长沙市大气中 PM2．5浓度分布的空间插值方法比较

摘要

采用普通克里金（ Kriging）、反距离加权（ IDW）、趋势面、协同克里金（ Co-kriging）插值和基于反距离加权的克里金插值方法对长沙市某时刻的PM2．5浓度空间插值比较，结果表明：反距离加权和趋势面插值的结果与实际情况相差较远；普通克里金和协同克里金插值虽能反映出PM2．5浓度分布一定的空间结构，但精度有待提高，且运行速度较慢；前4种方法效果不好的主要原因是由于长沙市PM2．5浓度监测站点稀少。而基于反距离加权的克里金插值方法结合了2种模型的优点，其结果较为准确地分析长沙市PM2．5浓度分布特征，误差远小于其他方法，且具有平滑性好、运行速度相对较快等优点。%The spatial interpolation methods of ordinary Kriging, inverse distance weighted, trend surface, Co-kriging and Kriging interpolation based on inverse distance to a power method were utilized for the comparison study on spatial interpolation of PM2.5 concentration in Changsha City.It was preliminary concluded that the results of inverse distance weighted and trend surface interpolation were far different from the actual.Ordinary Kriging and Co-kriging interpolation could reflect the spatial distribution of PM2.5 concentration, but the accuracy still needed to be improved and the running speed was slow.The main reason why the former 4 methods was ineffective was the lack of the air quality monitoring stations in Changsha City.Kriging interpolation method, based on inverse distance weighted and combined with the advantages of the two models, presented relatively accurate distribution characteristics of PM2.5 concentrations, and had much less error than that of other methods. This method also had the features of good smoothness and fast running speed.