Advisor: Tony Ives
Phylogenetic signal in ecological community responses to environmental variation.
Chapter 1 Phylogenetic measures of biodiversity: We developed a theoretical framework based on phylogenetic comparative methods to integrate phylogeny into three measures of biodiversity: species variability, richness and evenness. With these metrics we demonstrated that north-temperate lake fish communities comprised more closely related species than expected by chance. This pattern of phylogenetic community structure was driven by low pH lakes that contain clades of acid-tolerant fishes.
Chapter 2 Separating the determinants of phylogenetic community structure: The role of competition in forbidding similar species from co-occurring has long been debated. It is difficult to identify this repulsion of similar species since similar species share similar environmental requirements and hence show an attraction to communities where the requirements are met. To disentangle these opposing patterns, we used phylogeny as a metric of species similarities. Studying 11 sunfishes from 890 lakes, we first showed no phylogenetic structure in the community data. We then regressed sunfish presence/absence on environmental variables. Lakes of similar environment contained closely related fishes. After statistically removing the environmental effects, phylogenetic repulsion was apparent, with closely related fishes less likely to co-occur. Thus, both phylogenetic attraction, driven by environmental filtering, and phylogenetic repulsion, possibly caused by competition, simultaneously occurred and obscured one another in the overall phylogenetic structure of sunfish communities.
Chapter 3 Communities contain closely related species during ecosystem disturbance: Since closely related species likely share similar trait values they should have similar sensitivities to environmental disturbance. We hypothesized that communities undergoing disturbance comprise more-closely related species than undisturbed communities. We tested this hypothesis with zooplankton community data from 18 disturbed lakes and found that communities did indeed contain more-closely related species during, than before, disturbance. The average relatedness of species in reference lake communities did not change over time. Community change was greatest in disturbed lakes that were acidic, small and deep. We thus demonstrated that disturbance may skew the phylogenetic component of biodiversity, with the response dependent on the environmental context of the affected ecosystem.