PROCEEDINGS ARTICLE | October 29, 2005
Proc. SPIE. 5983, Remote Sensing for Environmental Monitoring, GIS Applications, and Geology V
KEYWORDS: Statistical analysis, Agriculture, Soil science, Remote sensing, Error analysis, Vegetation, Humidity, Analytical research, Climatology, Statistical modeling
Land degradation processes, which imply a reduction of the potential productivity of the land (e.g., soil degradation and accelerated erosion, reduction of the quantity and diversity of natural vegetation), result from a long history of human pressure upon land resources as well as from interactions between varying climatic characteristics and ecologically unbalanced human intervention. The north-west region outside of Beijing, is one of the most important regions where many departments invest most and pay most attention. The land degradation and other environmental problems in this region affect not only Beijing but also the surrounding area. This paper analyzed characteristics of land degradation actuality situation in the NW region of Beijing, based on TM (ETM) in 2002. The wind-eroded land was mainly distributed in north of Yin Shan Mountain. Due to degradation of grassland, the sandy land increased from 1991-2002, mostly distributed in the monitoring zone of Hunshandake sandy land. The water-eroded land was mainly distributed in monitoring zone of the south of Yin Shan Mountain and south of monitoring zone of Horqin sandy land. The salination-land was mainly distributed in lake surrounded area and the drainage basin of Sanggan River. And To better understand the drive forces of land degradation processes in study area, a multivariate spatial model associated with land degradation is found by the explanatory variables of Logistic multivariate regression model(LMR). The explanatory variables include wind speed, soil humidity, soil organic matter, NDVI, average precipitation, soil slope, et al. The value of the parameter estimated by model with their corresponding standard error, chi-square statistics, and significance probability are analyzed to find the driver of land degradation in studied area. And the high or low probability of land degradation is predicted. Finally, suggestions to the eco-environment construction of the studied region have been put forward.
