欢迎访问林业科学,今天是

林业科学 ›› 2014, Vol. 50 ›› Issue (3): 76-82.doi: 10.11707/j.1001-7488.20140311

• 论文与研究报告 • 上一篇    下一篇

稀少群团状植被自适应群团抽样适宜的单元大小

朱光玉1, 李新建2, 陈振雄2, 吕勇1   

  1. 1. 中南林业科技大学 长沙 410004;
    2. 国家林业局中南林业调查规划设计院 长沙 410014
  • 收稿日期:2013-05-16 修回日期:2013-12-22 出版日期:2014-03-25 发布日期:2014-04-16
  • 基金资助:

    国家自然科学基金青年项目(31100476);国家教育部科技发展中心青年教师基金项目(20114321120002);国家自然基金项目(30510103195);科技部社会公益研究专项(2005DIB5J142);中南林业科技大学青年基金重点项目(QJ2010004A)

Appropriate Sample Unit Size of Adaptive Clustering Sampling for Rare and Clustering Vegetation

Zhu Guangyu1, Li Xinjian2, Chen Zhenxiong2, Lü Yong1   

  1. 1. Central South University of Forestry and Technology Changsha 410004;
    2. Central South Forest Inventory and Planning Institute of State Forestry Administration Changsha 410014
  • Received:2013-05-16 Revised:2013-12-22 Online:2014-03-25 Published:2014-04-16

摘要:

样本单元面积大小不仅关系到自适应群团抽样调查的精度,而且影响其抽样成本,是影响自适应群团抽样应用推广的重要指标。以内蒙古磴口县乌兰布和沙漠边缘的稀少群团状典型植被(灌木和小乔木)为研究对象,提出以变动系数作为衡量抽样的效率指标,设计几种不同样本单元面积,通过500次重复模拟试验,从变动系数与样本单元大小的相关关系入手,对模拟结果进行比较分析,研究适宜的样本单元面积大小。结果表明:同一抽样总体,当初始样本量相同时,随着样本单元面积变大,变动系数将递减,渐渐趋于一个稳定的常数;对于分布稀少、群团状的灌木,进行自适应群团抽样调查时,其最小的适宜样本单元面积为100 m2,可以采用的适宜单元面积大小区间为100~200 m2;对于分布稀少、群团状的小乔木,进行自适应群团抽样调查时,其最小的适宜样本单元面积为200 m2,可以采用的适宜单元面积大小区间为200~300 m2

关键词: 乌兰布和沙漠, 稀少群团状植被, Hansen-Hurwitz估计量, Horvitz-Thompson估计量, 变动系数

Abstract:

Sample unit size is an important effect factor for using adaptive cluster sampling,because which can affect the precision and the costs of adaptive cluster sampling.To find the appropriate sample unit size for adaptive cluster sampling,variation coefficient was first used as the criterion for evaluating efficiency of adaptive cluster sampling,simulation sampling was processed based on the rare and clustering population,which included typical shrubbery and small arbor at Ulanbuh desert edge,and the relation of variation coefficient and sample unit size had been analyzed to search for the appropriate sample unit size for adaptive cluster sampling.Several conclusion had been gained: when the sample unit size was increasing,the variation coefficient would be decreasing to a small constant with the population and equivalent sample size; the minimum appropriate sample unit size was 100 m2 for the rare and clustering shrubbery,and the appropriate sample unit size might be 100-200 m2; the minimum appropriate sample unit size was 200 m2 for the rare and clustering small arbor,and the appropriate sample unit size might be 200-300 m2.

Key words: Ulanbuh desert, rare and clustering vegetation, Hansen-Hurwitz estimator, Horvitz-Thompson estimator, variation coefficient

中图分类号: