Patterns of community assembly and the relative roles of space, environment, phylogeny, and disturbance in explaining functional diversity in a temperate forest

Background/Question/Methods

Determining the mechanisms underlying the assembly of diverse communities continues to be a central goal of ecology. While taxonomic diversity constitute the basis of most research in plant community ecology and are key to understanding biodiversity assembly mechanisms, the growing interest in using functional and phylogenetic diversity to understand community assembly focuses on the ecological roles and evolutionary relationships of species, additionally capturing community-level changes along relevant ecological axes. This can be especially helpful in the context of communities recovering from disturbance. In this study, we aimed to discern whether patterns of functional and phylogenetic assembly and turnover differed from a random expectation in a temperate mixed mesophytic tree community. The study area (Powdermill Nature Reserve) of 485 blocks of 120mx120m, encompassed different forest ages and disturbance types within a continuous region of ~900ha.  Further, we analyzed if the variation in functional alpha and beta diversity was best explained by space, environment, disturbance and land-use history, phylogenetic diversity, or a combination of these variables. For four continuous (seed mass, maximum height, rooting depth, and wood density) and two categorical traits (C:N ratio and N fixation), we used standardized effect sizes derived from null models for two indices of functional diversity – Mean Pairwise (SES PW) and Mean Nearest Neighbor (SES NN) metrics to evaluate assembly patterns and conduct variance partitioning. 

Results/Conclusions

Results suggest deterministic, non-random patterns of species co-occurrence. Functional composition was largely clustered for SES PW_alpha, for traits considered together or individually (21-91% of communities), suggesting habitat filtering winnowing broad functional subsets. SES NN_alpha patterns (terminal functional groups) were less pronounced. Functional turnover showed a trend towards clustering, although <5% of results were statistically significant. For phylogenetic alpha dispersion, 60% of blocks were over dispersed, but only 3.5% were significant. For beta functional dispersion, < 1% of results were significant. Variance partitioning showed that space and environment explained the largest proportion of variation in SES PW_alpha, for all traits combined (22%) and for individual traits (12-18%), followed by environment (3-8%). Results were similar for SES NN_alpha, but lower variance was explained. Beta diversity was best explained by phylogenetic turnover in all cases (SES PW_beta– 19-62%; SES NN_beta– 24-77%). Disturbance explained at most 1% variance. Results imply that assembly of basal functional groups is non-random, but close relatives that are functionally very similar are unlikely to co-exist. Dispersal limitation may account for the role of phylogeny in explaining beta dispersion.