Table : Trapping cost assumptions

There is very limited data on the costs associated with private control. Given the scope of this review we focus on estimating the costs of baiting control. For simplicity we assume the per unit cost of private baiting is equivalent to the per unit cost of public baiting.

3.5.3Comparative effectiveness of ground baits versus traps

The analyses in previous sections of this report illustrated trends in baiting and trapping effectiveness with reference to currently understood measures – baits taken per baits laid and dogs trapped per traps set or checked. However, this does not enable us to appropriately compare the cost effectiveness in destroying dogs across the two measures.

With this in mind, we have created new relative measures of effectiveness of baits and traps in destroying wild dogs. These measures are:

baits taken by wild dog per bait night. This is the number of baits taken for one bait that is in the ground for one night. This measure assumes that around 2 per cent of baits laid are taken by wild dogs; and

Dogs trapped per trap night. This is the number of dogs trapped for one trap that is set for one night.

We estimate that the kill rate per 1000 bait nights is substantially lower than the kill rate per 1000 trap nights. Substantially more baits nights than trap nights are required for a given kill rate (Table ).

Based on the current performance outcomes, levels and costs of control of activities, an issue of concern is whether overall performance of the program could be improved by altering current operations and mix of control activities across zones.

The analysis in this section is presented in terms of impact on wild dog take. The reason for not using wild dog attacks or livestock deaths is that, using standard regression analysis, we have been unable to establish robust statistical relationships between alternative control measures and wild dog attacks or livestock deaths.

Moreover, while there is some evidence that higher levels of baiting and trapping will impact livestock deaths (e.g. higher levels of baiting and trapping in Regional Cluster 4 in mid to late 2015 is likely to have resulted in lower livestock deaths - Figure ), analysis of existing data sets (e.g. Dogbytes) at either an aggregate or regional level does not provide statistically significant relationships between key variables to be able to assess the relative performance of baiting or trapping.

The options presented in this section, while not reflecting all possible options, provide a broad overview of potential changes that could be made. Additionally, the options have been developed assuming that for a Wild dog controller there is:

at least one day per week allocated to setting traps;

at least one day per week allocated to baiting; and

one day per week allocated to ‘other’ activities which are not directly related to trapping or baiting (e.g. community consultation, other types of control measures).

Additionally, we have assumed that 2 per cent of baits laid results in a wild dog being killed.

3.6.1Options for change under current 72 hour trap inspections

There are a range of options that could be undertaken to improve the amount of wild dogs killed. These include increasing the amount of days allocated to baiting and expanding work days to include weekend work, noting that weekend work will increase costs.

With this in mind, we have defined potential future options for wild dog management as:

These options are further illustrated in Table .

Table : Options for change under 72 hour trap inspections

The impact of these options compared to the base case is illustrated in Table , Figure and Figure . These show that there are a range of options to achieve a higher level of dogs killed in order to achieve lower livestock deaths, although some options are costlier than others. Option 1A shows the impact of increasing baiting at the expense of trapping while Option 1B shows the impact of weekend trapping.

In comparing the options, both achieve the same overall increase in dogs killed even though Option 1B is three times the cost of Option 1A. However, we note that the difficulty with this result is that it assumes that baits have the same ability as traps to kill a wild dog.

Moreover, it may be the case that some dogs are willing to go past baits to get to sheep which they consider a greater ‘prize’. Moreover, an increase in baits laid may actually kill dogs that are not targeting sheep. Therefore, it is possible that overall livestock deaths may not fall or by as much as expected if a lower number of traps under Option 1A results in the dogs that are attacking sheep not being trapped. The current set of data that are available does not enable us to examine whether this is the case or not.

Figure : Impact on dogs killed per annum for alternative options under 72 hour trap inspections – relative to the base case

3.6.2Options for change under 48 and 24 hour trap inspections

Table : Options for change under 72 hour trap inspections

The impact of these options compared to the base case is illustrated in Table , Figure and Figure .

Some of the key conclusions from this analysis are that:

if there is no change to current working arrangements (defined as 24 and 48 hour base case assumptions) this leads to no change in costs albeit with a reduction in trapping effort and dogs trapped by traps. Additionally, under these scenarios wild dog controllers are not able to provide the current response timeframe in response to a dog attack.

increased baiting can be used to compensate for lower levels of trapping under 48 and 24 hour trap inspections (Options 2A and 3A) and has a low impact on cost. The potential downside of this option is that baiting may not be as successful in targeting the dogs that are killing livestock and cannot always be applied in a timely manner because of approval processes.

working weekends can enable traps to be in operation 7 days a week (Options 2B, 2C and 3B) under 48 and 24 hour trap inspections. This will achieve the current level of dog kills. The cost is much greater under 24 compared to 48 hour trap inspections. Additionally, the cost under 48 hours depends on whether working arrangements are able to allow for a flow from one week to the next such that the days of trapping rotates from one week to the next (Option 2B compared to Option 2C).

We note that there are a range of animal welfare considerations in the choice of options given that:

some control measures such as baiting and CPEs are considered to be more humane than trapping; and

shorter trapping inspection times are likely to be considered to be more humane.

Table : Impact on trapping/baiting levels and costs



Note: baiting and trap levels refer to baiting and trap nights.

Figure : Impact on dog take per annum for alternative options under 48 and 24 hour trap inspections – relative to the 72 hour base case

4.1Strengths

Dog reporting (sightings, dead dogs/kills, livestock attacks/deaths) is voluntarily undertaken in all states. Victoria is the only state that has compulsory reporting requirements. Reporting of methods of control is voluntary in most states with the exception of the delivery of poisons. Pattern of use of baits within regions is usually not recorded although Victoria has the capability (but not routinely used).

The collection, analysis and communication of compulsorily reported data and public Wild Dog Control Program data has substantially improved since 2012. Dedicated reporting apps such as Dogbytes allow real-time point-of-activity reporting. This real time spatial reporting into the central database only exists in Victoria.

These data are supported by novel tools such as genetic sampling of individuals and sophisticated mathematical analysis of fixed camera records which can provide estimates of the underlying dog population size and stability.

The recording of data has enabled the Departments to demonstrate at local and regional levels the scale and scope of wild dog control activities.

This has been beneficial in:

improving transparency of management activities as control data can be generated and explained to landholders;

improving landholder perceptions of the program as improving access to control data has led to improved knowledge of program activities;

improving planning and coordination of control activities as program managers are better able to design and coordinate program activities by gathering, interrogating and sharing data;

improved performance evaluation as data are used to generate quarterly program performance reporting;

improving reporting of incidents as there has been a substantial decline in reported livestock attacks; and

use of camera monitoring to identify problem dogs and observe their behaviour, enables staff to share information with impacted farmers and develop strategies to deal with the dogs involved with individual incidents.

Examples of research activities that have utilised the data include:

DNA surveys for development of Dingo Action Statement and other analyses;

site selection for remote ground baiting and aerial baiting;

monitoring of remote ground baiting to meet Commonwealth requirements for aerial baiting;

wild dog movement research;

comparison of trapping and baiting efficacy;

wild dog diet surveys;

Social Benefit Cost Analysis; and

parasite surveys.

The data have been used to develop evidence-based decision making and help improve operational design including:

• 
  creating a containment strategy for the proactive baiting program;
  
• 
  identifying where a strategic, specific and targeted need exists to apply for work outside the 3km buffer zone for the protection of livestock;
  
• 
  supporting applications for emergency control beyond the LPB during the year and not cited in the annual work plan;
  
• 
  determining where baiting or trapping would be most effective;
  
• 
  supporting applications for external funding;
  
• 
  locating wild dog control positions;
  
• 
  identifying where a control measure could be replaced by a contractor;
  
• 
  developing of the service offer described in the Wild Dog Management Zone work plans;
  
• 
  developing of individual WDC control targets and performance management;
  
• 
  continuous improvement where staff undertake tactical planning for challenging areas;
  
• 
  community engagement planning;
  
• 
  evaluating approved traps;
  
• 
  developing of strategic partnership with Southern Ark to identify complimentary baiting runs and reduce duplication; and
  
• 
  supporting policy development with actual field data.
  

The Dogbytes tool has a number of strengths including:

Being a valuable real time operations reporting and design tool where on ground control activities are recorded.

Substantially improving the completeness, accuracy and comparability of data over time.

Real time recording that improves accuracy and usefulness of the data, particularly the number, locations and timing of attacks and subsequent control measures implemented.

Over time the data collection method enables the construction of detailed time-series data.

Increasing the transparency of control activity thereby enables interrogation of controller management and quality of service.

4.2Gaps

There are a number of gaps with quarterly reporting:

Quarterly reporting is generated on a financial year basis but the program is centred on seasonal control activities. Financial quarter reporting dilutes analysis of seasonal program features.

There is limited knowledge of some key performance relationships given:

the length of time the program has operated with collection of comparable data;

gaps in intelligence and surveillance such as: no estimates or understanding of wild dog population; target and non-target take of some control techniques, limited understanding of wild dog behaviours in relation to bait take; and

limited controlled experimentation and piloting of program variables.

There are a number of data gaps associated with Dogbytes:

Data prior to 2012 is not comparable to data post-2012 — prior to 2012 data were manually recorded and are not consistent or easily comparable with data recorded from 2012 using electronic media. Advice received from DEDJTR is that there were substantial differences in the approach to data recording across field service staff under the manual systems operating prior to 2012.

While noting that the North West region has only relatively recently been brought into the program, there is limited in-field recording of program activity in the North West region in the Dogbytes system.

There does not appear to be a robust systematic method to collect information on the impact of baiting on native wildlife and other non-target species.

Aspects of private control are not recorded. As a consequence, data are generally more accurate for public program activity but incomplete for private management activities. This includes gaps in knowledge of the number, location and success of control measures on private land, particularly the timing and location and number of take of private baiting. There are also no data recording of other forms of private management including changes to farm management practices such as private fencing controls, animal husbandry and management practices and use of guard animals.

Definition weaknesses of some performance indicators. Our exposure to the Dogbytes database indicates some trapping measures are not appropriate indicators of trapping effort. For example, the number of traps set bears little resemblance to trapping effort since traps set are only recorded if the location of the trap is changed. A better measure of trapping effort is trap nights.

There are gaps in the way the data are accessed and used:

There is limited centralised and coordinated resources within DELWP and DEDJTR to systematically interrogate the database for the purposes of program research and development.

Data storage and analysis is not fully centralised. A number of aspects of program performance management for externally funded projects are collated external to Dogbytes. Quarterly reporting of performance is generated from a number of recording and evaluation systems.

Dogbytes is largely used as an operational planning and reporting tool. The database has been used in a limited way for strategic assessments. Strategic data assessment has been largely limited to the demands to service quarterly reporting to the WDCAC. For example, there does not appear to have been a systematic, applied and empirical or econometric analysis of the data.

Limited use of the data as an intelligence mechanism. For example, the Dogbytes database does not appear to be interfaced or integrated with other key databases to improve intelligence and to check veracity e.g. Property Identification Codes, National Livestock Identification System. In a number of instances, we understand there are gaps within the database of key data identifiers for some properties.

Management zones while important for coordinating control and managing community engagement in response to local attacks, are too small and too numerous to identify meaningful changes in program performance. In a number of cases the control activities of one zone are closely intertwined with those of others. There is a need to establish appropriate zonal aggregation in order to understand broad trends in program performance.

4.3Risks

There are a number of risks with the current arrangements for program data including:

Likely inefficiencies and gaps in running multiple data systems and processes to generate reporting, as is currently the case.

The focus of the platform on control activity recording can create a mind-set of reaction and micro planning. There should be a regular cycle of strategic review of seasonal data rather than focusing solely on quarterly performance data.

Financial season analysis dilutes robust interrogation of seasonal performance issues.

Incomplete data on target and non-target take may affect the program’s ability to undertake best practice management.

4.4Opportunities

There are a variety of opportunities to improve the collection and use of program data.

An important first step would be to improve the gathering of data on private and group control activities. The aggregation of individual and community baiting into long time periods and the lack of analysis of control outcomes limits the effectiveness of measuring and analysing private control activities and thereby understand the interaction of public and private control activities and their relative efficacy.

There are opportunities to improve the use of the database as an intelligence tool to improve the efficacy of the program. These could include

Ongoing examination of the relative impact of various control measures on dog attacks and livestock deaths – over time and for different regional zones. This would include developing new metrics that can be automatically calculated and assessed such as measures of control intensity at management zone level that can be used for planning and consultation purposes.

Assessing alternative resource management strategies — for example, examining the impact of higher levels of baiting and trapping in different regional zones through the use of pilots and careful interrogation of Dogbytes.

Analysing the impact of trapping and baiting on the frequency of dog attacks or livestock deaths as a longer time series of Dogbytes data is accumulated.

Extending these data tools to more comprehensively include or link it to community management data.

Integrating the tool with remote surveillance data to assess wild dog versus non-target take for alternative control measures.

Providing live access to the database for field staff so they can view previous reports and control actions in the vicinity of current attacks.

Providing access for the community to the database to record private control activities.

Note that these opportunities need to be weighed against the costs of making changes to the database and program and how they fit with the objective of the program.

5.1Conservation of biodiversity

The Victorian government balances its commitment to reduce the impact of wild dogs on livestock with its management and regulatory requirements around the protection of threatened species, and the need to minimise the impact of wild dog control on Victoria’s biodiversity. In particular, dingoes are a threatened species listed under the Flora and Fauna Guarantee Act 1988 and protected under the Wildlife Act 1975. A challenge in balancing wild dog control and dingo conservation is that dingoes are visually indistinguishable from wild dogs thereby making it impossible to ensure that they are not inadvertently destroyed via wild dog control programs. To address this issue, in 2010, an Order in Council was made under the Wildlife Act 1975 (and extended in 2013 for five years) . This Order declared that dingoes are unprotected wildlife in certain areas of the state, in particular on all private land in Victoria and on public land within three kilometres of any private land boundary across two regions of Victoria (livestock protection zones). These two regions relate to the eastern part of the state and a section of the north west of the state.

The livestock protection buffer provides a strong line of defence between public and private land and enables wild dog control to occur where it is needed to best protect livestock.

A benefit of the livestock protection zone is that it allows a targeted and focused approach to wild dog management as government and private landholders are able to cooperate to undertake wild dog management on both private land and public land.

Wild Dog Zone Management Plans allow government and community to work together to identify areas of strategic importance for wild dog control on public land, noting that authorisation requires wild dog control outside of the 3km zone to be targeted, specific and justified for livestock protection.

5.2Animal welfare

One of the key aims of the Wild Dog Control Program is to minimise attacks on livestock and the consequential animal welfare impacts on the livestock from such attacks. However, the welfare of the wild dogs is also an important consideration when undertaking any control program. Improving animal welfare is a priority of the government and this policy objective must be balanced with the objectives for the Wild Dog Control Program. In 2008, a review was conducted into the welfare outcomes of leg hold traps in Victoria; this review underpinned the regulatory changes made at that time to the Prevention of Cruelty to Animals Regulations 2008 which prescribe the type of traps that can be used and the conditions of use.
The Prevention of Cruelty to Animals Act 1986 also regulates the use of poisons and includes an exemption for use of any poison where done in accordance with specified Acts including the Catchment and Land Protection Act 1994 . This allows the use of poisons such as 1080 for control of wild dogs.

Animal welfare arrangements in these regulations have influence on the choice and implementation of control measures. Of particular relevance is the length of time allowed between trap inspections to minimise suffering once an animal is caught in a trap.

The Prevention of Cruelty to Animals Regulations 2008 provide that a trapped animal must not be left alive in a trap for more than 24 hours or, in the case of a large leghold trap for wild dogs, as otherwise approved by the Minister. A Ministerial approval to allow government agents to leave animals alive in wild dog traps for up to 72 hours in certain approved Victorian parishes has been in place since 2008. This approval process was included in the regulations to allow the Wild Dog Control Program time to move to a 24 hour time limit. We estimate the impacts on the program of specifying a maximum period for which animals may be left alive in a trap (24, 48 or 72 hours) in section 3.6 as the practical implication for the Wild Dog Control Program is that the type, size and design features of traps set must be checked within that period to confirm the absence or presence of an animal (targeted or non-targeted) in the trap.

5.3Chemical use

There are a variety of restrictions and requirements on the use of chemicals that necessarily influence the program. Some key considerations include:

regulation on the access and use of 1080 poison and manufactured baits for wild dog baiting which given resource constraints could reduce the access to baits and timeliness of baiting and increase the costs of baiting control;

conditions on the location, timing and use of poison without appropriate authorisation;

banning of some poisons on lethal trap devices on animal welfare grounds; and

CPEs are yet to be brought into wide spread use on public land.

