云冷杉针阔混交林单木枝下高和冠幅模型构建  

Modeling of height to crown base and crown width in spruce-fir conifer-broadleaf mixed forest

作　　者：李应涛 刘时良 孙海龙 王卫霞[4] 向玮 LI Yingtao;LIU Shiliang;SUN Hailong;WANG Weixia;XIANG Wei(State Forestry and Grassland Administration Key Laboratory of Forest Resources&Environmental Management,Beijing Forestry University,Beijing 100083,China;Guangxi Forestry Survey and Design Institute,Nanning,Guangxi 530011,China;Wangqing Forestry Bureau,Yanji,Jilin 133200,China;School of Forestry and Landscape Architecture,Xinjiang Agricultural University,Wulumuqi,Xinjiang 830052,China)

机构地区：[1]北京林业大学森林资源和环境管理国家林草局重点实验室,北京100083 [2]广西林业勘测设计院,广西南宁530011 [3]吉林省汪清林业局,吉林延吉133200 [4]新疆农业大学林学与风景园林学院,新疆乌鲁木齐830052

出　　处：《森林与环境学报》2022年第3期289-296,共8页Journal of Forest and Environment

基　　金：中央高校基本科研业务费专项资金资助(BFUKF202102);国家自然科学基金项目(32160366)。

摘　　要：以吉林省汪清县金沟岭林场云冷杉针阔混交林为研究对象,基于研究区内均匀分布的98块样地的2487株单木数据,按7:3的比例随机抽样分别用于建模和检验。对林内云杉、冷杉、红松、中生阔叶树种组和慢生阔叶树种组这5个树种(组)的单木枝下高和冠幅模型分别采用最小二乘法估计、极大似然估计和似不相关回归估计3种参数估计方法进行建模,筛选效果最佳的为最终模型,并分析模拟了树木大小、竞争、林分和立地等因子对枝下高和冠幅的影响。同时考虑树种耐荫性和空间分布结构差异,讨论不同树种枝下高和冠幅模型中影响因子差异。结果表明:采用似不相关回归估计建立的枝下高和冠幅的联立方程组模型得到的预测效果最佳,模型检验的修正决定系数(R^(2)_(a))最大,均方根误差(E_(RMS))最小,且平均绝对误差(E_(MA))相较于最小二乘法估计和极大似然估计降低了57%~58%。胸径、树高和单木竞争指数在所有枝下高模型中均极显著(P<0.001);树冠竞争因子是影响红松枝下高的重要因子(P<0.001);林分密度在除了云杉的其余树种枝下高模型中均显著相关(P<0.01);立地因子中仅坡向与冷杉枝下高显著相关(P<0.01)。冠幅模型中,胸径、枝下高和树冠竞争因子在所有树种中均显著相关(P<0.001);此外冷杉和阔叶树种的冠幅则与林分密度显著负相关(P<0.001),而云杉冠幅则与林分断面积和显著负相关(P<0.001)。在云冷杉针阔混交林中,树种耐荫性和单木相对大小影响其树冠的衰退速率,而林地立地质量影响单木冠幅的大小。The research objects in this study were 2 487 individual trees from 98 sample plots in natural spruce-fir forests in Jingouling Forest Farm, Wangqing County, Jilin Province. A random sample of 70% of the data was used for model calibration, and the remaining 30% was used for model validation. Three parameter estimation methods such as least square method, maximum likelihood method, and seemingly unrelated regression method were used to model the height to crown base (HCB) and crown width (CW) of 5 tree species (groups) in forests, such as spruce and fir, Korean pine, and 2 deciduous groups. A model with the best performance was selected as the final model for a given species. In addition, the influence of tree species, tree size, competition, and site on the model of HCB and CW was determined. Considering the differences in shade tolerance and spatial distribution structure of tree species, the differences in the influencing factors of HCB and CW models of different tree species were discussed. The results showed that the prediction effect of the simultaneous equation group model of HCB and CW established by the seemingly unrelated estimation method was the best, its adjusted determination coefficient and root mean square error of the model test were the best, and the mean absolute error was reduced by 57%-58% compared with that obtained from the least squares method and maximum likelihood estimation methods. Initial diameter at breast height (DBH), tree height, and competition index were found to be highly significant (P<0.001) in all the HCB models. Crown competition factor (CCF) was highly significant (P<0.001) in the HCB models of Korean pine. Stand density had significant effects (P<0.01) on the HCB of tree except for spruce. Site factors showed significant effects (P<0.01) on fir HCB models. DBH, HCB, and CCF were highly significant (P<0.001) in all the CW models of 5 tree species. Stand density was highly significantt (P<0.001) in the models of fir and 2 deciduous groups, but Spruce CW was negatively corre

关 键 词：云冷杉针阔混交林 枝下高 冠幅 竞争指标 似不相关回归估计 

分 类 号：S757[农业科学—森林经理学]