Tracey Kreplins

Context Wild dog impacts on agricultural and environmental systems in Australia are commonly managed through broadscale baiting using sodium fluoroacetate (1080). There has been growing evidence of bait avoidance in some populations of wild dogs throughout Australia, raising concerns about the efficacy of 1080 baiting. Bait avoidance can be a learned behaviour in a population, which could be caused by several factors associated with baiting programmes. One potential causative factor in developing bait avoidance is the ability to detect and respond to the toxin in the bait. The toxin 1080 is described as odourless and colourless to dogs, suggesting limited cues for its detection, but has not been formally tested.Aims This study used a trained detector dog to evaluate the detectability of 1080 to a dog and then the detectability in a variety of bait matrices. We also used a field trial to assess if 1080 dried meat baits were less likely to be taken than non-toxic baits in sites with a history of dog baiting or no dog baiting.Methods We trained a detector dog to detect 1080 odour and trialled this ability on different bait matrices. We used a field-based cafeteria-style trial to investigate the possibility of toxin detection by wild dogs.Key results We demonstrated that a trained dog could detect the presence of 1080, but detectability of the toxin when presented in different baits was variable and mostly greatly reduced. The field trials demonstrated no significant difference in bait take between 1080 and non-toxic baits by wild dogs in either a bait-na & iuml;ve or bait-exposed population.Conclusions These results suggest that, while 1080 is potentially detectable, factors other than its presence are responsible for bait avoidance in wild dog populations.Implications Wild dog management is heavily reliant on baiting with 1080 to reduce populations and thus reduce impacts on the environment and agriculture. The use of 1080 is unlikely to be the cause of bait avoidance and so where reduced uptake of baits by dogs is occurring, other factors need to be investigated and addressed.

Context
In Western Australia, there are three invasive predators that require management for agriculture and biodiversity protection, feral cats, wild dogs, and red foxes. These three predators often coexist in the same locality, suggesting potential efficacy gains can be made via simultaneous control. While Western Australian native species have evolved a high tolerance to poison baiting (1080), invasive predators have not. Therefore, landscape-scale baiting is commonly used for predator management.

Aims
Eradicat baits designed for feral cat control have also been known to be consumed and control wild dogs and foxes. In this trial, we aimed to evaluate Eradicat as an all-predator bait, determine if there is a preferential time for the use of the bait and assess non-target impact(s) of baiting.

Methods
We aimed to control all three predators on an agricultural property adjacent to a conservation reserve over 16 months with eight baiting events using Eradicat as an all-predator bait. Twenty one Reconyx camera traps monitored some of the baits deployed.

Key results
A total of 300 baits had a known outcome with minimal uptake by all three predators. Many issues were encountered when working on a smaller-scale including interference with farming activities and management, flash flooding, and non-target uptake of the Eradicat baits. Wetter than anticipated environmental conditions likely increased alternate prey availability, negatively impacting bait uptake.

Conclusions
More work is required to determine if Eradicat baits can be used as an all-predator bait.

Most human‐carnivore conflicts arise from the impact of predation on livestock. In Australian rangelands, considerable resources are allocated to constructing exclusion fences and implementing control measures to manage dingo populations for sustainable livestock enterprise. Assessing the effectiveness of these measures is crucial for justifying the investment. We used a replicated experimental design to examine the effect of landscape‐scale dingo‐proof exclusion fences (‘cell‐fencing’) on activity and population density of dingoes in the Southern Rangelands of Western Australia. We monitored dingo populations for 22–24 months across six study sites nested within a landscape of about 75,000 km2 and defined ‘fence level’ as the number of dingo‐proof fences enclosing each study site. We used camera trap capture rate (number of independent capture events per 100 trap nights) as a metric for dingo activity (including the availability of resources as other potential covariates), estimated dingo density using spatially explicit mark‐resight models, and tested the relationship between capture rate and estimated density of dingoes for each study site. Significant variation in both metrics was observed between sites and across time. Fence level and prey occurrence significantly influenced dingo activity. The annual mean dingo density estimate across study sites was below two dingoes per 100 km2 (i.e., 0.02 dingoes per km2; the maximum value believed to be compatible with small livestock) at only one study site in the first year, but it was higher across all sites during the second year of monitoring. Dingo activity correlated with estimated dingo density at only two sites, suggesting differences in dingo behaviour and detection across the six study sites. This study provides experimental evidence that camera trap capture rate is not a reliable method for assessing variations in the population size of dingoes. These results have implications for monitoring outcomes of dingo control programs across Australia.
This study addresses a critical question: How reliable is the activity metric (here we used camera trap capture rate) to assess variations in population size of dingoes across diverse habitats in Australia? Through experimental evidence, we demonstrate that activity metrics can lead to inaccuracies in assessing changes in dingo population size. Our findings highlight the need to assess dingo population size from identified individuals, which is particularly relevant for effective dingo control programs across Australia.

Being able to store toxic baits could allow livestock producers to target ‘hot spots’ of predator activity, supplementing or even replacing broadscale baiting, therefore avoiding potential risk of bait‐resistant populations. We compared 1080 dose recovered from dried meat baits (DMB; camel, horse and kangaroo) and sausage baits stored by different methods (shed, locked transport box, freezer, cryo‐vacuumed) to identify whether they were still lethal, and address concern that freezing/thawing could result in loss of the water‐soluble 1080 toxicant. We developed a bait collection method to halt microbial activity (preserving 1080 dose), allowing collection of baits manufactured under field conditions by five regional Recognised Biosecurity Groups according to their own schedule and methods. We assayed 351 baits (including 43 negative control baits, i.e., no toxicant), just over half (54%) of which were manufactured by us, using consistent meat masses, 1080 doses, and drying methods. For freshly manufactured baits, there was good consistency in bait manufacture, with no significant difference in recovered 1080. For stored baits, there was no effect of time on recovered 1080 for up to 2 months (maximum length of study), indicating different storage methods were equally effective in maintaining lethal baits. Manufactured sausage baits contained significantly more 1080 than DMB manufactured for this study, and showed substantial variability in recovered 1080 dose, which could reflect gun handling error. Storage in a locked transport box resulted in marked insect damage, likely rendering baits unattractive to target species. Freezing baits did not result in reduced 1080 dose. The majority (93%) of deployed/stored baits had a lethal 1080 dose recovered (13/192 baits were below the LD 50 for a 20 kg dingo). Dry shed storage is prescribed as best practice, but freezing baits does not reduce their toxic dose. Future testing for longer durations of storage would be beneficial.

Camera traps are widely used in wildlife research and monitoring, so it is imperative to understand their strengths, limitations, and potential for increasing impact. We investigated a decade of use of wildlife cameras (2012–2022) with a case study on Australian terrestrial vertebrates using a multifaceted approach. We ( i ) synthesised information from a literature review; ( ii ) conducted an online questionnaire of 132 professionals; ( iii ) hosted an in‐person workshop of 28 leading experts representing academia, non‐governmental organisations (NGOs), and government; and ( iv ) mapped camera trap usage based on all sources. We predicted that the last decade would have shown: ( i ) exponentially increasing sampling effort, a continuation of camera usage trends up to 2012; ( ii ) analytics to have shifted from naive presence/absence and capture rates towards hierarchical modelling that accounts for imperfect detection, thereby improving the quality of outputs and inferences on occupancy, abundance, and density; and ( iii ) broader research scales in terms of multi‐species, multi‐site and multi‐year studies. However, the results showed that the sampling effort has reached a plateau, with publication rates increasing only modestly. Users reported reaching a saturation point in terms of images that could be processed by humans and time for complex analyses and academic writing. There were strong taxonomic and geographic biases towards medium–large mammals (>500 g) in forests along Australia's southeastern coastlines, reflecting proximity to major cities. Regarding analytical choices, bias‐prone indices still accounted for ~50% of outputs and this was consistent across user groups. Multi‐species, multi‐site and multiple‐year studies were rare, largely driven by hesitancy around collaboration and data sharing. There is no widely used repository for wildlife camera images and the Atlas of Living Australia (ALA) is the dominant repository for sharing tabular occurrence records. However, the ALA is presence‐only and thus is unsuitable for creating detection histories with absences, inhibiting hierarchical modelling. Workshop discussions identified a pressing need for collaboration to enhance the efficiency, quality and scale of research and management outcomes, leading to the proposal of a Wildlife Observatory of Australia (WildObs). To encourage data standards and sharing, WildObs should ( i ) promote a metadata collection app; ( ii ) create a tagged image repository to facilitate artificial intelligence/machine learning (AI/ML) computer vision research in this space; ( iii ) address the image identification bottleneck via the use of AI/ML‐powered image‐processing platforms; ( iv ) create data commons for detection histories that are suitable for hierarchical modelling; and ( v ) provide capacity building and tools for hierarchical modelling. Our review highlights that while Australia's investments in monitoring biodiversity with cameras position it to be a global leader in this context, realising that potential requires a paradigm shift towards best practices for collecting, curating, sharing and analysing ‘Big Data’. Our findings and framework have broad applicability outside Australia to enhance camera usage to meet conservation and management objectives ranging from local to global scales. This review articulates a country/continental observatory approach that is also suitable for international collaborative wildlife research networks.

In Australia, livestock predation by dingoes (Canis familiaris) has contributed to what some livestock producers consider a dire situation for rangeland pastoralism, driving demand for cooperative regional-scale exclusion (‘cell’) fencing (i.e. pest-proof fences that encompass one or more individual properties) and landscape-scale predator control. The present case study predicted the effect of four cell-fences in the state of Western Australia (WA) on the gross margin of sheep (for meat or meat and wool) and cattle pastoral enterprises. We modelled the potential effects of the following four key variables: (1) four levels of commodity prices, (2) five levels of livestock weaning rate (based on livestock records collected 1985–1995; weaning rate is defined as number of lambs or calves that are born and survive to weaning, expressed as percentage of total mated females), (3) three predicted levels of time required to remove dingoes from within the fenced area, and (4) five levels of macropod (mainly kangaroo) response as competitive grazers, with a total of 3600 scenarios representing all combinations of these factors. Each scenario was assessed for profitability (i.e. net present value (NPV) over 25 years) and benefit of fencing (i.e. NPV compared with an unfenced enterprise of the same livestock type, region, and commodity prices). Finally, the benefit–cost ratio (BCR) of investment in cell fencing was calculated for each fenced scenario. The majority (67%) of scenarios representing continuation of current management (i.e. no cell fencing) returned a negative NPV (i.e. livestock enterprises were projected to make a loss). However, only 37.4% of cell-fenced scenarios returned a positive NPV, meaning that even with a cell-fence and successful removal of dingoes, the enterprise was still unlikely to be profitable. Only 43.4% of cell-fenced scenarios returned a BCR of cell fencing greater than one. Weaning rate following dingo removal was the most important factor determining return on investment for cell-fencing. Survival and reproduction of small livestock, particularly wool sheep, benefit most from cell-fencing, whereas cell-fencing and dingo removal did not result in greater profits for cattle enterprises. Running sheep for wool and meat within cell fencing coupled with removal of dingoes would maximise the likelihood of achieving a positive return on investment in cell fencing (although the enterprise may remain unprofitable overall); otherwise, unfenced enterprises affected by dingoes should run cattle as this will be more profitable.

Hybridisation between Australian dingoes and domestic dogs is a controversial area of interest and research. An ongoing canine sterilisation programme in rural and remote Western Australia provided an opportunity to assess the dingo ancestry of camp dogs and opportunities for hybridisation. Blood samples were collected from 345 individual community dogs at 21 locations. Dogs were screened using 23 microsatellite loci and ancestry percentage assigned using an iterative Bayesian assignment algorithm. A single individual was a dingo, 96% were domestic dogs and 3.5% were hybrids. Camp dog and dingo hybridisation in these areas is of little concern in terms of conserving dingo purity.

Wildlife management actions are increasingly contingent on acceptance by the broader public. Consequently, understanding factors that influence acceptability of different management options is important. In Australia, kangaroos (native large herbivores, Macropus spp., Osphranter spp.) are managed using lethal control for a range of context-specific reasons. We surveyed 1293 members of the Australian public to test whether acceptability of lethal control of kangaroos depends on the reason for the control, values towards wildlife (assessed using the Wildlife Values Orientation questionnaire), and knowledge about kangaroos. We also tested whether acceptability could be shifted by providing relevant information. Lethal control of kangaroos for biocentric reasons (i.e., agricultural protection, biodiversity conservation, animal welfare) was more acceptable than lethal control of kangaroos for anthropocentric reasons (i.e., human consumption, human safety). Acceptability was greater among survey respondents high on the domination orientation (i.e., valuing wildlife for human benefits), and lower among those high on the mutualism orientation (i.e., valuing wildlife for its own intrinsic value). Acceptability was also positively associated with knowledge of kangaroo ecology and management. Provision of information did not impact acceptability of lethal control for any reason except for human use, which, while not significant, showed promise that acceptability of lethal control for this reason could be influential to the public. Additionally, while the acceptability of lethal control varied widely among the Australian population, there was little evidence of polarisation, suggesting that large sectors of the public may be amenable to different perspectives.