Heard Island’s seabirds under scrutiny

A census of seabird populations at Heard Island during the 2000–01 summer has provided contemporary data on the distribution and abundance of breeding seabirds in the western two-thirds of the island. The survey, by Eric Woehler and Heidi Auman, collected census and GPS data for seabirds between Cape Arkona in the south-west, parts of the extensive vegetated areas of the Laurens Peninsula in the north-west, and as far as Gilchrist Beach in the north-east. The remainder of Heard Island will be surveyed on the next visit, currently planned for 2003–04.

Predators, such as cats and rats, have been introduced to almost every subantarctic island except Heard Island, and they are known to prey on seabirds. However, without information on seabird populations before human disturbance, it is impossible to fully estimate the damage caused by these predators. Heard Island provides a unique opportunity to understand the population trends and dynamics of subantarctic seabird populations undisturbed by introduced pests.

As seabirds come under additional threats such as long-line fishing, it becomes increasingly important to understand how their populations change over time. On this visit, most of the survey effort was directed towards those species for which long-term data already exist, because building on these data sets is the only way we can begin to understand the variability of long-lived species.

King penguins (Aptenodytes patagonicus) were first surveyed on Heard Island in 1948 and southern giant petrels (Macronectes giganteus) in 1951. For both these species we are now able to see very marked population trends — one species is growing in numbers very rapidly, the other is decreasing (see below). Population surveys may not provide the reason for these changes but without this information we would not even know they were happening. Attention was also given to rockhopper penguins (Eudyptes chrysocome) a species whose numbers are decreasing rapidly elsewhere in the subantarctic, with some populations falling by 90% or more in the last 50 years. In the absence of historical data on rockhopper penguins at Heard Island, it was important to establish reference colonies and obtain baseline population data in case Heard Island's populations of rockhopper penguins also decrease or are already on the decline.

Contemporary data permits an assessment of current population sizes and trends, provides fundamental data for management purposes, including the current revision of the Heard Island Management Plan, and enables the conservation status of several species to be re-assessed. The detailed knowledge of the distribution, abundance and trends of seabirds on Heard Island will also be used for planning future activities on the island, including visits by expeditioners with the Australian Antarctic Program and tourists, to ensure they do not cause harm to the wildlife.

Southern giant petrels

Macronectes giganteus breed on many subantarctic islands. The population at Heard Island was first studied by Max Downes in the early 1950s, and a complete island census was first obtained in 1951. During the 1987–88 ANARE, with Max’s assistance, the island’s breeding population was re-censused. With counts on only two occasions it is impossible to understand the natural variability of the population. However, this work alerted scientists to a halving of the breeding population — from approximately 3500 pairs in 1951 to 1700 pairs in 1987. Similar decreases have been documented in other breeding populations of southern giant petrels around the Southern Ocean. Southern giant petrels are known to be easily disturbed by human activities, and it has been suggested that intensive programs of banding of chicks undertaken at many breeding localities in the 1950s and 1960s may have contributed to the observed population decrease. However, the decrease on Heard Island, where there has been little banding, indicates that this is not the full story. The minimal human presence since the closure of the ANARE station in 1954 effectively rules out human disturbance as a contributing factor to the decrease observed since the 1950s. Southern giant petrels are also frequently caught in longline fisheries in the Southern Ocean and this may now be a contributing factor. Ironically, the lack of banded birds from Heard Island makes it impossible to identify which of the southern giant petrels caught by longliners are from this population. At the same time as the giant petrels were declining, the population of southern elephant seals, Mirounga leonina, at Heard Island has also decreased by 60%. It may be that similar or parallel processes affect both species and that research to understand one may provide insights to help understand population changes in the other species. Future research may show there are previously unknown links between the species. However, without simple count data these enormous changes in population sizes would have gone unnoticed.

