[关键词]
[摘要]
以常用HydroSHEDS void-filled DEM、SRTM、ASTER GDEM等五种DEM数据为研究对象,针对不同流域下垫面地貌形态,采用不同的地貌特征分析方法以及将流域真实水系和模拟河网对比分析等方法,研究了多源DEM数据的质量问题。研究发现ASTER GDEM数据主要问题是噪声大,HydroSHEDS void-filled DEM和SRTM数据的主要问题是存在模糊,噪声、模糊均会降低数据的精度和提取的模拟河网的精度,但从整体上来看,三种DEM数据提取的模拟河网的质量相当。
[Key word]
[Abstract]
This paper selects five different sources of DEM data—— HydroSHEDS void-filled DEM, SRTM and ASTER GDEM and so on to analyze the quality of different DEM data under different geomorphologic shapes by choosing different geomorphological characteristics analysis method and comparing the simulated drainage networks with the real drainage networks. The result shows that the main problem with ASTER GDEM DEM data is data noise and the other two DEM data both have some fuzzy data. These data quality issues will all effect the accuracy of the data and the river network extracted from them. But in general the accuracy of these digital drainage networks extracted from the three DEM data all have high precision.
[中图分类号]
[基金项目]
水利部公益性行业科研专项项目(201401034);国家自然科学基金重点项目(41030636;51420105014)