基于冠幅的杉木人工林胸径和树高参数化预估模型

Parametric prediction models of DBH and height for Cunninghamia lanceolata plantation based on crown width

为了研究杉木生长与形态变量的关系,提高模型预估精度,本文根据福建省65块不同密度、林龄分布于6~35年的杉木人工林样地测量数据,选用了Richards理论方程,分别构建了杉木胸径和树高生长模型。在此基础上,分析理论方程中参数与冠幅（CW）的关系,建立两者间的关系函数;使用参数预估法构造胸径和树高的参数化预估模型,并对原模型和参数化模型进行模型检验、残差分析,讨论了冠幅与林木生长之间的关系。结果表明：胸径和树高Richards方程中,参数a与CW呈正比,参数b与CW呈反比,c则与CW呈现出类似正态分布的关系;与传统的理论方程相比,参数化模型提高了杉木胸径树高预估的精度。本文所构建的模型能间接反映林分因子的状态,同时对于分析林木竞争、光合作用、模拟林木三维可视化等方面也提供了一种模式。

In order to clarify the relationship between Chinese fir growth and form variables and to improve the forecast precision of harvest,we analyzed data from trees located in 65 Chinese fir（ Cunninghamia lanceolata） plots of varying densities with stand age ranging from 6 to 35 years in Fujian Province.Richard＇s equation was adopted to develop diameter at breast height（ DBH） and tree height（ H） models of Chinese fir. On this basis,the relationship between parameters of the theoretical equation and crown width（ CW） were analyzed,and the parameterized harvest forecast model was developed using parameter estimates method through the parameter-CW functions. The model test and residual analysis were carried out for the original models and the parametric models. The relationship between the CW and tree growth is discussed. Results showed that in DBH and parameter a in Richard＇s equations of tree height were proportional to CW,parameter b was inversely proportional to CW,and parameters c and CW presented approximately normal distribution. Compared with the traditional theory equations,the parameterized models improve the forecast precision of Chinese fir harvest. The proposed models in this paper are convenient for popularization and application in the future and can indirectly reflect the state of stand factors. The proposed models also provide a mode for analyzing tree competition,photosynthesis,and simulating dimensional visualization of trees.