秦岭火地塘林区典型灌木生物量估算模型1）

摘要

以秦岭火地塘林区美丽胡枝子（ Lespedeza formosa）、小叶六道木（ Abelia parvifolia）、毛榛（ Corylus mandshurica）和三桠乌药（Lindera obtusiloba）4种典型灌木为研究对象，通过灌木生物量及其生长指标调查，建立了灌木基径（ D）、冠幅（ W）、基径平方和树高乘积（ D2 H）与其生物量之间的回归模型，并对回归模型的精度进行了检验，利用已建立的回归模型估算了秦岭火地塘林区典型灌木的生物量。结果表明：灌木生物量最优模型以三次多项式为主，少数为二次多项式和幂函数；根、茎和叶生物量最优模型采用的自变量分别为D、D2 H和W，全株生物量最优模型采用的自变量为D或D2H；回归模型的决定性系数（R2）大多数高于0．90，估计精度均在0．91以上，相对误差均小于20％，均方根误差均小于3．00；通过生物量最优模型估算得到秦岭火地塘林区典型灌木生物量为5017．403 kg· hm-2。%With four typical shrub species of Lespedeza formosa, Abelia parvifolia, Corylus mandshurica and Lindera obtusiloba at the Huoditang forest region in Qinling Mountain , we established a best-fit regression equations by using diameter at 10 cm height (D), crown (W) and D2H as independent variables.The best-fit models were applied to estimate the biomass of understory shrub of oak-pine mixed forest in Qinling Mountain.The majority of the best-fit regression equations are cubic equations , while the others are quadratic polynomial equations and power functions .D, D2 H and W were used as the inde-pendent variables of the best-fit regression equations for root, stem and leaf, respectively.The best-fit regression equations of the whole plant were built with D or D2 H as independent variables.The determination coefficient is higher than 0.90, the estimation accuracy is higher than 0.91, the relative error is not more than 20%and the root mean square error is less than 3.00.The biomass of shrubs is 5 017.403 kg· hm-2 by the best-fit regression equations in the Huoditang forest region in Qinling Mountain .