欢迎访问《中国农学通报》，
中国农学通报

中国农学通报 ›› 2012, Vol. 28 ›› Issue (21): 311-316.doi: 10.11924/j.issn.1000-6850.2012-1908

• 农业信息 • 上一篇    

区域墒情的空间分析方法与应用

李阿伦 杨卫中 卢娟   

  • 收稿日期:2012-05-20 修回日期:2012-05-30 出版日期:2012-07-25 发布日期:2012-07-25

Spatial Analysis Methods and Application of Regional Soil Moisture

  • Received:2012-05-20 Revised:2012-05-30 Online:2012-07-25 Published:2012-07-25

PDF (PC)

721

摘要：

为了探索区域墒情的分布规律，获得科学的采样方案和分析方法，以北京通州区3022个点的土壤含水量数据为研究对象，综合利用距离反比法、普通克里金法、协克里金法、随机模拟法4种空间分析方法对区域内6、10、15、20 cm土层含水量进行了统计分析。研究表明，20 cm土层含水量较其他土层稳定，更适合作为土壤墒情的预测指标。随机模拟法平滑效应小，产生的区域墒情分布图与其他种方法生成的分布图差异较大，仅适用于存在较大变异的区域墒情分析。普通克里金、距离反比法、协克里金法产生的分布图基本一致，可用于一般区域墒情分析，其中，考虑6 cm土层含水量的协克里金法较普通克里金法减少了40%~60%的方差，是区域墒情分析的最优方法。在此基础上，提出了一种充分利用表层土壤含水量值来提高墒情预测精度的采样方法。

关键词: 产仔数, 产仔数

Abstract:

In order to explore the distribution of the regional soil moisture and get the scientific sampling and analysis method, four spatial analysis methods including inverse distance weighted method, ordinary kriging method, co-kriging method and stochastic simulation method were used to analyze the moisture content at the depth of 6 cm, 10 cm, 15 cm and 20 cm on 3022 sampled data in Tongzhou district of Beijing. The study showed that moisture content of soil at depth of 20 cm was more stable than any other depth, and it was suitable to be a prediction index of soil moisture content. The stochastic simulation method had a lower smooth effect and the map generated by this method was different from that of the other three methods, thus it was only suitable for the moisture analysis of area which has a large variation. The maps generated by other three methods were almost the same and all of them could be used for moisture analysis of the general area .Co-kriging method considering the soil moisture contents at the depth of 6 cm was the optimal method and the variance was 40%-60% less than that of ordinary kriging. At last, a sampling method which can make full use of the surface soil moisture content was proposed to improve the preciosity of soil moisture content prediction.