Conditional Simulation for Resource Characterisation and Grade Control - Principles and Practice

Conditional simulation is a set of geostatistical tools that is becoming popular as powerful personal computers make the use of these tools less costly. The more familiar set of tools for spatial estimation has a number of varieties (eg Ordinary Kriging, Simple Kriging, Multiple Indicator Kriging, etc) each suited to particular solutions of the problem of estimation, for example predicting the average grades of blocks in a region from more or less sparse data. Similarly, conditional simulation has a number of varieties, discussed in this paper, which in various ways address the problem of characterising the variability of a deposit between the known data locations.Conditional simulation can provide a number of equiprobable images of a deposit, at a finer resolution than would be economic or practical to sample, and that can be constrained by various criteria (which may be geological, statistical and geostatistical), to ensure that the images reasonably represent the deposit. These images can then be used to assess the risk involved in various practical mining decisions. Examples are provided where conditional simulation has been used for validating resource estimate parameters, optimising grade control sampling grids, and for ore block optimisation for mining grade control.