Predator management with a single bait? Use of Eradicat for the simultaneous management of feral cats, red foxes, and wild dogs

Introduction

All three mammalian carnivores, feral cats (Felis catus), wild dogs (defined as dingoes, free-roaming dogs and their hybrids; Canis familiaris), and the red fox (Vulpes vulpes) are pest species in Western Australia (State Government of Western Australia 2007). These three invasive species significantly impact the viability of this state’s native biodiversity and foxes and wild dogs negatively impact livestock production industries, costing millions of dollars a year in impacts and funding for their control (Gong et al. 2009; McLeod 2016).

Baiting with 1080 is an important tool to manage feral cats (Algar and Burrows 2004), wild dogs (Thomson 1986), and foxes (Thomson et al. 2000). Western Australia benefits from the fact that native wildlife have a high tolerance to the 1080 poison as it naturally occurs in the Gastrolobium spp. (Mcilroy 1981; McIlroy et al. 1986). However, achieving bait uptake by the predators is often complex. For example, it has been noted for foxes (Allsop et al. 2017) and wild dogs (Kreplins et al. 2018) that long-term, unvaried, baiting practices can reduce the effectiveness of baiting programs. Compared with canids, feral cats’ sensory systems predispose them an enhanced ability and preference to hunt live prey over sourcing and consuming carrion (i.e. baits) (Stokeld et al. 2015; Paton et al. 2024). To achieve an effective outcome when deploying baits for any of the three predators, a range of factors such as bait drop timing, rate, and availability are highly important.

Ensuring that baiting practices are consistently ‘novel’ (e.g. varying baiting rate and bait type) and bait availability is high may aid in increasing bait uptake by the target species (Allen et al. 1989; Fleming 1996; Fancourt et al. 2021a). For example, baiting at a rate of 40 baits per km instead of 10 baits per km greatly enhanced the baiting program results for foxes and wild dogs (Thomson and Algar 2000; Ballard et al. 2020). Bait placement can also alter bait uptake by wild dogs (Fleming et al. 2001; Kennedy and Rose 2015). Similarly, varying bait type (Eldridge et al. 2000; Thomson and Kok 2002) or lures (Hunt et al. 2007; Moseby et al. 2011) may increase the effectiveness of a baiting program. Increasing bait uptake is an area of science that is evolving and is constantly being fine-tuned for successful management programs.

Eradicat baits were designed as an enhanced feral cat bait (Algar and Burrows 2004; Algar et al. 2011) to reduce the problem of feral cats’ lack of interest in traditional baits (i.e. Probait and dried meat baits). The Eradicat bait is comprised of kangaroo mince, chicken fat, and flavour enhancers that are highly attractive to feral cats (Algar and Burrows 2004). Eradicat has been anecdotally known to control foxes (Berry et al. 2014) and wild dogs (Fancourt et al. 2021b), as well as feral cats based on current research in Western Australia (Lohr and Algar 2020). Whilst the lethal dose of 1080 in Eradicat is sufficient to control all three predators, timing for on-ground management will impact control efficacy for different species. For example, feral cat baiting programs are most effective in winter when small prey items are less common and ambient temperatures are not high (Algar et al. 2007). Wild dog management is most effective in spring and autumn, which coincide with peak periods in activity (i.e. mating and whelping; Thomson 1992a). Fox control should occur all year around but is most effective in autumn when young are dispersing to find new home ranges (Marlow et al. 2015).

Bait availability and consequently control efficacy may also be impacted by non-target interactions with baits. Adding lures and increasing the appeal of the bait substrate may increase the curiosity of non-target species. For example, Australian ravens (Corvus coronoides) consumed more wild dog baits laced with fish oil than without the lure when fish oil was used on wild dog baits (Kreplins et al. 2018).

The aim of this pilot study was to determine: (1) the uptake of the Eradicat bait by three predators; (2) non-target interference; and (3) optimal timing of control programs.

Materials and methods

Given the legislation and policy for invasive species control (State Government of Western Australia 2007), a control site where no baiting occurred is almost impossible to locate in Western Australia. Instead, a longitudinal approach was taken to investigate the uptake of baits.

This longitudinal study ran from November 2020 to February 2022 across three neighbouring farms. Baiting occurred each month, except when large rainfall events occurred and estimates of predator activity were generated from camera trap images.

