Improving spatial arrangement of honeybee colonies to avoid pollination shortfall and depressed fruit set

Pollination shortfalls affect yield of many crops, and the use of managed honeybee colonies is a common practice for addressing the problem. However, colony density and arrangement strategies are not generally based on replicated scientific trials, so there is considerable uncertainty regarding effectiveness of different practices. We address this problem with experiments in almond orchards in south-east Australia, considering impacts on honeybee pollen foraging and fruit set. We examined the effect of distance from colony location on the depletion rate of pollen from flowers near (similar to 40m) and far (260-490m) from colonies in almond orchards. We assessed pollen loads of 950 flowers in total, collected at four times of day (9-9:30h, 11-11:30h, 13-13:30h and 15-15:30h), from 8 near:far pairs. We also surveyed fruit set on nearly 900 trees in replicated transects over two seasons to determine the effect of distance from colony when using large placements of colonies (similar to 120 colonies, N=581 trees), and effect of colony density when using smaller placements of colonies (N=313 trees). Flowers near colonies maintained an approximately constant mean pollen load over the course of the day, indicating that the rate of pollen released by flowers was matched by the rate of pollen collection by bees. Flowers far from colonies increased in pollen load over the course of the day, indicating relatively less pollen-collecting activity, so that by 15:30h they had, on average, 46% more than flowers near colonies. Fruit set declined with distance by 22% over 850m, consistent with the observation that pollen foraging declines over distance from colony. Fruit setalso declined with colony density from 46% at 68 colonies per ha to 33% at 28 colonies per ha. Synthesis and applications. Pollen-collecting activity is relatively low at flowers far from honeybee colonies, creating a risk of lost yield through underpollination. Fruit set declines if colony density is reduced below 68 colonies per ha. Pollination outcomes in terms of fruit set are improved when fewer colonies are used per placement (<100) and placements of colonies are arranged with shorter distances (<700m) between them, so that more trees are within 400m of colonies. Pollen-collecting activity is relatively low at flowers far from honeybee colonies, creating a risk of lost yield through underpollination. Fruit set declines if colony density is reduced below 68 colonies per ha. Pollination outcomes in terms of fruit set are improved when fewer colonies are used per placement (<100) and placements of colonies are arranged with shorter distances (<700m) between them, so that more trees are within 400m of colonies.