文献类型：期刊文献

中文题名：落叶松云冷杉林单木树高曲线的研究

英文题名：Individual Tree Height-diameter Curves of Larch-spruce-fir Forests

作者：曾翀[1] 雷相东[1] 刘宪钊[1] 赵理文[2] 杨英军[2]

第一作者：曾翀

机构：[1]中国林业科学研究院资源信息研究所,北京100091;[2]吉林省汪清林业局,吉林汪清133200

年份：2009

卷号：22

期号：2

起止页码：182-189

中文期刊名：林业科学研究

外文期刊名：Forest Research

收录：CSTPCD;;Scopus;北大核心:【北大核心2008】；CSCD:【CSCD2011_2012】；

基金：国家“十一五”科技支撑课题(2006BAD03A0802);国家自然科学基金(30371157、60073007)的部分内容

语种：中文

中文关键词：单木树高曲线;预测区间;容许区间

外文关键词：tree height curve ; predict interval; tolerant interval

分类号：S791.22

摘要：本研究以20块近天然落叶松云冷杉林为对象,基于4 309对实测树高-胸径数据,分树种(组)对31种常见的树高曲线进行了拟合。模型评价指标除考虑决定系数、均方误、平均误差、残差图外,还重点考虑模型的预测能力,即模型的预测区间和容许区间。结果表明:选出的树高曲线除落叶松和冷杉为线性模型外,其它均为三参数的Gompertzt和Logistic模型。研究给出了所选模型95%的预测区间及表示90%误差分布的容许区间,他们从统计上提供了模型将来用于预测的可靠性。
Thirty one height curves were calibrated and validated to predict individual tree height from diameter at breast height for larch, spruce and fir, Korean pine and deciduous groups in semi-natural larch-spruce-fir forests. The data came from 20 permanent sample plots with 4 309 observations, of which 80% were used for model development and 20% for model validation. Model selection was on the base of integrated evaluation of determined coefficient, mean square error, mean bias, residual diagnosis, prediction interval and tolerance interval. Results showed that Gompertzt and Logistic models were the best models for Korean pine, spruce and broadleaved tree species groups, and linear model for larch. Prediction intervals and tolerance intervals with 95 percent probability were given for different tree species groups. They provide statistical reliability on model prediction.