Identification of geochemical anomaly and gold potential mapping in the Sonakhan Greenstone belt, Central India: An integrated concentration-area fractal and fuzzy AHP approach

The present work combines Concentration-Area (C-A) fractal model with the Fuzzy Analytical Hierarchy Process (FAHP) on a GIS platform for mapping gold potential in the Sonakhan Greenstone Belt, India. A set of stream sediment geochemical data obtained over a part of the study area was used to delineate potential regions of gold mineralisation. In order to develop a suitable predictive model, the gold exploration target was taken as target (dependent variable) and concentration of pathfinder elements of gold such as Au, As, Ag, Hg, Sb and Se in the stream sediments were used as predictors (independent variables). The C-A fractal model was applied to the geochemical data of each element to decompose anomaly and background components of the spatial dispersion of element concentration in the study area. AHP combined with fuzzy set theory (FAHP) was used to determine the priority or, the weight of each evidential geochemical anomaly map. The Gold exploration targets are delineated by multiplying the weights with the respective fuzzy normalised geochemical anomaly map of the gold pathfinder elements and integrating the weighted geochemical evidential layers using fuzzy Gamma operator. The reliability of the outcome was assured by the coincidence of known gold mineralisation with the potential regions delineated in the final gold potential map. The areas with high potential are good targets for detailed exploration of gold in the region. The results highlight that (a) application C-A fractal model can effectively separate geochemical anomalies as it considers the spatial variation of the data unlike the conventional statistical methods which account only the frequency of the data, (b) the FAHP allows more flexibility in judgements of multiple decision makers in a group and reduces the inconsistency of the result which is essentially required in the knowledge-driven predictive mapping of mineral prospectivity.