Use of spatial analysis in mineral resource evaluation

Abstract

Methods of spatial statistics are useful for constructing contour maps of geological, geophysical, and geochemical features in aid of mineral resource appraisal. Statistically estimated regional trends, as well as local signals, can provide useful information on the occurrence probability of hidden mineral deposits. A specific problem of probabilistic mineral resource evaluation is to estimate the probability that a cell or group of cells belonging to a grid superimposed on a map contains mineral deposits of a given type. Typically, in regional studies the cell is square shaped measuring 10 or 40 km on a side. Numbers of deposits and quantities of metals or hydrocarbons are to be predicted from variables systematically measured for all cells. In this approach it is important to model the frequency distributions of the measurements as a function of cell size. Continuous and discrete random variables for cell data can be modeled in different ways using orthogonal polynomials. Each model provides guidelines for quantifying the information on rock types and mineral deposits displayed on geoscientific maps, and produces practical methods for treating these data.