Do thermal limits predict wild bee community response to urban warming?

Urbanization and climate change increase local and global temperatures. It is important to understand how these higher temperatures affect organisms, particularly those that provide ecosystem services such as pollination. We investigated effects of urban warming on wild bee communities. We hypothesized that bees with higher thermal tolerances would be more common in the hotter study sites and bees with lower thermal tolerances would be less common or absent from the hotter sites. We also predicted that bee thermal tolerance would depend on their nesting habitat, since cavity-nesting bees naturally experience more extreme temperature than do ground-nesting bees. To measure bee thermal tolerances, we heated individual wild bees of 16 species and recorded the temperature at which each bee lost its ability to right itself (critical thermal maximum, or CT_max). We also surveyed wild bee communities at 19 study sites along an urban temperature gradient. We then compared critical thermal maxima of the bees to their distributions among the landscape relating to heat and other environmental variables. We found that species’ CT_max predicted species distribution based on mean maximum temperature and impervious surface ground cover (1000 m radius from study sites). These data provide a foundation for understanding the effects of urban and global warming on wild bees.