Upward ant distribution shift corresponds with minimum, not maximum, temperature tolerance.

Rapid climate change may prompt species distribution shifts upward and poleward, but species movement in itself is not sufficient to establish climate causation. Other dynamics, such as disturbance history, may prompt species distribution shifts resembling those expected from rapid climate change. Links between species distributions, regional climate trends and physiological mechanism are needed to convincingly establish climate-induced species shifts. We examine a 38-year shift (1974-2012) in an elevation ecotone between two closely related ant species, Aphaenogaster picea and A. rudis. Even though A. picea and A. rudis are closely related with North American distributions that sometimes overlap, they also exhibit local- and regional-scale differences in temperature requirements so that A. rudis is more southerly and inhabits lower elevations whereas A. picea is more northerly and inhabits high elevations. We find considerable movement by the warm-habitat species upward in elevation between 1974 and 2012 with A. rudis, replacing the cold-habitat species, A. picea, along the southern edge of the Appalachian Mountain chain in north Georgia, USA. Concomitant with the distribution shifts, regional mean and maximum temperatures remain steady (1974-2012), but minimum temperatures increase. We collect individuals from the study sites and subject them to thermal tolerance testing in a controlled setting and find that maximum and minimum temperature acclimatization occurs along the elevation gradient in both species, but A. rudis consistently becomes physiologically incapacitated at minimum and maximum temperatures 2 °C higher than A. picea. These results indicate that rising minimum temperatures allow A. rudis to move upward in elevation and displace A. picea. Given that Aphaenogaster ants are the dominant woodland seed dispersers in eastern deciduous forests, and that their thermal tolerances drive distinct differences in temperature-cued synchrony with early blooming plants, these climate responses not only impact ant-ant interactions, but might have wide implications for ant-plant interactions.