Rationale
(Changed due to Criteria revision)
The regional population of Great-winged Petrel Pterodroma macroptera satisfies the criterion for Vulnerable under D2 (population with a very restricted area of occupancy or number of locations (typically five or fewer) such that it is prone to the effects of human activities or stochastic events within a very short time period in an uncertain future, and is thus capable of becoming Critically Endangered or even Extinct in a very short time period). However, it was felt that it was unlikely that stochastic events or human activities at the Prince Edward Islands would lead to this species becoming Critically Endangered or even Extinct within one to two generations although there was a possibility that such activities could result in this species being listed as Vulnerable or Endangered during that period. As a result, Great-winged Petrel was assessed as regionally Near Threatened.
Distribution
The Great-winged Petrel has a circumpolar distribution in the Antarctic zone (Murphy 1936) and is widespread in the Southern Ocean, breeding on Tristan da Cunha, Crozet Island, Western Australia, North Island (New Zealand), Kerguelen Island and the Prince Edward Islands (Warham 1956). Non-breeding birds disperse northwards into temperate and subtropical waters (Ryan 1997) being present in South African waters predominantly between November and March. The Great-winged Petrel is common in oceanic waters and over the shelf-break (Ryan 2005).
Population
The global population is estimated to be 450 000-700 000 pairs (Brooke 2004). The distribution of petrel burrows on Marion Island is not uniform (Schramm 1983) which makes it extremely difficult to extrapolate from small sample plots and obtain an accurate population estimate. Within South African waters, Crawford et al. (1991) estimated there to be 35 000 individuals in summer and 3 500 in winter. Confidence in the regional population estimate for the Prince Edward Islands population is low.
Population trend
The global population is suspected to be in decline owing to predation by invasive species and unsustainable levels of exploitation. Breeding success at Marion Island increased from 0-20% to 60-64% following cat eradication (Cooper et al. 1995). Confidence in this regional population trend estimate is medium.
Threats
The primary threat to Great-winged Petrels on Marion Island was predation by feral house cats, which caused the species to be listed as Vulnerable by Brooke (1984). van Aarde (1980) estimated that c. 48 000 Great-winged petrels were killed annually by cats at Marion Island. The species has been recorded being caught by longline and trawl fisheries in South African waters (Petersen et al 2009a,b). Being a winter breeder makes this species particularly susceptible to predation by introduced house mice Mus musculus (Wanless et al. 2007, 2012) although evidence of mouse impacts on this species are lacking. The unexpectedly slow recovery of the Marion Island population after cat eradication is believed to be due to mice preying on chicks (PG Ryan pers. comm).
Conservation
Underway
No conservation measures that focus specifically on the Great-winged Petrel are currently in place. However, its breeding grounds on the Prince Edward Islands are listed as a Ramsar Wetland of International Importance, and as a Special Nature Reserve under the Environment Conservation Act (No 73 of 1989) and managed under the Prince Edward Islands Management Plan 1996. The Prince Edward Islands Marine Protected Area was proclaimed in 2013.
Proposed
Eradication of mice from Marion Island is expected to have a direct benefit (improved breeding success through elimination of predation on chicks) and possibly indirect effects, through generally restoring Marion Island's ecosystem functioning. This might be mediated through recovery of all burrowing petrel numbers, which could then reduce the relative impacts of predation by Subantarctic Skua Stercorarius antarcticus on Great-winged Petrel adults and fledglings.
Research
* Confirmation of breeding at Marion Island remains to be achieved and this should be considered a major priority for the species. Identifying whether it breeds individually or in loose colonies in the region would indicate the scale of the task required to estimate the breeding populations.
* Virtually nothing is known about many basic demographic parameters for this species at the Prince Edward Islands. Knowledge of breeding success, recruitment parameters, longevity, breeding frequency, sex-based differences in parental investment, age/sex structure, would be useful for modelling population trends; currently all this information would have to be inferred from other areas or sister taxa.
* Levels of connectivity and meta-population dynamics between any Prince Edward Island colonies and elsewhere should be assessed. Movements of this species during and outside the breeding season are also unknown, including the geographical range of juveniles.
