Introduction
Studies examining bee communities within temperate forests have largely restricted sampling to the understory (Milam et al., 2022) with the presumption that most bees remain in this lower stratum. However, recent evidence indicates that bees are vertically distributed within temperate forests (e.g., Ulyshen et al., 2010; Urban-Mead et al., 2021), suggesting a potentially large knowledge gap in the ecological role of these important forest pollinators. Despite this revelation, research regarding the vertical distribution of bees and other pollinators within forests is further limited by the difficulty of sampling the high canopy (Cannon et al., 2021; Cunningham-Minnick et al., in press ). Current sampling methods reach into the canopy (e.g., Maguire et al., 2014; Ulyshen et al., 2010), but the canopy-aerosphere interface — a potentially ecologically important area for bees due to copious floral resources available — remains unexplored in temperate forests (Nakamura et al., 2017; Urban-Mead et al., 2021). Thus, our understanding of pollinator ecology within forests will remain incomplete until the distribution of forest bee communities along the entire vertical gradient of vegetation structure is documented. Moreover, if the current understanding of bee abundance and diversity patterns in forests are inaccurate, forest management recommendations for bee conservation may be biased or potentially misguided (Milam et al., 2022; Urban-Mead et al., 2021), further highlighting the importance of understanding the distribution of bee communities along the full vertical gradient of temperate forests, including the canopy-aerosphere interface.
Bees are expected to be spatially and temporally distributed throughout temperate forests in response to local resource availability. Studies have demonstrated that forest bee communities are diverse and vertically stratified on sun-exposed edges (e.g., Cunningham-Minnick & Crist, 2020), within forests near edges (e.g., Urban-Mead et al., 2021), and within the forest interior (e.g., Campbell et al., 2018; Milam et al. 2022; Ulyshen et al., 2010) when floral resources of the forest are available, as well as when they are not. Inferences and observations further suggest that bees will both forage on floral resources and nest at different vertical strata within forests (Cunningham-Minnick & Crist, 2020; MacIvor et al 2014; Russo & Danforth, 2017; Smith et al., 2019; Sobek et al., 2009; Urban-Mead et al., 2021; Wood et al. 2018). For instance, Smith et al. (2019) and Wood et al. (2018) found support through pollen analyses that forest bee communities rely upon floral resources of dominant tree species. Yet floral resources of herbaceous and woody species within temperate forests are typically limited to a spring and early summer phenology, which has been correlated to fewer late-season bees in the forest understory (Cunningham-Minnick & Crist, 2020). Alternatively, studies have found more bees in the forest herbaceous layer during spring and more bees in the canopy during the summer (Cunningham-Minnick & Crist, 2020; Ulyshen et al., 2010), suggesting that the distribution of forest bees may also shift out of the understory and into the higher vertical strata of the forest as the year progresses. However, no studies have examined the bee fauna in the aerosphere above the forest canopy. Thus, we undertook this study to determine the extent to which bees occupy the open air above the forest canopy, how the bee assemblages of this canopy-aerosphere interface compare in abundance, species richness, and composition with assemblages at other strata, and how these patterns change with seasonal phenology.