Search Documents

The process of resource estimation of large precious metal deposits is illustrated by reference to major deposits now being mined or developed by Placer.The estimation procedure commences with building an appropriate model of the mineralization and proceeds through steps of data validation and geostatistical studies. It is important that the geologists most familiar with a deposit be involved with all stages of the estimate, and also prepare a manual estimate used to validate the kriged resource.Resource estimates are major exercises taking many man-months of work. The most time consuming part of this is usually construction of an appropriate three dimensional geological model. The data base for the resource calculations must be complete and checked for accuracy. The most appropriate estimation method must be selected.

Jan 1, 1990

By Luke Evans

What are Best Practices? What are Bad Practices? Who Uses Best Practices? Why Bother? Acceptability Guidelines Published Best Practices Published Bad Practices Additional Best Practices Some Best Practices for Low Grade Deposits Interpretation Guidelines Big Bear Gold Deposit Example Conclusions Questions

May 1, 2013

By T. Flitton, R. Peattie

"The extreme depths and consequent expense of drilling and sampling the gold-bearing reefs of the Witwatersrand Basin have resulted in limited data being available for estimation of grade ahead of the mining face. There is, however, a wealth of information from mined-out areas of these deposits. This estimation challenge resulted in the development of a unique method of Mineral Resource estimation.AngloGold Ashanti (AGA) utilizes a technique termed macro cokriging (MCK), which allows for the integration of the limited advanced borehole data with the large chip sample data-sets from the previously mined-out areas by adopting a Bayesian geostatistical approach. The MCK process, in short, is the estimation of mixed support size data together with the application of four-parameter distribution models.The gold value estimation for the Carbon Leader Reef (CLR) on the AGA TauTona Mine is used in this paper as a case study of the process and to demonstrate the effectiveness of this technique through production reconciliation. The current method has been proven over the last 20 years and is now an established part of the Mineral Resource evaluation process within AGA.IntroductionOwing to the extreme depths of the goldbearing reefs at the AngloGold Ashanti (AGA) Witwatersrand operations and the prohibitive expense and time involved in drilling boreholes to the required depth, only limited drilling and sampling data is available ahead of the mining face. Whereas traditional estimation techniques are interpolative, within the Witwatersrand the estimation is primarily extrapolative, with the majority of the data being sourced from the mined-out areas. This significant challenge to estimation resulted in the development of a unique method of Mineral Resource estimation and evaluation."

Jan 1, 2016

By C Grobler, M Recklies, G Jiang, I Taylor, M Morgan, G MacDonald, L Smith

The resource estimation of lateritic deposits, such as Aurukun, presents specific issues related to the lateral changes in thickness and elevation of the various horizontal layers (or zones) within the deposit where the x and y dimensions are orders of magnitude greater than the z dimension. The objective was to develop a three-dimensional (3D) block model that retained the vertical and lateral variation inherent in deposits of this type to allow full optimisation of the production plan without prior selection of an economic portion (enriched bauxite layers) of the profile at the resource estimation stage. The Aurukun bauxite deposit is held by Chalco Australia Pty Ltd (Chalco) and forms part of the world-class Weipa bauxite province, Cape York Peninsula, Queensland, Australia. The bauxite is predominantly pisolitic with an upper boehmite-rich zone and lower gibbsite-rich zone. The bauxite consists of both in situ and reworked domains, is overlain by a thin soil horizon and has kaolinite clay as a transitional base. A portable infrared mineral analyser (PIMA) was utilised on site, in addition to elemental assay, on each drill sample to determine the mineralogy, in particular levels of boehmite, gibbsite, kaolinite and reactive silica. As part of the scope of work, a field program was undertaken using large diameter Boart Longyear sonic drilling to acquire detailed density measurements across the deposit to allow compliance with Joint Ore Reserves Committee (JORC) reporting requirements. Interpretation of the stratigraphy was conducted on a hole by hole basis with definition of the zone layering based on stratigraphy, geochemistry and mineralogy. The zone boundaries were then modelled as 3D surfaces and used to constrain a block model. The block model estimation used the ordinary kriging method, with the estimation done in unfolded space and then refolded, therefore maintaining the zones irrespective of thickness or orientation. The unfolding process converts the real-world positions for both blocks and informing samples to a scaled position relative to the roof and floor of each zone. Analysis of the data and variography were also undertaken in unfolded space. The x and y dimensions are unchanged, but the z dimension is a relative position. The effect of the unfolding and limits on number of samples per hole is to æpushÆ the informing sample search sideways but within the stratification, rather than vertically. This honours the strong vertical zonation within the bauxite profile. The JORC Code requires that the stated resource must have æreasonable prospects for economic extractionÆ. This is a qualitative rather than quantitative definition. A number of selection criteria to define a resource roof and floor were developed in consultation with the project engineers and owners to meet these criteria. The vertical selection of material to include in the resource highlights the variation in the development (and destruction) of the bauxite profile from area to area within the deposit.

