【作者】 杨自安；

【导师】 王润生；

【作者基本信息】 中国地质大学（北京） ， 地图制图学与地理信息工程， 2005， 博士

【摘要】 西部高寒山区是我国极具找矿潜力的重要成矿区域。快速勘查评价这些区域的矿产资源,变资源优势为经济优势,是带动西部地区经济发展的重要途径之一,但是西部高寒山区自然环境条件恶劣,找矿难度较大。可见,探索适合于西部高寒山区矿产资源勘查评价的技术方法,提高找矿效率,具有非常重要的现实意义。本文以青海省乌兰—都兰地区为研究区,采用多光谱遥感信息提取技术,从ETM图像中定量提取出与金属矿化有关的蚀变信息;运用多种化探数据处理方法,对水系沉积物地球化学测量数据进行处理分析,圈定出各类化探异常;基于多元地学信息间的套合和耦合关系,对遥感与化探信息进行综合处理及相关性分析;应用GIS空间分析技术,进行研究区主攻金属矿产的靶区定位预测与评价,初步形成了一套适合于西部高寒山区的、遥感与化探信息综合找矿定位预测的有效技术方法组合和工作流程。取得的主要研究成果如下: 1、以区域成矿地质概念模型、岩矿光谱机理及特征分析为基础,采用特征主成分分析、SVM 技术等,从多光谱遥感图像中定量提取出与金属矿化有关的蚀变信息;基于最优分割法和费歇尔准则,研究改进出一种新的异常分级方法—最优密度分割,避免了阈值选择和人为因素影响。2、在地球化学分区基础上,采用多重分形法,辅助以迭代法、聚类分析、主因子分析,对水系沉积物地球化学测量数据进行处理与分析,较真实地提示了各地球化学元素含量空间分布规律,圈定出各类化探异常;应用成矿能量和系统核度理论,优选出矿致异常,为综合找矿定位预测提供了重要信息。3、基于多元地学信息的套合和耦合关系,创新性地建立起综合分析模型,并采用相应的技术方法,结合自然景观环境和成矿地质背景分析,对遥感与化探信息的综合处理及相关性分析进行了较为深入的探讨。4、通过多元地学信息的找矿有利性和空间耦合度分析,优选出有利综合评价的地质变量,建立起综合预测量化模型;以网格单元和最小汇水盆地为地质统计单元,采用特征分析法、证据加权法,进行了研究区主攻金属矿产的靶区定位预测与评价,圈定出成矿远景区和找矿靶区,为研究区的进一步地质找矿工作提供了重要依据。本文所研制的遥感与化探信息综合找矿定位预测的技术方法组合和工作流程,对西部高寒山区全面开展矿产资源勘查评价,快速扩大地质找矿成果,具有较好的参考价值和实际应用意义。更多还原

【Abstract】 The high and cold mountainous areas in West China are metallogenic provinces of great ore prospecting potential, and exploring and assessing the mineral resources in the areas to turn resource advantages into economic advantages is one of most valuable approaches in terms of boosting the economy of West China. However, harsh natural environments hinder us to go there for ore prospecting. As a result, it is of great significance to explore techniques suitable for mineral resources prospecting and assessment in these areas to improve geologic prospecting efficiency. With Wulan and Dulan in Qinghai as the regions of interest, by adopting multispectral remote sensing information extraction techniques, alteration information related to metalliferous mineralization was extracted from ETM images; various data processing methods were employed to process the stream sediment data to delineate geochemical anomalies; based on the matching and coupling relationships between multivariate geosciences information, integrated processing and relevance analysis were conducted for RS information and geochemical prospecting information; by using GIS spatial analysis techniques, target area positioning predication and assessment were conducted for major metallic minerals in the areas of interest; and as a whole, a combination of effective techniques of comprehensive positioning prediction by integrating RS and geochemical prospecting information and its work flow were formed which are suitable for the high and cold mountainous areas in West China. Major findings are as follows: 1. Based on a geologic concept model for regional metallogery, rock-mineral spectra mechanism and characteristic analysis, by adopting principal component analysis and SVM techniques, alteration information related to metalliferous mineralization was extracted from multispectral remote sensing images; based on optimum slicing method and Fisher criterion, a new anomaly grading method, i.e., optimum density slicing method, was derived, which helps to avoid threshold selection and man-made influence. 2. Based on geochemistry zonation, by adopting multiple fractal method, iteration, cluster analysis, and principal factor analysis, stream sediment data were processed and analyzed, which revealed the spatial distribution of geochemical element contents and helped to delineate the geochemical anomalies; by employing metallogenic energy and a system core degree theory, anomalies related to mineralization was screened out, which provided important information for integrated positioning predication. 3. Based on the matching and coupling relationships between multivariate geoscience更多还原