Remote sensing-based structural and lithological mapping for prospecting polymetallic mineralization at Xiaoshan region, China

The Xiaoshan region of China is distinguished for its polymetallic ores affected by faults and geological interfaces. Geological assessments are difficult due to the uneven terrain and dense vegetation. To address this, integrated remote sensing imagery through GIS-based modeling, utilizing a weighted overlay tool, has emerged as a widely recognized approach for identifying structural features and geological units critical to mineral resource distribution. An automated lineament extraction technique was implemented using Landsat-8 data, revealing predominant NW-SE and EW-oriented structures. Furthermore, a hill-shade algorithm was employed on the SRTM-DEM data to delineate linear features. The Support Vector Machine (SVM) method was applied to the Sentinel-2 dataset to delineate rock mass boundaries based on Optimum Index Factor (OIF) and Principal Component Analysis (PCA) derived images, attaining an overall accuracy of 81% and a Kappa coefficient of 0.78. Thematic layers including lineament density, lineament intersection, distance to lineaments and faults, and lithology were re-classified and weighted based on their influence on mineralization. Consequently, this integrated methodology has revealed a substantial correlation between geological structures and rock interfaces, identifying promising zones particularly in the northern, western, and northeastern sectors within the Xiaoshan uplift at the interfaces of Taihua, Yanshanian, mylonitic, and Xushan outcrops, which had not been previously mapped. This study illustrates that the GIS-based modeling through a weighted overlay tool represents a valuable strategy for directing metallogenic predictions in the challenging terrains of the Xiaoshan belt. The findings provide crucial insights for future regional exploration and field investigations.