Research On Spatio-temporal Variation Law Of Vegetation Cover For Surface Subsidence In Mining Area

With the increasing exploitation of coal resources, ecological destruction andenvironmental pollution of the mining area become more and more serious，which posesa gerat threat to the sustainable development of mining area economy. Accordingly, itbecome an important research that solving the problem how to accurately grasp thespatial variation regularity of the surface subsidence for fully mechanized caving panelin different geological and mining conditions, grasp the surface cover spatial variationregularity in subsidence areas and the relationships between them, which will providetheory basis and technical support for the local and the similar mining areas in coalmining, mining damage control and improving ecological environment of mining areas.Seven production mines of LuAn group in Changzhi mining area were selected asthe research object. By applying Landsat-7ETM+, SPOT2/4, ERS-1/2satellite data,GTOPO30DEM, SRTM DEM,1:250000digital map, AutoCAD mining project planes,as well as the data of the surface movement observation stations, preprocess the remotesensing data and integrate multi-source data. This will be tested firstly by combinedapplying mining subsidence data, remote sensing features reflection and phase changeinformation and so on. The spatial variation monitoring for the surface cover index ofmining subsidence areas will be quantitative researched.1. According to the analysis for the different geological and mining conditions, theresearch region is divided into three kinds, namely thick surface soil thick bedrock(Tunliu, Changcun), thick surface soil thin bedrock (Sima), thin surface soil thickbedrock (Wangzhuang, Wuyang, Zhangcun). Respectively Wuyang, Sima, Tunliu minebe took for examples, and the analysis of measured data of the surface movement anddeformation under the condition of fully mechanized caving panel of3kinds of miningsubsidence models, the basic change laws of the different mining mode of groundsubsidence land destruction are obtained.2. Selected9typical panels within the time and space range of remote sensing data,the field data is proceeded with the multi-temporal remote sensing data on the surfacecover of mining subsidence area. Firstly, after the geology and mining condition oftypical panels were analyzed, the mining subsidence area and non-mining subsidence area were determined. According to the measured strata movement predictingparameters, mining subsidence prediction of the mining subsidence area weredone.According to7temporal satellite remote sensing images data based onatmospheric correction and band calculation of SPOT2/4from2004to2007, the surfacecover NDVI sampling average were collected and the NDVI change curves of9typicalpanels that pre and post subsidence were obtained by applying Fourier fit.3. The multitemporal NDVI of the mining subsidence area and non-miningsubsidence area was extracted in the front and back of mining subsidence areas mined.The temporal and spatial variation law of the surface cover effected by miningsubsidence was acquired. Research shows that vegetation index change has displayedmore time-space correlation with the typical faces mining subsidence.(1)The change of the surface NDVI has spatial correlation with mining area in theexperiment area. The maximum NDVI mean value of the forest canopy has the trend ofsteady growth year after year, which is average annual net increase of6.2%.This changefits the forest biomass accumulation law of semi-arid areas. The maximum of NDVImean value of the mining subsidence areas above the typical panels has increasing trendbefore mining, but the NDVI begin to decrease after mining. The nine typical panelshave the average annual decrease of7.2%, and the maximum annual decrease of37.3%.(2)The change of the NDVI has time correlation with mining area in theexperiment area. Vegetation index in mining areas is relatively stable before mining, butthe change of the surface vegetation index has certain lag after mining and shows adecreasing trend with time until steady, subsequently transfers to rising period. Itcoincides with the surface destruction stages of mining subsidence in the initial, active,decline and stable period.(3)Because of the different mining geological condition, the NDVI of differentmining areas also have obvious change. For the mining areas belong to thick surfacesoil thick bedrock, annual average decline rates of the NDVI peaks are12.2%,13.3%,19.2%respectively, with an average change of-14.9%. For the mining areasbelong to thin surface soil thick bedrock, average annual rate of change is-10.1%.These mining areas have relatively light damage. This is consistent with the differentgeological and mining conditions and the damage degrees. (4) The temporal and spatial variation on the ANDVI integration quantity of thetypical panels in the subsidence areas is studied. The results of experiment show that thesum of the integrated quantity ANDVI of9typical panels decreases15.1%per year inpast4years. It shows that there is causality between the mining subsidence and thecover growth state and the biomass accumulation weakened in coal mine subsidenceareas. For the mining areas belong to the thick surface soil thick bedrock, the intergralof the surface cover ANDVI average annual reduces12.7%. But for the mining areasbelong to thin surface soil thick bedrock, average annual reduces10.9%. Comparativelyspeaking, the surface cover damaged degrees of the type belong to the thick surface soilthick bedrock is more serious than the type belong to thin surface soil thick bedrock.