西南典型岩溶区土壤硒空间分布预测

发布时间：2018-01-06 19:38

  本文关键词：西南典型岩溶区土壤硒空间分布预测　出处：《农业工程学报》2016年22期 　论文类型：期刊论文

  更多相关文章： 土壤 硒 模型 化学行为 协同克里格 地理加权回归模型 空间分布预测 桂林永福

【摘要】：土壤硒精准预测和制图是富硒土壤资源开发利用和环境规划管理的基础。该文以西南典型岩溶区桂林永福百寿河流域为例,在分析影响土壤硒化学行为因子的基础上,通过野外样品采集和室内化学分析以及Arc GIS空间分析,获取了研究区相关地理环境因子和土壤属性因子数据。利用逐步回归方法选择土壤硒空间分布预测的辅助变量,使用协同克里格模型对非连续分布的辅助变量进行插值。在此基础上利用地理加权回归模型对土壤硒空间分布进行预测,同时以普通克里格插值结果作为参照。研究结果表明:使用地理环境因子和影响土壤硒化学行为的土壤属性因子可以提高土壤硒预测精度;协同克里格插值解决了辅助变量数据连续分布的问题;土壤硒的空间分布与地形和影响土壤硒化学行为的因子有关。
[Abstract]:The accurate prediction and mapping of soil selenium is the basis of the exploitation and utilization of selenium-enriched soil resources and environmental planning and management. This paper takes the Yongfu Beshouhe River Basin of Guilin as an example. Based on the analysis of influencing factors of soil selenium chemical behavior, field sample collection, indoor chemical analysis and Arc GIS spatial analysis were carried out. The data of geographical and environmental factors and soil attribute factors in the study area were obtained, and the auxiliary variables of soil selenium spatial distribution prediction were selected by stepwise regression method. The cooperative Kriging model was used to interpolate the auxiliary variables of discontinuous distribution, and the geographical weighted regression model was used to predict the spatial distribution of soil selenium. The results show that the precision of soil selenium prediction can be improved by using geographical environment factors and soil attribute factors which affect the chemical behavior of soil selenium. Cooperative Kriging interpolation solves the problem of continuous distribution of auxiliary variable data. The spatial distribution of soil selenium is related to the topography and the factors affecting the chemical behavior of soil selenium.
【作者单位】： 华中农业大学资源与环境学院;桂林理工大学地球科学学院;
【基金】：国家自然科学基金资助项目(41261082)
【分类号】：S153.6
【正文快照】： 邵亚,王毅伟,蔡崇法,杨顺华,张海涛.西南典型岩溶区土壤硒空间分布预测[J].农业工程学报,2016,32(22):178-183.doi:10.11975/j.issn.1002-6819.2016.22.024 http://www.tcsae.orgShao Ya,Wang Yiwei,Cai Chongfa,Yang Shunhua,Zhang Haitao.Prediction on spatial distributio，

本文编号：1389270