Jan 1, 2009

By Michel Dagbert, Mohammed Harfi

"Abstract - The Draa Sfar polymetallic deposit, at about 16 km NW of Marrakech in Morocco, is a massive sulfide zone at the contact between an underlying volcanic and an overlying clastics sequence. The mineralized body is a flat lens with schistosity and "boudinage," tilted to the west and with a strong dip to the east. The lens extends over a strike length of about 2 km and from surface to a depth of about 1 km, with thickness varying from 1 m to almost 30 m. It has been intersected by about one hundred surface drill holes and, in its south part, by about thirty underground fan holes from two exploration levels at depths of 300 m and 400 m. On these two levels, the massive sulfide zone has been channel sampled over a cumulative length of about 0.7 km. Results from drill and channel samples are used to estimate the resources of the orebody according to several methods: sections, polygons around intercepts on a longitudinal plan, grids on the same longitudinal plan with various distance weighting methods including kriging and, finally, a 3D grid with the same range of interpolators. A comparison of the various estimates delineates the important factors in deriving resources for this type of deposit. Those factors are mostly geological and related to the interpretation of the geometry of the two limiting surfaces of the zone. For the 3D gridding, the authors have devised a new and original method that constrains the interpolation of blocks from samples to a geometry of limiting surfaces. The resulting model shows internal grade variations that parallel fluctuations of zone thickness and orientation without having to re-orientate the search window in each block. Résumé - Le gisement polymétallique de Draa Sfar, situé à environ 16 km au NW de Marrakech est un horizon de sulfure massif au contact d’une formation volcano-sédimentaire de base et d’une formation détritique sur-jacente. Le corps minéralisé se présente sous forme d’une méga-lentille aplatie, schistosée, boudinée et déversée vers l’ouest avec un fort pendage vers l’est s’étendant sur une longueur de 2 km et une profondeur de près de 1 km avec une épaisseur variant de moins d’un mètre à près de 30 m. Il a été reconnu par plus d’une centaine de sondages de surface et, dans sa partie sud, par une trentaine de trous souterrains en éventail à partir de deux niveaux à des profondeurs respectives de 300 m et 400 m. Sur ces mêmes niveaux, des traçages ont permis de rainurer le minerai sur une longueur cumulée de 0.7 km. Les données de sondage et de rainurage ont servi à estimer les ressources du gisement selon plusieurs méthodes : sections, polygones sur longitudinale, grille sur longitudinale avec différentes pondérations en fonction des distances incluant le krigeage et enfin grille tridimensionnelle avec le même éventail de pondérateurs. La comparaison des estimés obtenus permet de mettre en évidence les facteurs importants dans l’estimation des ressources de ce type de gisement. Ces facteurs sont essentiellement géologiques et relatifs à l’interprétation de la géométrie des limites des corps minéralisés. Pour l’estimation sur grille tridimensionnelle, on a mis au point une technique originale qui calque l’interpolation des blocs à partir des échantillons sur cette géométrie des surfaces de contact. Le modèle résultant montre des variations internes de teneurs qui épousent les fluctuations d’épaisseur et d’orientation de la zone sans qu’il soit nécessaire de ré-orienter les fenêtres de recherche dans chaque bloc."

