不同龄组华北落叶松人工林径向生长模型构建

The Diameter at Breast Height and Age Growth Model of Larix principis-rupprechtii Plantations with Different Age Groups

本文以塞罕坝不同龄组华北落叶松人工林为研究对象,以非线性单木胸径生长模型为基础模型,考虑主要立地因子及林分密度对不同发育阶段华北落叶松胸径生长量的影响,利用哑变量方法构建包含不同龄组林分竞争指数及主要立地因子的非线性混合效应胸径生长模型。结果表明:由立地因子与立木胸径生长量进行相关性分析,确定每公顷株数、海拔、坡向、土层厚度是影响华北落叶松人工林胸径生长主要限制性因子,相关系数分别为:0.67、0.51、0.54、0.38;在不同龄组由于经营水平和作业方式不同,林分平均竞争指数与龄组呈负相关关系,不同龄组林分平均竞争指数分别为:6.26、2.17、0.61;以不同龄组林分平均竞争指数为哑变量,构建了包含影响立木胸径生长主要因子的非线性混合效应模型,不同龄组模型BIC值分别为-142.3、-109.4、-94.7;AIC值分别为:-145.8、-129.5、-113.6;-2 Log Likelihood值分别为:-463.8、-147.5、-131.6。在不同立地条件下,混合效应模型中固定效应参数能够反映大多数立木胸径生长量随年龄的变化总趋势,随机效应参数能解释不同立木年龄和胸径生长曲线的变异。因此,非线性混合效应模型提高了对胸径生长量的预测精度及适用范围。

This paper studied the Larix principis-rupprechtii plantations of different age groups and determined the nonlinear single tree diameter growth model as the basic model.Considering the influence of the main site factors and density on diameter growth of Larix principis-rupprechtii at different stages of development,a nonlinear mixed effects model of different age groups was established by using dummy variable method,including the competition index and main site factors of different age groups.The results showed the main limiting factors of the growth were density,elevation,slope and soil thickness by correlation analysis of site factors and tree DBH growth.The correlation coefficients were0.67,0.51,0.54,0.38;The competition index of the forest was negatively correlated with the age groups due to the difference of operation and management level.The average competition index of the different age groups were6.26,2.17,0.61;the nonlinear mixed effects model of different age groups including main site factors was established by dummy variables.The BIC value of different age group model were-142.3,-109.4,-94.7for BIC;-145.8,-129.5,-113.6for AIC;-463.8,-147.5,-131.6for-2Log Likelihood.Under different site conditions,fixed effects parameters can reflect most tree diameter growth with age change trend and the random effect parameters can explain the variation of the growth curve of different tree age.Therefore,the nonlinear mixed effects model improved the prediction accuracy and applicable range of diameter growth.