Sampling pollinators in the field
A. mellifera and Bombus spp. (hereafter honeybee and bumblebee, respectively) samples were collected from six winter squash farms in southeastern Michigan, USA (Appendix S1: Table S1) during two visits to each site between 26 July and 30 August 2016 during the peak squash bloom. The pollinator sampling described here includes a subset of the sites that were previously sampled in Fearon and Tibbetts (2021) and Fearon et al. (2022). In this study, we focus on V. ceranaeinfection in honeybees and bumblebees, while the prior studies examined links between the pollinator community composition and bee viral prevalence. Sites were at least 10 km apart to ensure that the pollinator communities were isolated from each other (Greenleaf et al. 2007). We only sampled on sunny days with windspeeds less than 2 m/s. To collect the bees, four 50 m transects were randomly placed at each field site. Three transects were placed in the field along the crop rows, while the fourth transect was placed along a field edge to sample bees foraging near native flowers and invasive weeds. All honeybees and bumblebees observed along the transect lines were collected using handheld nets or pan traps. Details on the trapping methods are included in Appendix S2 and Fearon and Tibbetts (2021).
All pollinator samples were stored on dry ice in the field, and later placed in a -80 ˚C freezer to maintain the integrity of the DNA for detection of V. ceranae infection. All bees were identified to species using the Discover Life key (Ascher and Pickering 2013). The collected bumblebee species were primarily Bombus impatiens , but also included Bombus auricomus , Bombus bimaculatus ,Bombus griseocollis , Bombus fervidus, Bombus pensylvanicus , Bombus sandersoni , and Bombus vagans at very low densities (< 8 individuals total). A. mellifera and B. impatiens were common at all six field sites.