The role of local conditions and carry-over effects from migration on the arrival-breeding interval in a long-distance migrant

Background/Question/Methods

As the climate continues to change rapidly, it is vital to study if species are able to acclimate fast enough to their new environmental conditions. While most focus on avian species has been on determining if spring arrival dates or egg-lay dates are adequately advancing; few studies have looked at the flexibility of the time from when a bird first arrives at breeding grounds until they lay their first egg, termed the ‘arrival-breeding interval’. The aims of this study are to determine how the length of the arrival-breeding interval in a long-distance migratory songbird (purple martin, Progne subis) is influenced by: (i) local conditions experienced during the interval, (ii) carry-over effects from spring migration, and (iii) breeding latitude. We will also determine (iv) whether the arrival-breeding interval influences breeding success. The dataset contains 124 purple martins located in 8 states/provinces throughout their eastern breeding range from Florida to Alberta. Arrival date, migration rate, and stopover duration were found using migration tracks derived from light-level geolocators. First egg-lay date and number of young fledged for each tracked bird were determined through nest monitoring at colonies. Local temperature and precipitation data were obtained from the closest weather station to each breeding colony.

Results/Conclusions

We found the arrival-breeding interval to vary widely, ranging from 3 to 63 days long with a mean and standard deviation of 28.0 ± 11.1 days. Preliminary results (using linear mixed models) indicate that when local average temperatures during the interval were warmer, the duration of the interval shortened (βTemp =-1.9 ± 0.4, CI =-2.6 to -1.1). There was no detectable effect of precipitation on the interval length. Faster migration rates (βMigration =0.02 ± 0.01, CI = 0.01 to 0.04) and longer stopovers during migration (βStopover =11.4 ± 5.2, CI =1.3 to 21.6) slightly resulted in longer intervals. Individuals living in southern colonies had longer intervals on average than individuals in northern colonies. While adult purple martins had longer intervals than subadults. The length of the arrival-breeding interval, however, did not have a significant effect on breeding success. Altogether this study highlights the flexibility within the arrival-breeding interval which may enable birds to shorten their interval with increasingly warmer springs; potentially reducing the impacts of phenological mismatch with peak prey abundance. However, the degree to which adjustments to the arrival-breeding interval may allow migrants to keep up with their rapidly changing environment could be further explored in future studies.