欢迎访问林业资源管理

林业资源管理 ›› 2016, Vol. 0 ›› Issue (4): 47-52.doi: 10.13466/j.cnki.lyzygl.2016.04.010

• 科学研究 • 上一篇    下一篇

森林面积不同抽样估计方法的无偏性及有效性分析与证明

葛宏立, 孟源源   

  1. 浙江农林大学 环境与资源学院;浙江省森林生态系统碳循环与固碳减排重点实验室,浙江 临安 311300
  • 收稿日期:2016-05-20 修回日期:2016-07-18 出版日期:2016-08-28 发布日期:2020-11-04
  • 作者简介:葛宏立(1960-),男,浙江诸暨人,教授,硕士,主要研究领域为森林资源监测。Email:jhghlhxl@sina.com
  • 基金资助:
    国家自然科学基金“遥感图像森林信息的膨胀-剔除方法研究”(41371411)

Analysis and Proof of Bias and Efficiency of Different Sampling Methods for Forest Area Estimation

GE Hongli, MENG Yuanyuan   

  1. School of Environment & Resource Sciences,Zhejiang A & F University,Zhejiang Provincial Key Laboratory of Carbon Cycling in Forest Ecosystems and Carbon Sequestration,Lin′an,Zhejiang 311300,China
  • Received:2016-05-20 Revised:2016-07-18 Online:2016-08-28 Published:2020-11-04

摘要: 森林资源连续清查是我国获取宏观森林面积的主要手段,采用的是系统抽样方法。用理论分析的方法对样地森林面积计量3种不同确定方法的偏性与有效性进行论证。结果表明:1)连续变量法是无偏的,且效率最高;2)点定法也是无偏的,但效率较连续变量法低;3)优势地类法在某些情况下是有偏的,当总体单元的森林面积比例均接近于0或1时,结果接近无偏;4)本文结论表明,在遥感大样地调查中选择连续变量法是最合理的。

关键词: 森林资源连续清查, 森林面积估计, 抽样估计, 无偏性, 有效性

Abstract: Continuous Forest Inventory (CFI) is the main method to get the China's forest area,which is conducted in a systematic sampling manner.In this paper,bias and efficiency of these three plot forest measurement methods will be proved by theoretical analysis.The result shows that:1.CV is unbiased,and it has the highest efficiency;2.PLB is also unbiased,but the efficiency is lower than CV;3.DLB is biased,however the population proportion of forest is almost 0 or 1,the result is nearly unbiased;4.CV method,which is unbiased and has high efficiency,should be chosen for remote sensing large-plot inventory.

Key words: continuous forest inventory, forest area estimation, sampling, unbiased, efficiency

中图分类号: