Comparative evaluation of the most common kriging techniques for measuring mineral resources using Geographic Information Systems

Abstract

This study evaluates and compares the performances of ordinary, simple, universal, and disjunctive kriging techniques for measuring mineral resources using geographic information systems. These techniques differed in terms of their relative complexities, underlying assumptions, and analytical goals. The methodology consisted of selecting the best kriging techniques, based on their performances in the validation stage, for estimating the depth, thickness, and mass of a mineral resource by interpolating four surfaces representing the surface topography, mineral resource grade, and elevations of the upper and lower surfaces of the mineral resource. The methodology was tested on a phosphate layer detected in the Al-Abiad Mine in Jordan. The total measured phosphate resources were found to be about 1.3 million metric tons. They existed at depths varying from 6.9 to 19.4 m below the ground surface and had thicknesses ranging from 0.2 to 2.3 m. However, it can be concluded that the more complex the spatial behavior of the surface, the more complex is the kriging technique needed to interpolate that surface. Moreover, the deeper and thicker the mineral resource and the denser the spatial distribution of the exploration boreholes, the larger are the relative accuracies of the chosen kriging techniques for interpolating the required surfaces and the greater are the relative accuracies of the estimated depth, thickness, and mass of the mineral resource.