| Recently, with the rapid development of global industrialization and urbanization, human demand for mineral resources is unprecedented urgent, and the serious shortage of mineral resources has become a huge bottleneck restricting the global economic development. The present situation of mineral resources is extremely severe, the proven metal deposits are mainly small ores or lean ores, and through long-term mining, the shallow mine resources are almost exhausted. So the focus of mineral exploration has to tilt toward deep prospecting, but greatly increasing the difficulty and cost of traditional mineral exploration. Therefore, the combination of new scientific technology such as computer, GIS and image processing, to explore a set of quantitative synthetic assessment system of mineral resources is of great practical significance to guide the prospecting direction of new ores and concealed ores, to relieve the global mineral resources crisis and to promote the development of social economy.This study chooses the typical region of gold mine in Zhejiang segment of Qinhang metallogenic belt as the study area WIth the full knowledge of ore-forming conditions and metallogenic mechanism of the study area, we extract multiple metallogenic information in the GIS platform with multiple geosciences data collected and organized previously, mainly including geological, geophysical and geochemical exploration, terrain and remote sensing data. And then use two kinds of mineral resources assessment technology, namely weight of evidence (WofE) method and support vector machine (SVM) method, for comprehensive assessment of gold metallogenic potential in the study area. Finally, preliminarily delineate the metallogenic prospective areas. The involved contents and achievements include:(1) With the full knowledge and summary of the ore-forming geological background, regional ore-controlling structures and the mineral characteristics of the shear zone type gold deposits in the study area, pre-process the multipal geological data of the geologic map, the geophysical map, the chemical anomaly map, DEM data and remote sensing images with the GIS platform, and dig into multiple geosciences data and extract multiple ore-forming favorableness.(2) In view of the characteristics of high vegetation coverage in the study area, abandon the traditional remote sensing prospecting technology focusing on the extraction of weak alteration information, and innovatively put forward the concept of "remote sensing environment unit". This concept indicates that multi-band remote sensing information reflects the comprehensive influence of geology, lithology, soil, climate and other factors in a certain environment unit, and the enrichment of mineral resources inevitably leads to the particular remote sensing environment. Therefore, with the use of decision tree classification technology and aiming at isolating ore-forming favorable environment units, extract the remote sensing environment factor.(3)Considering the characteristics of the mineral resoures in the study area, put forward the "mineral comprehensive assessment strategy of mineral resources with WofE and SVM model". On the base of optimizing mineral assessment factors according to the weight coefficient calculated from WofE model and the comprehensive assessment index of ore-forming favorableness put forward in this study. Then use the SVM model for comprehensive assessment of gold deposits in the study area, and use the assessment result with WofE model for mutual authentication. Finally, with multiple ore-forming factors, analyze the prospect result.Practice has proved that the study area is rich in geological exploration data and multiple metallogenic information, so the traditional way of artificial interpretation based on expert experience cannot take full advantage of a large number of geological information, and it is good to adopt the "knowledge driven" comprehensive assessment technology of mineral resources with GIS platform. When a single assessment strategy is not enough to meet the needs of the research, comprehensively take advantage of two different assessment models to authenticate to each other from two dimensions, thus ensure the accuracy and reliability of the results. At the same time, to make full use of multispectral remote sensing data information and to build "the remote sensing environment factor" is also an important supplement to traditional geological mining exploration and has profound significance to the periphery and deep prospecting. |
PDF Full Text Request