{"id":"F2026L00375","name":"Environment Protection and Biodiversity Conservation (National Recovery Plan for the Pookila (New Holland mouse) Pseudomys novaehollandiae) Instrument 2026","slug":"environment-protection-and-biodiversity-conservation-national-recovery-plan-for-the-pookila-new-holl","collection":"legislative_instrument","jurisdiction":"commonwealth","status":"in_force","isInForce":true,"actNumber":null,"makingDate":null,"administeringDepartment":null,"currentVersion":{"id":107982,"registerId":"commonwealth-F2026L00375-current","compilationNumber":null,"startDate":"2026-04-03","status":"InForce","reasons":null,"registeredAt":null},"sections":[{"sectionNumber":"Tables","sectionType":"part","heading":"Tables","content":"# Tables\n\nTable 1: Glossary of important terms as defined in this recovery plan.......................vi\n\nTable 2: List of acronyms used within the National Recovery Plan for the Pookila..............vii\n\nTable 3: Conservation status of the Pookila..........................................2\n\nTable 4: Groups with primary interest in Pookila recovery...............................3\n\nTable 5: Populations of Pookila...................................................9\n\nTable 6: Risk matrix..........................................................21\n\nTable 7: Strategies and actions to ensure the area of occupancy increases (Objective 1).........25\n\nTable 8: Strategy and actions to ensure the national population size is stable or increasing (Objective 2)......30\n\nTable 9: Strategies and actions to ensure genetic diversity is improved (Objective 3)...........32\n\nTable 10: Strategies and actions to ensure First Nations people are collaborators in management and recovery (Objective 4)......35\n\nTable 11: Strategies and actions to ensure community awareness and engagement with recovery efforts have increased (Objective 5)......36","sortOrder":0},{"sectionNumber":"Figures","sectionType":"part","heading":"Figures","content":"# Figures\n\nFigure 1: A captive-bred Pookila individual (Pseudomys novaehollandiae)....................5\n\nFigure 2: Modelled distribution of the Pookila........................................7\n\nFigure 3: Photographs of contemporary Pookila habitat................................13\n\nFigure 4: Camera trap and bait canister set-up to improve detection.......................38","sortOrder":1},{"sectionNumber":"Executive Summary","sectionType":"part","heading":"Executive Summary","content":"# Executive Summary\n\nThe National Recovery Plan for the Pookila (New Holland mouse) Pseudomys novaehollandiae is the first recovery plan to be developed for this Vulnerable listed threatened species. The plan guides conservation management across its range and identifies actions necessary to slow the decline and support the recovery of the Pookila so that its chances of long-term survival in nature are maximised.\n\nThe Pookila has a fragmented distribution across southern Queensland, northern and central coastal areas of New South Wales, central southern coastal Victoria, and north-eastern Tasmania including Flinders Island. Populations are typically coastal but in Queensland and New South Wales can occur up to 300 km inland. Habitats occupied include open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heathy understorey, and vegetated sand dunes.\n\nThe principal threats to the Pookila are:\n\n Loss, disturbance and modification of habitat\n\n Inappropriate fire regimes (too frequent, too large, too severe, or poorly timed)\n\n Climate changes causing:\n\n mortality from increasing frequency and intensity of extreme heat events\n\n habitat loss through increasing intensity and frequency of drought and/or flood events\n\n Excessive predation by invasive fauna species\n\n Loss of genetic diversity from having small and fragmented populations, and\n\n Increased exposure to novel disease.\n\nInteractions between threats must also be considered in managing the recovery of the Pookila.\n\nThe objectives of this plan are that within 10 years:\n\n- area of occupancy has increased\n- national population size is stable or increasing\n- genetic diversity is improved\n- First Nations people are collaborators in management and recovery, and\n- community awareness and engagement with recovery efforts have increased.","sortOrder":2},{"sectionNumber":"Glossary","sectionType":"part","heading":"Glossary","content":"# Glossary\n\nTable 1: Glossary of important terms as defined in this recovery plan\n\n| Term | Definition |\n| ------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| Area of occupancy (AOO) | AOO is a standardised measure that represents the area of suitable habitat known, inferred or projected to be currently occupied by the species. Under the IUCN definition, it is estimated using 2 x 2 km grids to enable comparison with the IUCN Criteria thresholds. It is not a fine-scale estimate of the actual area occupied. In some cases, AOO is the smallest area essential at any stage to the survival of existing populations of a species (such as breeding sites for migratory species). |\n| ESU | Evolutionarily Significant Units is a concept used by conservationists to strategically prioritise taxa or populations of a species for recovery purposes (for example, to target for captive-breeding), when resources are limited or when the granularity of genetic diversity is not sufficiently captured in the current overarching taxonomic status (Avise 1989; Moritz 1994). |\n| Fragmentation | Fragmentation leads to increased extinction risk for small and relatively isolated populations (in certain circumstances this may be inferred from habitat information). These small populations may become extinct, with a reduced probability of recolonization (IUCN, 2019). |\n| Gene flow | Gene flow is the movement of genes into or out of a population. Such movement may be due to dispersal of individual organisms that reproduce in their new populations, or through assisted translocations. |\n| Genetic diversity | Genetic diversity refers to the range of different inherited traits/genes within a species or a population. High genetic diversity would indicate that the population contained a wide variety of genes and traits which may aid the population in adapting to changing environments. |\n| Genetic rescue | Genetic rescue is the process of increasing population-level viability through the immigration of alleles in small, isolated and inbred populations/populations to reduce the extinction risk and restore genetic diversity (Whiteley et al. 2015). |\n| Important populations | Populations identified by the National Pookila Recovery Team as a priority target for conservation action and protection. |\n| Mini Safe Haven | A Mini Safe Haven (MSH) is a permanently fenced core area of the landscape that permits the two-way passage of a focal species, while being maintained to preserve a low-to-zero risk of key threats inside; allowing for in-situ learning and adaptation by the focal species to those threats at a rate and intensity of exposure determined by the animals themselves (Smith et al. 2023c). |\n| Population Viability Analyses (PVAs) | A process which utilises quantitative analytical tools to estimate the probability that a population, or collection of populations, will persist for a set time in a particular environment, thereby determining the extinction risk under a set of conditions. |\n| Specific Needs Assessments (SNAs) | SNAs utilise a structured decision support process for comparing the benefits and cost-effectiveness of various management strategies for threatened species/population/ecosystem recovery. |\n| Translocations | Conservation translocation is the human-assisted movement of organisms from one site for release in another. Translocations can either take the form of an introduction or reintroduction. The latter is the release of the animal within the species’ historic range from which it has disappeared. Therefore, a reintroduction aims to re-establish a viable population of the species within its historic range (IUCN/SSC 2013). Supplementation is the translocation and release of an animal into an existing population of conspecifics. |","sortOrder":3},{"sectionNumber":"Acronyms","sectionType":"part","heading":"Acronyms","content":"# Acronyms\n\nTable 2: List of acronyms used within the National Recovery Plan for the Pookila\n\n| Acronym | Name |\n| ------------ | ------------------------------------------------------------------------------- |\n| ANU | Australian National University |\n| CSIRO | Commonwealth Scientific and Industry Research Organisation |\n| DCCEEW Cwlth | Department of Climate Change, Energy, the Environment and Water Commonwealth |\n| DCCEEW NSW | Department of Climate Change, Energy, the Environment and Water New South Wales |\n| DEECA | Department of Energy, Environment and Climate Action Victoria |\n| DETSI | Department of the Environment, Tourism, Science and Innovation Queensland |\n| DNRE | Department of Natural Resources and Environment Tasmania |\n| EPBC Act | Commonwealth Environment Protection and Biodiversity Conservation Act 1999 |\n| GLaWAC | Gunaikurnai Land and Waters Aboriginal Corporation |\n| IUCN | International Union for the Conservation of Nature |\n| NPRT | National Pookila Recovery Team |\n| NPWS NSW | National Parks and Wildlife Service New South Wales |\n| NRM | Natural Resource Management |\n| RBGV | Royal Botanic Gardens Victoria |\n| RFA | Regional Forest Agreement |\n| TEK | Traditional Ecological Knowledge |\n| ZAA | Zoo and Aquarium Association, Australasia |\n| ZV | Zoos Victoria |","sortOrder":4},{"sectionNumber":"1","sectionType":"part","heading":"Background","content":"# 1 Background\n\n## 1.1 Introduction\n\nThe Pookila (New Holland mouse; Pseudomys novaehollandiae) is a small, burrowing, rodent native to south eastern Australia, that is patchily distributed from Tasmania to south-east Queensland. The species has had catastrophic declines that were recorded since the discovery of new populations in the 1970s. Isolated remnant populations are at high risk of extirpation and genetic inbreeding.