COS 118-7 - Surrounding landscape and floral resources influence pollinator community composition in oak-savannah fragments of British Columbia, Canada

Wednesday, August 8, 2012: 3:40 PM
Portland Blrm 258, Oregon Convention Center
Julie C. Wray1, Lisa A. Neame2 and Elizabeth Elle1, (1)Department of Biological Sciences, Simon Fraser University, Burnaby, BC, Canada, (2)Biological Sciences, Simon Fraser University, Burnaby, BC, Canada
Background/Question/Methods

Fragmentation of natural areas via urbanization has the capacity to disrupt interactions among plants and animals. The surrounding matrix of urban or semi-natural areas can sometimes support biodiversity in focal habitats of fragmented landscapes, but this can depend on availability of resources within both fragments and matrix, as well as species-specific traits and life history strategies. In particular differences in body size have been hypothesized to influence the susceptibility of a species to fragmentation and matrix quality; however these tests are often met with inconclusive results. We used data on bees collected from 19 oak-savannah fragments to ask whether the surrounding land-use type (Douglas-fir forest vs. urban neighborhoods) affected pollinator community composition, and whether differences in pollinator composition could be attributed to species traits (body size, specialization) or within-fragment resources. We used non-metric multidimensional scaling to visualize the similarity in species composition between sites, as well as t-tests and linear regressions to analyze the relationships between body size, surrounding landscape, and within-fragment floral resources.   

Results/Conclusions

Ordination revealed that oak-savannah fragments embedded in Douglas-fir forest matrix support different bee communities than those surrounded by urban neighborhoods. An examination of species lists suggests forest-associated bee species included floral specialists, bees with large body sizes, and those with earlier flight periods. Large bees with greater energy requirements could be responding to differences in floral resources between oak-savannah fragments embedded in forest vs. urban matrix. Forest matrix fragments had increased floral density and diversity and decreased bare rock compared to those in the urban matrix (t = 2.76, P = 0.01). Bees in forest matrix fragments were larger (t = 2.19, P = 0.04) and linear regressions showed that increases in floral resources resulted in an increase in the mean pollinator size at a site (R2 = 0.43, P = 0.002), as well as an increase in the relative abundance of large bees (R2 = 0.38, P = 0.005). In contrast, the relative abundance of small bees declined with increasing floral resources (R2 = 0.33, P = 0.01). Future research will investigate the characteristics of matrix habitat that influence pollinator community composition in oak-savannah fragments by sampling pollinators in matrix habitats in addition to natural remnants.