A geostatistical study of the Manhattan gold deposit

Abstract

Geostatistical reserve estimation techniques are divided between those which assume an underlying distribution function for the data and those that do not. This thesis applies a parametric technique, Lognormal Kriging, and a non-parametric technique, Probability Kriging, to the structurally controlled Manhattan gold deposit located in Manhattan, Nevada. The average grade of the deposit is suitable for bulk mining and heap leaching, however, substantial portions of the deposit are high grade mill ore. These high grade pods are small in size relative to the dimensions of the exploration drilling grid. Therefore, interpretation of the exploration data by other than a probabilistic method will give no indication of the presence of a high grade pod other than the average grade of a block. Probability kriging estimates a distribution function for each estimated block. The estimated distribution may be used to determine average grades and tonnages of blocks above specified cutoff grades.