罗江县农田土壤全氮协同克里格插值和采样数量优化研究

[目的]研究县域土壤全氮含量的空间分布和采样数量,为紫色土丘陵区采样提供参考。[方法]利用协同克里格法,以初始的1 777个土壤全氮含量数据为随机抽取的数据,分别随机抽取1 599、1 421和1 243个数据为目标变量,并以初始的1 777个土壤有机质数据为辅助变量,对四川省罗江县土壤全氮含量进行插值分析,从而利用协同克里格法对县域尺度下农田土壤全氮含量在不同样点数量下空间分布中的适用性进行评价。[结果]在相同取样数量下,全氮协同克里格法的均方根误差相对于普通克里格法降低0.019 6%～0.072 5%,预测值和实测值之间的相关系数提高0.69%～0.90%。利用协同克里格法,土壤全氮含量数据在缩减30%情况下,其估值精度高于1 777个样点下的普通克里格估值,且二者的分布图都具有较高的拟合度。[结论]协同克里格法是一种经济、精准的方法,可为县域土壤养分含量的空间分布提供基础信息。

Spatial Interpolation and Sampling Numbers of Concentration of Total Nitrogen in Cropland Soil of Luojiang on County Scale Using Cokriging

[Objective]The spatial distribution and sampling numbers of the concentration of total nitrogen(TN) on county scale were studied to provide the information for sampling points in purple soil hilly region.[Method]Cokriging was used for the interpolation of the concentration of TN in cropland soil in Luojiang County,Sichuan Province.A total of original 1 777 measured data of the concentration of TN in soil and 1 599,1 421 and 1 243 measured data by random were selected as target variable,original 1 777 measured data of soil organic matter(OM) content as auxiliary variable.The Cokriging interpolation method was used to evaluate the suitability of estimating the spatial distribution for the concentration of TN in soil on county scale under different sampling.[Result]Compared with the ordinary kriging under the same sampling numbers,the root-mean-square-error produced by Cokriging decreased by 0.019 6% to 0.072 5%,correlation coefficient between the predicted value and the measured value increased 0.69% to 0.90%.The prediction accuracy of original data reduced 20% to 30% using Cokriging,which was still higher than original 1 777 data using ordinary Kriging,and their interpolation maps were very similar.[Conclusion]Cokriging was shown to be a quite economic and accurate method which could provide more information for the study of the spatial distribution of the concentration of TN in soil on county scale.