文献类型：期刊文献

中文题名：内蒙古大兴安岭兴安落叶松和白桦地上生物量分配特征及模型研究

英文题名：Aboveground Biomass Allocation Characteristics and Models of Larix gmelinii and Betula platyphylla in the Greater Khingan of Inner Mongolia

作者：曹元帅[1] 冯素静[2] 哈斯巴干[2] 王建军[3] 张亮[2]

第一作者：曹元帅

机构：[1]国家林业和草原局华东调查规划院,浙江杭州310019;[2]中国内蒙古森林工业集团有限责任公司,内蒙古牙克石022150;[3]中国林业科学研究院林业科技信息研究所,北京100091

年份：2025

卷号：48

期号：5

起止页码：68-78

中文期刊名：内蒙古林业调查设计

外文期刊名：Inner Mongolia Forestry Investigation and Design

基金：内蒙古森工集团科研项目(NXLKJ[2021]005)。

语种：中文

中文关键词：兴安落叶松;白桦;生物量分配模式;可加性生物量模型;似乎不相关回归

外文关键词：Larix gmelinii;Betula platyphylla;biomass allocation pattern;additive biomass model;Seemingly Unrelated Regression(SUR)

分类号：S718.5

摘要：探究内蒙古大兴安岭兴安落叶松(Larix gmelinii)和白桦(Betula platyphylla)地上生物量分配模式,建立两个树种的可加性地上生物量模型系统,有助于为该区域的碳储量变化精准监测提供理论依据。本研究以兴安落叶松和白桦为研究对象,基于收获法获取的生物量数据,分析两个树种地上部分各组分(树干、树枝和树叶)生物量的分配特征及其随林木大小的变化规律。以胸径(D)和树高(H)为预测变量,幂函数为树木生长方程式,确定两个树种各分量的最优生物量模型,使用似乎不相关回归法(SUR)建立可加性模型,并引入权函数消除模型的异方差性。此外,采用五折交叉验证法检验模型的拟合精度。结果表明:随着林木的生长,兴安落叶松和白桦地上部分生物量中树干生物量所占的比例逐渐上升,树枝生物量和树叶生物量所占比例逐渐下降,其中白桦树叶生物量比例变化不显著。各组分以D为预测变量的生物量模型拟合效果较好(落叶松的R^(2)_(adj)为0.914~0.995,白桦的R^(2)_(adj)为0.933~0.984)且模型系数均显著。加入H后的D^(2)H双变量模型系统提升了两个树种的树干生物量模型拟合效果,但树枝生物量和树叶生物量模型拟合效果却有所下降。SUR模型的引入使得各组分生物量模型之间具有可加性。林木大小是兴安落叶松和白桦生物量分配的重要因素,两个树种各组分生物量占比有所差异,但整体均呈现为树干>树枝>树叶。本研究构建的可加性生物量模型系统拟合优度较高,能为内蒙古大兴安岭地区森林碳储量的准确估算提供数据支撑。
Exploring the aboveground biomass allocation patterns of Larix gmelinii and Betula platyphylla in the Greater Khingan of Inner Mongolia,and establishing an additive aboveground biomass model system for the two tree species,are conducive to providing a theoretical basis for the accurate monitoring of carbon stock changes in this region.Taking Larix gmelinii and Betula platyphylla as the research objects,this study analyzed the allocation characteristics of biomass in aboveground components(trunk,branch,and leaf)of the two tree species and their variation rules with tree size,based on biomass data obtained by harvesting methods.With diameter at breast height(D)and tree height(H)as predictive variables,and the power function as the tree growth equation form,the optimal biomass model for each component of the two tree species was determined.The Seemingly Unrelated Regression(SUR)method was used to establish the additive model,and a weight function was introduced to eliminate the heteroscedasticity of the model.In addition,the five-fold cross-validation method was adopted to test the fitting accuracy of the model.The results showed that with the growth of trees,the proportion of trunk biomass in the aboveground biomass of Larix gmelinii and Betula platyphylla gradually increased,while the proportions of branch biomass and leaf biomass gradually decreased;among them,the change in the proportion of Betula platyphylla leaf biomass was not significant.The biomass models with D as the single predictive variable showed good fitting effects for all components(R^(2)_(adj) of Larix gmelinii ranged from 0.914 to 0.995,and R^(2)_(adj) of Betula platyphylla ranged from 0.933 to 0.984),and all model coefficients were significant.The introduction of the D^(2)H bivariate model system(adding H as a variable)improved the fitting effect of the trunk biomass model for both tree species,but reduced the fitting effect of the branch biomass and leaf biomass models.The application of the SUR model ensured the additivity among the biomass models of different components.Tree size is an important factor affecting the biomass allocation of Larix gmelinii and Betula platyphylla;although the proportion of each component biomass differs between the two tree species,the overall order is trunk>branch>leaf.The additive biomass model system constructed in this study has high goodness of fit,which can provide data support for the accurate estimation of forest carbon stocks in the Greater Khingan Range Area of Inner Mongolia.