Nest Density of the Native Bumble Bee,Bombus nevadensisCresson (Hymenoptera: Apoidea), in an Agricultural Landscape

Abstract

Agricultural intensification, with its loss in nesting sites, is believed to be a major factor causing bumble bee declines (Bombus spp.) (Hymenoptera: Apoidea) (Kells and Goulson, 2003; Goulson et al., 2008). Many species nest in the ground in abandoned rodent burrows and, when natural habitats are converted to agricultural fields, these nesting sites are believed to be limited to the uncultivated field boundaries (Banaszak, 1983) and remnants of semi-natural habitats in the landscape (Saville et al., 1997). However, it is a challenge to locate bumble bee nests, and worldwide, only a small proportion of naturally occurring nests have ever been located (Svensson and Lundberg, 1977; Donovan and Wier, 1978; Fussell and Corbet, 1992). In earlier studies, artificial nests were placed outdoors for studies on nest behaviors but few were ever occupied and hence queen nest-searching behavior (zigzag flight trajectory, low over the ground, with stops to investigate potential nest sites) was used as an indicator of preferred nest sites (Svensson et al., 2000; Kells and Goulson, 2003). More recent studies (Osborne et al., 2008; O’Connor et al., 2012) have provided insights on bumble bee nests of species in the United Kingdom but there remains little information about species in other regions. Here, we report nest density of a North American native bumble bee species in the Willamette Valley, a rich agricultural area in western Oregon. In August 2011, during a native bee monitoring study, a remarkable abundance of Bombus nevadensis Cresson workers was observed in a blue vane trap (SpringStar) in a landscape dominated by seed fields of diverse crops including hairy vetch (Vicia villosa Roth), red clover (Trifolium pratense L.), and sunflower (Helianthus annus L.). On 24 August 2011, upon examination of the area around the trap (GPS: 44u44.1049, 123u15.8189) we observed B. nevadensis workers entering vole burrows in the stubble of a recently harvested annual rye grass (Poaceae; Lolium multiflorum Lam.) seed crop. The entrances (Fig. 1) were flagged and monitored until at least 5 individual B. nevadensis emerged or returned to the same hole, which we considered as evidence of the presence of a colony. In one afternoon, we recorded 11 colonies in a 178.3 m 3 32.8 m (55848.2 m) area on the western edge of the field. Hence, the estimated nest density of B. nevadensis was 18.8 nests ha. For further observations, vole burrows were excavated to expose the nests (Fig. 2). In all, seven nests were examined. These were located 1.0 to 10.8 cm below the surface. Some nest entrances were separated by less than 9 m. The colonies were small (maximum brood 5 46 immatures) (Table 1), and present in chambers 5 to 15 cm deep and 5 to 11 cm wide. Five nests were in the reproductive phase; they included the original queen (tattered wings; low body pubescence), new queens, males, workers, eggs and larvae. The remaining two nests only contained workers and pupae. In fall, the field was planted with meadowfoam, Limnanthes alba Hartw. ex Benth. and it was not possible to continue the study the following year. This is the first study to describe naturally occurring nests of B. nevadensis. In earlier reports related to this species (Hobbs et al., 1962; Hobbs, 1965), artificial nests constructed with plywood were used to examine nesting behaviors. In both studies, queens preferred to nest in underground (,10 cm below ground surface) rather than in above-ground hives. Hobbs et al. (1962) also noted that queens chose artificial nests placed in woodlands (14 nests) rather than those in prairies (3 nests). This is, thus, also the first report of B. nevadensis nests in an agricultural habitat. Bombus nevadensis belongs to the subgenus Bombias which includes one other closely related species, B. auricomus (Robertson). This subgenus exhibits unique colony development in that each egg and