基于DS-InSAR的尾矿库变形监测与分析——以巴西Brumadinho矿为例  

Deformation Monitoring and Analysis of Tailings Dam Based on DS-InSAR:A Case Study of Brumadinho Mine in Brazil

作　　者：吴昊 范洪冬[1,2] 郑春柳 刘久利 WU Hao;FAN Hongdong;ZHENG Chunliu;LIU Jiuli(School of Environmental and Spatial Informatics,China University of Mining and Technology,Xuzhou 221116,China;Key Laboratory of Land Environment and Disaster Monitoring,Ministry of Natural Resources,China University of Mining and Technology,Xuzhou 221116,China;Beijing Urban Construction Exploration&Surveying Design Research Institute Co.,Ltd.,Beijing 100101,China;Institute of Remote Sensing Satellite,China Academy of Space Technology,Beijing 100094,China)

机构地区：[1]中国矿业大学环境与测绘学院,江苏徐州221116 [2]自然资源部国土环境与灾害监测重点实验室,江苏徐州221116 [3]北京城建勘测设计研究院有限责任公司,北京100101 [4]中国空间技术研究院遥感卫星总体部,北京100094

出　　处：《金属矿山》2023年第3期169-176,共8页Metal Mine

基　　金：“十三五”国家重点研发计划政府间重点专项(编号:2017YFE0107100);国家自然科学基金项目(编号:51774270,41604005);江苏省自然科学基金项目(编号:BK20190645)。

摘　　要：尾矿库是一种人造高重力势能的危险源,一旦发生溃坝将会严重污染环境,并威胁下游居民的生命财产安全。然而,现有的尾矿库形变监测方法存在费时、费力、成本高等问题,为此,提出了一种基于分布式目标干涉雷达测量(DS-InSAR)的尾矿库形变信息提取方法,有助于实现尾矿库的高效率、高精度形变监测。该方法利用基于参数假设检验的FaSHPS(Fast SHP Selection)方法选取同质像元,通过相干矩阵的特征值分解提取主导散射体相位以实现相位优化,进而联合永久散射体(PS)获取尾矿库的时序变形信息。以23幅Sentinel-1B影像为数据源,利用该方法获取了巴西Brumadinho尾矿库时间序列地表形变信息。结果表明:该方法获取的尾矿库监测点密度是PS-InSAR的2.2倍,两者获取的形变速率具有较强的相关性,并且,尾矿库的形变都呈现出由坝顶中部向外扩散的趋势。将降雨量与形变量进行关联分析表明,降雨对尾矿库的变形影响显著,总体上表现为旱季形变量较少、雨季则相反的态势。Tailings pond is a dangerous source of artificial high gravity potential energy.Once the dam breaks,it will seriously pollute the environment and threaten the life and property safety of downstream people.However,the existing deformation monitoring methods of tailings pond have the problems of time-consuming,laborious and high cost.Therefore,this paper propose a deformation information extraction method of tailings pond based on distributed target interferometric radar measurement(DSInSAR),which is helpful to realize high-efficiency and high-precision deformation monitoring of tailings pond.This method uses the FaSHPS(Fast SHP Selection)method based on the parameter hypothesis test to select homogeneous pixels,extracts the dominant scatterer phase through the eigenvalue decomposition of the coherence matrix to achieve phase optimization,and then obtains the time-series deformation information of the tailings pond in combination with the permanent scatterer(PS).Taking 23 Sentinel-1B images as data sources,this method is used to obtain the time-series surface deformation information of Brumadinho tailings pond in Brazil.The results show that the density of monitoring points of tailings pond obtained by this method is 2.2 times that of PS-InSAR,and the deformation rate obtained by the two methods has a strong correlation,and the deformation of tailings pond shows a trend of outward diffusion from the middle of dam crest.The correlation analysis between rainfall and deformation shows that rainfall has a significant impact on the deformation of tailings pond.Generally,there are few deformation in dry season,but the opposite in rainy season.

关 键 词：尾矿库 分布式目标 INSAR 变形监测 

分 类 号：TD926.4[矿业工程—选矿] TP722.6[自动化与计算机技术—检测技术与自动化装置] P237[自动化与计算机技术—控制科学与工程]