\n\nDistribution is fragmented across southern Queensland (Qld), the northern and central coastal areas of New South Wales (NSW), central southern coastal Victoria (Vic), and north-eastern Tasmania (Tas) including Flinders Island (Burns et al. 2023a). While the species is typically restricted to coastal areas, it occupies areas up to 300 km inland in Qld and NSW.\n\nSince the species’ description in 1843, more than a dozen known populations have been lost and many more have declined. It has not been live trapped on mainland Tas since 2010 and in Vic, seven of twelve populations have become extinct since the 1970s (Burns et al. 2023b). Encouragingly, the Pookila was rediscovered on Flinders Island, Tasmania in 2021 and was last recorded in southern Qld in 2022. It has now been translocated into Mulligans Flat Woodland Sanctuary in the Australian Capital Territory (ACT), Yiraaldiya National Park feral predator-free area (NSW), reintroduced into the Cranbourne Royal Botanic Gardens (Vic) in 2024, and has numerous ongoing and successful captive-breeding programs.\n\nEuropean colonisation in Australia exposed the Pookila to compounding threats that caused population declines across its range. These threats include the direct destruction and fragmentation of habitat due to land clearing for agriculture, urbanisation, road construction, and forest plantations, as well as predation by introduced cats and European red foxes (DEWHA 2010; Burns 2019a). Introduced rodents may outcompete the Pookila under some conditions, and both introduced rodents and cats are vectors of novel diseases. Infection may also cause mortality and/or increase the Pookila’s susceptibility to predation (for example, toxoplasmosis spread by feral cats) (Lazenby et al. 2019; Burns 2019a).\n\n## 1.2 Conservation status\n\nThe Pookila is listed as Vulnerable under the Environment Protection, Biodiversity and Conservation Act 1999 (EPBC Act), Vulnerable (ACT, Qld, NSW, VIC) and Endangered (Tas) (Table 3).\n\nThe Pookila was first listed under the EPBC Act in the Vulnerable category on 11 August 2010. It was reassessed by the Threatened Species Scientific Committee in July 2024 retaining its Vulnerable category status under Criteria 1 and 2 (DCCEEW 2024b).\n\nCriterion 1 (A4b) was met due to the severe declines in population size estimated to have occurred in Tas and Vic and the significant declines suspected to have resulted from the impacts of the 2019 and 2020 bushfires in NSW, with the Committee concluding that the Pookila was subject to a population size reduction of at least 30% (DCCEEW 2024b).\n\nCriterion 2 B2ab(i,ii,iii,iv,v) was met due to the limited Area of Occupancy (AOO), low/very low number of locations (six), and estimated continuing decline in EOO, AOO, area of habitat, number of locations and populations, as well as the number of mature individuals (DCCEEW 2024b).\n\nTable 3: Conservation status of the Pookila\n\n| Listing | Conservation Status |\n| ---------------------------------------------------------------------------- | ------------------- |\n| Environment Protection and Biodiversity Conservation Act 1999 (Commonwealth) | Vulnerable |\n| Nature Conservation Act 2014 (Australian Capital Territory) | Vulnerable |\n| Nature Conservation Act 1992 (Queensland) | Vulnerable |\n| Biodiversity Conservation Act 2016 (New South Wales) | Vulnerable |\n| Flora and Fauna Guarantee Act 1988 (Victoria) | Vulnerable |\n| Threatened Species Protection Act 1995 (Tasmania) | Endangered |\n| IUCN Red List of Threatened Species | Vulnerable |\n\n## 1.3 Cultural and community significance\n\nThe Pookila currently occurs on the Country of the Awabakal, Banbai, Barunggam, Biripi, Bunurong, Boonwurrung, Bundjalung, Burrgati, Daingatti, Darkinung, Dharawal, Dharug, Eora, Geawegal, Gumbainggir, Gunaikurnai, Kuringgai, Gunaikurnai, Ngunnawal, Palawa, Wiradjuri, Wonnarua, Worimi and Yuggera Peoples.\n\nThe species has likely been known by numerous names in different Aboriginal languages across its geographic range. The name Pookila is derived from the Ngarigo word for ‘mouse’ – bugila (Braithwaite et al. 1995; Hercus n.d.), deviating somewhat from traditional Ngarigo pronunciation (Braithwaite et al. 1995). The assignation of that name was part of an initiative by the CSIRO to provide Indigenous Australian names for Australian rodents (Braithwaite et al. 1995). This was considered a way to avoid the undesirable associations with introduced vermin often attached to the words ‘rat’ and ‘mouse’, impeding conservation of native rodents due to negative public perception (Braithwaite et al. 1995). An alternative name could have been ‘tuwaranga’ which means ‘mouse’ in palawa kani, the language of the Tasmanian Aboriginal people (Tasmanian Aboriginal Centre, 2023), however, the name Pookila has been associated with this species since 1995 and is included on the Australian Government’s SPRAT database.\n\n## 1.4 National Pookila Recovery Team\n\nA National Pookila Recovery Team (the recovery team or NPRT) was established in 2021 with the aim of improving recovery efforts for the species at a national scale. A previous Pookila Recovery Team (1995 to 2002) was largely based in Vic. The current recovery team has representation across the species’ entire distribution. Beyond the core recovery team there are three sub-groups: captive breeding, habitat and fire management, and Vic management.\n\nThe role of the recovery team is to facilitate and encourage collaborations across jurisdictions, provide advice to partners and land managers, increase engagement, improve the public perception of the Pookila as well as the responsibility of advising and coordinating the implementation of recovery actions. The recovery team also seeks to facilitate opportunities for mutual learning and skill sharing with First Nations people. The membership of the recovery team includes species experts and land managers from relevant government agencies and non-government organisations and draws from a broader network of expertise from First Nations people, researchers, and community groups.\n\n## 1.5 Key partners\n\nImplementation of recovery planning is a collaborative effort involving a range of key partners and affected interests. The groups likely to have primary interest in recovery planning for the Pookila are listed in Table 4.\n\nTable 4: Groups with primary interest in Pookila recovery\n\n| Organisations | Involvement/ interest |\n| -------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| ACT Government | Management of the land of the first Pookila translocation site. Ongoing monitoring at the site and potential future releases in the area. |\n| Australian National University (ANU) | Captive-breeding program for NSW individuals. Ongoing monitoring and management in Mulligans Flat Woodland Sanctuary and potential future reintroductions in the ACT and other jurisdictions. |\n| Aussie Ark (AA) | Maintains an insurance population. Potential development of a captive-breeding program for the Pookila and future reintroductions into cat and fox free sites in NSW. |\n| Corangamite Catchment Management Authority (CCMA) | Managers of habitat under threat from P. cinnamomi. |\n| Department of Climate Change, Energy, the Environment and Water (DCCEEW, Cwlth) | The Pookila is a priority species in the Threatened Species Action Plan 2022-2032, which provides guidance and support in the design and delivery of projects for priority species. Recovery plans inform decision making under the EPBC Act referral and assessment process. |\n| Department of Energy, Environment and Climate Action (DEECA) | Primary agency involved in threatened species management on public and private land in Vic. |\n| Department of Natural Resources and Environment (DNRE) | Primary agency involved in threatened species management on public and private land in Tas. |\n| Department of the Environment, Tourism, Science and Innovation (DETSI) | Primary agency involved in threatened species management on public and private land in Qld. |\n| Flinders Island Aboriginal Association Incorporated (FIAAI) | Traditional Owners of Flinders Island population Tas. |\n| Friends of the Prom (FOTP) | Group established to support the management of Wilsons Promontory National Park. |\n| Gippsland Water (GW) | Manages a significant Vic population. |\n| Gunaikurnai Land and Waters Aboriginal Corporation (GLaWAC) | Joint managers of Lakes National Park, and Gippsland Lakes Coastal Park, and have Registered Aboriginal Party status of Providence Ponds Flora and Fauna Reserve in Vic. |\n| Land holders/Landowners | Consultation and engagement with all types of landowners will play a role in the success in meeting the recovery plan’s actions and objectives. |\n| Local councils | Community involvement projects and cooperation with management actions on a local level. |\n| Monash University (MU) | Storage and research of Pookila cell lines and genetic tissue. Supporting research of the genetic sequencing of the Pookila and subsequent projects surrounding genetic information. |\n| Moonlit Sanctuary Wildlife Conservation Park (MSWCP) | Managing a captive population of Pookila as well as community conservation and engagement activities. |\n| Forestry agencies | Actions and management regarding forestry operations and Regional Forest Agreements. |\n| NSW Department of Climate Change, Energy, the Environment and Water (DCCEEW NSW) | Primary agency involved in threatened species management on public and private land in NSW. |\n| NSW National Parks and Wildlife Service (NPWS NSW) | Implementation of appropriate management actions within parks and reserves in NSW, including reintroduction of the Pookila to establish new populations in NPWS-managed feral predator-free areas. |\n| Parks Victoria (PV) | Implementation of appropriate management actions concerning habitat within Vic parks and reserves. |\n| National Threatened Species Institute (NTSI) Priam Psittaculture Centre (PPC) | Host of a captive breeding program founded from northern ESU individuals. |\n| Royal Botanical Gardens Victoria Cranbourne (RBGVC) | Location for reintroduction to establish a new population. |\n| Tasmanian Aboriginal Land and Sea Council Aboriginal Corporation (TALSCAC) | Traditional Owners of Northeast population Tas. |\n| Trust for Nature (TFN) | Managing properties adjacent to occupied habitat. |\n| Turner Family Foundation (TFF) | Manager of Spicers Peak Nature Refuge south-east Qld habitat. |\n| University of Canberra (UC) | Involved in captive husbandry and dietary analyses. |\n| University of Southern Queensland (USQ) | Collaborative research project with the DETSI and the Turner Family Foundation to refine knowledge of distribution and ecological requirements in south-east Queensland. |\n| University of Sydney (USyd) | Analysis and advice on captive-breeding genetics. |\n| Wadawurrung Traditional Owners Aboriginal Corporation (WTOAC) | Traditional Owners of Anglesea population Vic. |\n| Woodlands and Wetlands Trust (WWT) | Community engagement, education and volunteer programs at Mulligans Flat Woodland Sanctuary. |\n| World Wildlife Fund Australia (WWF) | Potential partner for island translocation. |\n| Zoos Victoria (ZV) | Programs underway for monitoring, management, captive breeding, genetic rescue, reintroductions, research, studbook management and species coordination, community conservation and engagement. |\n| Zoo and Aquarium Association Australasia (ZAA) | Coordination of the NPRT, initial engagement/consultation with First Nations people, and assistance with studbook management and coordination. |\n\n# 2 Species information\n\n## 2.1 Taxonomy and description\n\nPseudomys novaehollandiae, family Muridae, is endemic to Australia and was first described in 1843 as Mus novaehollandiae (Waterhouse 1843). It was later transferred to the genus Pseudomys with the new name of Pseudomys (Gyomys) novaehollandiae (Thomas 1910). There is some geographic overlap and occasional evidence of hybridisation with its sister taxon, P. delicatulus (Delicate mouse), in the north of its range (Roycroft et al. 2024). However, the two taxa were found to be both genetically and morphologically distinct (Ford 2003; Roycroft et al. 2024).\n\nThe Pookila is a nocturnal, terrestrial, burrowing rodent. It is small, with a head-body length of approximately 65–90 mm and a tail length of 80–105 mm (Menkhorst & Knight 2011). The species’ dorsal coat is sandy brown with black flecks and the ventral coat is white (Figure 1). The tail appears bicoloured with a darker dorsal side and paler pink ventral side, and the feet are pink with fine white hairs (Figure 1).\n\nIndividuals in the south of the range are heavier than those further north. Tas individuals are the heaviest, weighing 19–28 g, followed by Vic at 18–‍25 g, then Qld and NSW at 12–20 g (Hocking 1980; Burns et al. 2023a). The species is most similar in appearance to the delicate mouse, which is generally smaller and more finely featured. It is superficially similar in appearance to the invasive house mouse (Mus musculus). It is distinguishable by its larger prominent eyes, larger rounded ears, lack of ‘mousy’ odour and absence of a notch on the upper incisors (note that examining dentition is not recommended for identifying live animals) (Seebeck et al. 2003; DEWHA 2010; Burns et al. 2023a; DCCEEW 2024b). The Pookila’s body shape is also thicker and more rounded compared to the house mouse, lacking an obvious neck (Burns 2019b).\n\nFigure 1: A Pookila (Pseudomys novaehollandiae) individual showing identifying characteristics: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.\n\n![Left: An image of a captive Pookila (New Holland Mouse) on a branch.\nRight: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.](image.004.png)Source: Left © Zoos Victoria (2023), Right and insets Kiarrah Smith (2025), Coexistence Conservation Laboratory, Australian National University.\n\n## 2.2 Distribution\n\nThe Pookila has a fragmented distribution across coastal areas of central-eastern and south‑eastern Australia (Figure 2). It occurs in southern Qld, the northern and central coastal areas of NSW, central-southern coastal Vic, and north-eastern Tas including Flinders Island (Burns et al. 2023b; DCCEEW 2024b). A large portion of its distribution occurs within areas with Regional Forest Agreements (RFAs), Indigenous Protected Areas and National Parks. In Qld and NSW, the Pookila occurs in the World and National Heritage Listed Greater Blue Mountains Area and the Gondwana Rainforests of Australia. It is known only from subfossil records in National Heritage Listed areas including the Australian Alps National Parks and Reserves, Ku-ring-gai Chase National Park, Lion, Long and Spectacle Island Nature Reserves and the Great Ocean Road region (Ford 2003; Smith et al. 2023c). While most modern detections of the species exist from coastal areas, it has also been located in areas up to 300 km inland in Qld and NSW. The species’ distribution is believed to have once been continuous, forming one large population throughout these areas.\n\nThe species’ historic distribution is inferred through the presence of subfossils as well as genetic information (Ford 2003; Breed & Ford 2007; Smith et al. 2023c). Its current distribution is attributable both to longer-term climatic and landscape changes, as well as a large population decline following European colonisation (Breed & Ford 2007; Theden-Ringl et al. 2018; Roycroft et al. 2021). Despite the ACT and surrounding region not being mapped within the current Pookila distribution (Figure 2), there is subfossil evidence of historical occurrence in nearby areas with similar habitat (Ford 2003; Smith et al. 2023c).\n\nThe extent of occurrence (EOO) for the species was estimated at 326,000 km2 and the area of occupancy (AOO) for the species was estimated at 694 km2 (DCCEEW 2024b). The AOO may have further contracted since this estimate, as nearly half of the species’ habitat was affected by the 2019 and 2020 bushfires. However, this includes approximately 30% of the species’ habitat that burned at low to medium severity (Legge et al. 2022). The species is likely to survive low to medium severity fires, but high severity fires could greatly reduce survivability (Wilson et al. 2018; Burns & Phillips 2020; DCCEEW 2024b).\n\nThe Pookila is associated with early to mid-stage vegetation succession following fire events (Woinarski & Burbidge 2016). Therefore, some recovery after the 2019 and 2020 bushfires is possible due to potential development of suitable habitat post-fire (DCCEEW 2024b).\n\nFigure 2: Modelled distribution of the Pookila\n\n![The modelled distribution of the Pookila or the Pookila habitat is known or likely to occur from south east Queensland along the northern and central east coast of New South Wales to costal regions in Victoria, predominantly to the east, all the way to Flinders Island and north and eastern Tasmania. The Pookila is also likely to occur 100-300 km inland in New South Wales. Pookila or Pookila habitat may occur in close proximity to these known or likely regions. Refer to subheading 'Species distribution' for more information. ](image.005.jpeg)\n\n> Note: Source: Base map Geoscience Australia; species distribution data Species of National Environmental Significance database.\n\n> Note: Caveat: The information presented in this map has been provided by a range of groups and agencies. While every effort has been made to ensure accuracy and completeness, no guarantee is given, nor responsibility taken by the Commonwealth for errors or omissions, and the Commonwealth does not accept responsibility in respect of any information or advice given in relation to, or as a consequence of, anything contained herein.\n\n> Note: Species distribution mapping: The species distribution mapping categories are indicative only and aim to capture (a) the habitat or geographic feature that represents recent observed locations of the species (known to occur) or habitat occurring in close proximity to these locations (likely to occur); and (b) the broad environmental envelope or geographic region that encompasses all areas that could provide habitat for the species (may occur). These presence categories are created using an extensive database of species observations records, national and regional-scale environmental data, environmental modelling techniques and documented scientific research.\n\n## 2.3 Population size and trends\n\nIn 2008, the national population size of the Pookila was estimated at fewer than 10,000 individuals (Menkhorst et al. 2008; DEWHA 2010) and was reported to be 8,000 individuals in the IUCN assessment by Woinarski and Burbidge (2016). In the assessment for listing under the EPBC Act, the maximum plausible population size prior to the 2019 and 2020 bushfires was estimated (using calculations of AOO and density) to be 34,700 individuals, with the post-fire plausible population range of 13,100–26,300 (DCCEEW 2024b). There was a reduction in the number of Pookila populations located in Vic (from 12 to 5; Burns 2019a), and in Tas (from 16 to 2; (Lazenby et al. 2019). However, there is limited information from which to infer similar trends across the NSW and Qld populations (Lazenby et al. 2019; Burns 2019a; DCCEEW 2024b). A study conducted over a 32-year period near the Tomago aluminium smelter in NSW indicated an overall declining population (SMEC 2015). Shorter-term studies conducted over 5 years recorded a possible decline at Wyrrabalong National Park but no evident trend over 7 years at Munmorah State Conservation Area (DCCEEW 2024b).