Aerial baiting for wild dogs: the impact on spotted-tailed quoll populations
Because wild dogs can cause severe damage to grazing enterprises, all land managers are legally required to control wild dogs on their land. To be effective, wild dog control programs must cover large areas, often over rugged and inaccessible terrain. In such circumstances, aerial baiting with '1080' (sodium monofluoroacetate) meat baits is the only feasible option.
Some native predators, particularly the spotted-tailed quoll (Dasyurus maculatus), are potentially at risk of being poisoned by these aerial baiting programs. The NSW Department of Environment and Conservation (DEC) is concerned about this risk, and has researched the impact of aerial baiting on spotted-tailed quoll populations. Field experiments were conducted in northern NSW in 2004 and 2005, and in southern NSW in 2005.
The results of this research are very important to a range of people and organisations. Because of this, the research program was guided by separate steering committees in northern and southern NSW.
In northern NSW, the Aerial Baiting and Quolls Research Steering Committee included representatives from NSW Farmers' Association, Armidale Rural Lands Protection Board, the Wildlife Preservation Society of Australia, and CSIRO, as well as government agencies such as DEC, the Department of Primary Industries and the Department of Lands.
In southern NSW, the Cooma Wild Dog Advisory Panel included representatives from the NSW Farmers' Association, Cooma Rural Lands Protection Board, DEC and the Department of Primary Industries.
Program of work
The steering committees agreed that the research would be carried out in two stages. The first experiment (stage 1) took place in 2004. This experiment sought to measure the impact of aerial baiting on quoll populations in areas that have a continuous and current history of aerial baiting. Stage 2 involved two more experiments, in 2005, which extended the research into areas that did not have a current history of aerial baiting.
The methods were similar for all of the experiments. Fresh meat baits dried overnight were injected with a standard dose of 1080 (6 mg per bait) and a bait marker, Rhodamine B (RhB). RhB temporarily stains the mouth and digestive tract of an animal that consumes bait. It is also incorporated into the animal's whiskers, should it survive after eating bait.
Prior to baiting, all trappable quolls in the study areas were trapped, tagged with identity transponders and fitted with radio-transmitters, which contain a motion sensor that signals the death of an animal.
For several weeks after the aerial baiting, all quolls were radio-tracked on the ground and from a helicopter. Where possible, carcasses of dead quolls were retrieved, subjected to a post-mortem examination, and inspected for residues of RhB. Tissue samples were taken for 1080 analysis. Subsequently, live quolls were re-trapped, the transmitters were removed, and samples of the quolls' whiskers were taken for RhB analysis.
The toxicity of 1080 baits decreases naturally over time, so a degradation study was used to monitor the rate of decreasing toxicity. This involved placing a number of baits in wire cages at the study site on the day of the aerial baiting. Samples were collected at regular intervals and analysed for 1080.
In winter 2004, a population of 31 radio-collared quolls was monitored following a routine aerial baiting program in Tuggolo State Forest and Nowendoc National Park, on the Northern Tablelands of NSW.
This area has had a continuous history of annual aerial baiting for over 30 years. The aerial baiting program is currently coordinated each year by the Department of Primary Industries and carried out by the Niangala and Barnard River Wild Dog Control Associations. Baits were deployed at a rate of 40 baits per kilometre.
Thirty-one quolls were monitored during the baiting. Of the radio-collared quolls that died, only one tested positive to 1080, and this animal died 23 days after baiting. By this time, the degradation study showed that the toxicity of the baits would have dropped to an average of 1.3 mg of 1080 (from its starting point of 6 mg on the first day of baiting). This is far below the expected lethal dose - 6.3mg of 1080 would be the lethal dose for 50 per cent of quolls of this size. Furthermore, the autopsy indicated that this animal had severe injuries that appear to have happened before its death.
After baiting, 35 quolls were sampled for RhB (including a number of non-collared quolls). Analysis of the quolls' whiskers showed that a further five quolls had eaten bait and survived, demonstrating that bait consumption is not necessarily fatal for quolls.
However, overall only 17 per cent of the quolls had eaten the baits. This was substantially less than in simulation trials conducted in southern NSW using non-toxic baits, in an area where aerial baiting had not been used in the past. In these simulation trials, 50-65 per cent of animals consumed bait. This suggests that in Tuggolo State Forest and Nowendoc National Park, quolls may have developed an aversion to baits over 30 years of regular exposure to aerial baiting.
Stage 2 - Northern NSW
Stage 2 of the research was conducted during late-autumn and winter of 2005. The first trial was conducted in the Styx River State Forest and parts of Cunnawarra National Park on the Northern Tablelands. Aerial and surface-laid baits had not been used in this area since 1993.
A total of 14 radio-collared quolls were subjected to aerial baiting. Fresh meat baits, prepared as described previously, were delivered at 40 baits per kilometre within Styx River State Forest and 10 baits per kilometre within Cunnawarra National Park (consistent with DEC practice elsewhere in the state). This was done in cooperation with the Jeogla Wild Dog Control Association and the Armidale Rural Lands Protection Board.
