研究论文
齐峰,王学军.环境空间分析中常用采样方法及其对空间结构表达的影响[J].环境科学学报,2000,20(2):207-212
环境空间分析中常用采样方法及其对空间结构表达的影响
- Sampling methods in environmental spatial analysys and their effects on spatial continuity display
- 基金项目：国家自然科学基金项目(49771068);国家教委跨世纪人才培养计划项目
- 作者
- 单位
- 齐峰
- 北京大学城市与环境学系, 北京 100871
- 王学军
- 北京大学城市与环境学系, 北京 100871
- 摘要：采用几种采样方法在不同的采样密度下分别在两套模拟数据集上进行模拟采样,并以实验半变异函数为结构分析工具探讨了采样方案对具有不同空间分布特征数据的空间结构表达的影响.研究发现,采样密度与采样方法的选取都对空间结构的表达有显著影响.采样密度的影响主要表现在,随着采样密度的减小即样本量的减小,实验半变异函数对空间结构信息表现的精确度下降,对不同的采样方法其精确度的下降有不同表现.此外,采样方法在空间结构的合理表达上也起着重要作用,本文涉及的6种采样方法各有优劣,在空间结构表达上表现出不同的特点.
- Abstract：Spatial structure analysis in describing the spatial continuity of regionalized variables is a fundamental component in spatial analysis. Effects of different sampling design parameters on the quality of estimates from spatially correlated data have been extensively investigated, while efficiency of sampling designs for displaying and describing spatial continuity, particularly in environmental research, has been given less attention. In this paper, the performances of several sampling designs were examined and their different effects on simulated data were compared. Sample variogram is used to describe spatial structure. Based mainly on visual comparison, results showed that both sample size and sample pattern are of significant influence on the sample variograms, therefore, each plays important role on practical sampling.
摘要点击次数： 2056 全文下载次数： 4406
