气温网格化方法对比分析

Comparison of Gridding Methods on Air Temperature

为获取区域气温最优网格化方案,采用反距离权重法、普通克里格法、径向基函数法和改进谢别德法对黑龙江省976个观测站气温资料进行网格化,并利用NCEP/GFS资料进行验证。结果表明:4种方法的气温误差空间分布存在差异,反距离权重法、普通克里格法的分布比较均匀,而径向基函数法、改进谢别德法存在误差大值点。4种方法均能反映气温误差随时间变化特征,夏季的误差小于冬季,反距离权重法最小,普通克里格法次之,径向基函数法再次之,改进谢别德法最大。研究结果可以为气温网格化方法的选取提供借鉴。

In order to obtain the optimal gridding methods and simulate the spatial distribution of air temperature in Heilongjiang Province,this paper selects inverse distance weighting(IDW),ordinary Kriging(OK),Radial Basis Functions(RBF),Modified Shepard’s Method(MSM)to grid based on data from 976 meteorological stations in Heilongjiang Province under the support of Surfer 8.0 platform,and compares the air temperature absolute error of each gridding method.Results indicate that the spatial distribution of the air temperature absolute error of the four gridding methods is different,and the air temperature absolute error distribution of IDW and OK is more uniform,and the air temperature absolute error distribution of RBF and MSM is not uniform.Those four methods can reflect the characteristics of the time variation of air temperature absolute error in Heilongjiang Province to some extent.The air temperature absolute error of each gridding method in summer is smaller than that in winter,and the air temperature absolute error of IDW has the lowest absolute error,followed by OK’s,RBF’s has slightly lower accuracy,and MSM’s has the worst accuracy.The research results can provide reference for the selection of temperature gridding methods in Heilongjiang Province.