CCAMLR

Breeding grey-headed albatross Thalassarche chrysostoma, tracked from Marion Island (Prince Edward Islands) during November–December 1997 and January–February 1998, showed a strong association with mesoscale oceanographic features, as identified by sea surface height anomalies, in the southern Indian Ocean. During incubation, most birds foraged to the north of the island, at the edges of anomalies created by the Agulhas Return Current in the Subtropical Convergence and the Subantarctic zones. In contrast, during chick-rearing all tracked birds foraged to the southwest of the island, at the edges of anomalies along the South-West Indian Ridge. Previous work in this area has shown that these anomalies are in fact eddies that are created as the Antarctic Circumpolar Current crosses the South-West Indian Ridge. Diet samples taken during the chick-rearing period showed a predominance of fresh specimens of the predatory fish Magnisudis prionosa and the squid Martialia hyadesi. Myctophid fish and amphipods Themisto gaudichaudii, both known prey of M. hyadesi, were also well represented in our samples. Diet samples taken from tracked birds showed birds feeding at edges of positive anomalies returning with fresh specimens of M. prionosa and M. hyadesi. Predatory fish and squid are thus presumably concentrated at these features. Eddies formed at the South-West Indian Ridge have also been shown to drift closer to Marion Island, within the foraging range of penguins and seals breeding on Marion Island. We therefore suggest that these mesoscale oceanographic features may be an important component of the ‘life-support’ system enabling globally significant populations of seabirds and seals to breed at the Prince Edward Islands.

We report on the population changes of five species of Procellariiform seabirds breeding at Marion Island: wandering albatross (Diomedea exulans), grey-headed albatross (Thalassarche chrysostoma), northern giant petrel (Macronectes halli), southern giant petrel (M. giganteus), and white-chinned petrel (Procellaria aequinoctalis). The four large surface-nesting species (wandering albatross, grey-headed albatross and the northern and southern giant petrel) show similar population trends over the last 16-18 years. All were stable or decreasing during the 1980s, followed by a recovery period during the early to mid-1990s. Recently, all species have once again stabilized or decreased in numbers. Population trends of wandering albatross at Marion Island were strongly correlated with other Indian Ocean populations, but were different from the Atlantic Ocean population. These similarities suggest a common cause and can be explained by changes in tuna (Thunnus spp.) longline fishing effort in the southern Indian Ocean. A recent increase in tuna longlining, as well as recent large-scale Illegal, Unregulated and Unreported (IUU) longline fishing for Patagonian toothfish (Dissostichus eleginoides) close to Marion Island, could be contributing to the recent decreases in some of these species. Adoption of mitigation measures and an effective means of dealing with IUU longline fishing seem necessary to reduce the potential impacts of longline fishing on populations of albatross and petrel.

Concerning numbers of Wandering Albatrosses Diomedea exulans are killed when they attempt to scavenge baited hooks deployed by long-line fishing vessels. We studied the foraging ecology of Wandering Albatrosses breeding on Marion Island in order to assess the scale of interactions with known long-line fishing fleets. During incubation and large chick-rearing, birds foraged farther away from the island, in warmer waters, and showed high spatial overlap with areas of intense tuna Thunnus spp. long-line fishing. During small chick-rearing, birds made shorter foraging trips and showed higher spatial overlap with the local Patagonian toothfish Dissostichus eleginoides long-line fishery. Tracks of birds returning with offal from the toothfish fishery, showed a strong association with positions at which toothfish long-lines were set and most diet samples taken during this stage contained fisheries-related items. Independent of these seasonal differences, females foraged farther from the islands and in warmer waters than males. Consequently, female distribution overlapped more with tuna long-line fisheries, whereas males interacted more with the toothfish long-line fishery. These factors could lead differences in the survival probabilities of males and females. Non-breeding birds foraged in warmer waters and showed the highest spatial overlap with tuna long-line fishing areas. The foraging distribution of Marion Island birds showed the highest degree of spatial overlap with birds from the neighbouring Crozet Islands, during the late chick-rearing and non-breeding periods. These areas of foraging overlap also coincided with areas of intense tuna longline fishing, south of Africa. As the population trends of Wandering Albatrosses at these two localities are very similar, it is possible that incidental mortality during the periods when these two populations show the highest spatial overlap could be driving these trends.

Sanctioned longline fishing for Patagonian toothfish around the Prince Edward Islands (a globally important seabird breeding site) commenced in 1996 following high levels of Illegal, Unregulated and Unreported (IUU) fishing. Independent fishery observers accompanied all but two sanctioned trips between 1996 and 2000, allowing a thorough excellent understanding of the impact of this fishery on seabirds. Overall, white-chinned petrels Procellaria aequinoctalis were by far the most frequently killed species. However, during the first year, when lines were set during the day and night, a significant number of albatrosses (particularly grey-headed albatrosses Thalassarche chrysocome) were also killed. Birds were caught almost exclusively during their breeding seasons and most birds killed were breeding adult males from the Prince Edward Islands. Albatrosses and giant petrels were caught almost exclusively during day sets, whereas catch rates of white-chinned petrels did not differ between day and night sets. Albatrosses were caught closer to the islands than white-chinned petrels. Most white-chinned petrel carcasses were hooked in their wings and bodies, whereas albatrosses were caught most frequently in their bills. Albatross carcasses also contained large numbers of baits, indicating that fishers may be losing significant potential revenue by allowing these birds to access baited hooks. Rates of seabird bycatch in the sanctioned fishery decreased from 0.19 birds/1000 hooks to 0.034 birds/1000 hooks during the time of this study. This was probably mainly due to stricter implementation of mitigation measures and a progressive movement farther away from the islands over the years. We estimate that the combined impact of sanctioned and IUU longline fishing around the Prince Edward Islands over the past four years, could have resulted in close to 7000 seabird mortalities and could have had significant impacts on the breeding populations of several seabird species breeding on the Prince Edward Islands. This was mainly due to high levels of IUU fishing during 1996/97.