A trait based framework to predict community shifts in response to changing disturbance regimes

Background/Question/Methods

Functional traits have been promoted as an organizing principal on which to build predictive community models, and several traits-based frameworks have been proposed to explain community structure using traits. In particular, Bayesian hierarchical methods have been advocated because of their ability to naturally incorporate the complex hierarchical relationships that are inherent in natural communities. However, application of these approaches to predict the dynamics of natural plant communities from the traits of their constituent species in a variable environment is largely missing.

Results/Conclusions

We use a Bayesian hierarchical model to show that the interaction between a single trait and three environmental drivers explains more than 84% of the variation in community functional diversity for a naturally assembled North American tall grass prairie over a 25-year period, as well as explaining more than 76% of the variation in relative cover for each of the 4 most abundant species. Further, our results identify factors of niche differentiation that primarily drive changes in community structure at broad spatial and temporal scales.