理解群落的动态变化是群落生态学的一个重要内容,但目前为止在较大尺度的森林群落中相关的研究还很少.基于本研究提出的一种群落动态的随机零模型,我们设计了一种度量群落动态变化的方法.在考虑群落内的物种组成、环境因子及空间因子可能对群落动态变化产生不同的影响的前提下,利用鼎湖山20 hm2大样地2次调查数据,采用方差分解分析了群落动态变化与以谱系结构为代表的物种组成、环境因子及空间因子3种因子之间的关系.结果表明,整体上空间因子是影响群落动态的主要因素.在小尺度上(〈30 m),环境因子有比物种组成更高的解释量;在更大尺度上,物种组成的解释量更高.随着空间尺度的增加,环境因子的解释量在40 m尺度达到最大(6.42%),增加到50 m后其解释力消失;与环境因子不同,物种组成对于群落动态的影响力从20 m的1.12%逐步增加到50 m的17.79%.研究发现,环境因子和群落的物种组成结构能够在不同的空间尺度上影响群落的动态变化.
Ecologists have been monitoring community dynamics worldwide with the goal of understanding the evolution of community ecology. However, few efforts have been made to explore the dynamics of a large-scale forest community in terms of phylogenetic structure and habitat conditions. This study was based on two successive census data from a 20-ha subtropical forest communities in Guangdong, China. Based on a null model, we obtained a temporal phylogenetic turnover under random conditions at four spatial scales by dividing the plot into different size of quadrats. The null model simulates the demographic processes occurring between census periods while also considering the variation in the ability of species to disperse and the variations in mortality among both species and size classes. Variation partitioning was employed to analyze the dynamics of the forest community and the effects of habitat conditions, phylogenetic structure, and spatial factors on the temporal turnover of quadrats at four spatial scales. The results showed that spatial factors played the most important role in regulating the temporal turnover. Habitat factors contributed more than phylogenetic structure to explaining the temporal turnover at smaller scales(〈30 m), but became less important at larger scales. As the scale increases from 20 to 50 m, the contribution of habitat reached its maximum of 6.64% at 40 m. However, the contribution by phylogenetic structure increased gradually from 1.12% at 20 m to 17.79% at 50 m. Our study suggests that community dynamics are influenced by both habitat conditions and phylogenetic structure of co-existent species at different spatial scales, which should provide new insights for our understanding of assembling rules of species coexistance.