Analysis And Quantitative Evaluation Of Soil And Water Loss Topography Factors In Small Watershed

Sponsored by joint Chinese-German technical project (2007DFB70200) and Shandong provincial natural science foundation (Y2008E10),a DEM of the Huangshui River watershed was constructed with Topographic data obtaining by digital contour lines on topographic maps. The optimum resolution was determined through analyzing information and data error of different horizontal resolution of the DEM (Digital Elevation Model) image. To evaluate terrain factors extracting from the optimum resolution DEM (OR-DEM), comparison with background real data was carried out. Through the complex analysis of the terrain factors withdrawing from OR-DEM and the land use status map, the law of variation (knowledge) of land use with terrain factors was acquired, and then it was used to supervise the classification of remote sensing images to improve the accuracy of automatic classification of remote sensing image. On this basis, the present situation of soil and water conservation on Huangshui River watershed were analyzed and evaluated through the Soil erosion model of quantitative remote sensing with meteorological, vegetation, and soil data. The main research contents and results are as follows:(1) Based on former research results,the paper synthesized the theories and methodology of the constructional methods of DEM and Digital Terrain Analysis,evaluated and analyzed the method of the DEM Topographic data acquisition, and then introduced the principle and algorithm of extracting terrain factors such as slope, slopes and slope length base on DEM. According to the technical requirement of this study, topographic data was collected through scanning contour map image recognition and DEM was set up in the form of Regular Square Grid (RSG) which was generated by TIN (triangulated irregular networks) interpolation. By analyzing information content and errors that are contained in different horizontal resolution of the DEM image to determine the best one. Evaluation of topography characteristics abstracting from the OR-DEM was examined by comparison with background real data. The results showed that this algorithm was better in Topographic expression, run-time speed and modeling efficiency, and 5m is the OR-DEM in this study. (2) Knowledge about the variation of land use with terrain factors was gained through the complex analysis of the topographic factors extracting from DEM and the present land-use map, and then it was used to land use classification information extraction, showing that it could greatly improve the precision of the land use information automatic extraction, proved by the total classification precision of 87.13 % and Kappa index value of 0.8511. The results showed that the methods which integrated a prior knowledge of GIS data into the classification of the new remote image, could promote the automation of GIS data update, and improve the accuracy of remote sensing automatic classification(3) Base on GIS and RS,this study distilled the rainfall factor,the soil erodability factor,the slope length factor, the slope gradient factor, the cropping management factor and the erosion control management factor from DEM, remote sensing images, the data of soil types and meteorological data,and obtained the soil and water loss situation map in Huangshui River basin in the soil erosion model of quantitative remote sensing. According to the sector standard of the People's Republic of China Standards for classification and gradation of soil erosion (SL190-2007-1), and the requirement of soil and water conservation technical specifications, the strength levels of soil and water loss of Huangshui River basin was achieved. Research results showed that slightly and slight erosion was the mainly type of Huangshui River basin. Serious erosion and above area was only amounted to 12.7% of the total, occurring in sloping fields and abandoned mining land where slope was greater than 15°in hills areas. Regardless of land use, soil conditions, vegetation, the erosion was not obvious in areas where slope is less than 5°. However, when the slope is greater than 15°, serious erosion and above was more likely to occur.