Costs of breeding are rapidly buffered and do not affect migratory behavior in a long-lived bird species

Life history theory states that individual fitness in one stage of life is conditioned by what occurred in previous stages. In migratory species, reproductive effort during breeding has often been found to influence body condition, molt schedule, self-provisioning and migration of individuals in subsequent seasons (i.e., carryover effects of breeding). However, there is a current uncertainty in understanding how long-distance migrants trade off among such energy-demanding activities (i.e., breeding, molting and migrating). To provide evidence to the field, we experimentally reduced the parental effort of a long-lived Procellariform, the Cory's shearwater (Calonectris borealis), by inducing failure at the incubation stage. Treatment and control birds were tracked during their subsequent migration by means of light-level and immersion loggers and sampled for six specific feathers (molted at different periods along the migratory cycle) upon the recovery of the loggers 1 yr later. Feathers were used to perform stable isotope analysis (SIA) and determine corticosterone levels (CORT). By these means, we evaluated the effect of breeding effort on migratory strategy, at-sea activity patterns, molt patterns, and levels of stress experienced by birds along the non-breeding period. We did not detect any difference between birds in the induced failure group and successful breeders in terms of spatio-temporal distribution: all birds shared common foraging areas throughout the study period and the timing of major phenological events did not differ. Failed birds significantly advanced their molt, as revealed by SIA and flying activity patterns. The stress levels of failed birds, inferred through CORT concentrations in feathers, were found to be consistently lower than in successful breeders, through the end of the breeding to the non-breeding period. Thus, we provide robust evidence that the costs of reproduction can be physiologically mediated from the breeding to the non-breeding period through molting schedules and CORT levels. However, we failed to detect clear effects on migratory behavior or subsequent breeding success, suggesting that costs of breeding in long-lived species may be rapidly buffered during the post-breeding period, as would be expected from life history theory.