卫星遥感技术在矿山历史遗留固废排查中的应用——以广东清远和江西崇义为例  

Application of satellite remote sensing technology in the investigation of historical solid waste in mining areas——A case study of Qingyuan in Guangdong Province and Chongyi in Jiangxi Province

作　　者：区杰泳 华菁 郑晓雄 童立志 孙中平 雷蕾 陈岩贽 王炜 OU Jieyong;HUA Jing;ZHENG Xiaoxiong;TONG Lizhi;SUN Zhongping;LEI Lei;CHEN Yanzhi;WANG Wei(South China Institute of Environmental Sciences,MEE,Guangzhou 510655,China;Chongyi Technical Service Station of Ganzhou Ecological Environment Technology Service Center,Ganzhou 341300,China;Satellite Application Center for Ecology and Environment,Beijing 100094,China)

机构地区：[1]生态环境部华南环境科学研究所,广州510530 [2]赣州市生态环境技术服务中心崇义技术服务站,赣州341300 [3]生态环境部卫星环境应用中心,北京100094

出　　处：《环境保护科学》2023年第5期10-17,共8页Environmental Protection Science

基　　金：国家重点研发计划项目(2021YFB3901105)。

摘　　要：矿山历史遗留固废精准排查是矿山污染治理的必要条件。卫星遥感技术广泛应用于固体废物排查在内的环境监测领域,但存在识别对象差异性和遥感数据质量影响排查全面性和准确性的问题。该研究基于矿山历史遗留固废特征构建遥感解译数据库,对广东清远和江西崇义2个研究区域开展单时相影像遥感解译、多时相影像遥感解译和人工解译修正,通过计算漏判率、总体精度,评价卫星遥感解译在矿山历史遗留固废排查工作中的应用效果,并对非目标图斑进行分类,优化了矿山历史遗留固废遥感解译精度评价参数。结果显示:单时相影像遥感结果漏判率较大,未能全面排查到研究区域内的历史遗留固废;经过多时相影像进行遥感解译人工解译修正的解译结果漏判率为0%,近似精度>95%;近似精度可作为后续矿山历史遗留固废现场核查的依据。Accurate identification of historical solid waste in mining areas is essential for the pollution management.Satellite remote sensing technology is widely applied in environmental monitoring,including solid waste identification.However,it faces challenges due to the target recognition difference and the data quality.This study constructs a remote sensing interpretation database based on the characteristics of the historical mining solid waste.In Qingyuan and Chongyi,single and multi-temporal image interpretations,followed by manual corrections,were conducted.By calculating the false negative rate and overall accuracy,the application effect of satellite remote sensing interpretation in the investigation of historical solid waste from mines was evaluated,and non-target patches were classified to optimize the evaluation parameters of remote sensing interpretation accuracy for historical solid waste from mines.The results showed that single-temporal image interpretations had significant omission rates and failed to comprehensively identify historical solid waste.In contrast,multi-temporal image interpretations followed by manual correction resulted in a 0%omission rate and an approximate accuracy exceeding 95%.This accuracy can serve as a basis for subsequent on-site verification of historical mining solid waste.

关 键 词：矿山污染 矿山固体废物 遥感解译 监测应用 

分 类 号：X506[环境科学与工程—环境工程]