摘要: |
通过AIC、BIC准则结合R2选择不同恢复阶段的最佳模型,用校正系数估计值的标准误差、平均系统误差和总相对误差值评价所建立模型的精确性,并用所构建的生物量模型对檵木群落不同恢复阶段群落地上生物量和檵木地上生物量的分配格局进行分析。结果表明:(1)运用5种模型进行回归分析,不同恢复阶段树干和地上生物量估测效果极佳,叶和枝生物量次之。选用模型Ⅳ:W= a+b(D2H)建立乔灌阶段地上生物量的最优模型,选用模型Ⅲ:W= a215;Db215;Hc 建立灌木阶段和小乔林阶段地上生物量的最优模型。(2)利用建立的檵木群落不同恢复阶段立木生物量生长模型对檵木群落的植被生物量进行估算,不同恢复阶段干生物量和地上生物量大小排列顺序为小乔林阶段>乔灌阶段>灌木阶段; 叶生物量和枝生物量大小排列顺序为乔灌阶段>小乔林阶段>灌木阶段。(3)檵木群落中作为建群种的檵木,其地上生物量大小排列顺序为乔灌阶段>小乔林阶段>灌木阶段,檵木生物量在檵木群落不同恢复阶段的群落生物量中所占比例在持续下降。这说明随着桂林岩溶石山檵木群落自然恢复演替的进行,生态系统运行的能量基础和营养物质来源随着群落向更高级的演替阶段发展,而檵木的建群种地位可能会被逐步替代而退居亚乔木层。 |
关键词: 檵木群落, 地上生物量, 估测模型, 不同恢复阶段 |
DOI:10.11931/guihaia.gxzw201801010 |
分类号:Q948.1 |
文章编号:1000-3142(2019)02-0161-09 |
Fund project:国家自然科学基金(31660197,31160156); 广西研究生教育创新计划项目(XYCSZ2018054)[Supported by the National Natural Science Foundation of China(31660197,31160156); Innovation Program of Guangxi Graduate Education(XYCSZ2018054)]。 |
|
Model construction and its above-ground biomass allocation at different restoration stages of Loropetalum chinense communities in karst hills of Guilin |
ZHANG Yajun1, MA Jiangming1, 2, 3*, SU Jing1, QIN Jiashuang1, MO Yanhua1
|
1. College of Life Sciences, Guangxi Normal University, Guilin 541006, Guangxi, China;2. Key Laboratory of Ecology of Rare and Endangered
Species and Environmental Protection, Ministry of Education, Guilin 541006, Guangxi, China;3. University Key Laboratory of Karst Ecology and
Environmental Change of Guangxi Zhuang Automomous Region(Guangxi Normal University), Guilin 541004, Guangxi, China
|
Abstract: |
The best model was selected by combining AIC and BIC criterions with R2, simultaneously, the accuracy of the estimation models of above-ground biomass were assessed using the correction coefficient, including standard error of estimated value(SEE), mean systematic error(MSE)and total relative error(TRE), and the distribution pattern of above-ground biomass of Loropetalum chinense and community in different restoration stages of L. chinense communities were analyzed with the established biomass model. The results were as follows:(1)Five models were used for regression analysis, above-ground biomass and trunk biomass had the best estimate effect, comparing with a lower estimate effect in leaf biomass and branch biomass. We chose Model Ⅳ to establish the optimum model for above-ground biomass of shrub to tree stage, Model Ⅳ was W= a+b(D2H). And we chose Model Ⅲ to establish the optimum model for above-ground biomass of shrub stage and smal tree stage, Model Ⅲ was W= a215;Db215;Hc.(2)Based the established optimum model, we estimated that biomass above-ground on the vegetation of L. chinense communities. Among different restoration stages, the leaf biomass and branch biomass order of size was shrub to tree stage > small tree stage > shrub stage, while the above-ground biomass and trunk biomass order of size was small tree stage > shrub to tree stage > shrub stage.(3)L. chinense as dominant species in L. chinense communities, the order of its above-ground biomass size was shrub to tree stage > small tree stage > shrub stage, and the above-ground biomass of L. chinense contributed to the above-ground biomass of different restoration stages showed a decreasing trend across the succession. This illustrates the energy base and nutrient source of ecosystem operation is progressing toward with the community moves to a higher stage of succession, and the status of edificator of L. chinense in different restoration stages of L. chinense communities may be replaced step by step, and backseat to the sub-tree layer. |
Key words: Loropetalum chinense communities, above-ground biomass, estimation models, different restoration stages |