Geographically weighted regression in mineral exploration: A new application to investigate mineralization

Geographically weighted regression (GWR) is an effective model for the investigation of spatially nonstationary relations among variables in the geographical and social sciences. GWR was introduced to the field of mineral exploration to further understanding of the location, controlling factors, and coupling mechanisms related to the triggering of mineralization—in other words, the where, what, and how. Previous studies reported that Cu and Au in a porphyry system present a paragenetic relation at different stages of mineralization, which can be an informative indicator in mineral exploration. As a successor, the current study further applies the GWR model to characterize the paragenetic relation between the ore-forming elements Cu and Au in the Duolong mineral district of Tibet, China, in a spatial scenario. Unlike the spatially varied ore-forming mechanism quantified by the regression coefficients of GWR, the coefficient of determination (R2) is discussed to verify the existence and to evaluate the strength of the paragenetic relation between Cu and Au, because regression coefficients can only inform the mutual influence between one and the other. Furthermore, the fractal and multifractal-based spectrum–area method is adopted to separate the GWR results into anomaly and background. Areas with GWR results that indicate the existence and intensity of a paragenetic relation are mapped as target areas for mineral exploration. The current quantitative recognition of mineralization represents a meaningful and useful extension to the application and interpretation of the GWR model.