土壤含水量空间变异性的频域分析方法  

Frequency-domain analysis of spatial variability of soil properties

作　　者：杨卫中[1] 王一鸣[1] 李保国[2] 

机构地区：[1]中国农业大学信息与电气工程学院,北京100083 [2]中国农业大学资源与环境学院,北京100094

出　　处：《水利学报》2011年第5期580-587,共8页Journal of Hydraulic Engineering

基　　金：国家"863"高技术研究发展计划资助项目(2006AA10Z264)

摘　　要：基于频域法分析理论原理,给出了基于二维信号采样与重建的频域法插值公式,对土壤特性的空间变异性进行表征。对一块麦田采用20m×20m的网格进行土壤含水量采样,利用频域法对15×15个采样值进行分析,结果表明含水量分布的有效带宽为0.01,说明空间变异主要由周期大于100m的频率分量组成;通过改变采样间距的"重采样"方法,从原始采样数据中抽取采样间隔为40m×40m和60m×60m样本数为8×8和5×5的两组数据,分别用频域法和Kriging法插值。比较两种插值结果的分布图和统计特征值,结果显示Kriging法平滑效应明显而频域法插值能更好的表现空间变异性,二者的插值精度基本相同,因此频域法优于Kriging插值。Frequency-domain analysis was employed in investigating the spatial distribution of soil properties.The concept of effective bandwidth was presented to express the spatial variability of soil properties quantitatively.Frequency-domain method interpolation formula based on the theory of two-dimensional signal sampling and reconstruction was presented.A measurement of soil moisture sampling value in a wheat field under 20m×20m sampling grid was carried out,and the frequency-domain method was used to analyze the 15×15 soil moisture sampling value.The results show that the effective bandwidth of the wheat field equal to 0.01,it means that the spatial distribution of soil moisture is mainly composed of the frequency components in the period of more than 100m.Then,re-sampling technique was employed in the original sampling data with changing sampling interval to 40m×40m and 60m×60m to extract two data sub-sets of 8×8 and 5×5.The sub-sets were used in frequency-domain interpolation and Kriging interpolation.Comparison of the soil moisture distribution profile and statistical characteristics of two interpolation method shows that the smoothing effect of Kriging method cover up some detail of the spatial variability,two methods are basically the same in interpolation accuracy,and frequency-domain method is better than Kriging method.

关 键 词：频域法 空间变异 地统计学 有效带宽 土壤含水量 KRIGING 

分 类 号：S126[农业科学—农业基础科学] S152.7