降水空间插值方法应用研究——以黄河源区为例

使用反距离加权法、张力样条函数法和普通克里格法对黄河源及其周围地区65个气象站点1990～2001年12年间月平均降水数据进行空间插值,并对模拟结果进行了交叉检验和站点检验.3个检验标准的检验结果表明:3种插值方法的结果都基本能反映黄河源区降水的大体空间分布规律,而普通克里格法略优于其他两种方法.研究结果表明:样本本身的空间分布在很大程度上影响插值精度,台站的密度对插值效果的影响也非常显著.给出了黄河源区年平均降水空间分布图.

Application of the spatial interpolation of rainfall——A case study of the headstream region of the Yellow River

Inverse distance weight, spline and ordinary Kriging method are respectively employed to interpolate the monthly averaged precipitation of 65 meteorologic stations in and around the headstream region of the Yellow River from 1990 to 2001. Then the results of the three interpolators are assessed under both cross validation and station validation and the three validating criterions indicate that all the three interpolators can generally reflect the spatial distribution of the precipitation of the headstream region of the Yellow River. And ordinary Kriging performs comparatively better than the other two methods. The results also imply that the spatial distribution of the samples can dramatically influence the accuracy of the interpolation, and the density of the stations can also sharply impact the result of interpolation. We present the spatial distribution of the annual average precipitation in the headstream region of the Yellow River.