Using feather corticosterone to assess the effects of non-breeding season conditions on breeding of Atlantic puffins (Fratercula arctica) and rhinoceros auklets (Cerorhinica monocerata)

In order to fully understand factors that affect animals during distinct parts of their annual cycle, it is important to consider that processes acting in one season may carry over to influence an individual’s success in the following season. Measuring conditions over multiple seasons and life history stages allows carry over effects to be identified and places an individual's current condition into a broader context. Corticosterone levels measured in blood reflect hypothalmic-pituitary-adrenal (HPA) activity in birds in response to challenges that threaten homeostasis. Circulating corticosterone is integrated into growing feathers and can provide physiological information about birds during times when they are unavailable for blood sampling. Here, we used a commercially-available enzyme immunoassay kit to measure corticosterone in alcid feathers, demonstrated the benefits of acetonitrile/hexane purification of samples, and showed that blood and feather corticosterone are biologically-meaningful, albeit non-identical, measures. We used our enzyme-immunoassay and purification method in tandem with stable isotope analysis to measure corticosterone and stable isotopes in feathers and blood collected from rhinoceros auklets Cerorhinca monocerata nesting on three widely-dispersed colonies during years with different oceanographic conditions. We found that individuals from different colonies could be distinguished by their δ15N and δ13C stable isotope values during, but not prior to the breeding season, and that corticosterone levels were consistent with this pattern. Furthermore, we found that rhinoceros auklets had significantly lower corticosterone levels in a year and on a colony assumed to have less favourable feeding conditions, which is opposite to results for other taxa. In a relative of the rhinoceros auklet, the Atlantic puffin Fratercula arctica, we found that egg mass increased in relation to female corticosterone and δ15N values in feathers grown in the months prior to breeding, indicating that physiological state of females prior to the breeding season can influence egg mass. In contrast, we found that pre-breeding corticosterone and δ15N values of rhinoceros auklet females were not correlated with egg mass or egg protein levels (pilot study). Overall our results support our hypotheses that corticosterone levels vary with environmental conditions and that differences in corticosterone levels during pre-breeding correlate with egg size in the subsequent season (breeding). However, our results also indicate that interpreting these relationships requires careful consideration of ecological and physiological characteristics of the individual or taxa in question