摘要
针对传统DRASTIC模型在参数选择与权重分配中存在主观性及区域适用性局限的问题,以焦作市平原区浅层地下水为对象,提出动态参数优化与主客观组合赋权方法。基于区域水文地质特征,将静态参数含水层介质替换为动态含水层厚度,以量化污染物稀释能力,并将包气带介质重构为水力阻力指标(通过渗透试验与达西定律表征污染物迁移时间),形成DRASTHC优化框架。参数敏感性分析表明,含水层厚度贡献率(18.7%)显著高于原介质类型参数(9.3%)。通过进一步融合层次分析法主观权重与熵权法客观权重,均等整合专家经验与数据离散性特征,有效降低了赋权偏差。基于73个井孔水样的Ca^(2+)和COD浓度数据,采用ROC曲线验证显示,优化模型精度显著提升(COD验证曲线下面积为0.549,Ca^(2+)验证曲线下面积为0.570,较原始模型提升了16.3%)。空间分析表明,焦作市地下水极高脆弱区占27.06%,集中分布于南部工业密集带及西北部包气带渗透性强区。研究通过参数动态化重构与组合赋权机制,解决了传统模型在冲积平原区的适应性缺陷,为地下水污染风险分区防控提供了科学依据。
Addressing the subjectivity and regional applicability limitations of traditional DRASTIC models in parameter selection and weight allocation,taking the shallow groundwater in the plain area of Jiaozuo City as the research object,a dynamic parameter optimization and subjective-objective combination weighting methods were proposed.Based on regional hydrogeological characteristics,the static aquifer medium parameter was replaced by dynamic aquifer thickness to quantify pollutant dilution capacity,and vadose zone medium was recon-structed into a hydraulic resistance index(characterizing pollutant migration time via permeability tests and Darcy′s law),forming the DRASTHC optimization framework.Parameter sensitivity analysis shows that the contribution rate of aquifer thickness(18.7%)signifi-cantly exceeds the original medium type parameter(9.3%).By integrating subjective weights from Analytic Hierarchy Process(AHP)and objective weights from the Entropy Weight Method(EWM),expert experience with data discreteness to reduce weighting bias was bal-anced.ROC curve validation based on Ca^(2+)and COD concentration data from 73 borehole water samples demonstrates a 16.3%accuracy improvement in the optimized model(COD the values of area under curve for COD and Ca^(2+)were 0.549 and 0.570,respectively).Spatial analysis reveals that extremely high vulnerability zones account for 27.06%,concentrated in the southern industrial belt and northwestern areas with high vadose zone permeability.Through dynamic parameter reconstruction and combination weighting,this research overcomes the adaptability defects of the traditional model in alluvial plains,providing a scientific basis for zonal prevention and control of groundwa-ter pollution risks.
作者
王庆超
王丽琳
刘承勇
Wang Qingchao;Wang Liin;Liu Chengyong(Ecological Environment Geo-Service Center of Henan Geological Bureau,Zhengzhou 450000,China;Henan Province Mining Ecological Environment Restoration Engineering Technology Center,Zhengzhou 450000,China;Henan Natural Resources Science and Technology Innovation Center(Collaborative Research on Mining Development and Utilization and Ecological Restoration),Zhengzhou 450000,China;Henan Haohong Foundation Engineering Co.,Ltd.,Zhengzhou 450000,China)
出处
《能源与环保》
2025年第8期6-14,共9页
CHINA ENERGY AND ENVIRONMENTAL PROTECTION
基金
国家自然科学基金项目(42102286)
河南省科技攻关项目(232102320274)
河南省自然资源科研项目(2022-3)。
关键词
地下水脆弱性
脆弱性评价模型
迭置指数法
熵权法
ROC曲线验证
组合赋权
groundwater vulnerability
vulnerability assessment model
overlay and index method
Entropy Weight Method(EWM)
ROC curve validation
combination weighting
