

10.2.2      Domaining

 

A major new aspect to the Mineral Resource estimation procedure for the 2004 model is the modelling of specific sedimentological domains within individual deposits. The domaining procedure uses a number of sedimentological characteristics to define the domain boundaries for each individual reef. These characteristics are listed here in order of decreasing significance:

 

·                  Palaeocurrent direction has been mapped throughout the mine and indicates a radial pattern with the primary flow direction from the northeast and east. Locally, the current direction can be directly linked to gold mineralization in the proximal (up stream) areas of the deposit such as Pepe, Mantraim and Teberebie. The domains for these areas are defined as a series of narrow strips striking from north-east in Pepe/Mantraim through to east in the Teberebie area.

 

·                  Footwall control plays a more significant part in delineating domains as the distance from the source increases in the more distal areas of the mine such as Akontansi. The footwall control occurs as a result of sub cropping footwall beds normal to the flow direction which has disrupted the flow and formed a series of ridges. The footwall control domains tend to strike perpendicular to the flow current direction and the majority of the domains at Akontansi strike to the north-west.

 

·                  Reef thickness increases to the west with a corresponding decrease in grade and it is essential to take the thickness into account so as to restrict the interpolation of grade into areas of increased thickness. T3 time units are identified from diamond drilling and used to help outline the individual domains.

 

·                  Gold grade and accumulation are used in the local estimation to delineate pay shoots within individual domains. In the distal areas at Kotraverchy, the grade plays a more significant role in the delineation of the domains used for local grade interpolation.

 

SRK considers the development of a detailed sedimentological model for a deposit such as Tarkwa to be appropriate and the application of these domains during the grade interpolation procedure is providing increased confidence in the local grade estimation. It is obvious from the work carried out both at Tarkwa and at South African Witwatersrand Basin gold mines that an understanding of sedimentological controls on gold mineralization are essential for the production of enhanced local grade estimates. As additional drilling is completed, it is essential that the current models are updated as there are large areas of the current domains with only wide spaced drilling informing the model. In addition, the newer areas of the mine such as Kotraverchy require drilling over a wider area in order to improve the resolution of the current domains and in certain areas it may be necessary to further sub-domain based on structural complexity.

 

29

 

10.3         Tonnage and Grade Reconciliation

 

10.3.1      Application of Tonnage Factor

 

Detailed reconciliation studies have been carried out since the commencement of mining in the mid 1990s. It has been apparent, since that time, that the Mineral Resource estimated from the exploration data is significantly underestimating the tonnage of ore, when compared to the results from the belt weightometers, by some 10% on average, although this figure varies up to 30% in areas of high structural complexity. The grade reconciliation with the exploration model is generally good and the result is an increase in contained metal in the grade control/production data when compared to the exploration model.

 

Since 2002, a detailed programme of reconciliation studies and re-modelling has been instigated by GFGL in an attempt to identify where the discrepancy is occurring in the resource estimation process. This process has included re-modelling of specific areas, as well as detailed checking of the equipment used for calculating tonnages, and the drilling and sampling protocols used for both exploration and grade control drilling.

 

The results of this study have resulted in a tonnage factor being applied to that portion of the Mineral Resource model which is informed by diamond drilling only, with a minimum spacing of 100 x 100 m.

 

Figure 10.1 summarizes the results of the monthly tonnage reconciliation between the original and the current exploration models and the belt weightometer for the two year period between July 2002 and May 2004.

 

The monthly variation plot shows that the old model produced a consistent underestimation of tonnage within a range of between 4% and 30% underestimation by the resource model and at an average of 9%. The new resource model shows a similar range between the minimum and maximum variation of between 4% overestimation and 19% underestimation when compared to the belt tonnage. However, it should be noted that the new model includes a 10% tonnage factor which was not included in the old model and when this factor is excluded, the two models show very similar values (6-29% underestimation). The period since October 2003 has shown a consistent underestimation of tonnage by the new model including the 10% factor, at an average of some 8% while for the initial period between July 2002 and September 2003 the average variation was only 2%.

