论文部分内容阅读
树高是林木的主要生长指标之一,也是森林资源调查中的一个重要指标;林木树高生长往往可以反映其所在林地的质量。利用40株福建柏样木调查数据,基于以胸径为自变量的福建柏树高一元线性估测模型,引入枝下高与冠幅指标,建立了多元线性回归模型,并比较了二者的估测效果。结果表明:利用相同的建模数据拟合的福建柏树高多元线性回归模型的决定系数R~2较一元线性回归模型高0.0309,剩余标准差S低0.0612,估测精度高0.0047,说明引入同样对树高具有显著影响的枝下高和冠幅2个生长指标后,单独以胸径为自变量的福建柏树高线性回归模型得到了优化,估测结果的稳定性更强。
Tree height is one of the main growth indicators of forest trees, and it is also an important indicator in forest resources survey. The growth of tree height often reflects the quality of the forest land in which it is located. Based on the survey data of 40 strains of Picea phellodendron in Fujian Province, a linear regression model was established based on the linear regression model of Picea koraiensis with DBH as the independent variable. The multiple linear regression model was established by using the branch height and crown index. effect. The results showed that the coefficient of determination R2 of Fujian cypress tree multiple linear regression model fitted by the same modeling data was 0.0309 higher than that of the univariate linear regression model, the residual standard deviation S was 0.0612, and the estimation accuracy was 0.0047, indicating that the same Tree height had significant effects on branch height and crown width of two growth indexes, and the DBH model was used as independent variable to improve the high linear regression model of Cypress tree, the result of the estimation was more stable.