多元统计在土壤重金属污染源解析中的应用  

The Application of Multivariate Statistical Analysis in the Pollution Source Recognition and Analysis of Heavy Metals in Soils

作　　者：林燕萍[1,2] 赵阳[1] 胡恭任[1,2] 苏光明[1] 

机构地区：[1]华侨大学环境科学与工程系,福建厦门361021 [2]中国科学院城市环境研究所环境与健康重点实验室,福建厦门361021

出　　处：《地球与环境》2011年第4期536-542,共7页Earth and Environment

基　　金：国家自然科学基金项目(21077036);中国科学院城市环境与健康重点实验室开放基金项目(KLUEH201004)资助

摘　　要：土壤中重金属污染及其防治愈来愈受重视,查明污染源是有效治理污染的前提。本文总结了近年来国内外聚类分析、主成分分析、因子分析在土壤重金属溯源中的应用。针对以往研究工作的不足和存在的问题,指出了今后土壤重金属污染源解析研究中的新思路:(1)GIS技术与多元统计相结合生成的污染专题地图,可以更好地确定污染物来源;(2)利用富集因子(EFs)可以深入区别重金属是人为污染源还是自然源,并且进一步验证多元统计分析的结果;(3)利用一些改进的方法如增强矩阵主成分分析(MAPCA)和正定矩阵因子分解(PMF),可以克服多元统计分析中的不足。More and more attention has been paid to the pollution and control of heavy metals in soils.It is very important to find the contamination source before effective treatment.The applications of cluster analysis,principal component analysis and factor analysis in research on heavy metal source recognition in soils are summarized in this paper.On the basis of the lack and the existing problems involved in the past studies,the key points of studying heavy metal contamination source recognition and analysis in the future are set forth as follows;(1) combining the GIS technology and multivariate statistical analysis to generate thematic maps of pollution,one can better determine the source of pollutants;(2) using the enrichment factors(EFs),one can determine the source of a single metal,which verifies the results of multivariate statistical analysis;and(3) taking the advantages of some improved methods such as matrix-augmented principal component analysis(MAPCA) and positive matrix factorization(PMF),one can overcome some limitations caused by multivariate statistical analysis.

关 键 词：土壤 重金属 多元统计 污染源解析 

分 类 号：X53[环境科学与工程—环境工程]