Corticosterone Reaction Norms to Temperature in Wild Birds: Implications for Understanding Endocrine Flexibility and...

The glucocorticoid mediated stress response plays a major role in coping with both gradual and rapid changes in environmental conditions and may be especially important when conditions depart from expectations. Conceptual models of endocrine flexibility suggest that individual flexibility, measured using reaction norms along an environmental gradient, might predict differences in the ability to cope with challenges. For example, differences in the speed or scope of acute endocrine responses might underpin coping ability. However, empirical results have been limited by the inability to accurately measure individual level endocrine reaction norms. Here, we took advantage of a database of corticosterone measures in 1727 individuals of 99 bird species sampled around the world to extend the concept of endocrine reaction norms to species differences. We first describe a global reaction norm for birds and then demonstrate species-specific differences in reaction norms for baseline corticosterone, maximum corticosterone, and the speed of corticosterone increase to both absolute temperature and to the difference between current and expected temperature. Overall, we found that in addition to changes in absolute corticosterone, the speed of the acute response increased when minimum daily temperature dropped below 0°C-2°C. In contrast, we found little evidence for increases at higher temperatures. We found a similar pattern when temperature was colder than expected given the location and date regardless of absolute temperature, but this effect was only seen for baseline corticosterone. Our models also consistently indicated that species differed in the shape of their corticosterone reaction norm to absolute temperature and temperature deviations. However, we did not have adequate data to fully characterize species-specific reaction norms. We suggest that the endocrine flexibility and reaction norm framework applied in a comparative context could help predict species sensitivity to changing climate, but that additional field data will be needed to fully test this idea.