Case studies: Rabbit eradication on Australia’s offshore islands

Rabbits have been introduced to over 800 islands around the world. They are present on 14 of the top 100, high conservation status offshore islands worldwide and have caused significant damage on islands where they have successfully established wild populations, including Australia. Where possible, their eradication is recommended. About 48 rabbit eradication attempts have been carried out globally, with around a 5% failure rate. In Australia, there have been 22 successful rabbit eradication programs on islands. Three examples presented here are:

  • Phillip Island off Norfolk Island, South Pacific
  • Cabbage Tree Island, Montague Island and Broughton Island off New South Wales
  • Macquarie Island, Tasmania.

To achieve eradication, a combination of techniques is often needed to overcome unique challenges such as rugged and isolated terrain, remoteness, effects on non-target species, and high costs. A number of other conditions must be met for an eradication program to be successful, including:

  • good planning, and extensive community and stakeholder consultation
  • commitment to finishing the program
  • making sure the entire population of the target species is exposed to control
  • removing the target species faster than they reproduce
  • preventing re-invasion.

Although challenging, eradication is often feasible on small islands and can be more cost- effective than long term sustained control. Aim: To permanently remove (eradicate) rabbits from islands with high conservation values, and attempt to restore biodiversity and natural processes as much as possible.

Example 1: Phillip Island (Norfolk Island), South Pacific

Rabbits, feral goats and feral pigs were introduced to Phillip Island during the early colonial settlement of nearby Norfolk Island. Introduced herbivores in particular, caused severe degradation of the ecosystem. Goats and pigs were shot or died out however rabbits survived. Some areas were fenced to exclude rabbits, and the vegetation within them survived and grew quickly. Wildlife, including Sooty Terns, also increased inside the exclosures.

In 1981, island managers (Parks Australia staff in consultation with Norfolk Island Council) attempted to eradicate rabbits using the highly virulent Lausanne strain of myxoma virus. Trapped rabbits were injected with the virus and released back into the wild. Some areas of the island were only accessible by sea or were not accessible at all, so European rabbit fleas carrying myxoma virus were introduced aerially using a bow and arrows shot from adjacent cliff tops. Fleas in a glass vial were fitted onto the steel arrowhead which broke on impact, releasing the fleas into the environment. Another inaccessible area was accessed by staff who abseiled and climbed onto the upper part of a 250 m cliff. Overall, this approach was very effective at reducing rabbit numbers however when the virus and fleas were no longer available, rabbits increased once again.

A new strategy was introduced in 1983 using a best practice approach. Poisoning with 1080 (sodium monofluoroacetate) oat baits was the main method used, supplemented by more intensive techniques such as trapping, fumigating, shooting and warren destruction. Because of the rough terrain and inaccessibility of many baiting sites, this was a long and difficult process. Nevertheless, rabbits were successfully eradicated by 1988.

Example 2: Cabbage Tree Island, Montague Island and Broughton Island, New South Wales

Successful rabbit eradication programs have been carried out by the NSW National Parks and Wildlife Service (part of the Office of Environment and Heritage) – with the help of collaborating agencies – on several islands off the coast of New South Wales, including Cabbage Tree Island (30 ha), Montague Island (82 ha) and Broughton Island (144 ha). These particular islands are small and heavily vegetated, and had been badly damaged by rabbit grazing. Nesting seabirds such as the endangered Gould’s petrel on Cabbage Tree Island were also at risk because of vegetation loss and environmental changes being caused by rabbits.

Rabbits were eradicated from Cabbage Tree Island using two biological control agents and poison baiting with brodifacoum. Myxoma virus was already present on the island, so in early 1997, European and Spanish rabbit fleas were introduced to aid its transmission and effectiveness. Although many rabbits died from myxomatosis, some remained unaffected and some infected rabbits recovered. After the natural outbreak of myxomatosis had subsided, rabbit haemorrhagic disease virus (RHDV) was deliberately introduced via injection of seven captured rabbits that were then released in August 1997. Trapping results showed that 55% of the remaining population of susceptible rabbits died from RHD within 3 weeks. In September 1997, 300 kg of poisoned bait (Talon ® 20P) was spread by helicopter across the island. All surviving rabbits died from poison baiting.