Site description

The study was located within the Shire of Perenjori on three farming properties that were located adjacent to Weelhamby Lake and only 20 km from the Western Australia State Barrier Fence (Fig. 1a). The State Barrier Fence is 97% impermeable to wild dogs but permeable to feral cats and foxes (T. L. Kreplins, unpubl. data). The camera transect was along the farm fence line within the paddocks where sheep and cropping were run. On the opposite side of the fence was a conservation reserve, which undergoes baiting and trapping all year round.

Fig. 1.

(a) Location of all the camera traps in the Northern Agricultural zone of Western Australia and (b) an example of the flash flooding that occurred several times during the study.

Results

A range of species were recorded on camera resulting in 23,173 activity events. A total of 48 feral cat, 12 wild dog, and seven fox activity events were recorded on camera during the project (Fig. 2). Feral cat activity on camera was constant throughout the study with peaks in July, August, and September 2021. Wild dogs were seen in December 2020, March, June, July, September 2021, and January and February of 2022. Foxes were only seen in February, April, May, and June in 2021 (Fig. 2). None of the individual wild dogs captured on camera were identified as the same individuals. Most of the feral cats were new individuals but there were five individuals that were seen four times, three times, and two times (three individuals).

Fig. 2.

The number of activity events for each predator (feral cats, wild dogs, and foxes) seen on camera.

Non-target animal activity was recorded throughout the study, including 8737 macropods (Osphranter rufa and Macropus fuliginosus), 6931 sheep, 3984 goats, 2815 emus (Dromaius novaehollandiae), 440 rabbits (Oryctolagus cuniculus), 142 Australian ravens, 24 Australian magpies (Gymnorhina tibicen), 20 Australian bustards (Ardeotis australis), three echidnas (Tachyglossus aculeatus), three varanids (Varanus gouldii), three grey butcherbirds (Carcticus torquatus), two alpacas (Lama pacos), and a single wedge-tailed eagle (Aquila audax).

The fate of 280 of the 300 Eradicat baits laid is known, with an average of 11.5 (range, 3–19) days until baits were taken. Bait uptake was low by the three target predators in this project. No wild dogs or feral cats consumed Eradicat baits. Two foxes ate an Eradicat bait. This occurred on the 17 April and 14 June 2021. Despite this, the three predators were present 52% of the time when the bait was also observed available on camera (54% feral cats, 58% wild dogs, 43% foxes; Fig. 3).

Fig. 3.

The opportunity each predator had to take a bait; the activity events by each predator, feral cats (a), wild dogs (b) and foxes (c) when a bait was also on camera.

A total of 18% of bait uptake can be accounted for by non-target animals, 15% by bird species, 1% by reptiles, and two baits were consumed by ants. Emus consumed six baits, Australian ravens consumed 30 baits, bustards consumed four baits, the grey butcherbird consumed three baits, and the Australian magpie consumed one bait. Varanids consumed four baits (Fig. 4).

Fig. 4.

The Eradicat baits consumed by emus, varanids, ravens, bustard, butcherbird, magpie, fox, and ants. Some baits were not taken, had an unknown outcome, were broken down by people or livestock trampling, and rainfall.

A total of 87 Eradicat baits or 29% of baits were not taken and were still on the ground at the successive baiting run. Fifty (17%) and 47 (16%) baits were unavailable to the predators as they were trampled by people and livestock and broken down by rainfall, respectively (Fig. 1b). These equated to 33% of baits being unavailable to the predators. Sixty one Eradicat baits (20%) of 300 of the baits had an unknown outcome due to cameras missing the animal removing the bait.

Discussion

To determine if Eradicat can be used as an all-predator bait, we undertook a small pilot study that also examined the best timing for the all-predator bait to maximise target uptake and reduce non-target bait removal. The disappointingly low uptake by the target species meant that many of our aims could not be achieved. However, the non-target uptake was low in comparison to other studies in Western Australia.

Bait uptake by all three predators was low with only two foxes consuming baits. There are several possible reasons for this outcome. This study site was originally chosen due to its high number of livestock losses to wild dogs and locality to native bushland for feral cat and fox activity. However, prior to our study, large investments in on-ground control were made to reduce the livestock losses in this area (Central Wheatbelt Biosecurity Association, pers. comm.). Therefore, the density of wild dogs and foxes on site was reduced, impacting our ability to investigate the use of Eradicat as an all-predator bait.

