扬子西缘小石桥新型锂矿的地球化学低弱异常信息提取及地质意义OA

Due to the complexity of geological processes and metallogenic systems,geochemical anomalies in exploration often exhibit weak and subdued characteristics,as well as complexities arising from the superposition of multiple distribution patterns.This limits the effectiveness of identification methods based on the normal distribution.To improve the recognition of weak and subtle anomalies,this study systematically compares various geochemical anomaly identification methods,using the Xiaoshiqiao new-type lithium deposit in central Yunnan as an example.To address the issues of weak interpretability and low convergence efficiency of Self-Organizing Maps(SOM),as well as the sensitivity of k-means++to initial cluster centers,a hybrid clustering algorithm combining SOM and k-means++is proposed for geochemical anomaly identification in the study area.Comparison with Gaussian Mixture Models,the concentration-area fractal method,and local singularity analysis shows that the hybrid method can effectively identify weak anomalies in high-dimensional data,and the delineated anomaly areas correspond well spatially with known lithium ore bodies.Fry analysis based on five sets of target anomaly points further reveals that lithium enrichment exhibits a north-south trend at multiple scales and is closely related to negative topography,suggesting that north-south-trending linear structures may be the main factors controlling the spatial localization of lithium deposits in this area.This understanding provides insights into the reinterpretation of the metallogenic mechanism of lithium deposits on the western margin of the Yangtze Block and offers a scientific basis for subsequent prospecting and exploration.