COS 1-6 - Exploring variation in local and landscape factors driving grassland bee communities in Colorado

Monday, August 7, 2017: 3:20 PM
E143-144, Oregon Convention Center
Adrian L. Carper, University of Colorado, CO, Collin J. Schwantes, University of Colorado, Boulder, CO, Stacy B. Endriss, Graduate Degree Program in Ecology, Colorado State University, Fort Collins, CO, Andrew P. Norton, Bioagricultural Sciences and Pest Management, Colorado State University, Fort Collins, CO, M. Deane Bowers, Ecology and Evolutionary Biology, University of Colorado, Boulder, CO and Mary A. Jamieson, Biological Sciences, Oakland University, Rochester, MI

Agricultural intensification for both food and fuel could have wide reaching impacts on native pollinator communities, especially in habitats vulnerable to human agricultural development such as grasslands. However, studies on pollinator communities, and the factors that drive pollinator community dynamics, are lacking from many agriculturally intensive regions. To explore how variation in local and landscape factors affect native bee communities in the most agriculturally intensive region of Colorado, we conducted a two-year survey of bees in 32 focal grassland sites located across five counties. We sampled bees using a combination of bee bowls, vane traps, and netting in 12 actively grazed rangelands and 20 Conservation Reserve Program (CRP) grasslands. Traps were deployed for 24 hours, once monthly (June-Sept in 2013 and May-Sept in 2014). We also measured plant diversity locally within sites and extracted agricultural land-cover from multiple scales surrounding each site. We explored how variation in local and landscape factors impacted bee communities using a combination of ordination and modeling techniques.


In total we captured 36,661 individual bees, spanning 50 genera, and 203 different species or morphospecies. Bee abundance varied between years with 2014 producing nearly 3 times more bees than 2013, driven mostly by the addition of a single sampling date in May of 2014 that produced 8,106 specimens. CRP plantings supported more bees than rangelands in both 2013 and 2014 (30% and 16%, respectively) and more bee species (24% and 9%, respectively), suggesting that grazing may be negatively affecting bees at our rangeland sites. Bee community responses to local and landscape factor were highly variable across sites and years. For example, total bee abundance appeared to be driven primarily by land cover diversity at 1,600m scales in 2013, but not in 2014. Responses were also guild-specific, with bees of different body sizes responding to landscape elements at different scales, roughly corresponding to size-specific foraging ranges. Although, local plant species richness and floral resources were highest in rangeland sites, we found weak relationships between floral communities and bee community dynamics. Ultimately, these results suggest that the responses of pollinator communities to agricultural intensification are complex and extremely variable, highlighting the value of large scale and long term monitoring studies.