RS与GIS支持下的土壤侵蚀自动判别模型——以武威市凉州区为例

An automatic discriminating model of soil erosion using Remote Sensing and GIS methods——a case study in Liangzhou district in Wuwei city

在分析土壤侵蚀成因和一般规律的基础上,选择土壤类型、植被覆盖度、地形、土地利用等最主要的土壤侵蚀影响因子,记录它们的匹配关系,建立知识库,并在ERDAS Imagine环境下建立基于栅格数据的土壤侵蚀判别模型.以甘肃省武威市为例,利用TM遥感影像、DEM及其它专题数据提取土壤侵蚀主导因子信息,利用模型对土壤侵蚀的类型和强度进行计算,并以2000年土壤侵蚀数据为参照进行精度检验,结果表明位置精度为75.46%以上,各类侵蚀强度的面积误差不超过16.08%.模型操作方便、运算快速,判别结果能够总体反映土壤侵蚀的空间分布.

Based on the ananysis on the causes and the general laws of soil erosion, soil type, vegetation, topography and land use were selected as the four key factors affecting soil erosion, the relationship between the indicators and the Classification and gradation of soil erosion was analyzed, the matching record was recorded in the knowledge base, and the soil erosion automatic discriminating model was established based on raster data. Taking Liangzhou district in Wuwei city of Gansu Province as an example, the fac- tors were extracted from TM image, DEM and soil map, Classification and intensity of soil erosion in the study area was calculated by the automatic Discriminating model. It is concluded that the model is scientific and efficient, which result contrasting with the investigation results of 2000 showed that the accuracy of Spatial distribution is more than 75% and the area error does not exceed 16.08%. The model is easy to operate and computing fast, the discriminating results can objectively reflected the distribution of soil erosion overall.