Characterizing The Mineralogical Variability Of A Chilean Copper Deposit Using Plurigaussian Simulations

Betzhold, J.
Organization: The Southern African Institute of Mining and Metallurgy
Pages: 10
Publication Date: Jan 1, 2000
Knowing more about an orebody makes it easier to exploit it profitably. Many orebodies are made up of different rock types, alteration zones or mineralogical ensembles: oxides versus sulphides or facies with different metallurgical properties for example. Being able to model the spatial layout of these different facies is vital to good mine planning and management. Kriging cannot be used to generate numerical models of orebodies with complex geometric patterns because it gives too smooth an image. Over-smoothing has the meaning of changing the natural selectivity. Conditional simulations are required to do this kind of model respecting the natural variability: honouring the natural selectivity, dilution and ore loss. Standard geostatistical simulations cannot duplicate many of the complex spatial relationships seen in the earth sciences, such as differently orientated facies, sets of conjugate veins or ore types where certain facies cannot touch. After a presentation of the case study, it will be shown why traditional methods for simulating indicators like the sequential indicator simulations, or the truncated gaussian method, cannot be used to simulate it. Plurigaussian simulations are then introduced and shown to be more powerful than the previous methods in their range of application. This is confirmed by the case study treated; a Chilean copper deposit, for which the plurigaussian simulations are seen to respect the different orientations of the individual facies as well as honouring the facies seen in drill holes, and respecting the proportion of each facies even if this varies in space (e.g. increases or decreases with depth). Key words: geostatistics, indicators, conditional simulation, plurigaussian method, copper variability.
Full Article Download:
(1767 kb)