Jan 1, 2002

By Binns MJ

Gold/silver orebodies of the Golden Plateau deposit, Cracow, Central Queensland, are currently being exploited by the Cracow Mining Venture. A geostatistical block model has been prepared to assist with long term mine planning. The Golden Plateau orebody, of inferred epithermal origin, is a multiply brecciated, silicified fault complex containing irregular, tabular and pipe- like oreshoots. Historically the lode has been stoped and open cut over a strike length of 750 metres at widths of up to 15 metres and to a depth of 250 metres. The Golden Plateau mine produced 1.55 million tonnes of ore at an average grade of 12 g/t Au and 13.5 g/t Ag over a 42 year period. In the preparation of the reserve model much reliance has been placed on recent diamond and reverse circulation drilling (71 holes in total). However a significant volume of data has been derived from historical level assays and underground diamond drilling. Over 3,000 level assays and 50 underground drill holes contributed to the database. Similar lognormal populations were observed for 1 metre composite grades from each major drilling campaign once corrections were made for core loss. Cumulative frequency plots revealed only one outlier which was cut to the population line. The population statistics from the historical level assays showed a significant difference from those derived from the drill hole composites. Level assay populations were cut to local drill hole populations before incorporation into the ore reserve model. The resulting data has yielded interpretable variograms, the parameters of which have been applied to the preparation of a kriged block model of the deposit.

Jan 1, 1987

By S. Chatterjee, S. Bandopadhyay

A spatial modeling technique based on the Bayesian neural network (BNN) is proposed. Incorporation of the Bayesian method for posterior probability of the output parameter helps calculate the uncertainty associated with the estimate. Parameters of the BNN model were selected using a grid search algorithm. The expected value and the variance of block support were calculated by Markov chain Monte Carlo sampling of the posterior distribution at discretized points within the block. The BNN model was validated by applying it to the Walker Lake data set and comparing the results with ordinary kriging results. The comparison revealed that the proposed BNN method performs marginally better than ordinary kriging. The variance map is less smooth than with ordinary kriging, and the proportional effect is less in the BNN-based model than in the ordinary kriging model. The developed BNN model was applied to the Coho block of a placer gold deposit in Nome, AK. Results reveal that the BNN model is globally biased for the Coho block and underestimates the sample mean. The grade volume curve was developed by calculating the block mean and block variance after discretizing the block into a number of points and applying the BNN model to each discretized point.

Jan 1, 2011

By Peter E. Halbach, Andreas Jahn

"Co-rich ferromanganese crust deposits exist on slopes, summits and platforms of seamounts and guyots throughout the global oceans, where deep-sea currents have kept the seafloor structures more or less free of sediment for millions of years. They form by direct precipitation from cold seawater under more or less oxic conditions and consist of thin layers (2 up to 25 cm thick) covering hard substrate rocks in water depths of 800 to 5000m (Halbach and Marbler, 2009; Hein et al., 2000). Besides Mn these crusts contain interesting concentrations of minor elements (Co, Ni, Ti and Cu) and trace elements (Mo, W, Te, Ga, Pt and REEs), thus they represent significant metal deposits and promise potentially valuable additions to the world’s metal resources.The prospects for mining of this marine type of metal deposits largely depend on the grades and the scale of the potential ore resources. However, only a few percent of existing seamounts and guyots have been mapped and sampled in detail, most of them in the Pacific Ocean. Therefore, a model to estimate the geological resources based on the global seamount catalogue of Wessel (2001; revised: Wessel et al., 2009) and the empirical parameters which control the metal composition and distribution was developed."

