中国农业科学 ›› 2017, Vol. 50 ›› Issue (21): 4138-4148.doi: 10.3864/j.issn.0578-1752.2017.21.008

• 土壤肥料·节水灌溉·农业生态环境 • 上一篇    下一篇

基于地理探测器的农田土壤重金属影响因子分析

李雨1, 2,韩平2,任东1,罗娜2,王纪华1, 2

 
  

  1. 1三峡大学计算机与信息学院,湖北宜昌 443002;2北京市农林科学院北京农业质量标准与检测技术研究中心,北京 100097
  • 收稿日期:2017-06-05 出版日期:2017-11-01 发布日期:2017-11-01
  • 通讯作者: 王纪华,Tel:010-51503488;E-mail:wangjh@nercita.org.cn
  • 作者简介:李雨,Tel:18230062821;E-mail:liyu1127641712@163.com。
  • 基金资助:
    国家公益性行业(农业)科研专项(201403014-04)国家高技术研究发展计划项目(2013AA102302)

Influence Factor Analysis of Farmland Soil Heavy Metal Based on the Geographical Detector

LI Yu1,2, HAN Ping2, REN Dong1, LUO Na2, WANG JiHua1,2   

  1. 1Computer and Information College of Three Gorges University, Yichang 443002, Hubei; 2Beijing Research Center for Agricultural Standards and Testing, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097
  • Received:2017-06-05 Online:2017-11-01 Published:2017-11-01

摘要: 【目的】研究土壤重金属与影响因子以及不同土壤重金属之间相关关系,为土壤重金属空间预测模型提供更加全面的辅助变量。【方法】利用地理探测器模型,结合空间插值技术对2014年湖南省湘潭县5个乡镇农田中5种土壤重金属Pb、Cd、As、Cr和Hg的空间分布变化与6种影响因子以及5种土壤重金属之间的相关性和交互作用进行研究。【结果】研究结果表明,GDP、平均温度和相对湿度对5种土壤重金属解释力较大(PD,H值均在0.5以上),土壤pH、土壤类型与高程对土壤重金属的解释力较小(PD,H值均在0.3以下),其中土壤类型对于5种土壤重金属的解释力最低(PD,H值均在0.1以下)。5种土壤重金属中,Cr对Cd的解释力最强(PD,H值达到0.95),As对Cd的解释力最小(PD,H值仅为0.20)。平均温度、相对湿度、GDP对土壤重金属的影响显著高于其他影响因子,而其他影响因子之间的解释力差异并不显著。各影响因子之间和5种重金属元素之间均具有相互增强或非线性增强作用。【结论】土壤重金属的空间分布是由多种影响因子共同作用的结果。通过地理探测器模型发现,GDP、平均温度和相对湿度等影响因子对研究区域内的土壤重金属的空间分布具有较强的解释力,这些影响因子可作为研究区土壤重金属空间预测模型的辅助变量。地理探测器模型可以对多种影响因子进行更加全面的分析,为土壤重金属空间预测模型的建立提供有效的依据。

关键词: 地理探测器, 土壤重金属, 影响因子, 相关分析, 交互作用

Abstract: 【Objective】To study the correlation between soil heavy metals and influencing factors and heavy metals in different soils, as well as to provide a more comprehensive auxiliary variable for soil heavy metal spatial prediction model.【Method】The spatial distributions of heavy metals Pb, Cd, As, Cr and Hg in five soils in five towns of Xiangtan County, Hunan Province were analyzed by using the geophysical model and the spatial interpolation technique. The correlation and interaction of spatial distributions of heavy metals and 6 factors, as well as that of five heavy metals, were studied.【Result】The results showed that Gross Domestic Product (GDP), the average temperature and relative humidity had a greater explanatory power to the five kinds of soil heavy metals to (PD, H values are above 0.5). Soil pH, soil type, elevation and soil heavy metals were less explanatory (PD, H values below 0.3). The soil type had the lowest explanatory power to five kinds of soil heavy metals (PD, H values were below 0.1). Among the five soil heavy metals, Cr has the strongest explanatory effect on Cd (PD, H = 0.95), and As is the least (PD, H = 0.20). The effects of average temperature, relative humidity and GDP on soil heavy metals were significantly higher than those of other Influence factors, while the difference of explanatory power between other influence factors was not significant. There are mutually reinforcing or non-linear enhancement effects between the Influence factors and between the five heavy metal elements. 【Conclusion】The spatial distribution of heavy metals in soil is the result of the interaction of multiple influence factors. Based on the geographical exploration model, the Influence factors, such as GDP, average precipitation, average temperature and relative humidity, have strong explanatory power for the spatial distribution of heavy metals of research area in soils. These influence factors can be used as the soil heavy metal space in the study area predictive model of the auxiliary variable. Geographic detector model can provide a more comprehensive analysis of various influence factors, and provide an effective basis for the establishment of soil heavy metal spatial prediction model.

Key words: geographical detector, soil heavy metal, influence factor, correlation analysis, interaction