Discrimination of gossans using principal components analysis of standardized data

Abstract The results of a study involving geology, geochemistry, mineralogy and exploratory statistical analysis of data for a set of gossans located in the south of Portugal are presented. The objective of the case study is to demonstrate how to complement data concerning the mineralogy and geochemistry of gossans with a multivariate statistical method — Principal Components Analysis of Standardized Data — in order to discriminate gossans according to their geological environment. Moreover, a test was carried out in order to assess how summary measures for characteristic variables of each gossan compare (in this case, means versus robust estimates as medians).