 

Additional factors which may be having an effect on the exploration tonnage estimate are:

 

·                  the density estimate;

 

·                  the duplication of reefs adjacent to reverse faults and thrusts, and areas of increased structural complexity which would not be picked up at the resolution of exploration drilling;

 

·                  conservative calculation of reef thickness during core logging when compared to RC estimates and visual grade control during mining;

 

30

 

·                  inaccurate modelling of the reef footwall which is then used as the basis for calculation of the exploration model thickness; and

 

·                  variations in topography due to use of aerial topographic surveys during exploration drilling and detailed surveys carried out during grade control drilling.

 

Figure 10.1: Monthly Tonnage Reconciliations July 2002 — May 2004

 

 

The results indicate that the new model, while providing a better reconciliation with the belt tonnage than the previous model, is still not providing a consistent reconciliation. It is likely that additional factors are influencing the tonnage distribution. Although SRK considers the factor to be appropriate given the current information SRK recommends that domaining exercises to delineate areas of higher and lower tonnage factors continue to be investigated.

 

10.3.2      Grade Reconciliation

 

Grade reconciliations of the new model and the old model when compared to the monthly average from the belt sampling indicate that the grade has not changed significantly between the two models and that both are providing a good reconciliation with the actual Run-of-Mine (“RoM”) grade.

 

Grade reconciliation for the RC drilling is indicating an underestimation of the grade. A recent study by D F Bongarcon has indicated that current sample size is insufficient and an updated RC sample protocol has been initiated. However, it is too early to state whether this has had any effect on the RC grade estimate.

 

10.4         Dilution Parameters

 

The dilution model was changed in 2001 after a reconciliation study suggested that the previous 40 cm skin at 0.0 g/t Au (80 cm total dilution) was overly conservative. The current

 

31

 

model applies a 30 cm hangingwall and 20 cm footwall (true thickness) skin to the reef wireframe at a grade of 0.15 g/t. In some cases where reefs lie very close together and are mined together, the dilution skin between the two is given an average grade of the adjacent reef. Dilution is applied to the resource model separately from application of the tonnage factors described previously.

 

10.5         Data Analysis

 

All assay data is captured electronically and in hard copy format from SGS laboratories. Basic statistics are produced after completion of wireframe modelling to check the validity of samples which lie within the geological model as well as the stationarity of the domains. 1 m composites are produced from the hangingwall intersection with a minimum of 40 cm allowed at the base to allow for shorter composite lengths. Statistics are calculated for vertical thickness, grade, and metal accumulation.

 

High grade cuts are performed based on a visual examination of the data distribution on QQ plots. In most cases, the number of high grade values is less than 1% and these are cut only for the purpose of semi-variogram modelling and are included in the sample population used for the final grade interpolation. Diamond core exploration data is combined with RC grade control data where available to provide as much information as possible for semi-variogram modelling. The modelling is carried out within individual domains and the domain boundaries are regarded as hard boundaries with no influence allowed from adjoining domains. In some cases, insufficient data is available to produce down hole semivariograms.

 

SRK considers the approach taken by GFGL to be appropriate, given the size of the database available. The process for dealing with high grade samples is acceptable given the relatively small number which could be classified as outliers. Furthermore, by excluding a small number of samples from the experimental semi-variograms the noise is significantly reduced and structural details are enhanced allowing more confidence in the final model.

 

10.6         Grade Interpolation

 

Grade is interpolated into 50 x 50 x 3 m blocks using a simple kriging (“SK”) algorithm. Local mean values are calculated for each individual reef domain using a 400 x 400 m moving window and a minimum of 20 composites. The boundary of the domain is treated as a hard boundary for calculation of the local mean. Search parameters are based on the individual semi-variograms produced for each reef in each of the domains. A dip domain model is created for each domain based on the strike and dip of the triangulation surfaces which make up that particular reef model within the domain.