\n\n## 2.4 Important populations\n\nThe Evolutionarily Significant Units (ESUs) concept is used by conservationists to strategically prioritise taxa or populations of a species for recovery purposes (for example, to target for captive-breeding), when resources are limited or when the granularity of genetic diversity is not sufficiently captured in the current overarching taxonomic status (Avise 1989; Moritz 1994). There are two recognised ESUs of Pookila, separated by more than 200,000 years of divergence (Burns et al. 2023b). The Southern ESU (Vic populations) is rarer, both genetically and geographically fragmented, and more threatened than the Northern ESU (NSW and Qld populations), which is comparatively genetically healthy and not at immediate risk of extinction (Burns et al. 2023b). Tas populations were not included in the Burns et al. (2023b) analysis due to the lack of data. However, these populations are expected to place either independently or within the Southern ESU and are considered to be at extremely high risk of extinction (Burns et al. 2023b). Given the Tas populations are likely closely related to the Vic populations, until further information is available these populations should be managed as part of the Southern ESU (Burns et al. 2024).\n\nPopulations in Vic and Tas are severely fragmented and are thought to make up less than 5% of Pookila numbers nationally (DCCEEW 2024b). In Tas, the species was known from 16 separate sites in 1998, but its status on the Tas mainland is in doubt with no recent records (Lazenby et al. 2019; DNRE Tas 2023). However, it was rediscovered on Flinders Island in 2021 (DNRE Tas 2023). In Vic, extensive surveys found the species in only 5 of the 12 parks and reserves in which it had previously been recorded (Burns 2019a; Burns et al. 2023b). The remaining populations are spatially isolated and at high risk of genetic issues without assisted migrations to support genetic rescue (Burns 2019a; Burns et al. 2023b).\n\nThe connectivity and extent of populations in NSW and south-east Qld is poorly understood. Genetic evidence indicates the isolation of these populations has occurred relatively recently compared to the isolation of populations in Vic (Burns et al. 2023a; Burns et al. 2023b; DCCEEW 2024b). The extant populations are in sites that are geographically distinct and unlikely to be connected by gene flow, as were the now locally extinct populations.\n\nThere is evidence that Pookila populations are strongly affected by rainfall. In NSW, the length of the breeding season, and thus reproductive output, has been shown to increase during times of above-average rainfall (Kemper 1976, 1980). Abundance in the Anglesea (Vic) population had a strong positive relationship with cumulative monthly residual rainfall (Lock 2005). Densities of this population peaked following 4 years of above-average rainfall and declined during below-average rainfall and drought conditions (Lock 2005; Wilson et al. 2017). There are also data to suggest that population fluctuations at Wilsons Promontory (Vic) were influenced by rainfall patterns (Wilson et al. 2005). All populations are considered important for the long-term recovery and survival of the species (Table 5).\n\n \n\nTable 5: Populations of Pookila\n\n| Region | Population | Status | Year of last detection | Year of last survey | Known or expected decline? | Genetic diversity (relative heterozygosity a) |\n| ------ | --------------------------------------------------- | --------------------------------------------------------------------- | ---------------------- | ------------------- | -------------------------- | -------------------------------------------------------------------------------------------------------------------------------- |\n| ACT | Mulligans Flat Woodlands Sanctuary | Not persisting, translocated | 2021 | 2023 | Known | Translocation trials with individuals from the Australian National University (ANU) captive-breeding colony (using NSW founders) |\n| NSW | Barcoongere State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Barrington Tops State Forest | Extant | 2001 | 2025 | Expected decline | Unknown |\n| | Booti Booti National Park | Extant | 2022 | 2022 | Unknown | Unknown |\n| | Broadwater National Park | Extant | 2024 | 2025 | Known | Low |\n| | Bunjalung National Park | Unknown | 2002 | 2024 | Known | Unknown |\n| | Bylong | Unknown | Unknown | Unknown | Unknown | Moderate |\n| | Carrai National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Chaelundi State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Dalmorton State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Doyles River State Forest | Unknown | 1984 | 1984 | Unknown | Unknown |\n| | Ellis State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Gibraltar Range State Forest | Unknown | 2015 | 2015 | Unknown | Unknown |\n| | Goobang National Park | Extant | 2025 | 2025 | Unknown | Moderate |\n| | Goulburn River National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Kiwarrak State Forest | Unknown | 2019 | 2019 | Unknown | Unknown |\n| | Malara State Forest | Unknown | 1999 | Unknown | Unknown | Moderate |\n| | Malara State Forest | Unknown | 1993 | 1993 | Unknown | Unknown |\n| | Manobalai Nature Reserve | Extant | 2024 | 2024 | Unknown | Unknown |\n| | Marara State Forest | Unknown | 2014 | 2014 | Unknown | Unknown |\n| | Marengo State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Mernot State Forest | Unknown | 2008 | 2008 | Unknown | Unknown |\n| | Mount Royal National Park | Extant | 2023 | 2023 | Unknown | Unknown |\n| | Mummel Gulf National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Munmorah/Chain Valley | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Myall Lakes | Extant | 2022 | 2022 | Known | Unknown |\n| | Nymboida National Park | Unknown | 2015 | Unknown | Unknown | Moderate |\n| | Olney State Forest | Unknown | 2004 | 2004 | Unknown | Unknown |\n| | Port Stephens | Extant | 2022 | 2022 | Unknown | Moderate |\n| | Ravensworth State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Royal National Park | Extant | 2022 | 2022 | Expected decline | Unknown |\n| | Styx River State Forest | Unknown | 2000 | 2000 | Unknown | Unknown |\n| | Watchimbark Reserve | Unknown | 2011 | Unknown | Unknown | High |\n| | Werrikimbe National Park | Unknown | 1984 | 2020 | Unknown | Unknown |\n| | Wollemi National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Wyrrabalong National Park | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Yiraaldiya National Park feral predator-free area | Translocated, persistence to be determined by post-release monitoring | 2025 | 2025 | Known | Translocation trials with individuals from the captive-breeding colonies at ANU and NTSI (PPC) using NSW founders |\n| | Yuraygir National Park | Extant | 2024 | 2025 | Expected decline | Unknown |\n| QLD | Crows Nest/ Ravensbourne/Helidon Hills | Unknown | 2022 | 2025 | Known | Moderate |\n| | Emu Creek (private property) | Unknown | 2022 | 2024 | Known | Unknown |\n| | Main Range National Park/Spicers Peak Nature Refuge | Unknown | 2022 | 2025 | Known | Unknown |\n| | Numinbah Valley | Unknown | 2005 | 2005 | Unknown | Unknown |\n| TAS | Beaconsfield | Unknown | 1978 | 2002 | Known | Unknown |\n| | Flinders Island | Extant in the north of the island, undetected in the south | 2025 | 2025 | Expected decline | Unknown |\n| | Freycinet | Unknown | 2003 | 2025 | Known | Unknown |\n| | St Helens | Unknown | 2009 | 2025 | Expected decline | Unknown |\n| | Waterhouse Conservation Area | Unknown | 2010 | 2025 | Known | Unknown |\n| | Wukalina/Mt William | Unknown | 2004 | 2025 | Known | Unknown |\n| VIC | Anglesea | Locally extinct | 1999 | 2021 | Known | Very low (from historic samples) |\n| | Cranbourne | Locally extinct | 1976 | Unknown | Known | Unknown |\n| | Dutson Downs | Extant | 2024 | 2025 | Known | Low |\n| | Gippsland Lakes | Extant | 2025 | 2025 | Known | Low |\n| | Hummock Island | Locally extinct | 1977 | Unknown | Known | Unknown |\n| | Langwarrin | Locally extinct | 1984 | Unknown | Known | Unknown |\n| | Mullungdung | Locally extinct | 1976 | 2017 | Known | Unknown |\n| | Providence Ponds | Extant | 2025 | 2025 | Known | Low |\n| | Reeves Beach | Locally extinct | 1978 | 2017 | Known | Unknown |\n| | The Lakes National Park | Extant | 2025 | 2025 | Known | Low |\n| | Tyabb | Locally extinct | 1975 | Unknown | Known | Low (from historic samples) |\n| | Wilsons Promontory | Extant | 2025 | 2025 | Known | Very low |\n\n> Note: a \\- Heterozygosity measures from Burns et al. (2023b).\n\n> Note: Source: Lazenby et al. (2019), Burns (2021a), DCCEEW (2024), NSW BioNet and unpublished data from members of the recovery team and Recovery Network.\n\n## 2.5 Habitat\n\nPookila habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heathy understorey, and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022) (see Figure 3). Historically, the species has been found across a broader array of habitat types (DCCEEW 2024b), and it does not occupy all plausibly suitable habitat at present (Burns et al. 2020; Smith et al. 2023a). Sandy soils suitable for burrowing are common across most occupied habitats and soft sandy substrates, while deep topsoils are thought to be optimal (Abicair et al. 2020). Within the northern parts of the species’ range, it also occupies sites up to 300 km from the coast and in subcoastal locations at higher elevation, where soil composition differs (Martin 1995; Van Dyck & Lawrie 1997; Vernes et al. 2006; Dixon 2014; Churchill et al. 2023).\n\nThe species is often, although not exclusively, found in areas with high floristic diversity with leguminous perennials (Wilson 1991; Seebeck et al. 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; DEWHA 2010). For example, the heathland element of the species’ habitat is represented by dry heathlands that are floristically diverse and even (a high number of plants that are all relatively abundant compared to each other) that grow on sandy soils with a high percentage of bare ground (Seebeck et al. 1996).