鸡公山自然保护区杉木人工林树高-胸径模型研究  

Study on height-D models for Cunninghamia lanceolata plantation in Jigongshan nature reserve

作　　者：周梦丽 王心楠 毕舒雯 张明浩 郭芳[1] 闫东锋[1] 姚建峰 陈涛 段世磊 ZHOU Mengli;WANG Xinnan;BI Shuwen;ZHANG Minghao;GUO Fang;YAN Dongfeng;YAO Jianfeng;CHEN Tao;DUAN Shilei(College of Forestry,Henan Agricultural University,Zhengzhou 450046,China;College of Computer and Information Technology,Xinyang Normal University,Xinyang 464000,China;Xinyang Academy of Ecological Research,Xinyang 464000,China;Henan Forestry Vocational College,Luoyang 471002,China;Henan Yuanzhi Forestry Planning and Design Co.,Ltd.,Zhengzhou 450003,China)

机构地区：[1]河南农业大学林学院,河南郑州450046 [2]信阳师范大学计算机与信息技术学院,河南信阳464000 [3]信阳生态研究院,河南信阳464000 [4]河南林业职业学院,河南洛阳471002 [5]河南省远志林业规划设计有限公司,河南郑州450003

出　　处：《河南农业大学学报》2025年第2期221-232,共12页Journal of Henan Agricultural University

基　　金：河南省科技攻关项目(242102320245);河南农业大学博士启动金项目(30500975);信阳生态研究院开放课题(2023XYQN04,2023XYZD02)。

摘　　要：【目的】杉木(Cunninghamia lanceolata)为典型的亚热带树种,鸡公山自然保护区为其分布的北缘,探究其胸径和树高关系,构建适宜的树高-胸径模型,为其林木生长及适应性经营提供一定的参考依据。【方法】以鸡公山国家级自然保护区的杉木人工林为研究对象,选择27个常用的树高-胸径经验模型,通过调整决定系数(Ra2dj)、均方根误差(J)、平均绝对误差(Q)、相对平均绝对误差(R)和Akaike信息准则(A)5个模型评价指标,筛选出5个最佳的基础模型,利用相关分析确定第1活枝高(Z)及冠幅(W)与树高(H)间的相关关系,并将其引入所筛选出的基础模型,构建15个修正模型,分析各修正模型间的拟合效果差异,确定适宜的树高-胸径模型。【结果】27个树高-胸径经验模型均可求解,M-10、M-16、M-17、M-22、M-25经验模型拟合效果较优。综合考虑模型的拟合及验证效果,M-10在基础模型中、引入Z的M-16(即M-16_Z)在修正模型中的拟合效果均最优,且两者相比,从0.6671(M-10)提高到0.7426(M-16_Z)。【结论】模型的适宜性可能受林分所处地区、林分类型等因素的影响,引入Z的M-16修正模型适宜量化鸡公山自然保护区杉木人工林树高与胸径关系。【Objective】Cunninghamia lanceolata is a typical subtropical tree species and Jigongshan nature reserve is the northern edge of its distribution.The relationship between its diameter at breast height and tree height was explored and a suitable height-D model was constructed to provide a certain reference basis for its tree growth and the adaptive forest management.【Method】Taking the Cunning⁃hamia lanceolata plantation in the Jigongshan National Nature Reserve as the research object,27 commonly used height-D empirical models were selected and 5 optimal basic models were screened using the five model evaluation indices,namely,adjusted coefficient of determination(R2a dj),root mean square error(J),mean absolute error(Q),relative mean absolute error(R)and Akaike information criterion(A).15 modified models were constructed by using correlation analysis to determine the correlation between the first live branch height(Z),crown width(W)and tree height(H)and then introducing them into the selected basic models.The differences in fitting effects among modified models were analyzed to determine the appropriate height-D model.【Result】27 height-D empirical models can be solved.The fitting effects of M-10,M-16,M-17,M-22 and M-25 have better fitting effects.Taking into account the fitting and validation effects of the models,the fitting effect of M-10 among the basic models and M-16 with Z(M-16_Z)among the modified models are both the best,and comparing to these two models,the fitting effect has increased from 0.6671(M-10)to 0.7426(M-16_Z).【Conclusion】Hence,the suitability of the model may be influenced by the factors such as the location and type of the stand.The modified model M-16 with Z is suitable for quantifying the relationship between tree height and diameter at breast height of Cunninghamia lanceolata plantation in Jigongshan nature reserve.

关 键 词：鸡公山 自然保护区 杉木 人工林 树高-胸径模型 模型评价 

分 类 号：S757.1[农业科学—森林经理学]