 

SK is carried out using all available data within the search radius, but the algorithm treats the domain boundary as a soft boundary and for blocks near the edge of the domain the search can include composites within a 50 m skin of the adjoining domain. This has the result of reducing the edge effect and therefore reduces any conditional bias for blocks around the edge of the domain. High grades are included in the kriging population used for grade interpolation. The data uses a combination of both RC and exploration composites. Post

 

32

 

processing is carried out for calculation of recoverable resources using a Selective Mining Unit (“SMU”) of 10 x 5 x 3 m and based on an assumption of a 25 x 25 m grade control drill spacing and a log normal grade distribution for a number of cut-offs ranging from 0.5- 2.0 g/t.

 

Block proportions are calculated using Datamine software and are cross checked against the physical wireframe volume and also against a block model produced using the Surpac software system. The results are within 0.01% of each other and are considered accurate. Each block is given a proportion of ore and waste to which a tonnage factor is then applied. The factor is also applied to the proportion of waste to take into account the increased tonnage of ore within the block.

 

10.7         Mineral Resource Statement

 

The Mineral Resource Statement in Table 10.3 represents the Mineral Resource at Tarkwa estimated by GFGL as at June 30, 2004. In SRK’s opinion the tonnage and grade estimates and classification are appropriate according to the SAMREC Code and CIM Standards.

 

Table 10.1: Tarkwa Mineral Resource Statement, June 30, 2004

 

Mineral Resource Classification		Mt		g/t		Koz
Measured
open-pit		200.5		1.5		9,610
surface stockpile		4.3		0.9		120
Measured sub-total		204.8		1.5		9,730
Indicated
open-pit		187.3		1.4		8,210
Indicated sub-total		187.3		1.4		8,210
Total Measured and Indicated		392.1		1.4		17,940

 

		Mt		g/t		Moz
Inferred
open-pit		3.3		1.5		160
underground		16.3		4.0		2,070
Inferred Total		19.5		3.6		2,230

 

Mineral Resources are reported after inclusion of a 10% tonnage factor for the volume of ore material in a block and a corresponding reduction is applied to the waste tonnage. Dilution is applied to the blocks and a pit optimization is carried out using a USD 400 per ounce gold price. After calculation of the pit shell the dilution is subtracted from the blocks and the Mineral Resource quoted is essentially the tonnage of the undiluted ore blocks which lie within the outline of the un-engineered pit shell.

 

The underground Mineral Resource estimate is also based on a gold price of USD 400 per ounce which equates to a cut-off grade of 3.7 g/t for Akontansi and 2.7 g/t for Kotraverchy.

 

33

 

11           MINING AND MINERAL RESERVES

 

11.1         Mining Operations

 

Mining operations at Tarkwa since July 2004 have been carried out on an owner operated basis. Prior to that, for four years, operations were carried out by a contractor AMS. This followed an owner operated cost assessment completed by GFGL, including a review of the capital and operating costs to be used for a base line comparison with contract mining. The review indicated that the most productive and cost efficient mining fleet was a combination of Liebherr R 994 excavators and CAT 785C dump trucks. Mining production requirements are now significantly higher than the previous throughput rate of 14.4 Mtpa in order to prepare for the increased material movement anticipated in January 2005 when the CIL plant is planned to be commissioned.

 

11.2         Mining Method

 

The mining at Tarkwa exploits seven distinctive areas, namely from west to east: Kotraverchy, Akontansi Ridge, Akontansi East, Pepe, Mantraim, West Hill, and Teberebie. The location of the mining areas is defined through the long term planning process. The boundaries of the pits are pegged out by survey. The area is cleared of bush and topsoil with a bulldozer. This material is later relocated for rehabilitation purposes. After clearing, RC grade control drilling is carried out, and the grade control geological models constructed. The short term plans and forecasts are updated with this information prior to the commencement of mining. From the highest point in the pit, material is free-dug or blasted to the first blasting reference level (“RL”). Fresh rock and transitional zones are drilled and blasted in 6 m lifts with excavation in 3 m flitches. Less weathered material is excavated without the requirement of blasting. Ore and waste is currently loaded by 8 hydraulic excavators (four Liebherr 994 and three Liebherr 984 and one Liebherr 994C). The current truck fleet consists of 24 rear-dump CAT 785C with a payload capacity of around 90 t and 2 CAT 773 rear dump trucks.

