Long-term multi-species demographic studies reveal divergent negative impacts of winter storms on seabird survival.

Understanding storm impacts on marine vertebrate demography requires detailed meteorological data in tandem with long-term population monitoring. Yet most studies use storm proxies such as the North Atlantic Oscillation Index (NAOI), potentially obfuscating a mechanistic understanding of current and future risk. Here, we investigate the impact of extratropical cyclones by extracting north Atlantic winter storm characteristics (storm number, intensity, clustering and wave conditions) and relating these with long-term overwinter adult survival of three long-lived sympatric seabirds which winter at sea-common guillemot Uria aalge, Atlantic puffin Fratercula arctica and razorbill Alca torda. We used multidecadal mark-recapture analysis (1970s-2020s) to estimate survival while correcting for resighting probability, combined with spatially explicit environmental data from geolocation-derived wintering areas, to determine the impact of different storm characteristics (i.e., number, intensity, duration, gap between storms, wave height and wind speed), as well as broad-scale climatic conditions (NAOI and sea surface temperature [SST]). All three species experienced rapid population growth over the study period. Guillemot and razorbill survival was lower during stormier winters, with an additive effect of summer SST for guillemots, and a negative interaction with population size for razorbills. Puffin survival was negatively correlated with winter SST, and the lowest puffin survival coincided with intense winter storms and a large seabird wreck in 2013/14. The number of days with wind speed >30 and 35 ms^-1 negatively impacted razorbill and guillemot survival, respectively, and puffin survival was higher when gaps between storms were longer. Our results suggest negative but divergent storm impacts on these closely related sympatric breeders, which may be compounded by warmer seas and density-dependence as these populations return to their previously much larger sizes. We tentatively suggest that frequent, long-lasting storms with strong winds are likely to have the greatest negative impact on auk survival. Moreover, we highlight the possibility of tipping points, where only the most extreme storms, that may become more frequent in the future, have measurable impacts on seabird survival, and no effect of NAOI.