Abstract
Animal foraging behaviour is largely determined by prey availability and environmental conditions, as well as inter- and intra-specific population structuring and reproductive requirements. For endangered seabirds such as yellow-eyed penguins (Megadyptes antipodes), understanding how these factors influence their marine distribution is critical for effective conservation management. The research presented in this thesis is the first comprehensive spatial analysis of the interactions between yellow-eyed penguins and their marine habitat. I aimed to quantify the space use and foraging range of yellow-eyed penguins in mainland New Zealand and to identify the oceanographic, biotic, and anthropogenic factors that might drive dispersal and occupancy in marine areas. Using a dataset of dive locations from 135 individuals collected since 2006, I compared marine home range size and predicted habitat suitability for adult yellow-eyed penguins across five distinct regions off the coast of the South Island by applying current mapping and modelling methods, including adaptive local convex hull estimators, fixed kernel densities, Maxent species distribution models, and overlap analyses. I also determined penguin distribution overlap with prey occurrence, commercial fisheries, and marine protected areas.
There were significant differences in range size associated with different colony locations and also in predictions of habitat selection, occurrence, and variable importance between geographical regions. Compared to their conspecifics from central breeding colonies, penguins breeding at the northern and southern extremes of their mainland distribution foraged across larger home ranges, with water current velocity and mean monthly turbidity being the most important variables to Maxent predictions, respectively. Although dispersal was limited by distance to the nearest breeding colony over the continental shelf, dissolved oxygen concentration was the most important variable in models predicting their South Island distribution and specifically across central regions adjacent to the highest number of breeding colonies. There was no clear distinction in foraging range size between the breeding, premoult, and winter seasons, as variation differed depending on breeding site. Although males had significantly larger home ranges than those of females, the home ranges of male and female penguins from most breeding colonies overlapped significantly.
As expected, the distribution of yellow-eyed penguins was largely influenced by predictors related to prey availability, and there was significant spatial overlap between penguins and their prey species (whose distributions were also modelled using Maxent), especially over the mid-continental shelf near key breeding colonies. However, there was also high overlap with commercial gillnet fishing intensity in these areas, increasing the risk of penguin bycatch. Penguins predicted to forage in areas of high trawling intensity, particularly to the north and south of their range, are more likely to be impacted by seafloor habitat and community disturbance due to bottom trawling. Although marine protected areas are a primary tool to reduce the effects of anthropogenic activities, including fishing, less than 1% of the mainland range of yellow-eyed penguins currently overlaps with a marine reserve, and a proposed network off coastal Otago will only cover 3.6% of their marine range. By testing previous hypotheses about the at-sea distribution of endangered yellow-eyed penguins and assessing threats to their survival, these findings can be used to inform ongoing conservation management decisions and marine spatial planning to prevent the predicted localised extinction of this endangered species.