 

Mining is highly selective. Backhoe excavators are used to select off waste from the ore, and vice versa, to an estimated accuracy of approximately 30 cm. Pit geologists supervise all digging. Ore material is either RoM, delivered to one of two primary crushers, or low grade, which is stockpiled close to the north primary crusher, dependent on grade. Waste material is hauled to the nearest waste dump.

 

Blasting currently utilizes, and will continue to utilize, relatively close patterns and small diameter holes, typically a 3.2 x 3.6 m grid with a hole diameter of 102 mm. Slightly larger diameter holes and an increased grid size will be utilized in the partially weathered material while a decreased grid size will be utilized in harder material. The small diameter holes are used to preserve, as far as possible, the integrity of the ore/waste contacts, and allow for visual identification of the zones by the pit geologists.

 

Currently, approximately 1.5 million bank cubic metres (“bcm”) of material is mined monthly (3.8 Mt). Total current RoM is approximately 1.2 million tonnes per month

 

35

 

(“Mtpm”), which is expected to increase to roughly 1.6 Mtpm after the commissioning of the CIL Plant.

 

The primary objective of the scheduling is to maximize production from the lowest cost per ounce pits, without neglecting the development of subsequent mining areas.

 

11.3         Optimization

 

An optimization was conducted that formed the basis for all mine planning requirements. The following parameters were used:

 

·                  owner operator mining, July 2004 to end of mine life;

·                  Heap Leach North (9.7 Mtpa) to end of mine life and Heap Leach South (6.3 Mtpa) to 2010;

·                  CIL milling (4.2 Mtpa) from January 2005 to end of mine life;

·                  gold price USD 350 per ounce; and

·                  starting projected topography and mining face positions at July 2004.

 

Tarkwa has recently introduced a mine scheduling software package (Xpac) together with a financial evaluation package (Xeras) that have improved the methodology and the detail whereby Tarkwa estimates the Mineral Reserves.

 

Tarkwa has derived updated cost and project parameters for the optimization of the Tarkwa Mineral Resources, generally based on the 2004/5 budgets. SRK has reviewed the updated unit costs for mining, processing and administration with those used for the 2003 Mineral Reserves and considers that these are appropriate.

 

Tarkwa has been consistent in the methodology for excluding certain capital costs from the optimization process, specifically the costs for extending the heap leach pads and for the replacement of mining equipment. This approach would satisfy a corporate objective to maximise the development of the Mineral Resources and the total gold produced and revenue.

 

11.4         Mining Operating and Capital Costs

 

The mining costs that have been used for optimization have been derived on the same basis as the 2004/05 operating budget was prepared. There has been a material change from the average mining costs for 2003, due to the implementation of owner mining with significantly lower mining costs than for contract mining. There has been a significant increase in the cost of fuel (USD0.52/l for 2004 compared with USD0.35/l in 2003), however this has been offset by a reduced cost for the Maintenance and Repair Contract (“MARC”) prices that were negotiated by GFGL.

 

Whilst the owner mining has demonstrated significantly lower unit mining costs than the contractor mining, the cost adjustment factors (“CAFs”) that have been derived for the owner mining are higher than those that have been derived from previous studies. This has had the impact of reducing the quantity of deeper Mineral Reserves in certain areas.

 

36

 

The mining operating costs prepared for the 2004/05 operating budget were developed using the best available estimates of unit costs and rates for the new owner mining operation. As this is the first year of owner mining there is no operating history to compare with. The mining operating costs can be compared with those developed for the BFS in December 2002. Differences are explained by the unit costs that were finally contracted with Caterpillar Inc. for the MARC, and for the increased costs for fuel. The changes in operating costs as each unit of mining equipment ages, have been reflected in the projected annual operating costs.

