Stochastic simulation applied to a copper synthetic dataset

Abstract

This article presents the development of a copper synthetic mineral deposit and its appropriateness for evaluating geostatistical methods since it is possible to compare the results obtained with the current values and those methods can be performed from any sampling of the deposit. The creation of this kind of model is essential to quantify the quality of geostatistical methods once it is impractical to do it in real mining sites – due to the fact that, even exhausting the mine and computing the ore of every single block, one won’t have the entire knowledge of the deposit. The synthetic model is based on a real copper mineral deposit model which presents geological and spatial statistics heterogeneity. The model reproduces a deposit with hydrothermal filling a reverse fault and bedrocks are meta-sandstones and shales folded. The spatial distribution was created by unconditional Sequential Gaussian Simulation. The sampling in the synthetic deposit was done by boreholes with separation of holes approximately 40 meters to the north and to the east. The direction and dip of the holes are approximately 270/45. The geostatistical analysis and stochastic simulation were calculated in this database and comparisons were made with respect to the deposit model. Comparisons were made by observation of descriptive statistics of the results, the sample, and the whole synthetic model. Moreover, the Q-Q plots were made to compare the distributions derived from simulation with the population distribution of the deposit. It was concluded that the experiment was satisfactory, because the results obtained by a known efficient simulation technique were consistent.