Study On Enviromental Change In Wansheng Mining Area Using Multi-source Remote Sensing Data

Environmental problem was the most severe test in today’s world, and the miningpollution made it even worse, had seriously affected the surrounding residents. Afterdecades of mining in the city, Chongqing Wansheng was encountered pollution anddepletion of resources, so the mining environment governance was very urgent.Due to the high speed, short cycle, Remote sensing technology had an obviousadvantage to quickly grasp environment of the status as a making detectiontechnology. Phase especially for multi-source remote sensing data with GIS, whichintegrated space, time, project information, has irreplaceable function in research onenvironment change type, distribution and variation analysis.This paper extracted the changes of mine environment related information in timeand space using two period remote sensing images of different spatial resolutionbased on spectral characteristics. And used these data combined with themathematical statistics method and GIS spatial analysis technology made acomprehensive research on the relationship between mineral development and landcover, environment change.The methods and results are as follows:（1）For inversion the heavy metal content in a mining concentrated area, ASDFieldSpec Pro FR was used to the soil spectrum reflectivity measurement, and sentparts of soil samples to do some chemical test at the same time. The chemical testshowed that the element fluorin（F）arsenic （As）, cadmium （Cd） and zinc （Zn） was inexcess of the standards. After some reasonable preprocessing with spectral data,potential pollution elements’spectral characteristics could be extracted by selectionanalysis, such as the continuum-removal, first derivative.Then correlation analysiswith chemical test results, and the best spectral characteristics for the three elementscould be obtained. The spectral characteristics of As, Cd and Zn wereR²320/R1755,R²260/R²210and R1920/R480, which were highly correlated to thethree actual content with Pearson correlation coefficient of-0.811,-0.703,-0.635. Toget the best inversion model of pollutant, this study used multivariate regression analysis on chemical element content and spectral characteristics reflected value,which had the highest multiple correlation coefficients should be chosen for that. Thestatistics showed the inversion model of F, As, Cd and Zn should be y=8.936x–4.274,y=265.0e^-3.69x, y=17603e^-12.2x, y=305.5e^-0.73x, which multiple correlationcoefficients were R²=0.696（F）,R²=0.744（As）, R²=0.555（Cd）, R²=0.621（Zn）. Put thethree model in corresponding band or layer stack of Aster image, processed byradiation correction、atmospheric correction and geometric correction, that the heavymetal content could be inversed directly.（2）Qualitatively extracted the distribution of vegetation pollution, which wasstressed by heavy metal and dust in environment, based on the theory of red edge"blue shifts" and a maturity dust pollution inversion model.（3）Visual interpretation on land cover, land occupation of mining, geologichazard from the two high-resolution remote sensing images, then the miningdevelopment triggered the changes of land, water resources and geologicalenvironment could be found by comparison analysis.（4）Based on “Pressure-State-Response” model, an ecological securityassessment indicator system had been established in this paper. And endow the stockprice diagram some special meanings, the quantitative relationship abouteco-environmental rangeability and it inside transformation, triggered by miningdevelopment could be directly illustrated. Then the analysis result would provide ascientific advice to mine rational development and environmental governance.