县域耕地土壤锌含量的协同克里格插值及采样数量优化  

Optimization of Spatial Interpolation and Sampling Size for Soil Zinc Content in Cultivated Land at County Scale Using Cokriging Method

作　　者：李润林[1,2] 姚艳敏[3,4] 唐鹏钦[3,4] 于士凯[3,4] 王德营[3,4] 

机构地区：[1]农业部兰州黄土高原生态环境重点野外科学观测实验站,甘肃兰州730050 [2]中国农业科学院 兰州畜牧与兽药研究所,甘肃兰州730050 [3]农业部农业信息技术重点实验室,北京100081 [4]中国农业科学院农业资源与农业区划研究所,北京100081

出　　处：《土壤通报》2013年第4期830-838,共9页Chinese Journal of Soil Science

基　　金：国家973计划课题“气候变化对我国粮食生产资源要素的影响机理研究”(2010CB951501-2)、“气候变化对西北春小麦单季玉米区粮食生产资源要素的影响机理研究”(2010CB951502);国家863计划子课题“农村抽样调查空间化样本抽选与管理系统”(2006AA120103)资助

摘　　要：耕地土壤锌含量的空间分布和适宜的采样数量研究,为农产品产地评价和认证、土壤环境质量调查提供帮助。以土壤有机质、pH、全氮、碱解氮、有效铁、有效铜、有效锰、高程数据为辅助变量,采用协同克里格插值方法,对吉林省舒兰市耕地土壤锌含量进行了插值精度分析,并对不同样点数量下协同克里格法在县域尺度耕地土壤锌含量空间分布研究中的适用性进行了评价。结果表明,相同采样数量下,土壤锌协同克里格法的均方根误差比普通克里格法低,拟合效果更好;辅助因子对土壤锌协同克里格插值的影响效果大小为:全氮>有效锰>高程>有效铁>碱解氮>pH>有机质>有效铜;当高程和有效铜作为辅助因子时,土壤采样数量可以减少20%,土壤锌协同克里格插值仍能满足精度要求。协同克里格作为一种更为精确和经济的空间插值方法。Achievements in the research on the spatial distribution and the suitable sampling size of soil Zn content in cultivated land could lay definite foundation for the evaluation and certification of regions producing agricultural products and for the investigation of land environmental quality. Using organic matters, pH value, total nitrogen, alkali-hydrolyzable nitrogen, available Fe and available Cu and available Mn in the soil and elevation data as auxiliary variables, the accuracy analysis of interpolation was conducted for soil Zn content in the cultivated land in Shulan City of Jilin Province in China, by adopting Cokriging interpolation method. Meanwhile, the applicability of Cokriging method at different sampling sizes was evaluated in the research on the spatial distribution of soil Zn content in the cultivated land at county scale. It was found that at the same sampling size, the root mean square error produced by the Cokriging was lower than that by ordinary Cokriging, with better fitting results. The effects of auxiliary variables on the Cokriging interpolation ranked as total nitrogen〉available Mn〉elevation data〉available Fe〉alkali-hydrolyzable nitrogen〉pH value〉organic matters〉available Cu. Using the elevation data and the available Cu as the auxiliary variables, the soil sampling size was reduced by 20%, and the accuracy on the Cokriging interpolation of soil Zn content was still satisfied. It is concluded that the Cokriging interpolation method as a more accurate and economic spatial interpolation method might provide more information and helps for the research on the spatial distribution of soil heavy metal content in the cultivated land at county scale.

关 键 词：耕地土壤 锌 协同克里格法 普通克里格法 采样数量 

分 类 号：S158[农业科学—土壤学]