The mechanisms explaining tree species richness and composition are convergent in a megadiverse hotspot

Understanding the drivers of species' geographic distribution and assembly of communities is one of the most intriguing questions in ecology and has become extremely important in face of global changes. This study aims to assess broad-scale patterns of tree species diversity and compare the importance of environmental correlates relative to purely spatial factors in structuring communities in the Brazilian Atlantic Forest. We compiled abundance data from 2948 species in 117 localities across the entire biome and obtained climatic and soil variables that we hypothesized to be environmental filters. We constructed correlograms for species richness and composition and assessed the relative importance of correlates using spatially explicit generalized additive models coupled with information-theoretic analyses. Variation partitioning was used to infer the relative importance of environmental- and spatial-based hypotheses. Species richness presented positive spatial correlation of similar to 435 km beyond which it became negative and again positive at the furthest distances, whereas compositional turnover showed positive correlation of similar to 435-555 km with decreasing similarity in increasing distances. Climatic variables were important factors correlating with richness patterns, with soil being less important. Species composition was significantly correlated with environmental variables and spatial constraints. The spatial component showed similar amount of explanation in statistical models compared to the environmental component. Our results suggest a combined contribution of environment, stochasticity, and historical-dispersion processes in patterns of biodiversity. More interestingly, a similar set of variables was the better correlates of species richness and composition, indicating convergence of driving factors for both descriptors of plant communities.