基于GIS的山西永定河流域降水空间插值方法分析

A Research on Shanxi Yongdinghe River Spatial Interpolation Methods Based on GIS

以山西省永定河流域20个国家基本气象站点43年平均降水数据为基础,运用ArcGIS软件地统计学模块进行空间降水插值试验,分别采用了反距离加权法(IDW)、径向基函数法(RBF)和克里格法(KRIG-ING)3种常见的空间内插方法,探讨了山西永定河降水的空间分布。分析发现,3种插值方法在分析区域降水空间分布方面各有优缺点,其均方根预测误差(RMS)的排列顺序为KRIGING<RBF<IDW,KRIGING在该研究区降水的空间插值中效果最好。

The experiment of investigation spatial interpolation, using the geo-statistical model of ArcGIS, is on the basis of annual average precipitation data of the last 43 years at 20 country basic stations in along reaches of Yongdinghe river Shanxi. Inverse distance weight tension （IDW）, radial basis functions （RBF） and Kriging （KRIGING） methods were applied. The result showed that the three methods could reflect the spatial distribution of precipitation of Yongdinghe and they have their pluses and minuses respectively. In this study, three methods for spatial interpolation of precipitation data were compared, the order of RMS is KRIGING 〈 RBF 〈 IDW. The method of Kriging is the most effective one in the whole analysis.