Maximising bait availability and bait consumption for invasive predators is very important, but difficult to achieve in practice (Moseby and Hill 2011). In this study, baits were present on average 52% of the time when one of the predators were also recorded on camera. This would indicate that the predators were sometimes choosing not to consume the baits. For example, feral cats encountered 26 of 300 baits that were deployed and did not eat a bait. Other studies have shown that 31 baits encountered by feral cats were consumed 16 times (Kreplins et al. 2018) but this was during drought conditions. The body condition of the feral cats on camera indicated they have substantial diet and it is possible that there as a high availability of alternate prey due to unprecedented rainfall during this study. Feral cats often choose to hunt live prey when abundant (Molsher et al. 1999; Comer et al. 2020). Similarly, there were seven opportunities where wild dogs did not consume Eradicat baits that were available to them. Other trials where Eradicat was deployed in wild dog localities (Fancourt et al. 2022), bait uptake was also low, indicating that even when a few baits are available to the target species other factors (e.g. live prey availability) may impact bait uptake.

There were only seven foxes seen on camera over the 16 months of the study. Foxes readily succumb to wild dog control programs in Western Australian and are rarely seen where wild dog control occurs regularly over extended years (Kreplins et al. 2018). Of the seven fox activity events on camera, there were only three occasions with baits available and two of these were eaten. This means that about two-thirds of the time a fox saw a bait, it was consumed. These results are from a low sample size but relate to the idea that foxes are the most likely invasive predator to readily consume baits, given they are scavengers (Marlow et al. 2015).

Optimal timing for the use of an all-predator bait remains to be determined. Whilst almost 50 feral cats activity events were captured on camera, enough activity to determine bait deployment timing, they chose not to eat baits. A peak in their activity occurred during winter, potentially coinciding with the unusual high rainfall season resulting in a higher prey availability. In contrast, wild dog activity was very low and not enough to determine timing for bait uptake. Foxes that did consume baits were most likely dispersing sub-adults in April (Coman et al. 1991) and adults potentially with young in dens in June; despite only having seven activity events on camera. Fox baiting historically can occur anytime throughout the year but there is a higher uptake when young disperse in late summer (Saunders et al. 2010).

Bait availability for the predator species would have been impacted by the bait uptake by non-target species (i.e. Australian ravens and Varanus spp.) as documented in other Eradicat trials (Hohnen et al. 2020; Fancourt et al. 2021b). The average number of days until bait was taken was 11.5 days. Other studies have recorded 57.6% of baits were removed within 14 nights (Fancourt et al. 2021b). These findings indicate that bait uptake by non-target species is fairly rapid, leaving fewer baits available to the predators. Dundas et al. (2014) also observed most baits being removed by non-target species for fox control the first night after deployment.

Landscape scale control programs can be very challenging. Landscapes undergo a range of management activities that interfere with control programs and environmental factors also influence baiting efficacy. For example, wet weather and flash flooding reduce the lethal content of 1080 baits (Fleming and Parker 1991; Twigg and Parker 2010) and 16% of baits in this study succumbed to flash flooding events making them unavailable to predators. Without rainfall, baits in this area could remain lethal for 6–8 weeks (Australian Pesticides and Veterinary Medicines Authority 2008), but with the sudden downpours experienced in this study, the baits would be non-lethal immediately. In addition, interference from machinery, people and livestock further reduced (by 17%) availability of baits to predators. Future trials should be extended to mitigate environmental variables and land holder activities.

This study occurred on three agricultural farms bordering conservation estate. The 21-km transect is likely to only occur over a single wild dog pack home range (Thomson 1992b) and around four feral cat (McGregor et al. 2015) and fox (Kobryn et al. 2023) home ranges. Potentially, this might mean fewer animals are present to take baits. However, given the high investment in predator management in the area, it is also likely that the site is a ‘sink’ for predators. We identified very few repeat images of known individual cats and wild dogs, suggesting that the predator population is transient and most likely being regularly removed by the varied control program. A larger scale trial would help alleviate issues around transient animals and encompass a larger number of predator activity events.

Bait removal or uptake studies are impacted by camera hardware and placement. A total of 24% of baits was removed by unknown species, not dissimilar to other studies such as Hohnen et al. (2020) and Fancourt et al. (2021b) who missed 29% and 78% of Eradicat bait takes using Recoynx cameras, respectively.

The use of Eradicat as an all-predator bait still needs further study in a location where there is a higher activity level of the target predator species. Other variables reducing the bait uptake by the predator species were the alternate prey, environmental variables (i.e. the likelihood of rainfall), and management activities in the region. Whilst predator control programs utilising baits must be adaptive to the conditions and other activities occurring in the landscape this trial indicates that there is potential in using a single bait for multiple species to increase efficiency when undertaking multi-predator management programs at landscape scales.