摘要
应用地统计学原理和方法 ,分析了夏玉米田间 90m× 90m(L区 ) ,内套一个 30m× 30m(M区) ,再内套一个 1 0m× 1 0m(S区 ) 3个尺度下土壤含水量和电导率的空间变异性。结果表明 ,各尺度间土壤含水量的标准差和变异系数比较接近而电导率的差别较大。在L和M尺度下土壤含水量在 3个置信度下呈正态或对数正态分布 ,在S尺度下只在 0 0 1置信度下呈对数正态分布 ;在L尺度下土壤电导率仅在信度为 0 0 5和 0 0 1时呈对数正态分布 ,在M尺度下仅在信度为 0 0 1时正态或对数正态分布 ,较高信度时呈对数正态分布 ,在S尺度下不呈正态或对数正态分布。各尺度下它们的空间结构可用球状模型描述 ,其变程分别约为 2 2～ 1 9 8m(土壤含水量 )和 2 5～ 1 7 2m(电导率 ) ,不同尺度间土壤含水量和电导率的空间变异结构函数差别较大 ,用套合结构模型能更好地描述不同尺度下土壤含水量和电导率的空间变异性。
Using geostatistics, spatial variability of soil water content and bulk soil electrical conductivity was analyzed in the L (90m×90m), M(30m×30m, nested in the L ) and S(10m×10m, nested in the M ) area of summer maize fields The results indicated that, no significant differences were in the standard error and coefficient of variation (CV%) of soil water content among the different scales, but such differences were in bulk soil electrical conductivity Water content was distributed normally or log normally in L and M, but it was only distributed log normally for 0 01 confidence level in S area Electrical conductivity was not distributed normally or log normally in S, but it was distributed normally or log normally for 0 01 confidence level, log normally for higher confidence levels in M, and log normally for 0 05 and 0 01 confidence levels in L area Semivariograms showed spherical models with the range of 2 2～19 8m and 2 5～17 2m for water content and electrical conductivity among different scales And it was suggested that nested models should be created when significant differences were showed among the models of different scales In addition, nested models were used to describe the spatial variability of soil water content and electrical conductivity among the different scales
出处
《植物营养与肥料学报》
CAS
CSCD
北大核心
2001年第3期255-261,共7页
Journal of Plant Nutrition and Fertilizers
基金
国家重点基础研究发展规划项目 (G19990 1170 9)
国家重点科技攻关专题 ( 95 0 0 1 0 2 0 9)
中德合作项目资助
关键词
土壤含水量
电导率
空间变异
不同采样尺度
套合结构模型
soil water content
bulk electrical conductivity
spatial variability
different scales
nested model
