Wednesday, August 4, 2010
Exhibit Hall A, David L Lawrence Convention Center
Rosemary L. Malfi, Environmental Sciences, University of Virginia, Charlottesville, VA and Neal M. Williams, Department of Entomology, University of California, Davis, CA
Background/Question/Methods Bumble bees (
Bombus spp.) are among the most important wild pollinators in temperate ecosystems in North America and Europe, and are believed to be vital to the functioning of the pollination networks in which they occur. Accordingly, evidence of their overall decline in Europe and more selective decline in the U.S. has raised concern about the long term persistence of many species. Human-induced changes in land use, including the loss of natural and semi-natural habitat and associated floral resources, are purported causes in several cases. Declines prompted an investigation of this potential trend in the urbanized landscape in and surrounding Philadelphia, PA. We surveyed the species richness and abundance of bumble bee communities in ten half-ha plots located in restored, managed meadows along a gradient of urban and suburban development in the Philadelphia metropolitan area from June 1 to August 15, 2006. In conjunction with collections, we measured floral density within each plot. We calculated the proportion of developed land at differing spatial scales ranging from 500 to 4000 m from survey sites. General linear models were used to test the effect of developed land, local meadow size, and floral resource density on overall bumble bee species richness and abundance.
Results/Conclusions Development did not affect species richness at any tested spatial scale. Bee abundance was best explained by a model that included the proportion of developed land at the 2500 m scale. In contrast to our expectations, total bumble bee abundance was significantly higher in plots with a higher proportion of developed land surrounding the site (F1,5 = 8.13, P = 0.04). Local floral density did not significantly affect bumble bee richness or abundance (F 1,5 = 0.93, P = 0.34), nor was local resource quality associated with development (r = 0.52, P = 0.12) . We hypothesize that gardens maintained in urban and suburban landscapes may provide an important and consistent food supply to bumble bees throughout the duration of their colony cycle. Forested habitats, which were the other dominate habitat type, may actually be relatively poor in bumble bee resources following closure of the canopy in late spring. This research indicates that restored meadows are excellent habitat for bumble bees and may promote pollination services in urbanized settings.