Spatial turnover of species composition in ecological communities may be driven by both deterministic (niche-based) and stochastic (dispersal limitation and drift) processes. Species turnover is expected to be non-random with respect to species’ functional traits under deterministic processes but functionally random under stochastic processes. Examining patterns of functional turnover, or functional beta diversity, may therefore be key to determining the relative roles of deterministic and stochastic processes in community assembly. We combine a functional trait database with a continental-scale dataset on tree species composition to examine spatial patterns of species and functional beta diversity and their relationships with edaphic and climatic factors.
Results/Conclusions
Species and functional beta diversity between communities both increased continuously with increasing geographic distance up to a scale of 2500 km, with functional beta diversity exhibiting stronger spatial structure. Species and functional beta diversity were best explained by climatic dissimilarity, indicating a primary role of deterministic environmental filtering in community assembly. Using null models, we find that functional beta diversity was less than expected given the observed species turnover at small distances (< 600 km) and greater than expected at large distances (> 1900 km). These results indicate that species turnover is non-random with respect to functional traits, as predicted by deterministic assembly processes. This study provides insights into the spatial turnover of community composition up to continental scales and demonstrates the utility of patterns of functional beta diversity for understanding community assembly processes.