水环境模拟中变异函数的稳健性分析  

Robust Analysis of Spatial Variogram in Water Environment Simulation

作　　者：牟向玉[1] 张征[1] 杨帆[1] 赵玉婷[1] 

机构地区：[1]北京林业大学环境科学与工程学院,北京100083

出　　处：《环境科学与管理》2008年第5期24-28,共5页Environmental Science and Management

基　　金：教育部科学技术研究重点项目(03028);北京林业大学振兴计划人才培养专项课题(200202013)

摘　　要：文章将稳健统计学(RS)引入地质统计学(GS),对廊坊市地下水现场采样数据中Cl-运用影响系数法和估计邻域法识别并处理特异值,取得了较好的稳健效果。在此基础上,用Cressie-Hawkins法、中位调节法两种稳健统计学方法系统的对进行特异值处理前后的数据作了处理和分析。结果表明,变量Cl-离子经ICM法处理特异值后再进行稳健计算方法中以MA法有较大的优势,而且获得稳健变异函数γR(h)及其估计量对于改善变异函数的结构性和理论模型的主要参数有明显效果。因此,稳健地质统计学的方法可以移植到水环境科学中应用,这对加强水环境空间变异性以及稳健变异函数的基础研究有着重要的应用价值和学术价值。This paper combined the Robust statistics with Geostatistic to identify and handle the outliers of sampling data of Cl^ - of underground water in Langfang city by Influencing Coefficient Method and Estimating Neighborhood Method, and better results were achieved. Based on what have been done, the sampling data of Cl^- before and after the identification and handling the outliers were systematically treated and analyzed by two robust statistical methods, Cressie - Hawkins Method and Median Regulation Method. The results show that the data of Cl^ - treated with the outliers by Influencing Coefficient Method and then analyzed by Median Regulation Method have more advantages, and obtaining the Robust variogram γR （h） and Kriging estimation are effective for improving the structure of variogram and parameters of its theoretical model Thus, the methods of Robust Geostatistics can be brought into the environmental science, which had high application and learned value to strengthen the spatial variability and the basic study of Robust variogram.

关 键 词：水环境 地质统计学 稳健统计学 特异值 稳健变异函数 

分 类 号：X52[环境科学与工程—环境工程]