Although stand age affects biomass partitioning and allometric equations, the size of these effects and whether it is worth incorporating stand age into allometric equations, requires further attention. We sampled a total of 90 trees for 10 Chinese fir (Cunninghamia lanceolata) plantations at seven stand age classes to obtain the data of tree component biomass using destructive harvesting. A multilevel modeling approach was applied to examine how stand age effects differ among tree components and predictor variables (diameter at breast height, DBH and tree height, H). Age class-specific allometric equations and the best fitting generalized equation that included stand age as a complementary variable were developed for each tree component. Large differences in both the intercept and slope for different stand age classes indicated that stand age affected allometric models. Branch and leaves were more sensitive to the environment and were the tree components most affected by stand age. Age class-specific allometric equations fitted well (R2 > 0.65, p < 0.001) using DBH and the combined form DBH2H as predictor variables. Including stand age as a complementary variable improved the fit of generalized allometric equations. Stem, aboveground and total tree biomass predicted by the multilevel model and generalized equation were comparable to the observed data. However, the multilevel model and generalized equations had a relatively low predictive capacity for branch, leaf and root biomass. These results could improve our capacity to evaluate carbon sequestration and other ecosystem functions in plantations.

中文翻译：

---

林龄对杉木人工林生物量分配及异速生长方程的影响

尽管林龄影响生物量分配和异速生长方程，但这些影响的大小以及是否值得将林龄纳入异速生长方程需要进一步关注。我们对 10 个杉木 (Cunninghamia lanceolata) 人工林的 7 个林龄等级的 90 棵树木进行了采样，以使用破坏性采伐获得树木成分生物量的数据。应用多级建模方法来检查林龄效应在树木组成部分和预测变量（胸高直径、DBH 和树木高度、H）之间有何不同。为每个树木组件开发了特定于年龄等级的异速生长方程和最佳拟合广义方程，其中包括林龄作为补充变量。不同林龄等级的截距和斜率的巨大差异表明林龄影响异速生长模型。枝叶对环境更敏感，是受林龄影响最大的树木成分。使用 DBH 和组合形式 DBH2H 作为预测变量，年龄等级特定的异速生长方程拟合得很好（R2 > 0.65，p < 0.001）。包括林龄作为补充变量改善了广义异速生长方程的拟合。由多级模型和广义方程预测的茎、地上和总树木生物量与观测数据相当。然而，多级模型和广义方程对枝、叶和根生物量的预测能力相对较低。