OOS 19-2 - Bee community structure across urban prairie fragments and gardens

Wednesday, August 9, 2017: 8:20 AM
Portland Blrm 253, Oregon Convention Center
Shalene Jha, Integrative Biology, University of Texas at Austin, Austin, TX, Kimberly M. Ballare, Integrative Biology, University of Texas, Austin, TX and John Neff, Central Texas Melittological Institute

It is estimated that by 2030, >80% of the global human population will live in cities. In response, urban agriculture has expanded dramatically in the past few decades and currently provides >15% of the global food supply. While insect pollination is critical for most flowering plants, the drivers of insect pollinator diversity and abundance in urban areas are not well established. Bees are one of the most effective pollinators and provide essential services to a variety of cultivated and wild plant species; however, they may be particularly sensitive to urbanization given their predominantly ground- and wood-nesting behavior. In this study, we investigate the how local and landscape composition influences ground- and wood-nesting bee communities across 20 gardens and 20 grassland fragments within central Texas urban landscapes. Bees were surveyed using a combination of pan and blue vane traps. Local habitat composition was quantified at the 50 x 50m plot scale via floral and ground cover surveys within each site, and landscape composition was quantified at the 2km scale using National Land Cover Database land use maps. Bee abundance and species richness were examined as function of local and landscape composition using linear mixed effects models.


We collected 12,440 bee specimens representing 172 species across all sites. We found that ground-nesting bees were significantly more abundant than wood-nesting bees at all sites, and they were significantly more abundant in sites with higher levels of bare ground. Relative abundance of the two nesting groups varied between gardens and grassland fragments with significantly more wood-nesting bees within grassland fragments compared to urban gardens. Additionally, within urban garden sites, we documented a highly significant positive effect of landscape-level (2 km scale) semi-natural habitat cover on the species richness and abundance of both nesting groups. In contrast, within grassland fragments, we found a small but significant negative effect of semi-natural habitat on species richness. Our results thus provide evidence that local and landscape composition interact to differentially influence ground- and wood-nesting bee communities within urban areas. Specifically, our results highlight the sensitivity of wood-nesting species to within-site land management and the importance of semi-natural habitat within the larger landscape for overall bee diversity and abundance in urban areas.