Jan 1, 2017

"Abstract - An estimation of resources of structurally complex gold lodes and stockworks represents a challenging task for geoscientists due to the complex geometry of the lodes and discontinuous grade. In the present case study, the resource of a gold stockwork has been estimated by uniform conditioning after the lode has been subdivided into geostatistically defined domains using the indicator probability model. The indicator probability model represents a powerful tool for structural interpretation of the mineralization and can be applied when traditional 3D volume constraining methods (wireframing) have failed to produce a unique, non-subjective model. For comparison, the resources of the same zone have been estimated without domaining. Results of the study clearly show that estimates obtained without domaining of the mineralization can yield a substantial error due to excessive smoothing of the grade.Application of uniform conditioning for modelling the grade distribution has a significant advantage over linear methods (e.g., ordinary kriging). Uniform conditioning enables estimation of local recoverable resources at different selective mining unit sizes that can be substantially smaller than the drilling density without creation of biases or over smoothed local values."

Jan 1, 2002

The development of the Australasian Code for Reporting of Identified Mineral Resources and Ore Reserves is a significan step towards informing the financial community and the public about the future profitability and risks of investing in resource companies. The Code is also important from the accountants' point of view because it provides them with a well-defined bash for assigning costs or attributing value to the assets of resource companies and hence improving their financial reports. Although reporting to the public is ultimately the responsibility of management, accountants, like geologists, are frequently required to provide information for inclusion in these reports. Jr the last ten years, there has been a considerable refinement in the reporting of mineral resources which can be attributed mainly tc the progress made towards developing the Code. It seems tha the classification and terminology that the geologists have developed are widely used in the annual reports. However, little progress has been made by accountants in improving the financia information. The purpose of this paper is twofold: (a) to outline briefl) some of the major problems in accounting and reporting fo; mining companies and; (b) to examine the potential contribution of the Code in solving those problems.

Jan 1, 1990

By M Stewart

In January 1994, Macraes Mining Company started a new phase of exploration to find additional resources along strike and down dip of identified resources for a mill that was being expanded from 2.1 Mtpa to 3.2 Mtpa. At 30th June 1997 a total of 456,784 m in 3,627 holes had been drilled at the Macraes Gold Project with approximately 325,000 m in 2,100 holes drilled in the latest phase of exploration. This necessitated the construction of resource estimates on a more frequent basis and resulted in the formation of a resource modelling team, assisted by Snowden of Perth. Since April 1995, 14 separate resource estimates have been completed. At Macraes, resource estimates are built using Medsystem/Minesight software. The geological interpretation is converted into a set of three dimensional wireframed solid models which are used to code drillhole composites and blocks in a block model. The interpolation method of indicator kriging is used to interpolate grade into the blocks. Classification into measured, indicated and inferred categories is on the basis of geological confidence and drill spacing. The resource estimate is reconciled against grade control polygonal estimates on a bench by bench basis and on a monthly basis. Grade control estimates are reconciled monthly to mill production. For the period 30 June 1995 to 30 June 1997, the resource estimates have under predicted tonnes by 4% and grade by 1% to under predict ounces by 5% at a 0.5g/t cut off. At ROM sulphide cutoff grade, the models have over predicted tonnes by 5%, under predicted grade by 5% to over predict ounces by 1%.

Jan 1, 1997

By J N. Moncrieff

The resource estimation process for the underground Mount Isa Copper Mine has undergone significant change. Block models are now created using a commercial software package as opposed to the in-house software that was previously used. With mineralisation occurring over a strike length of more than 4 km, Mineral Resources for all categories totalled 190 Mt at 2.7 per cent Cu at 31 December 2002. Estimation is by ordinary kriging with two separate models generated to cover the X41 and Enterprise orebodies respectively. The underlying Upper Basement Contact and the silica-dolomite alteration envelope are both used as hard boundaries to constrain the modelling. From a validated block model, grade shells are generated at the appropriate cut-off grade for a given area. The grade shells are depleted for mined voids and subdivided into 30 metre strike length blocks. Application of a SMU criteria and crown and pillar constraints results in failed blocks being culled from the resource inventory. Mineral Resource blocks are further subdivided into resource categories (Measured, Indicated, Inferred) by the Competent Person on the basis of geological confidence, taking account of geological and grade continuity, drillhole spacing and structural complexity (JORC, 1999). A geostatistical æconfidenceÆ category is incorporated into the block model as an additional tool for the categorisation process. The shapes are assessed in concert with a mining engineer to determine whether a reasonable expectation of mining potential exists and if not they are removed from the resource. The final categorised blocks constitute the Mineral Resource inventory and form the basis for conversion to Ore Reserves. The resultant Mineral Resources and Ore Reserves are reported on an æinclusive ofÆ basis.

