Abstract
Biological control programs provide simplified systems in which to investigate the influence of temperature on trophic interactions. We investigated the interaction between the invasive thistle Carduus nutans and its specialist herbivore, the biological control agent weevil Rhinocyllus conicus , along a temperature gradient spanning an urban-to-rural urbanization gradient in central Pennsylvania. As expected, more urbanized sites were warmer than rural sites. Nevertheless, C. nutans phenology was only slightly earlier at warmer sites, and the timing of observations of adult R. conicus on thistles did not shift earlier, leaving synchrony apparently unchanged. Despite unchanged phenological matching, seasonal patterns in weevil damage varied with temperature: early-appearing capitula at warmer sites were more heavily damaged than early capitula at cooler sites, yet overall a lower proportion of capitula was damaged at warmer sites. Rising temperatures may therefore reduce effectiveness of R. conicus for C. nutans control, as more capitula escape damage. Biological control systems and naturally occurring temperature gradients are useful systems in which to investigate climate change impacts, but a focus on the phenology of organisms’ presence or absence may not be sufficient; our study demonstrates that consideration of the interaction itself may be necessary.