2018 ESA Annual Meeting (August 5 -- 10)

COS 77-2 - Ecological filtering in scrub fragments reduces taxonomic and functional diversity of native bees, but does not homogenize bee assemblages

Wednesday, August 8, 2018: 1:50 PM
333-334, New Orleans Ernest N. Morial Convention Center
Keng-Lou J. Hung1,2, John S. Ascher3, Jessica A. Davids2 and David Holway2, (1)Evolution, Ecology and Organismal Biology, The Ohio State University, Columbus, OH, (2)Division of Biological Sciences, University of California, San Diego, La Jolla, CA, (3)Department of Biological Sciences, National University of Singapore, Singapore, Singapore
Background/Question/Methods

Predicting the long-term consequences of habitat alteration for the preservation of biodiversity and ecosystem function requires an understanding of how ecological filters drive functional and taxonomic biodiversity loss. Here, we evaluate the role of ecological filters in restructuring native bee assemblages inhabiting fragmented scrub ecosystems in southern California using a dataset of native bees (9,152 individuals representing 160 species or morphospecies) collected in 2011 and 2012 in San Diego County. We obtained a suite of functional traits for each bee species and compared bee assemblages in natural reserve and scrub fragment sites to test four predictions. We predict that ecological filtering should cause bee assemblages in fragment sites to exhibit the following characteristics compared to assemblages in reserve sites: (1) reductions in functional diversity beyond expectations under null models of random species loss, (2) shifts in taxonomic and functional assemblage compositions, (3) reductions in functional and taxonomic beta-diversity among sites, and (4) reductions in the proportional representation of range-restricted species.

Results/Conclusions

In support of the predicted importance of ecological filtering, bee assemblages in fragments lacked a suite of indicator species and functional groups found in reserves, exhibited distinct functional and taxonomic composition, and had reduced proportional representation of range-restricted species. However, contrary to expectations of strong filtering, the observed loss of functional diversity in fragments did not exceed that predicted under null models of random species loss, nor did it lead to taxonomic or functional homogenization of bee assemblages. Our results suggest that ecological filtering altered bee assemblages in habitat fragments even when such fragments contained well-preserved native plant assemblages, underscoring the importance of preserving large areas of natural habitat for the conservation of bees (especially range-restricted taxa) and their associated ecosystem functions. However, our finding that individual fragments harbored bee faunas that were taxonomically and functionally distinct from one another also highlights the value of fragments, in aggregate, for conserving biodiversity at the landscape scale.