Based on the support vector machine for LUCC research in Binchuan of Yunnan Province

摘要

Land use and cover change is the focus of environmental change research. We used the support vector machine classification method to extract the years 1995 Landsat TM, 2000 and 2005 LandsatETM+, and 2013 LandsatOLI four remote sensing data's LUCC types and evaluate the accuracy of extraction. Finally, use of land use transfer matrix system quantitatively described, simultaneous analysis of the area of LUCC spatial and temporal dynamic characteristics and the factors of driving force, in order to protect the valuable forest resources and continuing effective use land resources of Binchuan to provide a scientific basis for decision making. The result indicate that: the SVM classification method overall accuracy was 89.23, with a kappa coefficient greater than 0.7. From 1995 to 2013, rural and mining residential land generally increased, except 2000, and reaching the maximum in 2013, which is almost double than in 1995. This permits Binchuan, which for nearly 20 years, has always been committed to the city and rural development. The most obvious conversion land use types were unused land and cultivated land, with cultivated land showing a clear decreasing trend, and majority of conversions were to rural and mining residential land, and unused land mostly converted to grassland and rural and mining residential land. For woodland, it experienced an initial increase, then decrease, and then finally increased procedure. However the increase of woodland area is not large (it remained in stable condition), and this proved Binchuan has an emphasis on the protection of forest resources. The water area from 1995 to 2000 years showed a substantial reduction, but in the subsequent 10 years has rebounded. However water resources is still relatively scarce, suggesting the Government to strengthen the construction of water conservancy facilities and soil and water conservation and related work. The LUCC drivers of Binchuan are complex, but the human factor is the main driving fact- r, rapid population growth, and high-speed economic development are the fundamental factors that led to the massive building occupants, so there is a large number of rural and mining residential land.