Density dependence forces divergent population growth rates and alters occupancy patterns of a central place foraging Antarctic seabird

Density‐dependent regulation is an important process in spatio‐temporal population dynamics because it can alter the effects of synchronizing processes operating over large spatial scales. Most frequently, populations are regulated by density dependence when higher density leads to reduced individual fitness and population growth, but inverse density dependence can also occur when small populations are subject to higher extinction risks. We investigate whether density‐dependent regulation influences population growth for the Antarctic breeding Adélie penguin Pygoscelis adeliae. Understanding the prevalence and nature of density dependence for this species is important because it is considered a sentinel species reflecting the impacts of fisheries and environmental change over large spatial scales in the Southern Ocean, but the presence of density dependence could introduce uncertainty in this role. Using data on population growth and indices of resource availability for seven regional Adélie penguin populations located along the East Antarctic coastline, we find compelling evidence that population growth is constrained at some locations by the amount of breeding habitat available to individuals. Locations with low breeding habitat availability had reduced population growth rates, higher overall occupancy rates, and higher occupancy of steeper slopes that are sparsely occupied or avoided at other locations. Our results are consistent with evolutionary models of avian breeding habitat selection where individuals search for high‐quality nest sites to maximize fitness returns and subsequently occupy poorer habitat as population density increases. Alternate explanations invoking competition for food were not supported by the available evidence, but strong conclusions on food‐related density dependence were constrained by the paucity of food availability data over the large spatial scales of this region. Our study highlights the importance of incorporating nonconstant conditions of species–environment relationships into predictive models of species distributions and population dynamics, and provides guidance for improved monitoring of fisheries and climate change impacts in the Southern Ocean.