The Arctic Coastal Plain is one of the most important avian breeding grounds in the world; however, many species are in decline. Arctic-breeding birds contend with short breeding seasons, harsh climatic conditions, and now, rapidly changing, variable, and unpredictable environmental conditions caused by climate change. Additionally, those breeding in industrial areas may be impacted by human activities. It is difficult to separate the impacts of industrial development and climate change, however, long-term datasets can help show patterns over time. We evaluated factors influencing reproductive parameters of breeding birds at Prudhoe Bay, Alaska, 2003–2019, by monitoring 1265 shorebird nests, 378 passerine nests, and 231 waterfowl nests. We found that nest survival decreased significantly nearer high-use infrastructure for all guilds. Temporally, passerine nest survival declined across the 17 years of the study, while there was no significant evidence of change in their nest density. Shorebird nest survival did not vary significantly across years, nor did nest density. Waterfowl nest density increased over the course of the study, but we could not estimate nest survival in all years. Egg predator populations varied across time; numbers of gulls and ravens increased in the oilfields from 2003–2019, while Arctic fox decreased, and jaeger numbers did not vary significantly. Long-term datasets are rare in the Arctic, but they are crucial for understanding impacts to breeding birds from both climate change and increasing anthropogenic activities. We show that nest survival was lower for birds nesting closer to high-use infrastructure in Arctic Alaska, which was not detected in earlier, shorter-term studies. Additionally, we show that Lapland longspur nest survival decreased across time, which is concerning considering the continent-wide declines in many passerine species. The urgency to understand these relationships cannot be expressed strongly enough, given change is continuing to happen and the potential impacts are large.

We used generalized linear models and a logit link (Program MARK) to evaluate daily nest survival. We examined a hierarchical model set. Models of daily nest survival varied by year, guild (shorebird, waterfowl or passerine), habitat (wetland, moist, upland), initiation date, and incubation strategy (biparental or uniparental, shorebirds only), distance to infrastructure, area of infrastructure within a 2 km radius of the nest, and distance to the nearest road. We ran all main effects models, then combined ‘strong’ effects additively and multiplicably. We included one post-hoc model to further evaluate the relationship between spring conditions and incubation strategy where we held passerines and waterfowl constant but allowed shorebird daily survival rate to vary by individual year and incubation strategy (DSR_{year*strategy}). We compared models using AIC_c (Akaike’s Information Criterion corrected for small sample sizes), in which we considered the model with lowest AIC_c value to be the best-fitting and models with a ΔAIC_c < 2 that did not add to model complexity to be plausible (Arnold 2010, Burnham and Anderson 2002).

We calculated estimates of nest density (i.e., the cumulative number of nests found throughout the study plots divided by the total area of the plots (km²)) by year, guild, and incubation strategy (shorebirds only).

Program Mark