Effects of native prairie forbs on the foraging choices and recruitment behavior of honey bees (Apis mellifera)

Abstract

Recent increases in honey bee colony mortality have prompted many organizations to plant flowers to improve bee nutrition. However, there remain questions about which flower species to plant and how best to plant them. Honey bees and many non-native species of flowering plants arrived in North America with European colonists. To help restore diverse native species, many organizations would like to increase plantings of native flowers. In the Upper Midwest region, multiple organizations are working to reconstruct tallgrass prairie habitats, including native prairie forb species. However, it was unknown whether those species would attract and benefit honey bees. To inform future bee-friendly prairie planting projects, I conducted studies examining honey bee foraging choices. In Chapter 1, I gave honey bee colonies access to large, reconstructed prairies and recorded their recruitment behaviors (decoding and mapping waggle dance communications). I found that honey bee foragers mainly danced to advertise non-prairie flower patches and non-native flowers as profitable pollen sources, but seven native prairie taxa were also advertised as profitable pollen sources. At one site colonies became significantly more likely to advertise nectar sources in prairies in August/September. In Chapter 2, I gave honey bee colonies access to reconstructed prairies and concentrated plantings of native prairie species to further explore how access to prairies affects colony diet breadth. That study showed major contributions of non-native pollen sources, but colonies collected native prairie and native non-prairie sources as well, especially at the end of the season in most sites. The taxa most frequently collected tend to grow in dense patches, suggesting that planting density may have a strong effect on honey bee recruitment. In Chapter 3, I tested a recently-published method for mapping honey bee waggle dances using data collected during the study presented in Chapter 1. In Chapter 4, I used that new method to examine how the density of flowers in a patch affects honey bee recruitment behavior. This approach was novel as previous studies on recruitment used point-source sugar feeders. While my first attempt did not reveal significant preferences for more dense patches of flowers, it did highlight lessons for future experiments that manipulate variables at the flower patch level to better understand the factors that drive honey bee recruitment. Overall, these results provide insights into which species are most likely to attract honey bees to bee-friendly plantings in the Upper Midwest.