Coexistence in two tradeoff models: Limits to similarity arise when competitive ability or stress tolerance varies smoothly with seed size

Background/Question/Methods

A key goal in ecology is to identify the mechanisms maintaining trait diversity in communities. Ecologists aim to understand the influence of competition on community structure, and in particular the expected patterns in the distribution of traits along tradeoff axes that could enable coexistence. Most competition models lead to patterns of spacing in the distribution of species along niche axes, especially when the potential for environmental perturbations is considered. In this context, we study two models of competition and coexistence proposing specific tradeoffs as a possible mechanism underlying diversity in seed size in plant communities. The hierarchical competition-colonization model postulates a tradeoff between fecundity and ability to displace competitors, whereas the tolerance-fecundity model proposes a tradeoff between fecundity and tolerance to suboptimal environmental conditions. Though applied here to seed size, both models can in principle also describe diversity of other traits in plant, animal and microbial communities. Patterns of coexistence in these models have not been studied in depth, and questions of robustness in either model have not yet been addressed.

Results/Conclusions

We extend the models from a discrete set of fecundity values to a continuous range, and solve them to obtain the conditions on the relationship between mortality and fecundity or tolerance and fecundity leading to coexistence of species arbitrarily close in fecundity, i.e., tight packing of species along the fecundity axis. We assess the robustness of such densely populated axes to small changes on the mortality-fecundity or tolerance-fecundity relationship. Surprisingly, tight packing proves more robust than one might expect from classic models of competition, such as MacArthur and Levins' overlap model. Recent work, also to be presented at this conference, suggests that this may be due to these models' perhaps unrealistic assumptions of sharp transitions in competitive ability or stress tolerance with fecundity. In light of this, we also present modified versions of these models where competitive ability or stress tolerance varies smoothly with fecundity (or seed size), and repeat the aforementioned analysis on the new versions. We found that tight packing cannot be produced in the modified models.