文献类型：期刊文献

中文题名：基于空间结构调查的林分密度估计

英文题名：Stand density estimation based on the measurement of spatial structure

作者：王宏翔[1] 惠刚盈[1] 张弓乔[1] 李远发[1] 刘恩[1]

第一作者：王宏翔

机构：[1]中国林业科学研究院林业研究所/国家林业局林木培育重点实验室

年份：2014

卷号：25

期号：7

起止页码：1912-1918

中文期刊名：应用生态学报

外文期刊名：Chinese Journal of Applied Ecology

收录：CSTPCD;;Scopus;北大核心:【北大核心2011】；CSCD:【CSCD2013_2014】；PubMed;

基金："十二五"国家科技攻关计划项目(2012BAD22B03)资助

语种：中文

中文关键词：林分空间结构;距离法;密度估计;空间分布格局;角尺度

外文关键词：stand spatial structure; distance sampling; density estimation; spatial pattern; uniform angle index.

分类号：S758.5

摘要：利用林分空间结构抽样调查技术,采用测量抽样点到其第k株最近相邻木的距离(距离法)进行密度估计,分别选取k=4和k=6两种距离调查方法进行分析,并对Prodan、Persson、Thompson 3种不同密度估计方法的预估能力进行检验.结果表明:不同预估方法的预估能力受林木水平分布格局影响.Prodan法在均匀分布的林分中有较强的预估能力,随着分布格局聚集性增加会产生越来越大的偏差;Persson计算法在均匀及随机分布的林分中产生正偏差,但随着分布格局聚集性增加产生的相对误差接近0,预估能力增强;Thompson计算法对随机或接近随机分布的林分有较强的预估能力,而在均匀分布和聚集分布的格局中分别产生正偏差和负偏差.抽样点为49个时,选择6株木与4株木预估能力无显著差异,因此,密度估计可与选取4株相邻木的空间结构参数调查整合在一起.
This study estimated stem density by combining distance sampling with stand spatial structure investigation techniques. We tested the statistical performance of two investigative methods （ selecting the fourth- or sixth-nearest tree to the sample location） and three different density estimators （Prodan, Persson and Thompson）. Different spatial distribution patterns influenced the performanee of these estimators. Prodan＇s estimator was unbiased for uniform patterns, and it pro- duced increasing bias with increasing spatial clustering. Persson＇ s estimator produced consistent positive bias for uniform and random patterns, with the smallest bias for clustered patterns. Thomp- son＇ s estimator was robust for random or near-random patterns, and it produced positive and negative bias for uniform and clustered patterns, respectively. No significant performance difference was found between selecting the fourth- and the sixth-nearest trees with the same sample size of 49. Thus, we could combine distance sampling with spatial structure investigation techniques.