Jan 1, 2003

By C Boyle

Brockman 2 mine is located in the Pilbara of Western Australia and is operated by Rio Tinto Iron Ore. Mineralisation is hosted by synclines where there is structural complexity to focus fluid flow. Mineralisation is hosted by the Brockman Iron Formation. Mining is on 10 m benches and blastholes spaced at 6 - 8 m apart are geologically logged, sampled by shovel cuts, and assayed using X-ray fluorescence for ten chemical variables. Blast blocks are digitised by mine geologists to assign destinations of material within blasts. Resource evaluation drilling is on a spacing of 50 m × 50 m across known mineralisation and consists of recent reverse circulation drilling and older open hole percussion drilling. Sampling of resource evaluation drilling is by rotary splitters or riffle splitters, and assaying is by X-ray fluorescence for ten chemical variables. Ordinary kriging of resource evaluation produces a long-term resource model. This study examines whether mine planning would be improved by using a blasthole estimate to extrapolate grades into deeper benches and merging the blasthole estimate with the long-term model made using resource evaluation drilling. The two estimates are merged into one model using an inverse weighting based on kriging variance. However, reconciliation results in a test bench show the original estimate made using resource evaluation drilling more closely matches mine geology block outs than does the merged model. This is due to a local trend of more high grade and higher iron grades of high grade in the area. The blasthole estimate of deeper benches must use extrapolation and it is diffi cult to take local trends into account. In contrast, the estimation using resource evaluation drilling is mainly interpolation and is better able to handle local trends.

Jul 11, 2011

By S. Al-Hassan

The deposit has been exploited for Mn oxide and carbonate since 1916. Known reserves of MnO, however, are almost depleted and the survival of the mine will depend on the exploitation of the carbonate. Kriging, inverse distance weighting (IDW) and cross-sectional methods have been applied to the evaluation of the carbonate resources and reconciliation of the estimates obtained shows that kriging and IDW give almost identical block values. A comparison using conditional probability indicates that there is virtually no difference between the global reserve estimates derived from the cross-sectional and kriging methods. The physical causes of these similarities are highlighted and the problems of dealing with multimodal distributions are discussed together with the difficulties of semi-variogram modelling

Jun 19, 1905

By A E. Annels, G F. Johansen, B W. Cuffley

Nuggety gold-quartz reefs are characterised by localised, high-grades and in many cases coarse free-gold. Geometrical and grade complexities are often observed. The high-grade regions are generally erratic and have a low continuity; though may make the deposit economic. Viable grades are generally contained within discrete oreshoots, which are surrounded by barren/low-grade material. Drilling is an effective measure of geological continuity, however grade distribution can only be reliably obtained from underground development and bulk sampling. In general terms, only an inferred resource is likely to be estimated from surface drill data alone. Underground development, further drilling and probably bulk sampling will be required to delineate inferred and indicated resources. Closely spaced development with bulk sampling is likely to be the best way to determine a probable reserve. The resource evaluation process must driven by a clear geological model that should attempt to understand the controls to reef geometry and grade distribution. Geological interpretation has traditionally been undertaken in two-dimensions, though computer software is now being used to produce three-dimensional models. Grade interpolation has generally been undertaken using classical methods, typically weighted averages, zone of influence weighting and polygonal/sectional blocks. Tonnages can generally be calculated from diamond drill and development information with a reasonable degree of confidence. Grade is much more difficult to define with confidence because of its highly erratic and discontinuous nature. To account for variability it is recommended that a best estimate be reported within a grade range, as this more accurately reflects the uncertainties involved. Reporting should be undertaken within the framework of the JORC code, but should involve special consideration of grade reliability.

