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

Question

What conditions drive trait divergence during community assembly through environmental filtering, and why are some communities more trait-diverse than others?

Methods

An individual-based, stochastic, spatially explicit metacommunity simulation model produced data on species traits, spatially autocorrelated, nested, feedback-generated environmental factors, and resulting community composition. I quantified environmentally driven alpha trait divergence using the correlation r(RE) to measure the relationship between Rao functional diversity and environmental factors. Environmentally driven beta trait divergence was assessed through the correlation r(VE), involving environmental factors and the squared residuals (V) of a second-order polynomial regression of community-weighted means on environmental factors (E). Permutation tests, assuming independence between traits and species composition, were used to establish the significance of r(RE) and r(VE). Additionally, the method was applied to grassland and soil data collected in plots across southern Brazil. Both simulated and real data were analysed at two spatial resolutions.

Results

Significant r(VE) correlations were frequent with factor interactions incorporated in community assembly simulations, while r(VE) correlations mostly remained within expected Type I error range when factor interactions were absent. r(VE) was stronger than r(RE) at a finer spatial resolution and weaker than r(RE) when smaller community units were combined into larger units. r(VE) for specific leaf area (SLA) was related to soil variables, likely due to their interacting effects with total vegetation cover. When small plots were aggregated into larger units, r(VE) became non-significant, while r(RE) increased.

Conclusions

Environmentally driven trait divergence emerges during community assembly due to interactions between factors affecting the selection of individuals based on their traits. When the effects of these factors are spatially nested, including hidden, feedback-generated ones, trait divergence arises at the beta or alpha dimension, depending on the scale of the community units. This suggests that plant-to-plant positive or negative interactions, which can feedback on environmental factors and generate heterogeneity, do not necessarily lead to trait divergence if these factors do not interact.