University of Toronto, Department of Ecology & Evolutionary Biology Toronto, Ontario, Canada
Background/Question/Methods
Modern coexistence theory (MCT; Chesson, 2000) is increasingly appreciated as a framework with both theoretical and empirical utility in understanding species coexistence. Quantitatively formulated around the invasion criterion, MCT frequently partitions the invasion criterion into fitness differences (competitive inequivalence) and stabilizing differences (relative strength of species interactions). This partitioning allows simple analytical and graphical prediction of coexistence outcomes through MCT. However, it is unclear whether stabilizing and fitness differences summarize comparable outcomes across ecological communities and, in particular, how robust the coexistence predictions are in the presence of demographic stochasticity. Demographic stochasticity is expected to be important in small communities, which occur naturally in heterogenous landscapes and are increasingly common with habitat fragmentation. We used a simulation approach with two competition models which are commonly fit to empirical data, and asked how demographic stochasticity influenced coexistence outcomes across a range of stabilizing and fitness differences when communities differed in total size (number of individuals) and in their intrinsic growth rates.
Results/Conclusions
Stabilizing and fitness differences qualitatively predicted species coexistence fairly accurately across communities. However, communities varied greatly in the likelihood of coexistence even given identical stabilizing and fitness differences. This variation was explained by community size and finite rates of increase; when either of these values was low, coexistence was less likely.
Demographic stochasticity was detrimental for the invading species, although in small communities its negative effects on the invader were partially offset by fluctuations in the resident. In particular, demographic stochasticity skewed the equilibrium distribution of the resident population size, which reduced the average competitive effect.
Our results show that the predictive utility of stabilizing and fitness differences is vulnerable to variation in species growth rates and community sizes. Further, the importance of demographic stochasticity in nature may be mediated by competitive ability and order of species arrival, in addition to classically studied factors such as habitat size. Nonetheless, low density growth rates provided robust predictions of coexistence outcomes, suggesting that methods comparing these rates directly will produce more reliable coexistence predictions. These results demonstrate which ecological and demographic characteristics influence the applicability of coexistence theory, and deepen our understanding of coexistence and community assembly processes.
