地理学报 ›› 2009, Vol. 64 ›› Issue (3): 368-380.doi: 10.11821/xb200903012

• 研究综述 • 上一篇    

地理空间抽样理论研究综述

姜成晟1,2, 王劲峰1, 曹志冬3   

  1. 1. 中国科学院地理科学与资源研究所,北京100101;
    2. 中国科学院研究生院,北京100049;
    3. 中国科学院自动化研究所,北京100080
  • 收稿日期:2008-10-28 修回日期:2008-12-29 出版日期:2009-03-25 发布日期:2010-08-03
  • 通讯作者: 王劲峰,男,研究员,博士生导师。E-mail: wangjf@lreis.ac.cn E-mail:wangjf@lreis.ac.cn
  • 作者简介:姜成晟(1981-),男,博士,主要研究方向空间抽样理论、空间数据分析与建模。E-mail: jiangcs@lreis.ac.cn
  • 基金资助:

    国家自然科学基金(40471111;70571076);863 课题(2006AA12Z205)资助

A Review of Geo-Spatial Sampling Theory

JIANG Chengsheng1,2, WANG Jinfeng1, CAO Zhidong3   

  1. 1. Institute of Geographic Sciences and Natural Resources Research,CAS,Beijing 100101,China;
    2. Graduate University of Chinese Academy of Sciences,Beijing 100039,China;
    3. Institute of Automation,CAS,Beijing 100080,China
  • Received:2008-10-28 Revised:2008-12-29 Online:2009-03-25 Published:2010-08-03
  • Supported by:

    National Natural Science Foundation of China,No.40471111;No70571076;863 Program,No.2006AA12Z205

摘要:

抽样调查是地理研究、资源评估、环境问题研究和社会经济问题研究的重要手段。对于地理分布的各种资源, 由于调查数据往往具有空间相关性, 传统的抽样调查理论无法满足日益增长的空间抽样需求。空间抽样理论是对具有空间相关性的各种资源和调查对象进行抽样设计的基础。本文详细论述了空间抽样理论发展现状。首先介绍了空间抽样的产生和发展, 以及空间抽样所要研究的四个问题。然后介绍了基于设计的和基于模型的抽样统计推断方式, 以及它们适用的范围。最后本文详细论述了Kriging 理论在抽样理论的应用、前向、后向和双向样本布局方法和六种空间抽样样本优化选择标准。

关键词: 地理空间, Kriging 抽样, 抽样调查

Abstract:

Sample survey is the groundwork of the studies of natural resources, environmental problems and socio-economy. The geo-related characteristic of spatial data limits the application of classic sampling theory, which is essentially based on independent assumption. Spatial sampling theory is the foundation of sample survey of spatial related resources. Firstly, this paper introduces the history of spatial sampling theory and presents four main issues addressed by this theory. Then it reviews the theory and the applications of model-based and design-based statistics inference. Finally, this paper gives a detailed description of (1): Kriging theory application in spatial sampling; (2): forward and backward samples distribution methods, and the combination of the above two; (3): six criteria for optimization of sample selection.

Key words: geography spatial, Kriging sampling, sample survey