Measures of Uncertainty for Resource Classification
For many decades the mining industry regarded resource estimation and classification as a mere calculation requiring basic mathematical and geological knowledge. Often uncertainty associated with tonnages and grades were either ignored or mishandled. With initiatives to establish international standards for classifying mineral resources and reserves, it is important to establish the level of confidence in the results and correctly assess the error. Among geostatistical methods, Ordinary Kriging (OK) is probably the one most used for mineral resource estimation. It is known that OK variance is unable to recognize local data variability, which is an important issue when heterogeneous mineral deposits with higher and poorer grade zones are being evaluated. This study investigates alternatives for computing estimation variance from ordinary kriging weights that account for both the data configuration and the data values. These estimation variances are then used to classify resources based on confidence levels and their results are compared with those obtained by OK variance. The methods are illustrated using an exploration drill hole data set from a large Brazilian coal deposit. The results show the differences in tonnages within each class of resources when different measures of uncertainty are used.
KeywordsAnisotropy Covariance Pyrite Kriging
Unable to display preview. Download preview PDF.