Introduction
Bees are important pollinators of a variety of crops and wildflowers, with their ecosystem services being valued at \euro153 billion (Gallai, Salles, Settele, & Vaissière, 2009). Honey bees have been introduced for crop pollination around the globe, and when present, contribute to approximately 19% of all crop plant visits in a region (Hung, Kingston, Albrecht, Holway, & Kohn, 2018). Wild bees are also known to contribute to increases in fruit production for pollinator-dependent crops (Garibaldi et al., 2013), and similar environmental factors contribute to both honey bee and wild bee health and conservation (Evans, Smart, Cariveau, & Spivak, 2018). The taxonomic identity and relative abundance of floral resources collected by bee species can vary through time (Baum, Rubink, Coulson, & Bryant, 2004; Wood, Kaplan, & Szendrei, 2018). In addition, foraging strategies that facilitate the collection of resources, such as nectar and pollen, are variable among bee species (Rollin et al., 2013; Saleh & Chittka, 2007; Visscher & Seeley, 1982). To facilitate pollinator conservation, it is imperative to understand how foraging strategy influences patterns of resource collection.
Polylectic bees with an extended foraging season, such as honey bees and bumble bees, must navigate variable spatial and temporal resource availability challenges (Goulson, 1999). Pollen provides the primary source of macronutrients such as protein, lipids, sterols, and minerals, which are important for the growth and development of bee larvae and colonies (Avni, Hendriksma, Dag, Uni, & Shafir, 2014; Di Pasquale et al., 2013; Moerman, Vanderplanck, Fournier, Jacquemart, & Michez, 2017; Roulston & Cane, 2002). However, not all plant pollen is equivalent in the relative amount of nutrients or their ratios (Roulston, Cane, & Buchmann, 2000; Ruedenauer, Spaethe, Kooi, & Leonhardt, 2019). Generalist bees, therefore, have to collect pollen from a variety of plant taxa to maintain a nutritional balance that ensures appropriate growth and development (Filipiak et al., 2017), and to mitigate other stressors such as infection from parasites (Di Pasquale et al. 2013). Collecting pollen from a variety of plant taxa can be a challenge in high intensity agricultural landscapes, and low diversity in collected pollen can reduce colony growth rate (Hass et al., 2019). The foraging strategy of bees, such as the presence of dance communication for honey bees and the trapline foraging of bumble bees, may help bees navigate such challenges and ensure the collection of an adequate diversity of pollen resources in their diet.
Both honey bees and bumble bees are generalist central place foragers that must locate, collect, and return to the nest with a diverse array of resources to provision the colony. A major foraging strategy employed by honey bees is dance communication, whereby scouts and successful foragers transmit information regarding the distance and direction to profitable resource patches (Seeley, Camazine, & Sneyd, 1991; Visscher & Seeley, 1982; Von Frisch, 1967). This strategy facilitates large scale recruitment to profitable resources, such as blooms of monoculture crops (Rollin et al., 2013). In contrast, bumble bee individuals develop and follow trapline foraging routes among known resource patches, ‘majoring,’ while simultaneously monitoring for changes in resource availability through ‘minor’ routes (Bernd, 1979; Ohashi & Thomson 2009; Pasquaretta, Jeanson, Andalo, Chittka, & Lihoreau, 2017; Woodgate, Makinson, Lim, Reynolds, & Chittka, 2017). Trapline foraging by bumble bees is considered an optimal foraging strategy whereby individual bees learn, over successive foraging bouts the location of profitable resource patches and develop efficient routes that minimize distances traveled (Lihoreau et al., 2012; Woodgate et al., 2017). Based on these distinct foraging strategies, we expect honey bees to forage at monofloral resource patches, while individual bumble bees will disperse more evenly among all available resources, as has previously been observed in other agricultural systems (Rollin et al., 2013). We thus predict that bumble bees will be less flower constant relative to honey bees on individual foraging trips, where flower constancy is defined as the proportion of foraging bees that collect a single pollen morphotype during individual foraging bouts.
The frequency of flower constancy by individual bees can affect the richness and diversity of pollen collected at the colony level, along with the distribution of foragers among available resource patches. As bumble bees individually learn the location of rewarding flowers, and we expect individuals to collect from a greater number of plant types in single foraging bouts, we predict a greater richness of pollen collected by bumble bee colonies compared to honey bees. Furthermore, there is evidence indicating that bumble bees show preference for pollen resources, and that these preferences are sustained within and among colonies in a shared landscape (Saifuddin & Jha, 2014). Such preferences could lead to increased evenness among the collected resources, and hence to greater pollen diversity. In contrast, we expect most honey bee foragers to be recruited to resources based on their abundance in the landscape (Rollin et al., 2013; Visscher & Seeley, 1982), as opposed to demonstrating preferences, where a plant type would be visited more often than predicted based on its abundance. If we assume unequal abundance of plant types in the environment, we expect a lower evenness among collected resources, and together with lower species richness, lower pollen diversity for pollen collected by honey bees compared to bumble bees.
In the current study, we examined the pollen foraging behavior of honey bees (Apis mellifera ) and bumble bees (Bombus impatiens ) in a shared landscape. We collected pollen from returning foragers of each bee species over five time periods at each of three sites in a suburban-agricultural landscape. We quantified the frequency of flower constant foragers, and the richness and diversity of pollen collected by colonies. In addition, we examined bees’ preferences by comparing pollen collected to resources available. We tested the following predictions derived from the known foraging strategies for these two bee species: i) bumble bees are less flower constant relative to honey bees during a foraging trip; ii) bumble bee colonies collect a greater richness of pollen types relative to honey bees, along with greater pollen diversity and iii) honey bees forage randomly while bumble bees exhibit preferences for some plant types. Results are discussed in terms of differences in bee foraging strategies and how understanding such differences may be used to improve pollinator habitat and facilitate increased productivity in pollinator-dependent crops.