Seabird biologist Dr Rachael Alderman coordinates a long-term seabird monitoring program on Macquarie Island for the Department of Primary Industries, Parks, Water and Environment (DPIPWE), based in Tasmania. During the ‘Strategic Science in Antarctica’ conference she described the impact of poison baiting on giant petrels, during the rabbit, rat and mouse eradication project on the island. A 30% decline in the population of both northern and southern giant petrels has been observed, but there is optimism that populations will bounce back.
The introduction of alien species to subantarctic Macquarie Island began in the 1800s with the first human visitors, the sealers, to the island. The most conspicuous aliens, however, are the vertebrate pests, of which a variety have been introduced, deliberately and by accident, by sealers and the early research expeditions. Five species established feral populations on the island: wekas (Gallirallus australis), cats (Felis catus), European rabbits (Oryctolagus cuniculus), black rats (Rattus rattus) and house mice (Mus musculus). Collectively, these five have contributed to the extinction of endemic species (the Macquarie Island parakeet), the localised extinction of burrowing petrel species, and a decline in the abundance and breeding performance of a range of seabird species through predation and habitat degradation.
The Parks and Wildlife Service commenced management programs in the 1970s, with myxomatosis virus introduced to control rabbit numbers in the mid 1970s, eradication of wekas in 1989 and cats in 2000, and ad hoc baiting, trapping and shooting of rabbits and rodents.
Rabbit numbers have fluctuated over the years due to a range of factors. However, a recent peak in numbers at an estimated 150 000 in 2005, saw the island exposed to devastating grazing pressure. In response to lobbying, state and federal governments committed to funding the eradication of rabbits, rats and mice. The resulting Macquarie Island Pest Eradication Project (MIPEP) began in 2007.
The MIPEP plan involved aerial baiting of the island followed by on-ground hunting for survivors by specially trained dogs and their handlers (Australian Antarctic Magazine 23: 12–13, 2012). While this method has so far proved a success — with no rabbit and rodent sightings since 2011 — what was the cost to non-target species?
The aerial baiting posed two potentially significant impacts to native wildlife: disturbance of bird colonies by helicopters, and non-target poisoning, particularly of scavenging seabird species such as the northern and southern giant petrels, skua and kelp gulls.
The primary mitigation measure for both impacts was to conduct the aerial baiting between May and August, when the majority of seals and seabirds had left the island in between breeding attempts. This would not be effective for king penguins, which are present in large colonies year-round and have been known to stampede with fatal consequences in the past. Instead, specific over-flight protocols were developed for helicopter activities around the king penguin colonies.
The risk of poisoning non-breeding birds that would be present over the baiting period was recognised. However, the plan considered what was generally known about the spatial and temporal abundance and foraging behaviours of these susceptible species and concluded that population-level impacts of any mortality were likely to be limited and the risk was acceptably low.
In the first half of the aerial baiting phase in 2010, with less than 8% of the required bait dispersed due to weather, the first birds to to die of primary, secondary and even tertiary poisoning were detected. Due to delays in the baiting program, a decision was made to postpone further baiting until the following season.
Ad hoc searching over the next six months recovered 960 dead birds, including 323 giant petrels, 230 skuas, 385 kelp gulls and 22 ducks. These numbers are likely conservative as many carcasses would not have been detected if the birds had already been scavenged, or had died at sea. However, they are substantially higher than many anticipated.
For the giant petrels, listed as threatened under both our national and state legislation, this high mortality caused significant concern and triggered a review of the eradication and the non-target mitigation actions. It was widely agreed that the eradication needed to continue and two additional mitigation measures were identified.
The primary mitigation measure was the novel release of rabbit calicivirus, to reduce the rabbit population prior to baiting and subsequently limit the number of poisoned rabbit carcasses available to scavengers. When it was released in February 2011, the virus was spectacularly successful. By the end of April it had killed an estimated 90% of the rabbit population.
The secondary mitigation measure was to provide additional team members during the baiting phase, whose task was to systematically search for poisoned carcasses — both target and non-target species — and remove them.
By the end of the second aerial baiting attempt in 2011, in which all the bait was distributed, nearly 1500 birds were recorded dead; approximately 1.5 times more than in the previous baiting attempt. However, two things need to be remembered. Firstly, the search effort in 2010 was substantially less comprehensive than in 2011. Secondly, over 10 times the volume of bait was distributed in 2011 compared to 2010, but this was not matched by a proportionately similar increase in non-target mortality. Under these two criteria, the mitigation actions implemented for the 2011 aerial baiting were highly successful.
Nonetheless, at least 2500 birds died over the two baiting periods. 760 of these were threatened giant petrels, which has significant impacts on their populations.
Genetic analysis of samples collected from the giant petrel carcasses that were recovered on the island indicated that the majority were northern giant petrels (40:1 northern:southern). This is despite the fact that southern giant petrels are more abundant than northern giant petrels on Macquarie Island. We also found a very strong gender bias, with 80% of all giant petrels found on the island being male. Giant petrels form long-term monogamous pair bonds and it takes both birds of a pair to successfully rear a chick, so male biased mortality such as this exacerbates the effects on the population.
DPIPWE have been monitoring giant petrels breeding on Macquarie Island since 1994. Prior to the commencement of baiting, breeding populations of both species were increasing. Not surprisingly, a substantial decline of about 30% has been observed in northern giant petrel breeding numbers as a result of the baiting. Annual monitoring is continuing and it will take several more years before we can say how the population is responding conclusively. To date, there is no indication that the decline has continued beyond the initial mortality event and we can be cautiously optimistic that the increase will resume and numbers will return to those prior to the MIPEP baiting.
Monitoring of the southern giant petrels has provided some unexpected results. Although very few deaths of this species were documented, a decline in the breeding population of a similar magnitude as the northern giant petrels has been observed. A likely explanation, based on known intra-specific differences in foraging behaviour, is that mortality of this species was not detected because they died at sea. Ongoing monitoring is similarly required to understand how this population has been affected and if and when it will recover.
Rachael Alderman
Senior wildlife biologist, DPIPWE
