Species‐specific responses to warming alter community composition

Species are responding to global climate change in varied and nuanced ways. However, how species‐specific responses to climate change= affect interactions among species remains poorly understood. It is important to understand species interactions under potential climate change scenarios because those interactions can in turn alter community dynamics. In this study, we conducted two complementary experiments to examine how simulated warming might alter larval intraguild predation (IGP) rates and resulting adult assemblage composition in three species of North American dragonflies: Pachydiplax longipennis, Plathemis lydia and Libellula luctuosa. First, using both P. longipennis and L. luctuosa, we isolated interspecific and intraspecific pairs of larval dragonflies of different size differentials to determine how the size and species identity might influence IGP rates. In tandem, we conducted a year‐long mesocosm experiment with all three species to assess how simulated warming and heat waves influenced the resulting adult dragonfly assemblages. IGP trials revealed that P. longipennis individuals were much more likely to engage in IGP than L. luctuosa, regardless of size differential. In the mesocosm experiment, emerging adult assemblages were dominated by P. longipennis individuals, a pattern that was most pronounced in the control treatment. Our results indicate that while P. longipennis may be the competitively dominant species under current ambient conditions, warming may alter this dynamic and lessen the dominance of this species on the resulting assemblage composition. We investigated the effects of warming and temperature variability on community dynamics of larval dragonflies using mesocosm and intraguild predation (IGP) experiments. Pachydiplax longipennis had the highest survival under all temperature treatments and also dominated IGP trials. Our results indicate that species‐specific effects of climate change may influence community dynamics.