Mineral prospectivity modelling using singularity mapping and multifractal analysis of stream sediment geochemical data from the auriferous Hutti-Maski schist belt, S. India

Abstract

The regional stream sediment geochemical survey has become an all the more important exploration technique for delineating potential mineral zones to target exploration. However, the effective appraisal of pathfinder elements has remained a fundamental challenge in exploration geochemistry. In this regard, the singularity mapping technique based on modern nonlinear theory and fractal-multifractal geometry is a new efficient method to distinguish thresholds of anomalies based on stream sediment geochemical data. The auriferous Hutti-Maski schist belt in Eastern Dharwar Craton (EDC), S. India has been taken up for delineation future targets using local singularity analysis (LSA) of stream sediment geochemical data. Ordinary Singularity Mapping (OSM) and Spatially Weighted Singularity Mapping (SWSM) are employed for mapping singularity of gold pathfinder elements (Au, As, Ag, Hg, Sb, Se, Cu, and Zn). In SWSM, a distance-based fuzzy weighting method is used to estimate the spatial weighting factor with respect to ore controlling shear zones. The local isotropic singularity index and the local anisotropic singularity index are estimated for each of the eight elements. The Student’s t-statistics based on Weight of Evidence (WOE) model is used to determine the threshold singularity index value to separate geochemical anomaly from background. Principal Component Analysis (PCA) is applied to singularity maps of multiple elements derived using OSM and SWSM techniques. The first principal component (PC-1) scores estimated by the OSM-PCA and SWSM-PCA lead to comprehensive geochemical anomaly maps. Results of two different singularity mapping techniques are compared using Geochemical Mineralization Probability Index (GMPI), Concentration-Area (C-A) fractal analysis and Prediction-Area (P-A) plots. It is observed that the SWSM method results in a smaller target with higher predictive ability compared to OSM method that delineates a larger target area.