Apparent survival of an Arctic‐breeding migratory bird over 44 years of fluctuating population size

• Published in: Ibis (London, England) England), 2018-04, Vol.160 (2), p.413-430
• Main Authors: Wood, Kevin A., Nuijten, Rascha J. M., Newth, Julia L., Haitjema, Trinus, Vangeluwe, Didier, Ioannidis, Panagiotis, Harrison, Anne L., Mackenzie, Conor, Hilton, Geoff M., Nolet, Bart A., Rees, Eileen C.
• Format: Article
• Language: English
• Subjects: Age factors; Animal breeding; animal demographic rates; Bewick's Swan Cygnus columbianus bewickii; Birds; Breeding; Collars; Confidence intervals; Demographics; Dynamics; Emigration; Environmental changes; Food resources; global environmental change; herbivorous waterbirds; life history; mark recapture data; Migratory birds; Neck; Overwintering; Polar environments; Population; Population decline; Population dynamics; Population number; Survival; Variation; Waterfowl; Weather
• ISSN: 0019-1019; 1474-919X
• DOI: 10.1111/ibi.12521

Following increases in numbers during the second half of the 20th century, several Arctic‐breeding migrant bird species are now undergoing sustained population declines. These include the northwest European population of Bewick's Swan Cygnus columbianus bewickii, which declined from c. 29 000 birds on the wintering grounds in 1995 to 18 000 in 2010. It is unclear whether this decrease reflects reduced survival, emigration to a different area, or a combination of both. Furthermore, the environmental drivers of any demographic changes are also unknown. We therefore used an information‐theoretic approach in RMark to analyse a dataset of 3929 individually marked and resighted Bewick's Swans to assess temporal trends and drivers of survival between the winters of 1970/71 and 2014/2015, while accounting for effects of age, sex and different marker types. The temporal trend in apparent survival rates over our study period was best explained by different survival rates for each decade, with geometric mean survival rates highest in the 1980s (leg‐ring marked birds = 0.853, 95% confidence interval (CI) 0.830–0.873) and lowest in the 2010s (leg‐ring = 0.773, 95% CI 0.738–0.805; neck‐collar = 0.725, 95% CI 0.681–0.764). Mean (±95% CI) resighting probabilities over the study period were higher for birds marked with neck‐collars (0.91 ± 0.01) than for those marked with leg‐rings (0.70 ± 0.02). Weather conditions in different areas across the flyway, food resources on the winter grounds, density‐dependence and the growth of numbers at a relatively new wintering site (the Evros Delta in Greece) all performed poorly as explanatory variables of apparent survival. None of our 18 covariates accounted for more than 7.2% of the deviance associated with our survival models, with a mean of only 2.2% of deviance explained. Our results provide long‐term demographic information needed to help conservationists understand the population dynamics of Bewick's Swans in northwest Europe.