摘要
利用地统计学普通克里金法(Ordinary Kriging,OK)、简单克里金法(Simple Kriging,SK)、泛克里金法(Universal Kriging,UK)以及经验贝叶斯克里金法(Empirical Bayesian Kriging,EBK)分别对研究区土壤pH、全氮、有机质、速效钾和有效磷进行空间插值,通过插值结果分析不同土壤化学元素空间分布特征,利用交叉验证法定量分析不同克里金方法插值精度,确定最优空间插值方法以及最适宜半变异函数模型,定量定性分析不同空间插值方法对不同土壤化学元素空间分布特征的适用性。以遵化市某典型区土壤化学元素数据为例来验证不同插值方法对土壤化学元素空间分布特征的适用性分析,研究结果表明:普通克里金法插值后数据拟合性最优;不同土壤化学元素的空间分布差异性显著;土壤pH、有机质和速效钾选择简单克里金法插值的效果更好,土壤全氮和有效磷选择经验贝叶斯克里金法插值曲面更加平滑;以土壤有机质为例,选择精度最高的简单克里金法插值时,半变异函数模型选择指数函数空间插值精度最高。通过最优空间插值法研究土壤化学元素空间分布特征可以为地方农业健康稳定发展提供更加准确的科学依据。
Ordinary Kriging method(Ordinary Kriging,OK),Simple Kriging method(Simple Kriging,SK),Universal Kriging method(Universal Kriging,UK)and Empirical Bayesian Kriging(Empirical Bayesian Kriging,EBK)were used for spatial interpolation of soil pH,total nitrogen,organic matter,available potassium and available phosphorus in the study area,and the spatial distribution characteristics of different soil chemical elements were analyzed by interpolation results.Cross-validation method was used to quantitatively analyze the interpolation accuracy of different Kriging methods,the optimal spatial interpolation method and the most suitable semi-variation function model were determined,and the applicability of different spatial interpolation methods to the spatial distribution characteristics of different soil chemical elements was quantitatively and qualitatively analyzed.Taking the soil chemical element data of a typical district in Zunhua City as an example,the applicability of different interpolation methods to the spatial distribution characteristics of soil chemical elements was analyzed.The results show that the data fit best after the ordinary Kriging method interpolation.The spatial distribution of chemical elements in different soils is significantly different.The results of simple Kriging interpolation for soil pH,organic matter and available potassium are better,and the surfaces of empirical Bayesian Kriging interpolation for soil total nitrogen and available phosphorus are smoother.Taking soil organic matter as an example,when the simple Kriging method interpolation is selected with the highest accuracy,the exponential function spatial interpolation is selected with the highest accuracy by the semi-variogram model.The study of spatial distribution characteristics of soil chemical elements by optimal spatial interpolation method can provide more accurate scientific basis for the healthy and stable development of local agriculture.
作者
冯郑文
刘亚静
FENG Zheng-wen;LIU Ya-jing(College of Mining Engineering,North China University of Science and Technology,Tangshan Hebei 063210,China)
出处
《华北理工大学学报（自然科学版）》
CAS
2023年第4期7-18,共12页
Journal of North China University of Science and Technology：Natural Science Edition
基金
国家自然科学基金青年科学基金(42101393)。
关键词
空间插值
适用性
交叉验证
半变异函数
spatial interpolation
applicability
cross validation
semi-variogram
