【摘要】 基于广东省樟树、木荷和枫香3个阔叶树种共270株样木的生物量实测数据,采用多元非线性联合估计,按分量相加和总量控制两种方法,分别构建了一元(胸径)和二元(胸径、树高和胸径、冠幅)的相容性生物量模型,比较分析了这两种相容性生物量模型的拟合效果和预估精度。结果表明:(1)分量相加和总量控制构建相容性模型对3个树种拟合效果均较好,预估精度均较高,二者之间差别不大,但总体上分量相加模型拟合效果较优。(2)二元模型拟合效果和预估精度普遍优于一元模型,但加入第二个变量的异同,在各个分量中的表现不一。加入树高因子,显著改进树干和树皮的拟合效果;加入冠幅因子,显著改进树枝、树叶和地上部分的拟合效果。选择胸径和冠幅因子建立二元生物量模型可以获得较高的拟合精度。(3)基于最优独立模型构建的相容性模型拟合和预估效果最优,地上部分总量甚至优于最优独立模型。因此,各分量相加构建的相容性模型略优于总量控制。对于阔叶树种自变量的选取,二元模型加入冠幅因子的精度高于加入树高因子。更多还原

【Abstract】 Based on biomass data of 270 individuals of three broadleaved tree species in Guangdong,Cinnamomum camphora,Schima superba,and Liquidambar formosana,a multiple nonlinear joint estimation was adopted with components addition and total control methods to separately establish one-variable( diameter at breast height,DBH) and bivariate( DBH and tree height or DBH and canopy width) compatible biomass equations. The fitting goodness and estimated accuracy of compatible equations derived from these two methods were evaluated. The results showed that:( 1) The compatible equations separately established by components addition and total control methods performed well,with no significant difference in estimated accuracy between them.The model of components addition performed slightly better.( 2) Binary equations generally performed better than the unitary equation. Adding the second variable could produce different impacts for tree components. With tree height as the second variable,the binary model improved the fitting effect of stem-wood and stem-bark biomass. With the canopy width as the second variable,the model improved the fitting effect of branch,foliage,and aboveground biomass. For the compatible equation of broadleaved species,using the DBH and canopy width of trees could get better accuracy.( 3) Compatible model based on the second variable respectively from components model performed better,even with higher accuracy for total biomass estimation than individual model. The components addition was slightly better than total control in constructing compatible model. For the selection of second explanatory variable of broadleaved tree species,the binary model accuracy with canopy width factor was higher than that with tree height factor.更多还原