COS 98-4 - Quantifying the influence of weather conditions on behavioral contributions to reproductive attempts in birds of contrasting migration strategy

Friday, August 16, 2019: 9:00 AM
L007/008, Kentucky International Convention Center
Stephanie A. Cunningham1, Toryn L.J. Schafer2, Jay A. VonBank3, Christopher K. Wikle4, Bart M. Ballard3 and Mitch Weegman1, (1)School of Natural Resources, University of Missouri, Columbia, MO, (2)Department of Statistics and Data Science, Cornell University, Ithica, NY, (3)Caesar Kleberg Wildlife Research Institute, Texas A&M University - Kingsville, Kingsville, TX, (4)Department of Statistics, University of Missouri, Columbia, MO
Background/Question/Methods

Migratory birds make decisions during spring migration to fuel travel and subsequent reproduction. Yet individuals often contend with unfavorable weather conditions en route that hinder prior migration and breeding preparation. The North American midcontinent and Greenland populations of greater white-fronted geese (Anser albifrons) exhibit disparities in migration strategy and population trajectory. While the midcontinent population exceeds 1 million birds and is increasing, the Greenland population numbers 20,000 birds and has declined 43% over the past 25 years. The decline has been due to low productivity. To examine decision making that results in differential productivity, measured as breeding attempt or deferral, we fit 23 midcontinent and Greenland birds with neck collars and 15 Greenland birds with backpacks, each containing Global Positioning System-accelerometer technology. We identified 7 individuals from the midcontinent population and 13 from the Greenland population that initiated a breeding attempt. We used random forest models to classify acceleration data to create activity budgets throughout spring migration. We used Bayesian generalized linear mixed models to examine differences in proportion of time feeding during spring between individuals that subsequently attempted or deferred breeding attempts, and quantified the effects of local weather conditions on proportion of time feeding throughout spring migration.

Results/Conclusions

The proportion of time feeding during spring migration did not explain substantial variation in probability of a breeding attempt in Greenland birds, but the probability of a breeding attempt increased for Greenland birds as variability in the proportion of time feeding increased throughout spring. For midcontinent birds, the proportion of time feeding explained the most variation during the last step of migration as individuals arrived on breeding areas, whereby more feeding yielded a greater probability of a breeding attempt. Increased precipitation resulted in lower proportion of time feeding on arrival to breeding areas, and a significant interaction between population and precipitation suggests that proportion of time feeding decreased faster in Greenland geese. Temperature was positively related to feeding behavior in both populations on arrival to breeding areas, and the significant interaction between population and temperature suggested a more rapid increase in proportion of time feeding in midcontinent birds. These results suggest that midcontinent white-fronted geese showed a broader gradient of behavior leading to breeding attempts, and that local weather, and not only migration strategy, explains some variation in behavior. Quantifying decisions that promote productivity will be critical to improving conservation plans for these and similar migratory birds.