南京地区马尾松根径与树高非线性混合效应模型研建

Research on the Multivariate Nonlinear Mixed Effect Model of Different Ground Diameters and Heights of Pinus massoniana Lamb in Nanjing

基于南京地区56块20 m×20 m的方形样地的1 022株马尾松(Pinus massoniana Lamb)根径和树高数据,分别采用乘幂、对数和双曲线函数等一元非线性模型和多元非线性混合效应模型拟合了马尾松根径和树高的相关关系。结果显示,所有一元非线性模型的相关指数均小于0.5,适应性检验的系统误差均大于5%,平均误差均大于10%;多元非线性混合效应模型的相关指数为0.615 7,系统误差为4.58%,平均误差为9.07%。从而表明多元非线性混合效应模型可较为精确地拟合马尾松根径(d0)和树高的相关关系。

Related mathematical models of ground diameters and heights were established with monadic nonlinear models and multivariate nonlinear mixed effect models based on the data of 1022 Pinusmassoni- ana Lamb trees from 56 20m × 20m square plots. The correlation indexes of all monadic nonlinear models are less than 0. 5, systematic errors are more than 5 % and relative residual standard deviations are more than 10%. The correlation index, systematic errors and relative residual standard deviations were 0. 6157, 4. 58% and 9.07%. The results suggest that the relation between ground diameters and heights of Pinus massoniana Lamb trees can be precisely fitted with multivariate nonlinear mixed effect models.