The Spatiotemporal Variation Of Landscape Pattern And Ecological Rehabilitation In Coal Mining Area Based On GIS

With the global environment change research to strengthen and earth observationtechnology development, regional land use change has increasingly become the earth systemscience and global change research new topic. From a regional perspective on the changes ofland use pattern has become the focus of the study. Landscape pattern analysis is the land usepattern change and ecological environment change is an important method to study, contributeto the study of landscape pattern and ecological process in the relationship. As a result of coaldemand continues to grow, underground coal mining area number, original soil, vegetationecological system was severely damaged, the original natural landscape can be reduced tofragments, natural landscape types is reduced, the landscape has undergone tremendouschanges, and appeared in a series of ecological and social problems.The study on the landscape ecology theory and method as the main theoretical basis,integrated geographic information system, remote sensing technology, through thequantitative analysis of each type of landscape changes and qualitative analysis of mininglandscape pattern evolution characteristic and rule to deepen our understanding of thelandscape process understanding, establish the spatial pattern and landscape process mappingbetween prediction of the future trend of landscape pattern. In the landscape pattern on thebasis of qualitative and quantitative analysis, use landscape ecology and ecological protectionand other relevant principle, exploration mining landscape ecological reconstruction strategy,so as to provide scientific basis for exploring the sustainable development of mining area. Theresults of the discussion shows that:In view of the six remote sensing image select the best fusion band, after imagepreprocessing, enhancement and fusion effect analysis, using supervised and unsupervisedclassification method based on remote sensing images for interpretation and informationextraction, obtained in1986-2010six phase mining landscape type maps.The landscape pattern in different scale existence size effect problem, through spatialanalysis module for converting vector data is converted from30m to80m nine different sizesof raster image. Through the landscape index of particle size sensitivity to changes incombined with the actual situation to determine the40m for this study is the most suitablestudy size. Transfer matrix four aspects of the landscape types of space-time evolution of a detailedanalysis by the number of landscape types, structure, rate of change and landscape types.Study period woodland area of greatest change reduced3887.18hm2, cropland increased3301.06hm2, waters increase360.99hm2the industrial land area increased169.96hm2, thesmallest towns land area change, increase55.17hm2. Arable land of the study period, mainlyconversion of forest land, accounting for5.17%of the conversion area; woodland mainly toarable land, accounting for38.81%of the conversion of forest land area; mining area of themain flow of the waters and woodlands, accounting for14.15%and34.19%respectively,while the mining area the main source of towns and woodland, respectively, accounted for11.68%and9.38%of the mine area of origin.For five landscape types in the plaque level, the different characteristics of eachlandscape type changes. At the landscape level, the choice of the better correlation of theseven landscape index analysis of the trends of the overall landscape of the study area. Thestudy period, NP from685to831, ED16.24/ha to13.39/ha, PC from98.92%to298.78%;diversity indices generally showed a decreasing trend after the first increase; landscape as awhole broken0.2, landscape structure is complete, but on the whole, the trend is rising yearby year.1986-1995increased from0.108to0.135, in2010, rose to0.190.By analysis of the type of CA-Markov Model predict landscape change the obviousadvantages of using the model predicted in2005on the basis of landscape pattern landscapesimulation maps in2010. Analog diagram diagram with the actual status of2010compared tothe analysis of simulation and prediction accuracy of81.24%, higher prediction accuracy.Therefore, the landscape pattern in2015is forecast, and analyzed the future trend of themining area landscape.