Niche Overlap and Diet Breadth: Can bumble bee (Bombus spp.) foraging behavior reveal species vulnerabilities?

Abstract

Within the last two decades, notable shifts have taken place within the bumble bee (Bombus spp.) communities of North America: some species have experienced substantial, rapid declines, while others appear to be stable or even experiencing range expansion. Studies from the UK and Europe implicate the loss of natural habitat and associated floral resources in overall bumble bee declines and species losses. Research also indicates that loss of host plants disproportionately impacts species with narrower diet breadths, and that these species are more likely to be long-tongued (presumably due to the reduction of long-tubed floral hosts). In contrast, species with more generalized diets have been shown to be more abundant. In this study we examined resource utilization by the bumble bee community at Blandy Experimental Farm in Boyce, VA in order to evaluate (1) whether the diet of a potentially declining long-tongued species (B. fervidus) overlaps strongly with a more competitive, expanding species (B. impatiens), (2) whether the potentially declining species specializes on host plants relative to the other bumble bee species in the community, and (3) if bumble bee species abundance is associated with diet breadth or the relative abundance of preferred host plants. We surveyed and quantified all mass flowering species in two large meadows on a weekly basis from June 1 to August 15, 2010. Collections of bees were then carried out on these plants for a minimum of one person-hour per week per plant species. For each bee species, Simpson’s index was used to assess diet breadth.Six bumble bee species were observed at Blandy Experimental Farm in total. The potentially declining and expanding species (B. fervidus and B. impatiens) possessed the narrowest and widest diet breadths respectively. In addition to possessing specialist tendencies, B. fervidus had little niche overlap with other species in the community (i.e. it visited plants other species were not visiting). Although B. impatiens had the widest diet breadth and was also the most abundant species (50% total collections), diet breadth, overall, was not associated with abundance (Spearman’s rho = 0.37, p = 0.46). However, we found that species abundance was positively correlated with abundance of most preferred host plant (Spearman’s rho = 0.97, p = 0.0012). The results of this study indicate that resource limitation may put B. fervidus at risk of decline when preferred host plants are reduced or lost, whereas B. impatiens may be resistant to extirpation due to its “super-generalist” tendencies.