5.3.1Regulation on the access and use of 1080 on public land.

As a Schedule 7 poison, a person must have appropriate certification to access and use 1080; wild dog controllers and private landholders must be trained and authorised which involves users undertaking an approved course and being subject to ACUP program auditing. Community baiting is undertaken on a seasonal basis which allows relatively long lead times to undertake appropriate training. The Wild Dog Control Program provides a range of opportunities to improve bait access this includes, supported access to ACUPs training, bait delivery days and the provision of fresh and manufactured baits. As reported in Report 1 a large number of landholders have undertaken ACUP training as part of the program. The costs of this training have been paid for by the Baiting Coordinator Project.

There have also been improvements to the access of 1080 baits through changes to arrangements enabling mobile bait manufacturing and to access to baits from nearby Local Land Councils in NSW.

5.3.2Conditions on the location, timing and use of poison

Baiting can only occur with appropriate authorisation and in defined areas. For baiting on public land, approval for the laying of baits must be provided by the Land Manager. This approval has a lead time in the order of three months. The pre-approval process is designed to ensure the appropriate alignment of baiting activity with other land management activities and public land uses in the area. Our consultations indicate that pre-approvals and legal requirements regarding neighbour notification of baiting activities take in the order of 3 months. This can have implications for the flexibility to undertake baiting operations on an as needs basis (and could shift effort to less suitable control measures) or respond to changes in program policy and implementation.

There are circumstances under which blanket baiting approvals for some areas are granted that enable immediate baiting in specific locations. One approach could be to define blanket areas and require pre-approvals by exception such as areas of shared public amenity.

Private baiting activity is restricted on public land. The Wild Dog Control Program has introduced the ‘Over the Fence’ program and the broader Baiting Coordinator Project to, in part, help enable collective private control on public land. These programs have contributed to an increase in the level of private control activity on public land, although in comparison to the baiting Coordinator Project only a relatively small number of landholders participated in the Over the Fence program.

5.3.3Lethal trap devices (LTD)

Under the Victorian Prevention of Cruelty to Animals Regulations 2008 (POCTA regulations), a LTD is defined as a device attached to a trap that contains a lethal substance for the purposes of causing the rapid death of trapped target animals through ingestion of the substance.

The provisions of the POCTA regulations are designed to reduce the period of suffering by a trapped animal. Victoria has the Code of Practice for Wild Dog Control and has endorsed, the National Model Code of Practice for the Humane Control of Wild Dogs, which specifies that where leg-hold traps cannot be checked at least once daily, a lethal toxin must be applied to the jaws of the trap or a lethal trap device must be used (noting there is Ministerial approval to leave an animal alive in a trap up to 72 hours in certain approved parishes).

Strychnine laced traps are allowed in some jurisdictions and are used primarily in more remote applications where it is not feasible to regularly inspect traps. Strychnine, like 1080, is Schedule 7 poison and its use is restricted by licence. Consistent with a number of other jurisdictions, Victoria does not allow the use of Strychnine laced traps. A decision was made in 2013 not to progress the introduction of strychnine LTD’s in Victoria because it is not species specific and its mode of action is not considered humane.

We note there is a national approach to the cessation of strychnine laced traps.

New LTDs that are target specific, humane and effective are under development. One particular LTD for wild dogs and foxes uses PAPP as the toxin, which will bring about the death of the trapped animal in a timely and humane manner. The Proof of Concept phase has been completed but the field studies are ongoing. The registration application to the federal regulator APVMA is being prepared. The APVMA anticipate that a 12 to 18 month period will be required to approve the registration, once it is submitted.

Any subsequent introduction into Victoria, is likely to require amongst other things, amendment of regulatory framework and the development of Codes of Practice to ensure the safe, effective and humane introduction and application of this technique

5.3.4Canid pest ejectors

Canid pest ejectors (CPEs) have been widely used for many years in some United States jurisdictions and more recently by the NSW Parks and Wildlife Service as part of pilot program. In 2015, APFVMA registered CPE for use on private land and are expected to become widely used in Victoria and other jurisdictions as a wild dog control measure in the near term.