 

The mining capital costs were developed in accordance with the mining equipment life as contracted for the MARC, and from the estimated annual usage of each unit of equipment. Additional mining capital cost has been provided for relocation of infrastructure as these are impacted by the expansion of the pits.

 

Table 11.1 illustrates the historical production for the past three fiscal years (July to June) at Tarkwa. Note that historical costs are based on contractor mining.

 

Table 11.1: Tarkwa Historical Production and Cash Costs

 

Production		Year ended June
History		Units		2004		2003		2002
Production
Ore Mined		(kt)		17,164		16,067		14,630
Head Grade		(g/t)		1.43		1.46		1.58
Waste Mined		(kt)		43,987		27,521		28,986
Strip Ratio		(koz)		2.6		1.7		2.0
Tonnes treated		(kt)		16,000		15,210		14,914
Yield		(g/t)		1.1		1.1		1.1
Gold Produced		(koz)		550		540		544
Total Cash Costs		(USD/oz)		230		194		171

 

11.5         Stockpiles

 

The stockpiles as included in the Mineral Reserve comprise surface low grade mineralization accumulated from the mining operations since the start of mining at Tarkwa. The opening balance of Mineral Resource for this mineralised material (4.3 Mt at an average grade of 0.88 g/t) is based on the truck count for tonnage and grade control data for the grade of stockpile material placed.

 

SRK notes that a modifying factor of 95% tonnage recovery has been applied to the stockpile to convert it to a Mineral Reserve. SRK considers that this factor is appropriate.

 

11.6         Cut-off Grade

 

The heap leach cut-off grade and the CIL milling cut-over grades that have been applied to each of the deposits, have been based on the operating costs assumptions as used for the optimization. These include provision for royalty and head office management fees, operating costs for mining, processing and overheads / sundries, but have excluded any provision for extending the heap leach pads and for replacing mining equipment, that GFGL has considered to be capital costs. Further optimization of the cut-off / cut-over grades to

 

37

 

improve cash flows is carried out during the production scheduling as per the planning methodology used at Tarkwa.

 

11.7         Engineered Pit Designs

 

The optimization has been constrained in certain locations due to infrastructure constraints on the development of the open cuts. Where infrastructure can be relocated, the capital cost for this has been considered during the scheduling process. The CIL plant has been considered to be a fixture, and the optimization has been constrained to preclude relocation of this plant.

 

Engineering the pit cutbacks and thereby considering and including access ramps has reduced the average slope angles in certain of the deposits, from those average angles used for the optimization. With the benefit of the designs for the 2004 strategic plan, the average slope angles could be modified to more closely represent the influence of access ramps.

 

11.8         Geotechnical Constraints

 

SRK has previously recommended a range of geotechnical measures including the adoption of mapping and monitoring and the reduction of all slope angles by 5°. Tarkwa has now implemented these proposals, following training by SRK GFGL now carries out ongoing geotechnical assessment.

 

Generally, for long term slopes, there is a 12 m batter height and 8 m berm width, with overall slopes of 36-45° in the upper, weathered strata and 55-65° in the lower, fresh strata. If a ramp exists in the wall, the overall slope angle is effectively reduced. LoM plans are considered achievable with the above slope configurations.

 

Geotechnical staff has been assigned to collect geotechnical data for analysis and to advise on mine design. Wall monitoring has become routine and a preliminary series of piezometric boreholes have been drilled in order to monitor water movements. Appropriate software has been purchased to analyse wall failure potential and on-site users have been trained to use it in order to advise the mine planners. The data is regularly reviewed on-site (at least twice a year) by external geotechnical engineers in order to ensure that designs meet accepted standards. The maximum pit for the selected gold price as defined by the Whittle optimization, including mining dilution, forms the basis of the LoM pit design.

