A fractal filtering technique for geochemical and geophysical data processing in GIS environments for mineral exploration

Regional geochemical and geophysical data are two types of basic data sources for mineral exploration. These data usually contain the accumulative effects generated by multiple geological processes. Decomposing the mixed data into anomaly components caused by mineralization or mineralization-related geological activities from background components due to the normal geological processes is a fundamental task for exploration geologists and the basis for geochemical and geophysical data processing techniques.; The objective of this thesis is to investigate the proper technologies for extracting information from regional geochemical and geophysical data for the purpose of mineral prediction. Fractal filtering technology (S-A) that we recently developed on the basis of multifractal theory has been systematically investigated and demonstrated as an information extraction technology that can separate a complex geochemical or geophysical field into particular components with distinct scaling properties. The method involves multifractal filters defined on scaling properties of 2-D power spectra.; Applications of the fractal filtering method to arsenic (As) geochemical concentration values of lake sediment geochemical samples from southwestern Nova Scotia, Canada, have demonstrated that the method can effectively extract the anomaly components with anisotropy from the background with significant variation, for which most traditional statistical geochemical data processing methods and frequency-based regional geophysical data processing methods usually do not work well. The decomposed components with clear geological meanings can be used to estimate mineral resource potentials for the turbidite-hosted Au deposits in the study area.; The S-A has been successfully applied to identify specific rock units on airborne radiometric eU/K and eU/eTh images by removing the high value “noise” and low value background components. It has been confirmed by GIS spatial analyses upon geology, Sn-W-U mineralization and multi-element geochemical anomalies that the anomalies extracted from the two ratio images are capable of delineating Sn-U-W related late stage intrusions and alteration zones.; A multivariate analysis, Principal Component Analysis (PCA), was applied to the decomposed anomalous components obtained by using the S-A for Au, As, W, Sn and Sb geochemical maps, which improved the analysis result significantly in comparison with applying PCA to the bulk values.; The patterns decomposed by the S-A method with different ranges of scales and orientations are useful for identifying the hierarchical spatial relationships among mineralization-controlling factors. Integration of the three components obtained by applying the fractal filtering method to the As geochemical map provides a clear view how the multiple stage and scale mineralization processes finally led to the enrichment of precious metallic element Au and where to find the high potential districts for exploration. Integrating the anomaly components obtained from radiometric eU/K and eU/eTh images and geochemical elements F, Li, Rb, Nb and Th shows that the two types of anomalies may indicate the locations of Sn-W-U deposits.