Like all forms of wild dog control, CPEs require skilled use in order to optimise their efficacy. CPEs are likely to improve the overall efficacy of wild dog control as they:

require less inspection than traps;

limit non-target species take, noting that several threatened species in Victoria will be at risk from CPEs, including dingoes;

prevent the opportunity for target and non-target species to cache baits;

have similar variable costs to ground baiting; and

have similar fixed costs to traps and higher fixed costs to buried baits.

We model the application of CPEs to the program in section 7.

5.4Workplace health and safety

Workplace health and safety arrangements affect the design and implementation of the Wild Dog Control Program. Aspects of workplace health and safety concerns overlay with restrictions on chemical use and the use of some control technologies.

Control measures involve at times physically and mentally demanding work. Operations in Wild dog Management Zones require significant travel of wild dog controllers to remote areas where they work in isolation from other staff.

As a result, DELWP has developed a custom and practice of wild dog controllers calling supervisors to update their status (safety check) three times a day to discuss field operations. To operate staff in the field on weekends could require the provision of similar supervision services or greater flexibility in the practice and mechanisms for safety checks and field reporting. As such, workplace health and safety arrangements have limited the use of casuals over weekend periods. We examine the impact of using casuals on weekends in section 3.6 when investigating 24 and 48 hour trap inspection times.

Another important consideration is the custom and practice that wild dog controllers can only inspect the traps that they have personally set. Under animal welfare legislation the person setting the trap is liable for any third party management of the trap. For that reason, DELWP has decided it is not appropriate to have someone else check and deal with trap content. This limitation has the practical impact of preventing casual and contract workforce operating traps for short or inter-dispersed periods. As such, these staff only be used for baiting activities or where applied for an extend period of time to trapping activities. The impact of casual and contract workforce flexibility on the program design are modelled in section 3.6.
6. Capitalising on operational efficiencies

Key Task: Analyse current efforts to capitalise on operational efficiencies between private and public sector programs, initiatives and organisations, to assess whether mutually beneficial outcomes are being realised, to inform whether there are opportunities to build on this approach

The Wild Dog Control Program has achieved substantial gains by increasing the level of private baiting control and the coordination and timing of that effort. The program has achieved substantial and sustained improvement in the level of seasonal baiting by private landholders. The timing of the baiting is tightly clustered to align with seasonal wild dog population pressure points and the timing and location of public effort.

The Baiting Coordinator Project has been the central program mechanism along with zone management groups that operate under the project to achieve these outcomes; this program was discussed in detail in Report 1.

Considerations for public trapping arrangements mean that proactive control is secondary to reactive control due to labour costs. Containment is more effective with a greater balance of proactive control. This can only be leveraged by increasing the private control effort on public land.

Based on the known costs, practicality of management, levels of likely take, and alignment with wild dog population pressure points, increasing the level of coordinated and targeted private baiting on private and public land is the most effective approach to improving operational efficiency between private and public sector programs.

Given limited public resources and operating constraints to address wild dog control there are opportunities to enhance wild dog take by increasing the levels of complementary private provision of baiting on public land and potentially increasing other complementary strategies on private land.

Private control efforts that are coordinated are likely to increase the effectiveness of public control. The Baiting Coordinator Project should continue in some form. In the absence of leverage, external funding and/or additional public funding it may be necessary to reduce public control measures in order to provide the resources for the ongoing support of community led seasonal baiting programs.

There are opportunities to extend the reach and capability of the program by continuing to improve access to baits, streamlining and reducing red tape involved with private baiting programs. This could include:

reallocation of some budget from reactive control to support increased capability of private landholders to undertake coordinated baiting – including training, extension and outcome reporting;

continuing to improve provision of data to inform control activities; and

continuing to stream-line and improve access to fresh baits and poisons.
7.Optimal mix of approaches

Key Task : Recommends the optimal mix of approaches, tools and technologies, current or otherwise, to be used in Victoria, considering the efficacy, cost effectiveness, return on investment, animal welfare issues and impact on off target species

7.1Background

In this section we examine the impact on a range of comparative performance indicators of alternative control measures. This analysis is undertaken with reference to the trapping scenarios analysed in Section 3.6. The options are also extended to include operations only involving trapping and operations only involving baiting.

Comparative performance indicators include:

likely dog take;

likely non-target take;

costs of control; and

cost effectiveness of likely dog take.