The Yukon-Kuskokwim Delta of Alaska is a globally important region for numerousavian species including millions of migrating and nesting waterbirds. Climate change effectssuch as sea level rise and increased storm frequency and intensity have the potential to impactwaterbird populations and breeding habitat. In order to determine the potential impacts of theseclimate-mediated changes, we investigated both short-term and long-term impacts of stormsurges to geese and eider species that commonly breed on the Yukon-Kuskokwim Delta. Todetermine short-term impacts, we compared nest densities of geese and eiders in relation to themagnitude of storms that occurred in the prior fall from 2000–2013. Additionally, we modeledgeese and eider nest densities using random forests in relation to the time-integrated flood index(i.e., a storm-specific measure that accounts for both water depth and the amount of time theflooding occurred on the landscape) for four modeled storms (i.e., 2005, 2006, 2009, and 2011),as well as other environmental covariates. To determine long-term impacts, we modeled geeseand eider nest densities using random forests in relation to the annual inundation index (i.e., atime-static storm footprint calculated based on the time-integrated flood index for seven modeledstorms and their annual return rate), as well as other environmental covariates over a longer timeframe (1985–2013). We failed to find any short-term or long-term impacts of storms on nestinggeese and eiders, with storm magnitude, the time-integrated flood index, and the annualinundation index explaining little of the variation in geese and eider nest densities. Rather, otherenvironmental variables such as distance to coast appeared to be more influential to both annualand long-term nest densities. The sampling design and the limited availability of inundationprojections may have precluded us from finding a storm effect if one existed. For example, themonitoring design in which plots were surveyed once for waterbird nests then specifically notrevisited the next five years precluded a more focused assessment comparing spatial distribution of nest densities immediately before and after specific storms. Additionally, the temporal andspatial scale of the nest density and storm surge data may have been inadequate to detect trendsif they existed. Future studies should implement more targeted sampling designs to determine ifthe apparent lack of an effect is real or simply reflects limitations of the sampling design, as wellas investigate other demographic parameters (e.g., clutch size, nest success, fledgling success)that may be more impacted by storm surges.