\n\nThe species has been identified as preferring early successional habitat, at some sites with peak abundance approximately 3–7 years post-fire and 7–9 years post-sand mine rehabilitation (Braithwaite & Gullan 1978; Fox & Fox 1978; Fox & Fox 1984; Twigg et al. 1989; Wilson 1991; Wilson et al. 2018). However, research in Vic suggests that the relationship between the Pookila and fire is complex, and local densities may be high in areas that have not been affected by fires for over 60 years (Burns and Phillips 2020). Similarly, surveys conducted on the NSW central coast have found high densities in long-unburnt vegetation (F Ford 2023, personal communication 8 August). Season, fire severity and extent, predation pressures, and climatic conditions following fire, all affect habitat suitability more strongly than a linear effect of time-since-fire (Burns & Phillips 2020). Furthermore, it is likely that it shelters in burrows during times of fire. This behaviour may be one of the attributes that allows the Pookila to persist in situ through some fire events (Friend 1993; Burns & Phillips 2020).\n\nThe Pookila has been considered a ‘habitat specialist’ (Braithwaite & Gullan 1978), but fossil record observations, and a strategic translocation and assessment of microhabitat use, suggests that it had broad habitat preferences (Smith et al. 2023c). Remains of the species are common in most bone deposits in south-eastern Australia that has been examined, and these deposits occur across a broad range of historical vegetation types and altitudes (Ford 2003). The species’ perceived habitat specialisation may have been further challenged by the Pookila translocation into a predator exclusion zone in the ACT as part of the Mulligans Flat-Goorooyarroo Woodland Experiment. The box gum and grassy woodlands comprising the predator exclusion area are not the type of coastal habitat often associated with contemporary records of the species (Abicair et al. 2020). However, recent detections of the species in grassy open forests and woodlands in Qld means that these non-coastal habitat types are still represented in the Pookila distribution.\n\nThe currently skewed occurrence (and perhaps specificity) of the Pookila towards coastal areas may in part be due to the interaction among threats, such as habitat clearing, altered fire regimes and predation pressures by introduced species such as cats (Felis catus) and European red foxes (Vulpes vulpes). However, it is worth noting that individuals sourced from the ANU breeding colony in the Mulligans Flat population are significantly underweight compared to their founding populations (13.6 g vs. 17.5 g) (Abicair et al. 2020), and that the population may not be self-sustaining. Together, these attributes may indicate that the habitat there may now be suboptimal for the Pookila. However, it may be important to pursue translocations in a diversity of habitats (including those that may be suboptimal) to build the species' resilience to environmental change and to restore ecological functions (Smith et al. 2023a).\n\nFigure 3: Photographs of contemporary Pookila habitat\n\n![These grouped 6 images demonstrate the diversity in the Pookila's habitat throughout the species' range. Refer to the footnote of this collective image for descriptions on habitat per image in addition to subheading 'Habitat' for more information.](image.006.png)\n\n> Note: Contemporary habitat in which the Pookila has been detected. Depicted are: a) grassy woodlands at Cunninghams Gap in Qld (photo: Ian Gynther); b) woodland with a grass and sedge understorey at Mt Hyland, NSW (photo: Anna Lloyd); c) coastal heathland at Munmorah in NSW (photo: Barbara Wilson); d) woodland with a heath understorey at Anglesea in Vic (photo: Barbara Wilson); e) coastal heathland at Loch Sport in Vic (photo: Phoebe Burns); and f) coastal heathland with a woodland overstorey on Flinders Island in Tas (photo: Rosemary Gales) (Source: DNRE Tas 2022).\n\n### 2.5.1 Habitat critical to the survival of the Pookila\n\nAll potential habitat should be considered important to the Pookila’s persistence and recovery, this includes any:\n\n Areas within the species’ modelled distribution;\n Areas where the species previously occurred but has since become locally extinct;\n Patches of suitable habitat that may not be currently occupied by the species but have the potential to be occupied by natural dispersal; and\n Areas that could be restored for reintroduction.\n\nAll areas that contain habitat that serves one or more of the following ecological functions have the potential to be habitat critical to the survival of the Pookila:\n\n Foraging habitat: vegetation with appropriate density and diversity to be able to support a population, such as native vegetation with seasonal abundance of edible flowering, fruiting, and seeding plants. Vegetation cover is important for the species to be able to forage without increased risk of predation. This habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heath understorey and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022).\n\n Breeding habitat: substrate for the species to be able to create burrows, or alternatively, ‘surface nests’ in low-lying vegetation that is suitable for successful reproduction and rearing of young. Soft sandy substrates, with deep topsoils are thought to be optimal (Abicair et al. 2020) but not exclusive. The Pookila mostly occupies coastal areas that provide this substrate. However, it also occupies sites 100–300 km from the coast and areas with higher elevation where soil composition varies. For example, the presence of sandy soils is highly likely in Goobang National Park (NSW NPWS 2001), whilst in the NSW ranges and Northern Tableland areas, soil may vary with a more dominant loamy or quartz soil base where the Pookila occurs or may be present (NSW OEH 2023. pers comm 22 November).\n\n Habitat for the long-term maintenance of the species: including areas that are not subject to prolonged drought or at risk of extreme droughts in future (such as potential climate refugia), and habitat that is maintained through periodic small-scale burns to provide appropriate foraging habitat and connectivity for the species.\n\n Habitat for the maintenance of genetic diversity: includes all current habitats where the species is extant as they contain populations necessary to achieve maximum genetic diversity and increase resilience of the species against future threats. There are two recognised ESUs: the Southern ESU (Vic populations) and the Northern ESU (NSW and Qld populations). Tas may yet be identified as a third ESU or may cluster within the Southern ESU. Therefore, habitat across the species’ range is necessary to maintain the national genetic diversity of the species.\n\n Habitat for translocation: can be identified using habitat modelling followed by ground-truthing at the local-scale and can include locations that have accounted for future changes due to climate change regimes as well as the exclusion of invasive predators.\n\nRefining what constitutes habitat critical to the survival of the Pookila is a priority action in this recovery plan. Until those key characteristics are further clarified, all areas that contain habitat that serves any of the above five ecological functions have the potential to be habitat critical to the survival of the Pookila. Habitat that supports multiple ecological functions and/or a higher number of the ecological attributes identified within the descriptions provided above, increases the likelihood that the habitat may be critical to the survival of the Pookila.\n\nGiven that the species has declined markedly in parts of its range and populations may rapidly change in size, failure to detect the species during surveys does not indicate the absence of potential habitat critical to its survival.\n\n## 2.6 Ecology\n\nThe Pookila often nests underground in communal burrow systems in well-drained soils. Individuals have also been recorded sheltering under low growing shrubs in surface nests (Kemper 1980, Lazenby et al. 2007). Burrows vary in length, depth, and complexity, ranging from shallow surface nests to intricate tunnel systems (Kemper 1981). Of three burrows that have been excavated, the largest was 5 m in length with a maximum depth of 60 cm (Kemper 1981). The species is nocturnal, and with a significant proportion of its time spent foraging above ground, it is exposed to a high risk of predation by native predators including owls and introduced predators such as the cat and the European red fox.\n\nTen individuals were radio-collared in one area of Tas for approximately 2 weeks, yielding information on nightly movements (Lazenby 1999; Lazenby et al. 2007). Over a 2 ha grid, 9 individuals moved up to 206 m in one night, and up to 37 m in one hour (Lazenby 1999; Lazenby et al. 2007). Pookila home ranges have been known to overlap (Lazenby 1999; Lazenby et al. 2007). The species’ home range is approximately 0.5–1.8 ha, increasing as habitat quality decreases (Lock 2005; Burns et al. 2023a).\n\nThe generation length of the Pookila is between one and two years (Kemper 1980). Females reach sexual maturity at approximately 13 weeks and live 2–3 years (Kemper 1976; Kemper 1980). Males reach sexual maturity at approximately 20 weeks and longevity remains unclear (Kemper 1976; Kemper 1980). During good conditions in NSW, the average litter size for the species has been previously recorded as 4.6, with a range of 1–6 young, and with females usually producing 1–2 litters per year, or 3–4 in exceptional conditions (Kemper 1976; Kemper 1980). The gestation period is around one month (32 days) and young are weaned at 3–4 weeks of age (Kemper 1976; Kemper 1980).\n\nThe breeding season varies depending on location and rainfall, with breeding occurring in NSW from late winter to early summer, while in Vic and Tas the breeding season begins in mid-spring and ceases in late summer (Kemper 1980; Burns et al. 2023b). The impact of rainfall is likely related to its influence on resource availability, which influences variation in birth rates, recruitment, and adult survival (Woinarski et al. 2014). Energy requirements are likely to be high during breeding, particularly for females, and if dietary resources are inadequate during this time, females or their young may perish (Lock & Wilson 2017). Hence, the Pookila demonstrates fluctuating population dynamics where extended breeding seasons follow periods of higher rainfall (Lock & Wilson 2017; Burns 2021). Pookila abundance reduces during periods of drought due to resource shortages and likely failed litters (Lock & Wilson 2017; Burns 2021).