Phylogeny-coexistence relationships in annual plant communities: disentangling the role of species interactions, environment, and biogeography

Background/Question/Methods

Evolution not only shapes the form and function of all species, but also how they interact with each other and their environments. Because closely related species are often ecologically similar, theory predicts that competitive exclusion occurs more frequently between recently diverged species. In nature, however, these patterns are often weak at best or are not well-explained by underlying trait variation, in part because the confounding effects of multiple processes cannot be teased apart using field-based observational data. Here, we combine mathematical and experimental approaches to explicitly test key hypotheses on the ecological implications of evolutionary processes, including the role of species interactions (niche vs. fitness differences), biogeography (sympatric vs. allopatric evolution), and the environment. We tested these ideas using a greenhouse experiment with Mediterranean annual plant species that are native to California [n = 20] or Spain [n = 10]. To estimate niche differences, species were grown in pairwise combinations of differing phylogenetic distances at six relative frequencies. By contrast, fitness differences are not frequency dependent, and were estimated by comparing the seed production of species when grown without competitors. This experiment was replicated across two soil moisture environments, to evaluate the role of local conditions on shaping phylogeny-coexistence relationships.

Results/Conclusions

Our preliminary results with nine of 30 species suggest that close relatives do compete more intensely, as predicted by theoretical models. In all cases, niche differences increased with phylogenetic distance, with switching from near-neutral (0.10 ± 0.02; mean interaction slope ± SE) to niche-based dynamics (0.42 ± 0.06) as communities shifted from congeneric to conordinal species pairs [Note: an interaction slope of 0 indicates that frequency-dependence is absent]. These relationship, however, were much stronger when species pairs evolved sympatrically, with seed production decreasing by 22 to 65% when Californian species were paired with allopatrically- versus sympatrically-evolving species. The effect of soil moisture on species interactions, by contrast, was weak compared to the effects of phylogenetic relationships and biogeography. When grown in the absence of competition, fitness differences varied five orders of magnitude across species, with evidence of phylogenetic conservatism only between sympatrically evolving species. In sum, our work to date demonstrates that both niche and fitness differences are phylogenetically conserved across species, leading to the opposing predictions that phylogenetic relatedness both promotes and precludes species coexistence. However, we will use these results to parameterize annual plant models to definitively test the influence of phylogenetic relationships on coexistence outcomes.