| With the developing of geological field exploration and scientific research,the amount of geological data and maps are also in the explosive growth.Along with the coming of the era of big data,the methods and thinking modes of big data have begun to combine with various disciplines.It also provides an unprecedented good opportunity for geology,and makes it possible to take a full advantage of the accumulated datum and results for the last couple years.This study has chosen a new brand ore concentration area called NalingXuancheng Mine Area,which possesses a prospecting prospect for the prediction of deposit and locates in the middle and lower reaches of the Yangtze River.This paper systematically collected the existing datum of gold deposits named Yaojialing Zn-Au polymetallic deposit and Tongshan-Qiaomaishan Cu-S-W deposit,which is representative in this area.According to the previous three-dimensional solid model of Yaojialing Zinc-Gold Polymetallic Deposit and the method experiment.A complete 3D model of Tongshan-Qiaomaishan Copper-Sulfur-Tungsten Deposit,analysis of metallogenic prognostic indicators and metallogenic prognosis were carried out.It chose the suitable methods to respectively restored the three dimensional geological structure of Yaojialing Zn-Au polymetallic deposit and Tongshan-Qiaomaishan Cu-SW deposit.By using three dimensional Euclidean Distance Field Analysis,Bayesian Probability Analysis and three dimensional Uplift and Depression Morphology Analysis,we completed a three dimensional space analysis on two representative ore deposits and dug out some favorable metallogenic elements.Finally,based on the prediction model of evidential weight method,we finished the exploration of metallogenic prediction and put forward the prospecting target area.The following achievements have been made in this paper:(1)By comparing two kinds of modeling methods,we found that explicit modeling requires the modelers to have a deep understanding of the data in the research area,and it's more suitable for supporting the fine modeling of geological entities in the dataintensive area.The disadvantage of explicit modeling is that the model is susceptible to human factors and has great uncertainty.While implicit modeling,is more suitable for data-sparse areas.Due to its less dependence on modeling skills and more convenience on modeling method,it is less affected by human factors.(2)The results of three-dimensional spatial analysis show that the outputs of Pb,Zn,Au and Cu in the Yaojialing Zn-Au polymetallic deposit are closely related to the Carboniferous and Permian Strata,Cryptoexplosive Breccia,especially Breccia Porphyry and Breccia Marble,and the outputs of gold and copper are also related to Breccia Limestone.In Tongshan-Qiaomaishan Cu-S-W deposit,the output of copper and sulfur minerals is closely related to Carboniferous,Wutong Fm and Diorite,while the output of gold and copper is related to Brecciated Limestone.The analysis results are in rough agreement with the geological ore deposits surveyed by traditional geological ore deposit science.(3)Three-dimensional metallogenic prediction shows that the relevant modeling,data mining and evidence weighting mineralization prediction methods used in this paper have meanings.The quantitative analysis of mineral deposits used in this paper has strengthened the objectivity of geological data use,but its accuracy as a mineralization prediction needs to be improved.The three methods of quantitative analysis of the data used herein are effective,but more can be tried.In this paper,the evidence weight method is used to predict mineralization.Although the prediction results are useful,they can be verified by other methods. |
PDF Full Text Request