Eradication programs on Montague Island and Broughton Island aimed to remove rabbits as well as rodents. RHDV was already present on Montague Island, probably due to natural spread, but was deliberately introduced via carrot bait to Broughton Island. In both cases, biocontrol was followed by two aerial bait drops using brodifacoum (Pestoff 20R cereal bait) at a rate of 12 kg per hectare. Rabbits were successfully eradicated on both islands.

Example 3: Macquarie Island, Tasmania

Damage caused by warren building on Macquarie Island. Image: Australian Antarctic Division

On Macquarie Island, 1500 km off Tasmania, rabbits had severely damaged the landscape and impacted on native plants and wildlife. Parts of the island had been degraded by erosion and soil loss after major plant species were significantly over-grazed by rabbits (right). In 2006, two landslides partially caused by rabbit grazing and warren building killed penguins and damaged visitor boardwalks on the island. Grazing impacts also affected many of the island’s seabirds, including burrowing petrels, which rely on tussock vegetation cover for shelter and nesting. Because the island is listed as a World Heritage site, eradication of rabbits (and rats) was the highest conservation priority for the reserve.

In 2004, Macquarie Island’s managing agency the Tasmanian Government Parks and Wildlife Service (PWS), in collaboration with the Australian Government and the Australian Antarctic Division (AAD), began the largest pest eradication program of its kind in the world to date. Although vertebrate pest management has been carried out on the island since the 1960s, rabbits have persisted and numbers have fluctuated over time. Their success is thought to be a result of:

  • resistance to the myxoma virus, which was introduced in the 1970s
  • a high abundance of food
  • the removal of feral cats in the late 1990s – although there was some disagreement about the effect this had on rabbit numbers.

A major obstacle in the early stage of the project was a delay in the commitment of funds to carry out the program. Thorough planning was needed to overcome some of the anticipated challenges, such as the island’s remoteness, the long lead time required for training the rabbit detection dogs, and obtaining the various permissions needed to perform the eradication activities. During the implementation phase, other challenges included poor weather conditions and the lack of vessel availability to ship staff to the island.

At first, RHDV was ruled out as a possible control method on Macquarie Island due to the cold, wet conditions. However RHDV was reconsidered after the first round of aerial baiting resulted in a number of non-target deaths in 2010. A strategic virus release was carried out in February 2011 in an attempt to reduce rabbit numbers as much as possible before the next aerial baiting phase, and to minimise the risk of secondary poisoning from toxic rabbits by scavenging non-target birds. Virus-laden grated carrot was distributed by hand at selected high rabbit density sites. Some rabbits were also captured and injected with the virus, then released. RHDV spread rapidly around the island, even though release sites were kilometres apart.

After biocontrol, cereal baits (Pestoff 20R) containing a small dose of brodifacoum (0.002%) were distributed aerially across the whole island, targeting rabbits and rodents. Bait was spread from buckets slung underneath helicopters flying along predetermined routes, with an 80 m-wide swath, which was overlapped by 50% to ensure accurate bait coverage at the prescribed rates.

Although most poisoned rabbits died in their burrows, some rabbits died above ground. To reduce the risk to scavenging birds as much as possible, a field team searched for and buried any dead rabbits they found, and removed carcasses of birds that had died from primary or secondary poisoning. The field teams also carried out follow-up control using specialized techniques such as shooting, fumigation, baiting, rabbit detection dogs and trapping.


Rabbits were successfully eradicated from each of the NSW offshore islands and Macquarie Island using biological control followed by baiting. One example of a positive outcome has been regeneration of the rainforest understorey and an increased density of herbs and grasses on Cabbage Tree Island.

The release of RHDV at 15 high rabbit density sites on Macquarie Island was also successful at killing many rabbits, and successfully reduced non-target mortality island-wide. Two rounds of aerial baiting were completed across the entire island as of June 2011 and no rabbits have been sighted since December 2011.

