3种空间插值方法在道路尘中Pb监测数据统计处理中的应用比较  

Comparison of Three Spatial Interpolation Methods for Statistical Processing of Lead Monitoring Data in Road Dust

作　　者：田雷[1] 董德明[1] 魏强[1] 房春生[1] 

机构地区：[1]吉林大学环境与资源学院,长春130012

出　　处：《吉林大学学报（理学版）》2011年第5期964-968,共5页Journal of Jilin University:Science Edition

基　　金：吉林省科技发展计划项目(批准号:20040406)

摘　　要：以福建省龙岩市道路扬尘监测分析中的Pb元素质量浓度为例,选取地统计学中具有代表性的反距离权重法、径向基函数法、普通克里格法3种空间插值方法,进行污染物浓度的空间插值,通过对插值结果进行交叉验证和评价,对这3种插值方法进行对比研究.道路扬尘样本共56个,其中44个样本用于空间插值,其余12个样本用于结果验证.结果表明,普通克里格法的平均预测误差为-0.129 6,均方根预测误差为24.19,与其他两种方法相比,在检验标准上有明显优势,故普通克里格法对于描述污染物质量浓度的空间分布趋势效果最佳,而其他两种方法对Pb元素的空间分布插值精度较差.Taking the concentrations of lead in the monitoring and analysis data of the road dust in Longyan city as an example,we made spatial interpolation study for pollutant concentration using three types of representative interpolation methods in geostatistics,inverse distance weighted（IDW）,radial basis function（RBF） and ordinary Kriging（OK）.By validating and appraising the result,we compared the three interpolation methods.Among the total 56 samples,44 road dust samples were used for interpolation,and the rest 12 samples were used for validation.Compared with those of other interpolation methods,the ME of OK is-0.129 6,the RMSE of OK is 24.19.OK is apparently better than IDW and RBF for describing the error parameters.The results of comparison indicate that OK shows the best result in characterizing spatial distribution trend of pollutant concentration;IDW and RBF interpolate methods show less accuracy for the spatial distribution of lead elements.

关 键 词：空间插值 环境监测数据 统计处理 

分 类 号：X820.3[环境科学与工程—环境工程]