DEM地形综合方法对水文结构信息提取精度的影响

Influences of DEM Generalization Methods on Extracting Accuracy of Hydrological Structure Information

以巢湖流域1∶5万DEM为基础数据,采用3种指标(应用河网套合差、子流域套合差、沟壑密度)对5种DEM地形综合方法(稀疏采样法、样条函数插值法、三次卷积插值法、二次曲面拟合法、结构化综合法)所构建的粗分辨率DEM所提取的水文结构信息的精度进行了分析评价。结果表明,5种地形综合方法构建的粗分辨率DEM提取的水文结构信息精度由高到低的顺序为:二次曲面拟合法>结构化综合法>三次卷积插值法>样条函数插值法>稀疏采样法,但其精度总体上差异不大。

1 ： 50 000 DEM of Chaohu Basin was selected as base data for this study. The extracting accuracy of hydrologic characteristics was evaluated individually for five DEM generalization methods（nearest neigh bor, bilinear interpolation, cubic convolution, structured generalization, quadric surface fitting） based on three parameters including matching error of river networks, matching error of sub-watershed and gully den- sity. The results show that the extracting accuracy of hydrological structure information based on the low resolution DEM followed a descending order of the five terrain generalization methods as： quadric surface fit- ting structured generalization cubic convolution bilinear interpolation ~ nearest neighbor, with very subtle accuracy difference.