Foraging distribution of a tropical seabird supports Ashmole's hypothesis of population regulation

Many animals reproduce in large aggregations, which can vary in size from dozens to millions of individuals across species, time and space. The size of breeding colonies is a complex trade-off between multiple costs and benefits to an individual's fitness, but the mechanisms by which colony size affects fitness are still poorly understood. One important cost of breeding in a large colony is the spatial constraint in resource use due to the need to regularly return to a central location. Large aggregations, like seabird breeding colonies, may therefore deplete food resources near the colony, forcing individuals to travel farther to find food, which may ultimately limit their reproductive output and population size. This hypothesis, proposed in 1963 by Ashmole for tropical oceanic islands, has so far not been tested at tropical seabird colonies, where food availability is less predictable than in colder waters. We compare the foraging distribution of a common tropical seabird, the masked booby Sula dactylatra, breeding on two islands in the South Atlantic that differ in the size of the breeding seabird community by 2 orders of magnitude, but are surrounded by similar oligotrophic waters. Foraging trips from the island with the smaller colony were on average 221 km (61 %) and 18.0 h (75 %) shorter because birds from the smaller colony rarely spent the night at sea and foraged on average 64 km (46 %) closer to the colony. Energy expenditure was significantly lower, and nest survival higher (47 vs. 37 %, n = 371) on the island with the smaller colony. These results are fully consistent with the predictions from Ashmole's hypothesis and indicate that competition for food around tropical oceanic seabird colonies may indeed be a limiting factor for populations. Identifying important feeding areas for seabirds based on their foraging range may need to account for colony size of both the target and potential competitor species.