Exploring the ecological role of native and introduced plant species through network and phylogenetic analyses

Understanding community dynamics is a key challenge in ecology. Close relatives face challenges to coexistence: ecological similarity due to shared ancestry may lead to strong competition and gene flow may lead to species collapse. Habitat shifts may maintain coexistence in plants through induction of phenological shifts and increased opportunity for reproductive isolation. The first chapter of this dissertation explores whether flowering time is a conserved trait, and whether habitat segregation is associated with divergence in flowering time. There was significant phylogenetic signal in flowering time and habitat use. Habitat and flowering time shifts were strongly correlated, even after accounting for phylogenetic non-independence. After accounting for habitat, there was no remaining phylogenetic signal in flowering time. These results suggest that broad patterns of floral phenology may be driven by niche conservatism in habitat use, which in turn has a large influence on flowering time. Correlated habitat flowering time shifts like those observed here may promote reproductive isolation of close relatives with sympatric ranges.;The impact of introduced species on community dynamics and local biodiversity has been the subject of much debate. However, the effect of introduced species on higher trophic levels, such as herbivores, arthropod predators, and pollinators has received relatively little attention. The second chapter of this dissertation explores whether native and introduced composites play similar ecological roles by asking whether native composites host more diverse herbivore, arthropod predator, and pollinator communities, and whether native species experience greater costs or benefits from their associated arthropod communities than introduced species. Native plants host more diverse herbivore communities than do introduced species, but not more diverse predator or pollinator communities. Native plant species also experience greater seed damage but not less pollen limitation than introduced plant species. These results suggest that native and introduced species do not play similar ecological roles.;Species that play major roles in community dynamics are often the targets of conservation efforts. Network tools may be one method of identifying key species, and may allow comparisons of plant species' roles in different trophic interactions. Additionally, phylogenetic patterns in network structure may provide a way of predicting patterns of species interactions. Chapter three of this dissertation looks at fifteen plant species in Asteraceae at Bodega Marine Reserve and the arthropod community associated with their floral structures. Network and phylogenetic approaches were used to compare structure in herbivore, predator, and pollinator networks. While all networks are found to be nested and modular, plants varied in their network role between interaction types, perhaps because different processes structure different networks. Close relatives are likely to interact with similar herbivore communities, but not similar predator or pollinator communities. While phylogeny affects patterns of interaction between plants and herbivores, plant phylogenetic relatedness does not predict the dependence of those arthropod species on each plant species. Key plant players for one arthropod community may not be key players for another, suggesting that taking into account multiple trophic interactions is necessary for identifying target species for conservation.