Habitat use by bumble bees (Hymenoptera: Apidae: Bombus spp.) in New Zealand

Pollination is a valuable ecosystem service in which wild insect pollinators play a major role. Currently, managed pollination of crops relies almost entirely on one insect species, the honey bee Apis mellifera L. Interest in alternative pollinator species has increased in recent years with declines in honey bee and wild pollinator populations in Europe and North America. Bumble bee colonies are managed successfully for greenhouse pollination of tomatoes; however, field management of bumble bees has not been proven to be reliable and cost-effective. Habitat manipulation has been suggested as a low input means of increasing bumble bee populations but so far has not led to a reliable method of improving bumble bee numbers. Improving the scientific basis of habitat manipulation and the provision of artificial nest sites on farmland may help increase bumble bee populations in the vicinity of a crop. Eighty, four-unit nest boxes have been situated on Kowhai Farm at Lincoln University for four years, and occupancy by bumble bee queens has been recorded during that time. The results collected during the four years are summarised and used along with appropriate literature to suggest how the bumble bee nest boxes on Kowhai Farm could be used to investigate the factors influencing bumble-bee queen nest site selection. Twenty commercially-produced Bombus terrestris (L.) colonies were placed at 20 field sites. Sites were divided into two treatments, based on the predominant habitat within 10 m of the colony. The landscape surrounding the colonies out to a radius of 500 m was divided into one of four habitat categories: flowering crop, non-food crop, pasture, and 'other'. The proportion of each habitat category was calculated for each colony. Colony performance was measured by a productivity index. The performance of B. terrestris colonies was highly variable and no significant habitat effects were observed. Highly variable microsatellite loci have been used to differentiate nest mates from non-nest mates in several social insect species. Such markers could be used to investigate bumble bee forager movement within a landscape. Twelve microsatellite loci isolated in the bumble-bee B. terrestris were tested for applicability to New Zealand populations of Bombus hortorum (L.). Three loci could repeatedly produce informative gels. Regression relatedness was calculated between B. hortorum individuals collected from three naturally founded colonies. The three useful loci provided sufficient information to distinguish between related and unrelated workers. A high proportion of the workers collected from within nests appeared to be unrelated to each other.