CLASSIFICAÇÃO DE RECURSOS MINERAIS USANDO A VARIÂNCIA DE INTERPOLAÇÃO

Abstract

This paper presents a proposal for mineral resources classification using the interpolation variance. This alternative for measuring the uncertainty associated with the ordinary kriging estimate allows characterization of the local dispersion of data, as well as recognition of the proportional effect when it is present. Therefore, the interpolation variance presents the main characteristics for quantification of uncertainty and consequently it can be used for mineral resources classification. The tolerance error, calculated with the interpolation variance, is used to establish certainty classes, according to a classification model widely accepted in the mineral industry. The proposed model was tested for Pb and Zn resources classification for Canoas 2 Mine. According to the results, the classification provided by the interpolation variance is more reliable than that provided by kriging variance. It is important to note that the tolerance error leveis adopted for mineral resources classes will not allow classification of ali mineral deposits, mainly those that present high natural variability. However, the proposed methodology quantifies the tolerance errors, which are important for the knowledge of the deposit, as well as for mine planning.