Jan 1, 2000

By Matt Manson

Renard?1-10 kimberlite pipes and ?Lynx-Hibou?kimberlite dyke system within ?Foxtrot?property 50/50 JV with SOQUEM, Stornoway operator 2001: Discovery of the Renard kimberlite pipes 2003: Discovery of the Lynx-Hibou kimberlite dykes 2006-2007: Bulk sample program 2008: First 43-101 Resource and Preliminary Assessment 2009/2010: Updated 43-101 Resource

May 1, 2010

By Ian Selby

Resource management is a key factor in the operation of a successful marine aggregates business; it is integral to the development of sustainability, satisfying increasing commercial, quality, and environmental performance standards. Resource management is based on the combination of; (i) a thorough evaluation and understanding of the resources on the sea bed and (ii) a dredging plan targeted to permit the effective and efficient extraction of the resources whilst mitigating impacts and leaving the sea bed ready for recolonisation.

Jan 1, 2004

By W. G. Deeks

CANADIANS are world leaders in marketing metal and mineral products. Noranda Sales Corporation Ltd. is an international metal and mineral product marketing company operating from Toronto and London, England, where it operates a wholly owned sales subsidiary and is a 50% partner in Rudolf Wolff and Company Ltd., a charter member and the largest metal broker on the London Metal Exchange. Noranda Sales employs 97 people, of whom 52 are university trained and professionally competent in international resource product marketing. We market for 19 companies or divisions in the Noranda Group and 39 companies for whom we act as agent, consultant and purchaser of resource production. Our business in 1976 was 1.5 billion dollars carried out in 45 countries and covering 24 products and materials. Today, ownership of most of the mining industry in Canada has transferred back to Canadians, and this is said with the knowledge that a significant part of the Canadian industry has been Canadian owned and managed from its very early days. Noranda Mines Limited is such an example. It was born in 1922 as the result of a copper-gold showing at Noranda, Quebec, discovered by Ed Horne, a prospector, financed with a grubstake by a syndicate mainly composed of New York businessmen. Shortly after the discovery, when the shares of the company had risen, the syndicate was sold to Canadian interests and, since that time, Noranda has been dominantly Canadian owned and certainly Canadian managed. Today, over 90% of Noranda's shareholders are Canadian, who hold an equivalent percentage of the stock.

Jan 1, 1977

By R Dimitrakopoulos

The development of mineral resources is based on a spatial model of the orebody that is only partly known from exploration drilling and associated geological interpretations. As a result, orebody models generated from the available information are uncertain and may require the use of stochastic conditional simulation techniques. Multiple-point methods have been developed for petroleum reservoir modelling, enabling reproduction of complex geological geometries. This paper considers a multiple-point approach to capture the uncertainty of the lithological model. The implementation of the method at the Yandi channel iron ore deposit in Western Australia and performance characteristics are discussed. The uncertain lithological model translates into considerable grade-tonnage uncertainty.

Jan 1, 2007

By C K. Boafo, J Visser

Obuasi Mine in Ghana, with more than 100 years of mining history, developed from a surface operation into a deep >2000 m underground gold mine. The challenge is to extract value from narrow veins left after extensive mining in the wider ore zones of the mine. Since the merger of AngloGold and Ashanti Goldfields in 2004 the mine strives to achieve optimum extraction through an integrated Mineral Resource management process from geological modelling, resource evaluation to mine design and scheduling. The shear hosted gold deposit displays complex secondary shears of variable thickness and continuity. There is an ongoing process of changes taking place in the way that geological modelling and resource estimation is done. Recent developments include the use of computerised systems that support the whole resource management chain from resource modelling to reserve optimisation. The change in thought and technology now allows for targeting narrow veins previously ignored in optimisation and mine design since moving into mechanised mining. The increased knowledge and confidence in the resource models now allows for flexibility in mine design, selection of mining method and scheduling and extraction of the reserves. The focus area for change is the way the orebody is evaluated and extraction designed. Since 2005 major changes were made to the process. This paper considers the historical approach of resource management. It looks at the individual process changes and qualifying the way ahead.

Jan 1, 2008