Southern Plateau Land-use Change And Land Use Space Distribution Pattern Simulation

Land resources are the basis for human survival and development. China is one of the most affected countries of land resources destruction. The southern region of the Loess Plateau is one of the worst regions of land resources destruction in China. With dense population and industrial concentration in this region, the implementation of converting cropland to forest and grassland policy in the 1990s, the western development policy in the early 21st century, and the rapid development of urbanization all caused a more significant impact to land use in the area. Analysis of land use change and its driving mechanism, and simulation of the spatial pattern of land use of the area, can not only reveal the spatial distribution of land resources, but also provide guidance for the rational utilization and effective management of land resources, to provide reference for other similar areas.The research program was mainly supported by the major project (2009JJD770025)called Human and Natural Factors Research in the Process of Land Resources Evolution of the Southern Region of the Loess Plateau, which belongs to the humanities and social science research base of Ministry of Education in China. Based on the data of RS images, DEM data, various thematic data, statistics, and fieldwork data, supported by the technology of RS and GIS, with the software of Erdas Imagine9.2, ArcGIS9.3, SPSS 18.0, File Convert, these processes that remote sensing data were interpreted, data extraction and classification, data grid processing, spatial overlay analysis and computation, data format conversion, regression modeling and testing, etc., were achieved, then further analysis of spatial and temporal evolution of land use of the whole region in large scale and the typical area in the small scale was made, and the driving mechanism to land use change was revealed. On this basis, the spatial patterns of land use were simulated with the relevant software and models. The main conclusions are as follows:(1)It was found that the greatest change in the number of land resources occurred in farmland and construction land in the southern region of the Loess Plateau during 1980 and 2005, and the annual rate of change in construction land was also the largest. The order of the proportion of all the land use types was always that:farmland>grassland>woodland> construction land>water area>unused land. Increase in the proportion of construction land was most, while the other types changed a little. During this period, the main transfer of land use was that the farmland converted into construction land, the farmland converted into grassland, grassland converted into farmland, and woodland converted into grassland.(2)After the analysis of land use change of Qianhe basin between 1980 and 2005 was made, it was found, in time, that the area of all the land types in this basin changed, and the farmland presented decreasing trend all the time, while other types of land use increased or decreased in different years, but construction land was mainly a growing trend. The structure of land use did not changed greatly. In general, farmland and grassland were the larger proportion, followed by woodland, water area and construction land was both smaller proportion. In space, some area of various types of land use could converted into others or transfered from others. Comprehensive analysis of the transfer direction of each period, the main transfer of land use was that the farmland converted into woodland or grassland, and the transfer between woodland and grassland, between farmland and construction land.(3)With the software of SPSS 18.0, the Binary Logistic Regression model between each land use type and its driving force was established of the southern region of Loess Plateau and the typical area to simulate the spatial pattern of land use.The ROC curves was used to test the simulation results, and it was found that the simulation results was very good. The ROC value of all land use types in the southern region of Loess Plateau was that:farmland was 0.870, woodland was 0.850, grassland was 0.779, water area was 0.918, construction land was 0.847,and unused land was 0.984. The ROC value of all land use types in Qianhe river basin was that:farmland was 0.942, woodland was 0.953, grassland was 0.894, water area was 0.973, and construction land was 0.979. Compared to the southern region of the Loess Plateau, the simulation result of Qianhe river basin in the small scale was a higher degree of agreement. But the simulation results in some area was not consistent with the actual situation.(4) The impact of different types of driving factors to land use spatial distribution was different found by the analysis of Binary Logistic regression. The greatest impact factors to the spatial distribution of land use in the southern region of the Loess Plateau were GDP and slope factors, and the most influential factors to the spatial distribution of land use of Qianhe river basin were GDP and major transportation routes.The main innovations of the thesis are as follows:(1)In order to avoid the impact of the spatial autocorrelation of the data in large scale to the regression results, the random sampling method was used to randomly selected 5% number of the sample into the regression analysis.(2) Previous studies focused on the simulation of dynamic change in the number of land use, the study focused on the simulation of the spatial patterns of land use by Binary Logistic Regression analysis, and in the simulation process, considering comprehensively the natural, social and economic driving factors of land use Change, the impact mechanism of land use spatial distribution was analyzed.