Two of the radio-collared male quolls died, 29 and 34 days after baiting, but no 1080 residue was found in either carcass. The first of the animals that died appeared to have been killed by a large predator. The quoll had massive injuries to the thorax, with tissue bruising, indicating that it was attacked prior to death. Its spine was severed and a wild dog was seen close to the kill site, suggesting that a dog was responsible. The cause of death of the second animal is uncertain, but it died during a prolonged cold and wet period, possibly from exposure.
RhB analyses of samples from 19 quolls, 5 feral cats and 1 wild dog indicated that 13 quolls (including the two dead males) and 2 cats had eaten 1080 baits and survived. Most quoll whiskers showed more than two bands of RhB (up to six bands, in one female), implying that the quolls had eaten baits more than once. However, the RhB bands were faint, possibly because quolls had eaten only part of a bait or regurgitated some of it. Nevertheless, bait consumption was considerably higher (68 per cent) than that recorded during the previous year (17 per cent).
Stage 2 - Southern NSW
The second trial in 2005 was conducted within the catchment of the Jacobs River in southern Kosciuszko National Park. A total of 16 radio-collared quolls were potentially exposed to aerial baiting at a rate of 10 baits per kilometre.
Only one quoll death was recorded over the period of monitoring, and it did not test positive to 1080. Of 18 whisker samples collected from live quolls after baiting, six (32 per cent) tested positive for RhB.
Consistent with the results from the northern trial in 2005, most RhB bands were faint. Five of the six animals that tested positive for RhB had single bands present and the remaining animal had two bands.
The quolls' bait consumption was higher per capita than in the stage 1 research (at the regularly baited site in northern NSW). However, it was less than in Styx River State Forest and parts of Cunnawarra National Park (the study area in stage 2), which had a similar history of no recent aerial baiting.
Other research in Queensland
Concurrently with the DEC research, the Queensland Department of Natural Resources and Mines has been measuring quoll mortality during surface baiting programs for wild dogs.
In trials conducted between 2002 and 2005 in southern Queensland, 76 radio-collared quolls were exposed to surface-laid baits spaced 500 metres apart. The researchers recorded only two 1080-related deaths, with an additional six radio-collared quolls dying from unknown causes.
Implications of this research
These results indicate that aerial baiting had a minimal impact on the quoll populations studied. Quoll deaths after 1080 baiting were much lower than was predicted by previous non-toxic trials - both in DEC's three aerial baiting trials during 2004 and 2005, and in the Queensland research during 2002-05. It appears that quolls eat toxic baits. However, in this study, most of the quolls that consumed 1080 baits survived.
The consistency of these results across sites indicates little difference in the response to aerial baiting between distant quoll populations. The observed low mortality rates due to 1080 poisoning are apparently not caused by an adaptation of quoll populations to repeated exposure to aerial baiting, because low mortality rates were also observed in quoll populations in areas that had no recent history of aerial baiting.
We are unsure of the reasons for this lower-than-expected mortality rate in quolls in the wild. It may be that quolls have a higher resistance to 1080 than would otherwise be predicted on the basis of laboratory-based trials. Quolls may also regurgitate baits, or eat only part of a bait.
While individual quolls may die from 1080 baits, this research suggests that aerial baiting is unlikely to have an impact on quoll populations as a whole. In fact, aerial baiting which suppresses local fox and dog populations may benefit quolls in an area.
However, it is possible that even a few deaths caused by aerial baiting may have a significant impact on a small population of quolls that has already been weakened by factors such as drought, habitat fragmentation or disease. It is also possible that 1080 may have other impacts on quolls that survive after eating baits - such as infertility and birth defects. No information is available on these impacts.
Future approaches to wild dog control
After discussing the results and other published information, the steering committees agreed that aerial baiting can now be considered as an additional wild dog control technique where appropriate. However, the committees encourage the use of an integrated approach, using a range of techniques such as ground and aerial baiting, trapping, shooting and exclusion fencing.
All aerial baiting programs need to comply with the Pesticide Control (1080 Wild Dog Bait) Order 2002, as required under the Pesticides Act 1999. Also, all new aerial baiting programs in areas with spotted-tailed quolls must be referred to the Australian Department of Environment and Heritage. This is a requirement of the Commonwealth Environment Protection and Biodiversity Conservation Act 1999.
The steering committees identified the need for further research to improve quoll conservation. Priority should be given to:
- Investigating the demographics of quoll populations and their high turnover rate. The results from these studies can be used to predict the potential impact of disturbance events on quoll populations.
- Researching the interaction between native and introduced predators and their prey. 1080 dog baiting can impact upon the populations of some predators, potentially causing a positive or negative response in other species
This research was funded primarily by DEC, with additional funds being provided by the NSW Department of Lands. The NSW Department of Primary Industries, the Armidale and Cooma Rural Lands Protection Boards and the Niangala, Barnard River and Jeogla Wild Dog Control Associations provided in-kind support.
Page last updated: 26 February 2011