Fractal Modeling of Geochemical Mineralization Prospectivity Index based on Centered Log-Ratio Transformed Data for Geochemical Targeting: a Case Study of Cu Porphyry Mineralization

This work aims to investigate the geochemical signatures of the Cu porphyry deposit in the Dalli area using the geochemical soil samples. At the first step, the geochemical data was opened using the Centered Log-Ratio (CLR) transform method. Then those outlier samples that reduce the accuracy of the geochemical models were detected and removed using the Mahalanobis Distance (MD) method. We applied the Principal Component Analysis (PCA) and Geochemical Mineralization Prospectivity Index (GMPI) methods on the cleaned transformed geochemical dataset. The PCA method identified five principal components (PCs), from which PC1 including Cu, Au, and Mo, are specified as the mineralization factor (MF). The GMPI approach can improve the multivariate geochemical signature in geochemical mapping. Hence, the GMPI values of the samples were calculated based on the score values of MF (Cu, Au, Mo). The results convey that the large values of GMPI (MF) (Cu, Au, Mo) strongly correlate with the quartz diorite porphyry rocks and potassic alteration zones. The GMPI (MF (Cu, Au, Mo)) index was modeled using the Concentration-Number (C-N) fractal method. The C-N fractal model identified four geochemical populations based on the different fractal dimensions. The geochemical anomaly map of GMPI (MF) (Cu, Au, Mo) was delineated using these classified populations. The obtained promising areas were validated adequately by more detailed exploration works and deep drilled boreholes as well. The Cu-Au mineralization potential parts are appropriately mapped by this hybrid method. The results obtained demonstrate that this scenario can be adequately used for geochemical mapping on local scales.