\n\n### 2.6.1 Diet and foraging\n\nThe Pookila is an omnivore and feeds on a variety of seeds, leaves, fungi, and invertebrates, exhibiting seasonal and locality-dependent dietary variation (Cockburn 1980; Norton 1987; Wilson and Bradtke 1999; Braithwaite and Gullan 1978). The species plays an important ecological role in dispersing seed and fungi in scats (Wilson & Bradtke 1999; Seebeck et al. 2003; Burns et al. 2023b).\n\n \n\n# 3 Threats\n\n## 3.1 Historical causes of decline\n\nThe distribution of subfossils and the genetic structure of extant populations indicate that the historic distribution of the Pookila extended more continuously throughout parts of coastal eastern Australia, from south-east Qld extending down to the northern parts of Tas and up to 300 km inland on the Australian mainland (Ford 2003; Breed and Ford 2007; DEWHA 2010; Burns et al. 2024; Smith et al. 2023c). The species has since experienced significant declines in distribution and population size and is now distributed in fragmented patches across this area (Figure 2).\n\n## 3.2 Current known and suspected threatening processes\n\nCurrent threats and their prevalence and intensity are described for the national Pookila population below. It is important to note that different populations may be experiencing different combinations of threats in varying intensities. There is significant interaction between threats.\n\n### 3.2.1 Habitat loss, disturbance and modification\n\n#### Habitat clearing\n\nLand clearing for agriculture, infrastructure, road, and housing development continues to place Pookila habitat at risk of further loss and fragmentation. As very little targeted and systematic monitoring of Pookila populations has occurred across the range in NSW (DCCEEW 2024b), the species is at particular risk of losing habitat where presence has not yet been confirmed.\n\nIn 2019, 87% of the Australian population lived within 50 km of the coast (cited in Clark et al. 2021). Coastal development is a continuing, increasing pressure with very high impacts on coastal environments (Clark et al. 2021) that is sprawling beyond established urban areas to regional coastal areas (Infrastructure Australia 2020).\n\nWhen habitat is cleared, disturbed, or modified, the risk and magnitude of impact to adjacent extant habitat patches increases, along with the interactive capability of threats (Fischer & Lindenmayer 2007; DNRE Tas 2022). For example, land clearing for agriculture can increase the carrying capacity of native and introduced herbivores within the landscape, leading to increased grazing and soil impact pressures on adjacent habitat. This is exacerbated when fire results in a further increase in grazing pressure and movement as herbivores actively seek to consume newly emerging vegetative shoots (DAWE 2022). With an increasing trend of habitat loss due to land clearing, the likelihood that fragmentation has occurred and will continue to occur is very high.\n\n#### Habitat modification due to Phytophthora cinnamomi infestation\n\nPhytophthora dieback – caused by the root-rot pathogen Phytophthora cinnamomi – is listed as a Key Threatening Process under the EPBC Act (DOEE 2018) and has been observed across the national range of the Pookila (Liu et al. 2013; Wilson et al. 2015; Burns et al. 2020; DCCEEW 2024b). The pathogen is spread through the movement of soil – including on footwear, tools, vehicles and machinery – and in water (DOEE 2018). While the impacts of P. cinnamomi on Pookila populations and habitat have not been investigated directly, the species is likely to be at significant risk. A range of plant species within Pookila habitat have been identified as susceptible to infection by P. cinnamomi (see McDougall 2005).\n\nPhytophthora dieback causes the loss of plant species richness (including keystone species), reduction of structural complexity (Wilson et al. 1994; Garkaklis et al. 2004; Laidlaw & Wilson 2006; Annett 2008; Armistead 2008; Dundas et al. 2013, 2016; Davis et al. 2014; Wilson et al. 2020), loss of food resources (for example, seeds, leaves, nectar, pollen) and suitable refuges from weather and predation (Laidlaw & Wilson 2006). The impacts of Phytophthora dieback can also extend to invertebrate communities and thus impact insectivorous fauna (Cahill et al. 2008), reducing availability of this food resource.\n\nSignificant negative impacts from Phytophthora dieback on small mammals and their habitat has been previously documented (Newell & Wilson 1993; Lilith 2002; Laidlaw & Wilson 2006; Swinburn et al. 2007; Annett 2008; Armistead 2008; Dundas et al. 2013; 2016), and long-term studies have identified localised extinctions of species and mammal community declines (Zhuang-Griffin 2015; Wilson et al. 2017; 2020).\n\nPriority Protection Areas for Phytophthora dieback management have been identified for the Otway Ranges, which was informed by habitats occupied by EPBC Act listed species, including the Pookila (Walshe et al. 2021; 2022; Wilson & Garkaklis 2023a; 2023b).\n\n#### Habitat modification due to native and introduced weeds\n\nChanging grazing and fire regimes have led to habitat loss by encroachment of coast tea tree (Leptospermum laevigatum) in some Pookila habitat (Burns 2019a; Seebeck et al. 2003). This is particularly evident at Wilsons Promontory National Park and in other coastal Vic sites along Ninety Mile Beach where dune succession has given rise to dense thickets of coast tea tree, such as at Reeves Beech. Weed species are of growing concern in other Pookila habitat, for example, in Lakes National Park. Invasive weeds include Asparagus asparagoides (bridal creeper) and Lycium ferocissimum (boxthorn).\n\n### 3.2.2 Fire regimes that cause declines in biodiversity\n\n‘Fire regimes that cause declines in biodiversity’ is listed as a Key Threatening Process under the EPBC Act (DAWE 2022). Altered fire regimes, such as those resulting from changes to seasonality, frequency and severity of fire weather conditions, are causing declines and loss of biodiversity and ecosystem function (DAWE 2022). The Pookila and the habitat the species relies on have a complicated relationship with fire.\n\nUnder certain circumstances, the species can peak in abundance in the years following a fire as resources become more readily available due to regeneration of vegetation, with such areas described as ‘early successional’ (Fox & McKay 1981). Furthermore, the Pookila has been reported to have relatively high survival during and immediately following bushfires, which is also presumably attributed to its ability to shelter in burrows (Friend 1993; Burns & Phillips 2020). However, this response appears to be more common following small-scale, patchy, low-severity burns and is likely reliant on above-average rainfall in the months and years following the fire (Burns & Phillips 2020).\n\nLarge-scale, high-severity fires have the capacity to extirpate entire populations, as occurred in 1983 on Ash Wednesday at Anglesea (Wilson et al. 2018). Fires being too frequent, too large, too severe, or poorly timed may lead to direct loss of individuals and/or sudden or gradual loss of habitat through reduction of habitat patches of suitable successional age, size, distribution, and plant diversity that the Pookila requires for availability of food and shelter from predators.\n\nSimilarly, prolonged exclusion or poorly timed application of fire may encourage coast tea tree encroachment and loss of favourable food and habitat species. Also, introduced predators may target recently burnt areas, compounding the increased risk of predation from loss of vegetative cover (Hradsky et al. 2017). Planned burning may also threaten the species if conducted during or immediately prior to drought, at too high an intensity, or across a widespread area (Wilson et al. 2018).\n\nThe increasing scale, severity, and frequency of fires under a changing climate pose a risk to the Pookila and its habitat. In 2019 and 2020, catastrophic bushfires impacted a large area of eastern Australia with varying severity (Legge et al. 2021). It was estimated that over 45% of the Pookila distribution was burnt, with a third of this area burning with high severity (Legge et al. 2021). The overall population of Pookila was predicted to have declined by 24% by 2021 (one year after fire), with the species suffering an estimated 75% decrease in abundance in severely burnt areas in that same period (Legge et al. 2021).\n\n### 3.2.3 Climate change: Increasing number of hot days, prolonged drought and changes to precipitation patterns\n\nIncreasing frequency, intensity, and duration of drought across much of the species’ range (CSIRO & BoM 2020) may lead to further catastrophic declines and local extinctions. Extreme heat events may exacerbate the effects of drought and could lead to mass mortality events as individuals dehydrate and fail to thermoregulate but this would require further research to confirm whether behaviours would mitigate this threat. The projections for the NSW climate by 2050 include an increase in minimum and maximum temperatures, an increase (double) in the number of hot days per year (days greater than 35°C), and longer fire seasons, with around a 40% increase in ‘very high’ fire danger days (CSIRO & BoM 2021).\n\nExtended periods of low rainfall reduce food availability for Pookila. Recent periods of drought have caused demonstrable crashes in Pookila populations. The local extinction of the Anglesea population is in part attributed to the Millennium Drought (Lock & Wilson 2017), and the species is yet to recover from the 2019 Gippsland drought (Burns 2021).\n\nFire–drought interactions may have a significant impact on the Pookila with potential compounding effects. This is evident from the 2019 and 2020 bushfires, which followed a period of significant drought in 2017 to 2019 (BoM 2020; DPIE 2020).