Diet and breeding productivity in European Shag (Gulosus aristotelis): insights from two Portuguese colonies
Vieira, B.; Goncalves, D.; Oliveira, N.
Show abstract
Climate change and anthropogenic pressures are reshaping marine food webs, altering prey availability and affecting top predators. The European Shag (Gulosus aristotelis), a coastal demersal seabird, provides a valuable model for examining environmentally mediated dietary variation, given its trophic plasticity and capacity to adjust prey use according to local availability, while also allowing assessment of potential demographic consequences. This study investigated spatial and temporal variation in diet at two Portuguese colonies (Berlengas and Arrabida) between 2016 and 2024 and assessed long-term reproductive productivity at Berlengas. A total of 468 regurgitated pellets were analysed, and diet composition was quantified using the Index of Relative Importance (IRI). Generalised additive models were applied to assess environmental, spatial, and period-specific effects on diet composition, while reproductive productivity was modelled in relation to prey biomass. Diet variation was primarily explained by environmental predictors, including sea surface temperature, chlorophyll-a concentration, and zooplankton, whereas year per se had no significant effect, indicating environmentally mediated bottom-up effects. Spatial differences between colonies reflected contrasting prey field structures, and period-specific patterns suggested increased specialisation during breeding. Higher biomass of sandeels (Ammodytidae) was positively associated with reproductive output, whereas shifts toward lower-energy prey were associated with reduced productivity. These findings demonstrate that environmentally driven dietary change has measurable demographic consequences, underscoring the importance of bottom-up processes in shaping seabird population dynamics and informing conservation strategies under ongoing climate change.
Matching journals
The top 8 journals account for 50% of the predicted probability mass.