Objective based geochemical anomaly detection—Application of discriminant function analysis in anomaly delineation in the Kuh Panj porphyry Cu mineralization (Iran)

Abstract

A common objective method for anomaly detection in geochemical exploration is target delineation by discriminant function analysis. Discriminant analysis (DA) is a multivariate statistical technique that classifies each observation into a specific group based on observed predictor variables and predefined groups. In the present study a new approach is considered for geochemical anomaly identification employing DA and “real” pre-defined “anomaly” and “background” data set. The anomalous and background samples are identified based on presence or absence of mineralization in depth; so, this method is introduced as “objective approach”. In order to classify surface geochemical samples into anomaly and background, assays of core drillings in the Kuh Panj porphyry Cu mineralization are used. They are classed as anomaly if the presence of mineralization is proven and are labeled as background if the absence of mineralization is confirmed in cores. Stepwise Linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) are utilized to achieve discrimination functions. For the test data used to generate the models, both LDA and QDA methods have led to perfect classification however cross validation has shown 84% and 74% total correct classification for LDA and QDA respectively. Outcomes of this research have demonstrated that LDA can effectively be employed as an objective method for geochemical anomaly identification if available information from geology and geochemistry of target area are employed and utilized. It is also shown that the definition of anomalism in geochemical exploration can be improved remarkably by this approach.