基于DEM的广东省平均气温空间插值Ⅱ——研究结果

Interpolation of average temperature based on the DEM in Guangdong provinceⅡ——the results

以广东省为研究区域,利用广东省及周边的60个气象站点多年月平均气温数据,根据气温垂直递减规律引入数字高程模型对气温的空间插值结果进行修正。结果表明,基于数字高程模型的插值修正方法可以提高气温空间插值的精度,其中Spline方法的平均绝对误差由0.49降到了0.32,均方根误差由0.57降到了0.36;反距离权重法法的平均绝对误差由0.47降到了0.31,均方根误差由0.56降到了0.35;OK方法的平均绝对误差由0.48降到了0.33,均方根误差由0.58降到了0.38。在6种插值方法中,考虑了数字高程模型影响的反距离权重法误差最小,用于广东省气温的空间插值效果最好。

Monthly mean temperature in Guangdong province was interpolated using the data from 60 meteorological stations in Guangdong province and surrounding area. The interpolation were revised using digital elevation model （DEM） based on the vertical temperature lapse. Results show that the interpolation accuracy was improved by introducing DEM into interpolation. Mean absolute error （MAN） and Root-mean-square error （RMSE） of Spline method decreased from 0.49 to 0.32 and from 0.57 to 0.36, respectively. Those of IDW method decreased from 0.47 to 0.31 and from 0.56 to 0.35 respectively. Those of OK method decreased from 0.48 to 0.33 and from 0.58 to 0.38 respectively. The IDW method revised by DEM is of the highest accuracy for monthly mean temperature in Guangdong province.