One of the oldest debates in ecology concerns the extent to which population dynamics are regulated by density-dependent factors, such as competition, versus density-independent factors, such as climate.
Answering this question is a prerequisite for predicting the effect of climate change on plant communities. However, current approaches often focus exclusively on the direct effects of climate, ignoring species interactions. To address this issue, we used three decades of demographic data on eight Kansas prairie perennial forb species to test the relative importance of climate (precipitation and temperature) and competition variables (abundance of key functional types) in explaining observed population dynamics. We fit a series of Bayesian hierarchical models describing survival and recruitment as a function of either climate or competition variables, or climate and competition variables together. Comparisons of residual deviance across these models show that for all but one species climate variables were better predictors of individual survival than competition variables. For the density of new recruits and total individuals, however, competition variables were as or more important than climate for three species. For most species, the best models included both climate and competition covariates. Our results emphasize that the persistence of plant populations will depend on both the direct effects of climate change and indirect effects mediated by shifts in community composition.