Singularity theory and methods for mapping geochemical anomalies caused by buried sources and for predicting undiscovered mineral deposits in covered areas

In this paper, it is shown that recently developed nonlinear singularity theory and methods can be used to recognize weak but complex geo-anomalies for the prediction of the presence of mineral deposits in areas covered by deserts, regolith or vegetation. The theory and models of singularity and generalized self-similarity developed in the context of multifractals are proposed for analyzing weak anomalies caused by buried mineralization. These anomalies can be extracted from geochemical stream sediment data and can be used for the prediction of the Fe and Sn mineral deposits of skarn and hydrothermal types in Gejiu, Yunnan and Eastern Tianshan, Xinjiang, China. Significant portions of these areas are covered by vegetation, desert or regolith. The concepts and models of local singularity and generalized self-similarity are utilized to demonstrate that the singularity index, which is the exponent of the power-law associating density with scale (area or volume) of geochemical anomaly, determines an essential dimensional property of geochemical anomaly that is independent of its geometrical scale. Singularity values calculated by means of the local singularity analysis method are capable of enhancing weak geochemical anomalies caused by deeply buried sources. The principles and methodologies proposed in this paper are demonstrated by a case study of predicting the presence of Fe mineral deposits in Eastern Tianshan, China. The singularity analysis methods, in conjunction with combined low-pass and high-pass filtering methods, were successfully applied to process regional stream sediment geochemical maps, gravity map and aeromagnetic map with the aim of extracting weak anomalies revealing locations of mafic volcanic rocks, felsic to intermediate intrusions, skarn and hydrothermal alterations in the study areas. These geological features are genetically associated with skarn and other hydrothermal iron deposits in the area. A modified fuzzy weights-of-evidence method with a correction for conditional independence of evidence was applied to integrate the evidence and to create a posterior probability map. The results show that weak geochemical anomalies caused by buried mineral deposits in the considered areas are significantly enhanced and extracted from variable backgrounds. Approximately 30% of significant target areas delineated by the integrated geo-anomalies in the study areas had been previously ignored. (C) 2012 Elsevier B.V. All rights reserved.