Trait-divergence in plant community assembly is generated by environmental factor interactions

Abstract Question: I investigate the conditions driving trait divergence during community assembly through environmental filtering. Methods: I used an individual-based, stochastic, spatially explicit metacommunity simulation model to generate data on species traits and community composition. This was based on spatially autocorrelated, nested, feedback-driven environmental factors. Alpha trait divergence was quantified using the correlation r(RE) between Rao functional diversity and environmental factors. Beta trait divergence was assessed through the matrix correlation r(VE) between the local contribution (V) of each community to beta trait diversity and environmental factors (E). Permutation tests, assuming no relation between traits and species composition, established the significance of r(RE) and r(VE). Additionally, I applied the methodology to grassland and soil data collected in plots across southern Brazil. Both simulated and real data were analysed at two spatial resolutions. Results: Beta trait divergence was frequently observed when factor interactions were incorporated in community assembly simulations, while it remained mostly within expected Type I error limits when factor interactions were absent. Beta trait divergence was stronger than alpha trait divergence at a finer spatial resolution, and vice-versa when smaller community units were combined into larger units. Beta trait divergence for specific leaf area (SLA) showed a significant association with grassland soil variables, likely arising from observed interacting effects between soil variables and total vegetation cover. When small plots were aggregated into larger units, beta trait divergence disappeared, while alpha trait divergence increased. Conclusions: Trait divergence emerges during community assembly due to interactions between factors affecting the selection of individuals based on their traits. If these factors are spatially nested, including hidden, feedback-driven ones, trait divergence can manifest at the beta or alpha dimensions, depending on the scale of the community units. This provides a simpler and more general explanation than considering plant-to-plant positive or negative interactions, the effects of which may actually be mediated by environmental factors.Submitted for publication in the Journal of Vegetation Scienc