What worked

  • Each program had a clear objective to eradicate rabbits.
  • Careful planning and execution of the eradication strategy, with an adaptive management approach to changing situations (eg there was extensive peer review of the planning stages of the Macquarie Island project, to maintain international best practice for aerial operations targeting both rabbits and rodents).
  • Wildlife breeding seasons limit the time available for aerial baiting so good planning and timing of operations is vital to the program’s success.
  • Committed teams, with local knowledge of the island, who understood the eradication objective and worked towards that goal.
  • Learning from previous eradications worldwide, sharing knowledge and building local capacity by adopting proven techniques.
  • Having the Australian Government’s commitment to fund the multi-year Macquarie Island project from the start allowed project managers to secure contractors and suppliers as required, without having to wait for money to become available through annual funding rounds.
  • In all cases, biological control was highly effective at reducing the rabbit population in the early stages of eradication – but only where it was used in conjunction with other methods.
  • A skua scavenges a rabbit that has died from RHD on Macquarie Island. RHD is harmless to other wildlife and presents no risk of secondary poison to scavengers. Image: Australian Antarctic Division
    Using RHDV on Macquarie Island resulted in fewer rabbits being left to die from baiting, which meant a reduction in secondary poisoning among non-target species (Figure 3).
  • Hunting dogs were used to flush out surviving rabbits on Broughton Island and Macquarie Island. Specially trained dogs were taught to detect rabbits but not to harm native animals such as penguins or seals.
  • The time put into training dogs to specific standards for rabbit detection and non-target aversion had positive results (ie no dogs killed non-target species).

What didn’t work

  • Due to the remote location of Macquarie Island, access was only by ship. Logistical issues in 2010 meant that the eradication team did not arrive on Macquarie Island until late May. This delayed baiting until June, and meant that staff were forced to work with shorter daylight hours and were faced with unsuitable weather conditions for aerial baiting. Ultimately, this led to an unsuccessful baiting round in the time available and costs to defer the project by 12 months.
  • Although it was expected, the impact of baiting on Macquarie Island’s non-target species was considerable. Several hundred birds (including threatened species) died as a result of secondary poisoning from eating bait or poisoned rabbit carcasses after the first round of baiting.
  • No formal monitoring was undertaken on Phillip Island, so assumptions were made about the early impact of rabbits, the size of the rabbit population and the costs of the eradication programme.
  • Limited funding and access to Phillip Island along with extremely difficult terrain and inaccessible areas meant that new approaches were needed. Finding skilled staff, rock climbing, landing through surf in inflatable boats and high seas were some of the unusual challenges faced. It is unlikely that these techniques would be used in future due to the high safety risks involved.
  • The long-term response of flora and fauna on Australian islands after a rabbit eradication program is largely unknown (eg the removal of rabbits on Bowen Island resulted in an increase of exotic kikuyu grass).
  • The degree of innovation and responsiveness required in vertebrate pest eradications (especially large scale projects) is often stymied by government processes under which most of these projects are managed, leading to time-consuming and sometimes unproductive outcomes. Good planning and risk management can be a way of overcoming this issue.


It is evident that if rabbits continued to exist and cause damage on these islands, some seabirds and unique plant species would no longer be able to survive there. Removing rabbits (and other vertebrate pests) is expected to provide significant long-term protection and benefits to island ecosystems.

Suitable technology and methodology is now available for considering rabbit eradications on larger islands, and to help progress conservation goals of restoring biodiversity. On Macquarie Island, the short-term impact of poison baiting on non-target bird species is expected to be significantly outweighed by benefits to native species, especially if rodent eradication is also successful. Ongoing monitoring of islands from which rabbits have been eradicated will be crucial to determine the long-term effects of rabbit removal on these sensitive environmental systems.

Author Invasive Animals CRC
Year 2014
Publisher Invasive Animals CRC
Pages 4 pp
ISBN/ISSN PestSmart code: GENCS3
Control method Integrated Pest Management
Region Australia - national

GENCS3 PestSmart Case study: Rabbit eradication on Australia’s offshore islands   [460kb PDF]