基于贝叶斯最大熵的成矿元素含量空间统计预测--以南天山波孜果尔一带为例  

Spatial Statistical Prediction of Ore-forming Element Contents Based on Bayesian Maximum Entropy:A Case Study of the Boziguer Area in the South Tianshan Mountains

作　　者：高川 樊敏敏 郭小刚 吴国照 连国栋 Gao Chuan;Fan Minmin;Guo Xiaogang;Wu Guozhao;Lian Guodong(Changan University School of Earth Science and Resources,Xi'an 710054,China;Gansu Province Nonferrous Metals Geological Exploration Bureau Baiyin Mineral Exploration Institute,Baiyin 730900,China)

机构地区：[1]长安大学地球科学与资源学院,陕西西安710054 [2]甘肃省有色金属地质勘查局白银矿产勘查院,甘肃白银730900

出　　处：《广东化工》2025年第5期78-81,共4页Guangdong Chemical Industry

基　　金：重点实验室开放基金(YSMRKF202206)。

摘　　要：地球化学元素统计预测是地质勘探的重要基础,传统的地统计学方法克里金法仅对数据进行统计插值,无法融合地质图等各类辅助信息,往往与实际地质体并不耦合。贝叶斯最大熵方法具有较大的改进潜力,但尚未经过实际的验证。本文选取南天山波孜果尔矿区一带,以该区1∶20万化探数据为硬数据,1∶20万地质图等为软数据,验证贝叶斯最大熵方法在地球化学元素空间统计预测的可行性;以克里金插值为对照组,剖析该方法的有效性和先进性。结果表明,该方法能够整合地质信息,元素浓集中心与已有地质体之间有更强的空间相关性。因此,该方法在化探领域的应用是可行的,能更加精细的反应元素含量在空间中的变化情况;但是,融合辅助信息过程,计算量较大,尚需进一步改进和完善。Statistical prediction of geochemical elements forms a crucial foundation for geological exploration.However,traditional geostatistical methods such as Kriging are limited to statistical interpolation of data,unable to integrate diverse auxiliary information such as geological maps,and often lack coupling with actual geological bodies.The Bayesian maximum entropy method offers significant potential for improvement,but its practical application remains unverified.This study focuses on the Boziguoer mining area in the South Tianshan Mountains,utilizing 1∶200,000 geochemical exploration data as hard data and 1:200,000 geological maps as soft data to evaluate the feasibility of the BME method for spatial statistical prediction of geochemical elements.Kriging interpolation is employed as a control group to assess the validity and superiority of the BME method.The results demonstrate that the BME method effectively integrates geological information,revealing stronger spatial correlations between element concentration centers and existing geological bodies.Therefore,this method is feasible for application in geochemical exploration and provides a more refined depiction of spatial variations in elemental content.However,the integration of auxiliary information and the associated computational workload remains challenges,requiring further refinement and optimization.

关 键 词：贝叶斯最大熵法 克里金法 空间预测 化探数据 元素含量 地球化学 

分 类 号：TQ[化学工程]