Long-term population recovery of king penguins

The sealers on Heard Island during the 19th century made use of king penguins (Aptenodytes patagonicus) as food and fuel. While no accurate population data were ever collected, it is believed that the impact of these sealers on the local colonies was so great that they were nearly wiped out. In 1948, during the first ANARE to Heard Island, a small breeding colony of king penguins was recorded at Pageos Moraine, near the Atlas Cove station. Since then, most visits to Heard Island have made attempts to census the breeding colonies of king penguins scattered around the coastline. With seven colonies known in 1992, the population has exhibited remarkable growth (>20% annually), doubling every five to seven years (see plot). Preliminary results from the current survey indicate that this rapid increase is continuing, and that the breeding population at Heard Island shows no sign of reaching its limit. All king penguin populations at other breeding localities in the subantarctic for which long term data are available also show this dramatic increase. The causes for this rapid, sustained and widespread increase remain unknown.

The Heard Island cormorant is more common than thought

A previously unknown breeding locality of the Heard Island Cormorant (Phalacrocorax nivalis) was discovered in early November 2000. The species was listed as ‘vulnerable’ under IUCN criteria due to its endemism (breeding only at Heard Island) and small known breeding population, estimated at between 90 and 200 pairs. Three colonies were known: Red Island in the northwest — about 30 pairs; Saddle Point, on the central north coast — about 80 pairs; and Stephenson Lagoon on the northeast coast — about 100 pairs. The sites are difficult to access and had never before been counted simultaneously so an accurate estimate of the total breeding population has been difficult to determine. Biologists were puzzled by observations of more than 600 roosting birds, which is more than the known breeding population. On 2 November, a large colony of 1000 nests was discovered at Cape Pillar, on the remote and rarely visited southwest coast of Heard Island. The cormorant colony was on the western periphery of a macaroni penguin Eudyptes chrysolophus colony (about 25,000 pairs), and as such, very easy to overlook on aerial photographs or aerial inspections. A series of overlapping oblique (ground) photographs were taken to enable a more accurate count, and to supplement the ground counts. Approximately 100 to 200 roosting cormorants were also present at the fringes of the nesting areas. A revised estimate of total breeding population will be made at the end of the season on Heard Island following visits to all colonies. The population has almost certainly always been there but has been previously overlooked. The discovery that the population is larger than previously believed may not change its IUCN classification as 'vulnerable' as the breeding population is found only at Heard Island. However, the larger population means that the species is more likely to survive in the long-term.

The mysterious migrations of Antarctic terns partially explained

At Bird Island (33°50’S 26°17’E) in Algoa Bay off Port Elizabeth in South Africa, there is a site that is globally important for Antarctic terns (Sterna vittata). More than 10,000 terns roost there nightly during the winter months. Biologists have not known where these birds breed, so between July 1998 and September 2000, two South African ornithologists captured and banded more than 1,000 Antarctic terns and marked 600 of these with bright yellow leg bands to make them easier to spot. There is a small breeding population of Antarctic terns on Heard Island, estimated at less than 100 pairs. They are absent from Heard Island for the winter months and South Africa had previously been suggested as a potential wintering area, although there was no hard evidence to support this.

On 31 December 2000, one of six Antarctic terns feeding in the surf and sitting on the shoreline of Atlas Cove was seen to have a yellow band on its left leg. On 2 January 2001, two of approximately 80 Antarctic terns feeding in the surf at Corinthian Bay, approximately 1km east of Atlas Cove, were observed to have yellow bands, and another bird was seen to have a metal band only. A search at Jacka Valley, approximately 7km from Atlas Cove on 15 January 2001, and a known breeding locality for terns, was successful. Of 20 nest scrapes located in the colony, two belonged to birds with bands. Heard Island is approximately 4300km from Bird Island, South Africa, and these sightings provide the first evidence of migration by breeding Heard Island Antarctic terns to South African wintering areas.

Eric Woehler, Heidi Auman & Martin Riddle
Human Impacts Research Program,
Australian Antarctic Division