内蒙古新巴尔虎右旗多金属矿区扬尘风积物遥感监测方法  

Remote sensing monitoring method for dust and wind accumulation in multi-metal mining area of Xin Barag Right Banner,Inner Mongolia

作　　者：宋国策 张志 SONG Guoce;ZHANG Zhi(College of Geophysics and Spatial Information, China University of Geosciences(Wuhan), Wuhan 430074, China)

机构地区：[1]中国地质大学(武汉)地球物理与空间信息学院,武汉430074

出　　处：《国土资源遥感》2020年第2期46-53,共8页Remote Sensing for Land & Resources

基　　金：中国地质调查项目“全国矿山开发状况遥感地质调查与监测”(编号:DD20190511)资助。

摘　　要：内蒙古新巴尔虎右旗甲乌拉—查干铅锌银矿区地处呼伦贝尔草原腹部,当地半干旱的气候使得该矿区尾矿库、固体废弃物堆、矿石堆等易于产生扬尘,对周边草场产生污染。采用传统的化学采样和光谱分析调查矿区扬尘污染精度虽高,但费时耗力,而利用时序遥感方法监测矿区扬尘污染则比较便捷。本研究利用2018年GF-1卫星数据完成了矿区开发占地信息提取,在分析研究区的风场和最佳扬尘观测月份基础上,使用2000年、2005年、2010年、2015年和2018年获取的5期Landsat卫星数据,采取“扬尘风积物-植被-水体和阴影”端元分解模型和人工干预的半自动剔除道路干扰方法,对矿区扬尘风积物进行遥感监测;相较于归一化差值植被指数(normalized difference vegetation index,NDVI)分析法,该方法在考虑植被光谱信息的同时也兼顾扬尘的光谱信息,使监测结果更为客观。比较5期遥感图像扬尘风积物提取结果发现,截止到2018年,矿区1 km缓冲区内扬尘风积物污染面积扩大到190.57 hm^2,其中2000—2010年间和2010—2018年间的年平均增长面积分别为14.74 hm^2和0.64 hm^2。监测结果表明矿区采取的防治和治理措施可以明显改善扬尘风积物污染问题,但是随着矿区进一步开发,生态修复治理也应及时跟进。The Awula-Chagan lead-zinc-silver mine in Xin Barag Right Banner of Inner Mongolia is located in the abdomen of Hulunbuir grassland.Its semi-arid climate makes tailings ponds,solid waste piles and ore piles easily generate dust,polluting surrounding grassland.The traditional chemical sampling and spectral analysis investigate the high precision of the mining area but they are time-consuming and labor-intensive.It is convenient to use the time-series remote sensing method to monitor the dust pollution in the mining area.In this paper,GF-1 satellite data in 2018 were used to extract the information of the mining area in the study area.Based on an analysis of the wind field and the best observation month in the study area,the authors used the five-phase Landsat satellite data to adopt the end-element decomposition model of“dust accumulation-vegetation-water and shadow”,and used the method of semi-automatic elimination of road interference by manual intervention to remove the effects of roads.Compared with NDVI index analysis method,the proposed method considers the vegetation spectral information and takes into account the spectral information of the dust,thus making the monitoring effect more objective.A comparative study of 5 remote sensing image aeolian dust extractions found that,as of 2018,the mining area 1 km buffer aeolian dust contamination area expanded to 190.57 hm^2,of which annual average growth area in 2000—2010 and 2010—2018 were 14.72 hm^2 and 0.64 hm^2,respectively.The monitoring results show that the prevention and control measures adopted in the mining area can significantly improve the pollution of dust and wind accumulations;nevertheless,with the further development of the mining area,ecological restoration and management should also be conducted in time.

关 键 词：矿区 扬尘风积物污染 线性光谱混合模型 NDVI 时序监测 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]