Effects of Habitat Restoration on Wild Bee Communities.

摘要

Declines of honeybees and certain native bees have been reported over the past decades. To avoid losses of bee diversity and pollination services, bee conservation and restoration ecology must find common ground. The current approach to bee conservation consists largely of small-scale habitat enhancement. Restoration, while mainly botanically focused, may function as the 'tide that raises all ships.' Alternatively, techniques employed in restoration may function as disturbances from the perspective of bees (e.g., prescribed fire may reduce resources for bees that nest in wood). I conducted a meta-analysis in Chapter 1 to investigate bees' responses to restoration and specific management actions. Overall, restoration was beneficial in terms of bee abundance and richness. In Chapter 2, I report results from a field study in a chronosequence of prairie restorations in northeastern Illinois. Though unrestored abandoned agricultural fiends, restored prairies, and remnant prairies has similar bee abundance and taxonomic diversity, community composition substantially differed by site type. Analyses of bees' traits revealed bee communities of restorations were more differentiated than expectations based on null model predictions. My results indicate bee communities were not systematically filtered by disturbances associated with restoration, rather patterns appeared to be driven by stochastic processes, i.e., founder effects. In Chapter 3, I further investigated my findings from a phylogenetic perspective. Phylogenetic data are increasingly used in ecological studies on the basis that species' similarity in trait space is correlated with evolutionary relatedness. I found key ecological traits in bees showed significant phylogenetic signal (clustering), indicating that phylogeny could be used to inform understanding of bee communities. Bee communities exhibited random phylogenetic structure based on metrics of alpha diversity. Metrics of beta diversity revealed significant phylogenetic clustering at the site level in unrestored and restored prairies. My results based on functional and phylogenetic analyses suggest bee community assembly in restored grasslands can be attributed to stochastic, rather than deterministic events, consistent with founder effects, not environmental filtering. In order to best conserve wild bee communities, proximity to source populations should be considered in the planning stages of new restoration projects.