Foraging ecology and management of Bombus spp. (Hymenoptera: Apidae) in agricultural landscapes

Bumble bees (Bombus spp.) are important pollinators of agricultural crops such as red clover (Trifolium pratense L.) in New Zealand. However, there is a lack of knowledge about Bombus foraging behaviour and management at the landscape scale. This thesis evaluated habitat manipulation (i.e., provision of nest sites and floral resources) and shifting bumble bee colonies to a crop to enhance bumble bee numbers. Bumble bee flight distances from the nest and responses to different sized forage patches were also investigated. Habitat manipulation involved placing 80, four-unit bumble bee domiciles around 16 field margins, half of which had been sown with Phacelia tanacetifolia Benth. The effects of the flowers on domicile occupancy could not be determined because most plots were destroyed by plant competition and grazing animals; however, in areas with high naturally occurring floral diversity, domicile occupancy was higher. Occupancy rates over the three years of the study were 0.31, 4.06 and 8.12 % and the main occupant was B. hortorum L. (67.5 %). There was a positive association between domicile occupancy in one year and nest founding in the next. To supplement forager numbers, four commercially-obtained B. hortorum nests were shifted to a 4 ha red clover 'Pawera' seed crop. Foragers were marked with fluorescent powder. Reobservation rates within the crop ranged from 4-15 %. There was a virtual absence of marked foragers within 10m of the nest. Pollen analysis showed that most (85 %) pollen collectors were visiting the red clover crop exclusively, although some were flying 200 m from the nest to forage on other species. The ratio of marked to unmarked B.hortorum foraging on the crop was 1:43, suggesting that adding nests contributed little. An experiment designed to evaluate whether bees foraged close to the nest showed there was no difference in the number of B. terrestris L. on pots of P. tanacetifolia at 20 and 200 m from the nest. However, very low numbers of experimental bees were involved. More bees were observed on the P. tanacetifolia when the nest was downwind of the plants suggesting that upwind foraging took place; further experiments are needed to confirm this. Patches of potted Lavandula x intermedia Lois. were created to test the effects of patch size on the numerical and functional responses of B. terrestris. The number of B. terrestris visiting the lavender increased with patch size but this was less than proportional. The number of flowers visited per foraging bout increased with patch size but the proportion of available flowers visited decreased. Combining these two responses, the net visitation per flower was independent of patch size. Pollinator management in New Zealand is discussed. It is suggested that other bee species in addition to Apis mellifera L. should be used in New Zealand. Bumble bees are most promising alternative pollinators but the commercial stocking of field crops has not yet proved to be cost-effective. Habitat manipulation is therefore the recommended management option, because of the low inputs and the potential long-term benefits.