Sandra M Rehan

Elucidating bee response to urbanization is essential to promoting pollinator diversity in cities especially considering such landscapes are projected to expand to support future global populations. To determine how bee community composition and plant–pollinator interactions respond to urbanization, 29 sites representing three urban categories (high, medium and low urbanization) were monitored biweekly from May through early October in Toronto, Canada. Bees were collected passively using pan and blue vane traps as well as actively using aerial nets and vacuums to compare community structure and plant–pollinator networks among urban categories. Functional traits such as dietary breadth, behaviour, nesting substrate and native or non-native status were also examined to determine how landscape influences bee community assemblages. In total, 5477 bees, comprising 26 genera and 164 species, were represented in this study. The urban landscape was largely supportive of species within the family Apidae and Halictidae as well as ground nesting, native and generalist species. Overall, community composition was affected by urban landscape characteristics such as percent tree cover and impervious (i.e. paved and built) surface surrounding sites; however, bee richness and abundance were significantly influenced by plant richness and not by landscape variables. A total of 3267 interactions were observed throughout the study region with characteristics of plant–pollinator networks remaining consistent along the urban gradient with a few floral host plants such as Solidago dominating interactions in certain urban contexts. These results provide important land use and floral host information for targeting pollinator conservation and habitat restoration.