空间回归分析在土壤属性预测制图中的应用  

Study on Predicated Mapping of Soil Property Based on Spatial Regression Analysis

作　　者：刘晓冰[1] 程道全 刘鹏飞[1] 宋轩[1] 陈杰[1] 

机构地区：[1]郑州大学水利与环境学院,郑州450001 [2]河南省土壤肥料站,郑州450002

出　　处：《土壤》2013年第3期533-539,共7页Soils

基　　金：国家自然科学基金项目(40971128)资助

摘　　要：以河南省孟津县为研究区,选取坡度、高程、地面曲率和复合地形指数(CTI)作为表层土壤缓效钾含量空间预测的环境协变量,系统探讨了空间回归分析技术在土壤属性预测制图中的应用。结果表明:土壤缓效钾的空间自相关距离阈值约为10 000 m,与坡度、高程和地面曲率存在显著相关性;尽管空间回归模型的预测精度和普通回归模型相近,但前者可以更加准确地表征土壤缓效钾的空间分布格局及空间分异细节特征。In this paper Mengjin County of Henan Province we selected as the study area, slowly available K contained in topsoil was spatially predicated by the method of spatial regression analysis with the terrain characteristics, included slope, elevation, curvature and CTI. The results indicated that the slowly available K had a distance threshold of spatial autocorrelation around 10 000 m and a significant correlation between slope, elevation, curvature and soil available K. Although output precision of the spatial regression model was quite similar to that of the classic regression ones, the spatial regression model displayed an evident advantage to reveal further the spatial distribution pattern and variation characteristics of soil property.

关 键 词：土壤属性 土壤缓效钾 环境协变量 空间回归模型 土壤预测制图 

分 类 号：S159-3[农业科学—土壤学]