\n\n### 3.2.4 Invasive fauna species\n\n#### Predation by cats and European red foxes\n\nFeral predators can have rapid and catastrophic impacts on populations of threatened rodents (Tuft et al. 2021). While there is limited direct evidence of the consumption of Pookila by feral predators (records in dog scats; Hoskins 1997; Davis et al. 2015; Lock & Wilson 2017; MidCoast Council NSW 2024. pers comm 13 February), it is recognised in the Threat Abatement Plan for Predation by Feral Cats 2024 that the Pookila is likely highly susceptible to predation by cats (and/or in some cases by foxes) (DCCEEW 2024a) and that cats and foxes can significantly impact population numbers (Seebeck et al. 2003; Burns 2019a).\n\nLike many other Pseudomys species, Pookila are docile and display poor awareness of exotic predators (P. Burns, pers. obs.). In such species, selective predation can occur (as reported, for example, on the broad-toothed rat, Mastacomys fuscus; Green 2002), leading to a disproportionate impact of predation on the vulnerable species even when co-occurring species (for example, the house mouse) are at greater abundance. Furthermore, communal burrowing and small local population densities mean that predation events at burrow entrances can cause rapid local declines (Burns 2019a).\n\nThe relative impact of predation by feral cats and foxes on the Pookila is not yet understood. Based on factors such as small body size, lower breeding output, and occupation of open habitat, some researchers have identified the Pookila as having a high threat rating from feral cat predation (Dickman 1996; Stobo-Wilson et al. 2021). Radford et al. (2018) reported the cat and fox threat to Pookila to be approximately the same based on expert opinion, although with weak evidence. While foxes are currently absent from the Tas mainland and Flinders Island (DNRE Tas 2022), and fox control is in place in other parts of the species’ range, effective cat control in open, non-insular systems beyond the local scale is not yet possible and potentially counterproductive (Lazenby et al. 2015; Lazenby et al. 2021; see Predator management).\n\nThe density and predatory impact of cats and foxes can be altered by competitive interactions with other predators. For example, reduced numbers of foxes via fox control can increase cat density, alter cat behaviour, and ultimately increase the impact of predation by feral cats (Risbey et al. 2000; Rees et al. 2023). Tasmanian devils have been shown to alter the behaviour and abundance of feral cats (Hollings et al. 2014; Lazenby & Dickman 2013), and cats have a much higher predatory impact on small mammals compared to devils (Hollings et al. 2014). The decline of devils caused by Devil Facial Tumour Disease (DFTD) is likely to have increased the predatory impacts of feral cats in Pookila habitat on the Tas mainland.\n\n#### Impact of hooved feral animals including deer, goats, horses, and pigs\n\nThe negative impacts of hooved feral animals on vegetation health and soil erosion are well documented (Keith & Pellow 2005; Bengsen et al. 2017; Wills et al. 2023; Burns et al. 2021). This is also recognised within the Draft threat abatement plan for competition and land degradation by unmanaged goats (DCCEEW 2024c), the Threat abatement plan for predation, habitat degradation, competition and disease transmission by feral pigs (Sus scrofa) (DOEE 2017), and the National feral deer action plan (Government of South Australia 2023). Where these species co-occur with the Pookila, there are risks of habitat degradation through trampling, wallowing, and overgrazing (Burns 2021). Overgrazing pressures may increase in post-fire environments, particularly in heathy vegetation types preferred by the Pookila (Heaton et al. 2022), compounding post-fire food shortages and predation risk, while impeding regrowth of suitable habitat and food species (DAWE 2022). The effects of hooved feral animal control programs on Pookila occurrence and abundance have not been directly studied, but based on vegetation response studies, control programs are expected to benefit the Pookila and its habitat, particularly around fire events.\n\n#### Competition/predation from introduced rodents\n\nWhile experimental trials have shown the Pookila to be competitively superior to the invasive house mouse (Fox & Twigg 1991; Haering & Fox 1997), there is growing evidence that house mouse plagues may supress Pookila recovery following periods of drought where resources are subsequently scarce (Burns 2021). Rattus rattus and R. norvegicus (black and brown rats, respectively) may also out-compete the Pookila for resources and prey on adults and juveniles, although this requires further research.\n\n#### Baiting of rabbits and invasive rodents\n\nThe susceptibility of the Pookila to 1080, Pindone, and rodenticide has not been tested. However, such poisons are lethal at low doses to some other Pseudomys species and are expected to be lethal to the Pookila (McIlroy 1982; Eisler 1995; Fisher 1999). Running a lethal dose trial on the Pookila is not appropriate given the species’ threatened status and ethical considerations, so it is safest to apply the Precautionary Principle and assume the species is likely susceptible to 1080, Pindone and rodenticide. Rabbit poison baiting is usually conducted with a carrot or oat bait, both foods readily consumed by the Pookila (Prosser et al. 2007), and it is possible that rabbit-baiting programs have contributed to the decline of the species in some areas (Fisher 1999; Burns 2019a). When finding the balance between protecting Pookila habitat by controlling rabbits and potentially poisoning individual Pookila, it is recommended to entirely avoid poison baiting in areas where Pookila presence is known. This is particularly important for at-risk populations; however, it may also apply to locations where Pookila presence is uncertain but these areas are adjacent to suitable habitat and/or a known population (within a 5 km radius). Rabbit and invasive rodent population control measures should use alternative methods. Control measures should be carefully considered depending on the area of control and the potential side effects for native species. Currently, this threat is considered to be high in Wilsons Promontory in Vic, not applicable in Qld or Tas, and unknown in NSW (DNRE Tas 2023).\n\n### 3.2.5 Small and fragmented populations: Low genetic diversity and inbreeding\n\nHistorical and contemporary habitat degradation and destruction, and the species’ very low dispersal ability have left the Pookila highly fragmented across its distribution, with some populations genetically isolated. This severe fragmentation leaves populations at higher risk of extinction, as a single catastrophic event can more easily impact the total area occupied by a population (Garner et al. 2020).\n\nFurthermore, genetically isolated populations are at higher risk of inbreeding depression. Inbreeding reduces genetic diversity, which reduces adaptability to changing environments and anthropogenic impacts. Inbreeding depression also increases species’ vulnerability to disease while also allowing potentially detrimental traits to surface in the population (Keller 2002). Furthermore, inbreeding depression can cause reduced fecundity and further genetic health issues (for example, chromosomal abnormalities) (Keller 2002).\n\nVic populations (Southern ESU) of the Pookila have relatively low heterozygosity (genetic diversity) and are at risk of inbreeding, with no natural gene flow possible between localities given lack of dispersal (Burns et al. 2023b). Therefore, the Southern ESU is considered an ‘at-risk ESU’. Populations across NSW retain greater heterozygosity and are genetically distinct from those in Vic, although animals in Broadwater National Park appear to have undergone a genetic bottleneck (Burns et al. 2023b). At the time of publishing this recovery plan, two historic mainland Tas samples and a small number of contemporary samples from Flinders Island have been sequenced and are being analysed. Additional Tas samples are being sought through survey work. Genetic mixing within Vic populations has been recommended and is being pursued, whereas no immediate action is required in the Northern ESU (NSW and Qld) (Burns et al. 2023b; Burns et al. 2024) since it is more genetically healthy than the Southern ESU. The recovery team has developed a National Pookila genetic management strategy (Zoos Victoria) and will continue to update it when new information is forthcoming.\n\n#### Disease\n\nThere is limited understanding of the potential prevalence or impact of disease on the Pookila. In general, screening of native rodents for pathogens has been very limited and little information is available on symptomology (Thompson et al. 2010). However, feral cats and invasive rodents carry diseases that could impact the Pookila. In particular, Toxoplasma gondii, a virulent parasite spread by cats, is likely present through much of the species’ former range (Dickson 2018). Previous experimental trials in other Pseudomys species demonstrated rapid, high mortality rates after infection with T. gondii, hence this parasite is likely to also increase the mortality of the Pookila (Burns 2019a).\n\n## 3.3 Threat prioritisation\n\nEach of the threats outlined in this recovery plan has been assessed to determine the risk posed across the national distribution of the Pookila using a risk matrix (see Table 6). The risk matrix considers the likelihood of an incident occurring and the consequences of that incident and uses a qualitative assessment drawing on both peer-reviewed literature and expert opinion.