基于DEM的海南岛平均气温空间插值研究

选取海南岛18个气象站点1979～2008年30 a的逐月气温资料，结合数字高程数据（DEM），采用回归反距离权重插值法、回归克里格插值法、回归样条插值法和回归趋势面插值法等多种混合插值方法对研究区平均气温的空间分布进行了插值研究，通过5个检验站点的插值误差来分析4种不同混合插值方法，并对平均气温空间分布栅格图进行了比较分析。结果表明：回归反距离权重插值法误差最小，效果最好。

Spatial Interpolation of the Average Air Temperature in Hainan Island Based on DEM

The mean monthly air temperature of 18 meteorological stations from 1979 to 2008 was analyzed. A variety of interpolation methods were used to study the mean air temperature spatial distribution based on DEM. Different interpolation methods such as the combination of stepwise regression and inverse distance weighing interpolation method, the combination of stepwise regression and ordinary kriging interpolation method, the combination of stepwise regression and spline interpolation method, the combination of stepwise regression and trend interpolation method were analyzed by the prediction accuracy verified by 5 validated stations, and gird maps of mean air temperature spatial distribution were carried out a comparative analysis. The result demonstrated that the combination of stepwise regression and inverse distance weighting interpolation method had the highest prediction accuracy, and was better than others.