均值变点分析法在最佳集水面积阈值确定中的应用  

Determination of Accumulation Area Based on the Method of Applying Mean of Change-Point Analysis

作　　者：常直杨[1] 王建[1,2] 白世彪[1,2] 张志刚[1] 

机构地区：[1]南京师范大学地理科学学院,江苏南京210023 [2]南京师范大学虚拟地理环境教育部重点实验室,江苏南京210023

出　　处：《南京师大学报（自然科学版）》2014年第1期147-150,共4页Journal of Nanjing Normal University(Natural Science Edition)

基　　金：国家自然科学基金(40871010)

摘　　要：利用DEM数据提取河流地貌参数在地貌学和水文学研究中具有重要意义,其中集水面积阈值的准确确定是关键环节,然而,目前对集水面积阈值的确定存在随意性和主观性.为了探讨如何确定提取河网的最佳集水面积阈值,本文以青藏高原东缘杂谷脑河流域为例,基于SRTM-DEM数据,在利用ArcGIS水文分析模块计算不同集水面积阈值条件下河网密度参数的基础上,通过均值变点分析法获取河网密度值变化的拐点,并计算最佳集水面积阈值.研究表明:随着集水面积阈值的改变,河网密度逐渐降低,利用均值变点分析法所确定的最佳集水面积阈值为8.1 km2,该结果与1∶50万水系图相比,主干上吻合,支流上更具真实性.该方法可以较好地确定最佳集水面积阈值,可为其它地区水系提取提供参考.To gain the threshold of catchment area is the critical step of extracting drainage network from DEM. But this index determination were still subjective and random with no theoretical support in most studies. Taking Zagunao Drainage Basin as an example,we calculated the drainage densities of different accumulation areas with SRTM-DEM data by using the hydrology model in Arcgis followed by gaining the optimum accumulation area with the method of mean of change-point. The result indicates that the drainage density correlated with accumulation area negatively and the 8. 1 km2 optimum accumulation shows more authenticity compared with the geology map of 1：500 000. It means that this kind of way can be applied to determine the accumulation area.

关 键 词：杂谷脑河流域 河网密度 集水面积阈值 均值变点分析法 

分 类 号：P542[天文地球—构造地质学]