应用Kriging算法实现气象资料空间内插

使用Kriging算法对海平面气压和气温等气象观测资料进行空间内插，实现数据网格化。通过GRADS绘制等值线图与人工分析的天气图以及Surfer软件的Kriging方案的结果进行对比，并对比分析多个统计量，验证自制软件结果的好坏。结果表明：对于气象物理量场，变异函数应采用球面模型；对量级较大的海平面气压值应进行缩小化处理；对于常规观测资料，球面模型半径应取10个经纬距以上。

Application of Kriging Algorithm to Spatial Interpolation of Weather Data

The Kriging algorithm was used to realize spatial interpolation of weather data. Comparisons were done for the results of the Kriging algorithm,the Surfer software,and the manually analyzed,as well as for the analysis results of several statistics. The comparisons show that the spherical mode should be used in the variation function for weather data. The values of sea-level air pressure should be compressed to reduce errors in the Kriging algorithm. The radius of the spherical model should be ten longitudes/latitudes or more for the conventional data.