包含密度因子的杉木可变指数削度方程研制

Studies on variable-density and variable-exponent taper equation for Cunninghamia lanceolata plantations

采用位于江西大岗山的杉木密度试验林中183株树干解析数据,以Kozak(2004)削度方程为原型,探讨了包含密度因子的杉木可变指数削度方程构建过程及效果。研究表明:1)b10/sd为最优的包含密度的指数表达形式;2)引入密度因子后,无论是非线性回归还是混合效应模型,削度方程的拟合精度均有所提高;3)应用混合效应模型对包含密度因子的削度方程进行预估时较非线性回归模型更优;4)构建的杉木可变密度可变指数削度方程在树干中部预估值的平均绝对偏差较低,而在树干基部和顶端的预测效果较差;5)树干削度随林分初植密度的增大而降低,降低的程度随着密度的增大而减少。

Trunk analysis data from 183 trees in Cunninghamia lanceolata stands with different planting densities at Dagangshan of Jiangxi Province,Kozak′s variable-exponent taper model was adopted to build the variable-exponent taper equations including density factor for C.lanceolata plantations.The results showed that the function b10/sdwas the best form while adding planting density to the exponent of Kozak′s taper model.After including density factor,the taper equation had higher modeling precision than taper equation without density factor,and the predictive accuracy of the model was improved by including random-effects parameters.The variable-density and variable-exponent taper equation built for C.lanceolata trees had low bias while predicting diameters at middle of trunk,but the prediction effect is relatively poor at the base and top of the trunk.The trees have more taper at lower than at higher stand density.However,the difference in bole diameter between trees at lower and higher stand densities diminishes as stand density increases.