 

11.9         Scheduling

 

The schedule for the LoM has been determined for the following assumptions / constraints:

 

·                  limiting the total mining capacity to 100 Mtpa;

·                  closing the South Plant with effect from 2011; and

·                  during the cut-off optimization, limiting the tonnage of low grade stockpile being accumulated.

 

GFGL acknowledges that these specific constraints could be further optimized. GFGL has also identified a number of other strategic options that have the potential to further improve the project returns and these are planned to be investigated by GFGL during 2004 and 2005.

 

38

 

11.10       Mineral Reserve Statement

 

The Mineral Reserve Statement in Table 11.2 represents the Mineral Reserve at Tarkwa estimated by GFGL as at June 30, 2004. In SRK’s opinion the tonnage and grade estimates and classification are appropriate according to the SAMREC Code and CIM Standards.

 

Table 11.2: Tarkwa Mineral Reserve Statement, June 30, 2004

 

		Mt		g/t		Koz
Proved
open-pit		199.8		1.3		8,560
surface stockpile		4.1		0.9		120
Proved sub-total		203.9		1.3		8,680
Probable
open-pit		147.7		1.3		6,050
Probable sub-total		147.7		1.3		6,050
Total Proved and Probable		351.5		1.3		14,730

 

The Mineral Reserve estimates are based on a gold price of US D350 per ounce. The Mineral Reserve is based on heap leach, CIL and owner mining cost. Proved Mineral Reserves are derived from Measured Mineral Resources and Probable Mineral Reserves from Indicated Mineral Resources.

 

SRK has reviewed the methodology used by GFGL at Tarkwa for the estimation of the 2004 Mineral Reserve. The generally low grade for the Tarkwa deposits makes the Mineral Reserve particularly sensitive to changes in the resource models, operating costs and the gold price. The increase in the Mineral Reserve between 2003 and 2004 is consistent with changes in the resource models and the gold price. For the planning methodology applied by GFGL, and for the operating cost assumptions as stated by GFGL, SRK considers that the derivation of the Mineral Reserve at Tarkwa to be appropriate and conform with the guidelines and definitions of the SAMREC Code and CIM Standards.

 

Table 11.3: Tarkwa Mineral Reserve Sensitivities

 

Gold Price		(USD/oz)		315		333		350		368		385		400
Proved and
Probable		(Mt)		297.2		330.6		351.5		382.9		398.1		417.2
Reserve
Average Grade		(g/t)		1.3		1.3		1.3		1.3		1.3		1.3
Contained Gold		(Moz)		12.5		13.8		14.7		16.0		16.6		17.4

 

GFGL has produced a series of Mineral Reserve sensitivities (Table 11.3) that provide an indicative estimate of the Mineral Reserves at a range of gold prices. These estimates are derived by factorising pit shells optimized at each gold price.

 

39

 

12                                        METALLURGY AND MINERAL PROCESSING

 

12.1           Metallurgy

 

12.1.1        Weathering, density, porosity

 

The degree of weathering of the gold-bearing conglomerates at Tarkwa is critical to the viability of the mine due to its effect on the gold extraction. The weathering has affected rocks near the surface by oxidation with consequent clay alteration, which in turn affects porosity, density and metallurgical characteristics of the deposit. From the outset the mine has sought to model the weathering alteration characteristics of the deposit. Initially, a qualitative visual estimate was used to gauge porosity, which was subsequently updated to a wetting test where the quantity and rate of water absorption was measured. The procedure currently in use is a direct measurement of the porosity percentage, or void space, through an immersion technique. 6,000 porosity and bulk density determinations have been made on 5 cm core samples taken every 50 cm in the various reef units and in some of the intervening quartzite beds from 1,070 drill holes at Pepe, Mantraim, Akontansi and Kotraverchy.

 

The weathering profile, hence porosity, of the ore varies from surface as this relates to the fluctuations in ground water levels, which generally vary from 1 to 40 m. Faulting, fracturing and primary lithologic permeabilities modify the depth to porosity relationship. The mine uses a manual method of contouring the different porosity values (P Codes). The weathering zones are shown in Table 12.1.

