Application of spatial information fusion techniques in GeoDAS for mapping Sn mineralization associated intrusions in Gejiu ore district, Southern Yunnan, China

Geochemical, geophysical, and geological anomaly mapping techniques are common prospecting methods in mineral exploration. A great number of methods and techniques have been developed over the past two decades to identify and integrate anomalies from multi-source data. Geographic Information Systems (GIS) have shown a noticeable efficiency in data management and information extraction in support of mineral exploration. This research demonstrates the application of recently developed spatial information fusion techniques provided in GeoDAS GIS for mapping Sn mineralization-associated intrusions in the Gejiu ore district, Southern Yunnan, China, which is facing the problem of a mine resources crisis. The information fusion process created a new model connecting intrusions related anomalies by fusing direct and indirect relationships between anomalous information. For the first time the method practically shows a map of intrusions both on surface and buried at depth which provides geo-information to support future mineral exploration. Datasets currently used include stream sediment sampling, gravity and aeromagnetic data at 2000 m spatial resolution and remotely sensed ETM+ data at 30 m spatial resolution. At this time, information fusion aims to: (1) delineate the spatial distribution of intrusions on the surface and buried at depth; (2) summarize a feasible procedure for integration of spatial information from multi-source and multi-scale datasets in the Gejiu ore district. A series of newly-developed spatial information fusion techniques implemented in GeoDAS GIS and used in this study include: 1. for single source data, (1) the singularity mapping technique is applied to process geophysical data, and (2) the Principal Component Analysis (PCA) technique is applied to geochemical data to delineate areas of influence of intrusive rocks; 2. For multi-source data, PCA is applied to combine the information about intrusions from multiple sources, including (1) gravity and aeromagnetic data, and (2) geophysical and geochemical data; 3. for multi-source and multi-scale data, Spatially Weighted Principal Component Analysis (SWPCA), a new version of the PCA, is used to integrate information from geochemical and geophysical data with remotely sensed ETM+ data at 30 m spatial resolution; 4. The student's t test in the Weights of Evidence method is applied to test the spatial correlation coefficient between the results and the mapped intrusive rocks from a 1:200,000 scale geological map of the study area. PCA or SWPCA integration results obtained from multi-source and multi-scale datasets well demonstrate the spatial distribution of intrusions. Validated results will be used to evaluate the irregularity of intrusions by a perimeter-area (P-A) fractal model.