中文摘要：

空间尺度是地学研究的基础性问题,河流长度的提取依赖于空间尺度,遥感与地理信息系统技术为河流长度的空间尺度研究提供了强有力的工具。以SRTM DEM为数据源,应用最近邻方法对原始数据重采样得到不同分辨率的DEM数据,利用D8算法对不同尺度数据分别提取了钱塘江水系,并利用分形几何方法研究了河流长度的空间尺度。结果表明,D8算法准确地提取出了不同尺度下的钱塘江水系,且钱塘江及其支流具有明显的分形特征,钱塘江及其支流的提取长度,与以往的文献长度基本吻合,最后对误差来源及不确定性因素进行了分析。

英文摘要：

Spatial scale is the basic problem in geography study,and the extraction of river length depends on the spatial scale.Remote sensing and geographic information system technology provide a powerful tool for studying the spatial scale length of the river.Taking SRTM DEM as data sources,we use the nearest neighbor method to resample the original data to obtain DEM data under different resolutions,adopt D8 algorithm to extract data for Qiantang River system under different scales respectively,and study spatial scales of river length by fractal geometry.The results show that D8 algorithm can accurately extract Qiantang River system under the different scales;the Qiantang River and its branches have obvious fractal features.The extracted lengths of Qiantang River and its branches basically are in conformity with relative literature,and the error sources and uncertain factors is analyzed as well.