Trans-equatorial migration and non-breeding habitat of tropical shearwaters: implications for modelling pelagic important bird areas

Declining prey availability drives many seabirds to migrate following breeding. While long-distance, latitudinal migrations are common in temperate-breeding species (including temperate-breeding Procellariiformes), regional dispersal or longitudinal migration is more common in tropical-breeding species. We used geolocators to track adult, tropical-breeding wedge-tailed shearwaters Ardenna pacifica from the Great Barrier Reef, Australia, through a ~6000 km migration to non-breeding grounds in Micronesia. This lengthy, trans-equatorial migration was similar to that undertaken by temperate-breeding Procellariiformes, but contrasted with patterns previously observed for tropical-breeding species. However, the oceanographic characteristics of tropical non-breeding habitats differed significantly from those of temperate sites. Core-use habitat had high sea-surface temperatures, very low wind speeds and low primary productivity, features normally associated with poor foraging habitat. However, activity was strongly linked to positive sea-level anomalies, indicating the presence of anti-cyclonic eddies at foraging sites. Such eddies are often associated with oceanic fronts and are known to aggregate micronekton and facilitate sub-surface predator feeding. Consequently, our results suggest that eddies, frontal activity and feeding associations with sub-surface predators enhance prey availability to non-breeding shearwaters beyond levels expected based on standard indices of primary production. This is the first tropical study to simultaneously assess the full set of oceanographic features considered important for modelling pelagic Important Bird Areas (IBA). Our findings demonstrate the need for IBA modelling in the tropics to go beyond standard indices of productivity by including measures of frontal activity and assessments of biological interactions. Consequently, this study provides a framework for better predicting candidate Marine IBAs throughout tropical regions