| Elucidating the mechanisms underlying plant community assembly is a keystone of community ecology, ecosystem ecology and conservation ecology. Environmental filtering (EF) and negative density dependence (NDD) shape community assembly, but their relative importance is poorly understood. Furthermore, whole life cycle studies on community assembly are still lacking, owing to long life spans of most trees. Nevertheless, spatial structure can provide insight into the long-term effects of community assembly processes. In this dissertation, we provided population- and community-level studies on spatial structure to unravel the relative importance of EF and NDD across different life-history stages.A conceptual framework was proposed at the population scale based on effective gene dispersal and population spatial genetic structure (SGS), which argues that the SGS of a cohort can be deduced exactly from the effective gene dispersal kernel as long as the population is at drift-dispersal equilibrium, and thereby that the transition of SGS between two life stages reflects the ecological and biological processes underlying the difference between effective gene dispersal patterns of these stages. We applied this framework to the population of Cyclobalanopsis glauca on Dajinshan Island, China. In this population, the average of effective gene dispersal distance estimated by SSR-based parentage analysis and by spatial autocorrelation was 131.0 m and 129.7 m, respectively. Seedlings were simulated by individual-based spatial explicit models based on the effective pollen and seed dispersal kernels estimated by parentage analysis. There was no significant difference between the observed and simulated SGS of seedlings, which could be predicted by the framework. Furthermore, adults were simulated under scenarios of EF, NDD and neutral processes, respectively. The simulated SGS of adults was inconsistent with the observed pattern unless EF dominated, indicating the dominant role of EF during the seedling-to-adult transition in this population.To test the community-level consequences of NDD and EF across different life-history stages, the nearest-neighbor pairs (NNPs) were analyzed in a tropical rain forest (BCI, Panama) and a subtropical evergreen broad-leaved forest (Tiantong, China). The proportion of conspecific NNPs was higher among individuals with 1cm ≤ DBH≤10 cm than between individuals with 1cm≤ DBH≤10 cm and those with DBH> 10 cm in both communities, indicating the dominant role of NDD during early life-history stages. Moreover, NDD did not result in a community compensatory trend, but shaped species abundances in local communities, as suggested by the significantly negative relationship between the strength of NDD effects and species abundance. The proportion of conspecific NNPs was higher among individuals with DBH≥ 10 cm than between individuals with 1cm≤DBH<10 cm and individuals with DBH≥ 1 cm, suggesting the dominant role of EF during late life-history stages in both communities.Given that niche can be phylogenetically conserved, i.e. close relatives are more similar in niche, both NDD and EF can shape community phylogenetic structure. To investigate the phylogenetic pattern of NNPs in the Tiantong forest, we sequenced 366 samples from 152 co-occurring woody plant species at two DNA barcode loci (rbcL+matK), and reconstructed well-resolved community phylogenies, from which genetic distances among species could be estimated. The average of genetic distances within NNPs was higher than expected under neutral patterns for individuals with 1cm≤ DBH≤10 cm, whereas the average of genetic distances within NNPs did not show any significant deviation from expectations for individuals with DBH≥ 10 cm. These results indicated that, phylogenetic repulsion caused by NDD, and phylogenetic attraction driven by EF, occurred during the early and late life stages respectively, and obscured one another in the raw phylogenetic structure of adults.In conclusion, our population, community and phylogenetic studies highlight NDD and EF acting at different temporal scales in shaping community structure. The mechanisms underlying community assembly, as well as our simulations of SGS, are more than academic questions, but also have critical implications for conservation and restoration, especially when communities disassemble following perturbations such as climate change and habitat fragmentation. In addition, our studies are attempting, as well as calling on other ecologists and biologists, to bridge different ecological disciplines and thereby provide potentially novel perspectives and insights into ecology. |
PDF Full Text Request