 

Table 12.1: Ore Weathering Characteristics

 

				%						Range (%)
Highly Weathered		Beta-1		>5		5		<2.57		>90
				4-5		4		2.57-2.59		86-90
Weathered		Beta-2		3-4		3		2.59-2.62		82-86
				2-3		2		2.62-2.65		75-82
		Gamma		1-2		1		2.65-2.67		60-75
Highly Competent		Delta		0-1		0		2.67-2.70		<60

 

The ore is a free milling conglomerate with negligible sulphide content. In terms of the less heap leach amenable ore types, the gravity concentration recoverable gold component varies from 30% to 70%. The optimum grind size for stirred tank leaching corresponds to a P80 of 75 µm. The ore contains a mild preg-borrowing component necessitating the use of a CIL type of circuit. In this configuration, based on laboratory tests and modelling, a recovery in excess of 95% in a leach residence time of 24 hours is believed to be readily achievable.

 

12.1.2        Methodology for quantitative prediction of metallurgical recovery

 

Using the P Codes and the geological block model reef zones, the gold dissolution may be accurately predicted for the heap leach operation. The average relative density values of the zones, together with the predicted gold dissolutions, for the Pepe, Mantraim/Teberebie and Akontansi/Kotraverchy deposits are shown in Table 12.2 below.

 

40

 

Table 12.2: Heap Leach Dissolution Predictions

 

		Pepe				Mantraim/Teberebie				Akontansi/Kottraverchy
P Code		Average  Relative  Density		Gold  Dissolution %		Average  Relative  Density		Gold  Dissolution %		Average  Relative  Density		Gold  Dissolution %
0		2.68		60		2.68		50		2.67		50
1		2.65		67		2.65		60		2.65		60
2		2.6		79		2.61		75		2.61		75
3		2.59		85		2.59		82.5		2.60		82.5
4		2.58		88		2.58		86		2.59		86
5		2.44		91		2.45		91		2.47		91

 

12.2           Mineral Processing

 

12.2.1        Introduction

 

Tarkwa currently utilizes conventional heap leach techniques to recover gold. Operations consist of two separate heap leach circuits, the Tarkwa plant (North Plant) and the Teberebie plant (South Plant). The North Plant was commissioned by GFGL in 1998, while the South Plant was originally commissioned in 1992 by Teberebie Goldfields Limited and, subsequent to GFGL’s purchase in August 2000, was re-commissioned at a cost of USD1 M. The two plants each have multiple stage crushing processes combined with agglomeration and a combined capacity of approximately 1.2-1.3 Mtpm.

 

The BFS was based on incorporating the then current 14.4 Mtpa heap leach operation, with a newly constructed 4.2 Mtpa CIL Plant and owner mining. The addition of the CIL Plant complements the existing heap leach operation by providing processing flexibility and with a means to target those ores with a much lower P Code that are less amenable to heap leaching.

 

12.2.2        Heap Leach Process

 

For LoM planning, heap leach production is now based on 815,000 tpm at the North Plant and 530,000 tpm at the South Plant. This is similar to the 12 month average achieved in F2004.

 

Crushing & Agglomeration

 

The crushing facilities at the North Plant were commissioned by GFGL in March 1998 at an original design throughput of 1,350 tonnes per hour (“tph”). The plant is currently operating at a mean rate of 1,400 tph with surge rates up to 1,700 tph and an overall availability of 85%. The 3-stage plant currently crushes to a P80 of 12.5 mm. The crushing facilities at the South Plant currently operate at a mean rate of 800-850 tph. The 4-stage plant crushes to a finer product than the North Plant with a product P80 of 9 mm.

 

Agglomeration at the North Plant is carried out in three agglomeration drums with a feedrate of approximately 800,000 tpm (436 tph per drum). Agglomeration at the South Plant is carried out through a single agglomeration drum which has a nominal rate of 480,000 tpm (784 tph) but is operating above this. The South Plant drum is currently operating towards

 

41