\n\nTable 6: Risk matrix\n\n| Consequence (x) / Likelihood (y) | Not significant | Minor | Moderate | Major | Catastrophic |\n| -------------------------------- | --------------- | ------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------- |\n| Almost certain | Low risk | Moderate risk | Very high riskPredation by European red foxes | Very high riskHabitat clearing (past/present)Fire regimes that cause declines in biodiversityIncreasing number of hot days, prolonged drought and changes to precipitation patternsPredation by cats | Very high risk |\n| Likely | Low risk | Moderate riskImpact of hooved feral animals including deer, goats, horses, and pigs | High riskHabitat modification due to Phytophthora cinnamomi infestationHabitat modification due to native and introduced weeds | Very high riskHabitat clearing (future) | Very high risk |\n| Possible | Low risk | Moderate riskCompetition/predation from introduced rodentsBaiting of rabbits and invasive rodents | High riskLow genetic diversity and inbreedingDisease | Very high risk | Very high risk |\n| Unlikely | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n| Unknown | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n\nRisk matrix key\n\n| Low risk | Moderate risk | High risk | Very-high risk |\n| -------- | ------------- | --------- | -------------- |\n\nDefinition of likelihood and consequence categories\n\nConsequence is a measure of the severity of the impact of the threat on the survival of the species.\n\n- Catastrophic – affecting survival, reproduction or essential movement of individuals in >80% of the population or across >80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Major – affecting survival, reproduction or essential movement of individuals in 50-80 % of the population or across 50-80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Moderate – affecting survival, reproduction or essential movement of individuals in 20-50% of the population or across 20-50% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Minor – affecting survival, reproduction or essential movement of individuals in 5-30% of the population or across 5-30% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Not significant – affecting survival, reproduction or essential movement of individuals in <5% of the population or across <5% of the distribution within the next 3 generations or 10 years, whichever is longer.\n\nLikelihood is a measure of how regularly the threat impacts the species.\n\n- Almost certain – >90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Likely – 66-90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Possible – 33-66% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unlikely – <33% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unknown – probability that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer, could be 0-100%.","sortOrder":5},{"sectionNumber":"1.1","sectionType":"part","heading":"Introduction","content":"## 1.1 Introduction\n\nThe Pookila (New Holland mouse; Pseudomys novaehollandiae) is a small, burrowing, rodent native to south eastern Australia, that is patchily distributed from Tasmania to south-east Queensland. The species has had catastrophic declines that were recorded since the discovery of new populations in the 1970s. Isolated remnant populations are at high risk of extirpation and genetic inbreeding.\n\nDistribution is fragmented across southern Queensland (Qld), the northern and central coastal areas of New South Wales (NSW), central southern coastal Victoria (Vic), and north-eastern Tasmania (Tas) including Flinders Island (Burns et al. 2023a). While the species is typically restricted to coastal areas, it occupies areas up to 300 km inland in Qld and NSW.\n\nSince the species’ description in 1843, more than a dozen known populations have been lost and many more have declined. It has not been live trapped on mainland Tas since 2010 and in Vic, seven of twelve populations have become extinct since the 1970s (Burns et al. 2023b). Encouragingly, the Pookila was rediscovered on Flinders Island, Tasmania in 2021 and was last recorded in southern Qld in 2022. It has now been translocated into Mulligans Flat Woodland Sanctuary in the Australian Capital Territory (ACT), Yiraaldiya National Park feral predator-free area (NSW), reintroduced into the Cranbourne Royal Botanic Gardens (Vic) in 2024, and has numerous ongoing and successful captive-breeding programs.\n\nEuropean colonisation in Australia exposed the Pookila to compounding threats that caused population declines across its range. These threats include the direct destruction and fragmentation of habitat due to land clearing for agriculture, urbanisation, road construction, and forest plantations, as well as predation by introduced cats and European red foxes (DEWHA 2010; Burns 2019a). Introduced rodents may outcompete the Pookila under some conditions, and both introduced rodents and cats are vectors of novel diseases. Infection may also cause mortality and/or increase the Pookila’s susceptibility to predation (for example, toxoplasmosis spread by feral cats) (Lazenby et al. 2019; Burns 2019a).","sortOrder":6},{"sectionNumber":"1.2","sectionType":"part","heading":"Conservation status","content":"## 1.2 Conservation status\n\nThe Pookila is listed as Vulnerable under the Environment Protection, Biodiversity and Conservation Act 1999 (EPBC Act), Vulnerable (ACT, Qld, NSW, VIC) and Endangered (Tas) (Table 3).\n\nThe Pookila was first listed under the EPBC Act in the Vulnerable category on 11 August 2010. It was reassessed by the Threatened Species Scientific Committee in July 2024 retaining its Vulnerable category status under Criteria 1 and 2 (DCCEEW 2024b).\n\nCriterion 1 (A4b) was met due to the severe declines in population size estimated to have occurred in Tas and Vic and the significant declines suspected to have resulted from the impacts of the 2019 and 2020 bushfires in NSW, with the Committee concluding that the Pookila was subject to a population size reduction of at least 30% (DCCEEW 2024b).\n\nCriterion 2 B2ab(i,ii,iii,iv,v) was met due to the limited Area of Occupancy (AOO), low/very low number of locations (six), and estimated continuing decline in EOO, AOO, area of habitat, number of locations and populations, as well as the number of mature individuals (DCCEEW 2024b).\n\nTable 3: Conservation status of the Pookila\n\n| Listing | Conservation Status |\n| ---------------------------------------------------------------------------- | ------------------- |\n| Environment Protection and Biodiversity Conservation Act 1999 (Commonwealth) | Vulnerable |\n| Nature Conservation Act 2014 (Australian Capital Territory) | Vulnerable |\n| Nature Conservation Act 1992 (Queensland) | Vulnerable |\n| Biodiversity Conservation Act 2016 (New South Wales) | Vulnerable |\n| Flora and Fauna Guarantee Act 1988 (Victoria) | Vulnerable |\n| Threatened Species Protection Act 1995 (Tasmania) | Endangered |\n| IUCN Red List of Threatened Species | Vulnerable |","sortOrder":7},{"sectionNumber":"1.3","sectionType":"part","heading":"Cultural and community significance","content":"## 1.3 Cultural and community significance\n\nThe Pookila currently occurs on the Country of the Awabakal, Banbai, Barunggam, Biripi, Bunurong, Boonwurrung, Bundjalung, Burrgati, Daingatti, Darkinung, Dharawal, Dharug, Eora, Geawegal, Gumbainggir, Gunaikurnai, Kuringgai, Gunaikurnai, Ngunnawal, Palawa, Wiradjuri, Wonnarua, Worimi and Yuggera Peoples.\n\nThe species has likely been known by numerous names in different Aboriginal languages across its geographic range. The name Pookila is derived from the Ngarigo word for ‘mouse’ – bugila (Braithwaite et al. 1995; Hercus n.d.), deviating somewhat from traditional Ngarigo pronunciation (Braithwaite et al. 1995). The assignation of that name was part of an initiative by the CSIRO to provide Indigenous Australian names for Australian rodents (Braithwaite et al. 1995). This was considered a way to avoid the undesirable associations with introduced vermin often attached to the words ‘rat’ and ‘mouse’, impeding conservation of native rodents due to negative public perception (Braithwaite et al. 1995). An alternative name could have been ‘tuwaranga’ which means ‘mouse’ in palawa kani, the language of the Tasmanian Aboriginal people (Tasmanian Aboriginal Centre, 2023), however, the name Pookila has been associated with this species since 1995 and is included on the Australian Government’s SPRAT database.","sortOrder":8},{"sectionNumber":"1.4","sectionType":"part","heading":"National Pookila Recovery Team","content":"## 1.4 National Pookila Recovery Team\n\nA National Pookila Recovery Team (the recovery team or NPRT) was established in 2021 with the aim of improving recovery efforts for the species at a national scale. A previous Pookila Recovery Team (1995 to 2002) was largely based in Vic. The current recovery team has representation across the species’ entire distribution. Beyond the core recovery team there are three sub-groups: captive breeding, habitat and fire management, and Vic management.\n\nThe role of the recovery team is to facilitate and encourage collaborations across jurisdictions, provide advice to partners and land managers, increase engagement, improve the public perception of the Pookila as well as the responsibility of advising and coordinating the implementation of recovery actions. The recovery team also seeks to facilitate opportunities for mutual learning and skill sharing with First Nations people. The membership of the recovery team includes species experts and land managers from relevant government agencies and non-government organisations and draws from a broader network of expertise from First Nations people, researchers, and community groups.","sortOrder":9},{"sectionNumber":"1.5","sectionType":"part","heading":"Key partners","content":"## 1.5 Key partners\n\nImplementation of recovery planning is a collaborative effort involving a range of key partners and affected interests. The groups likely to have primary interest in recovery planning for the Pookila are listed in Table 4.\n\nTable 4: Groups with primary interest in Pookila recovery\n\n| Organisations | Involvement/ interest |\n| -------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| ACT Government | Management of the land of the first Pookila translocation site. Ongoing monitoring at the site and potential future releases in the area. |\n| Australian National University (ANU) | Captive-breeding program for NSW individuals. Ongoing monitoring and management in Mulligans Flat Woodland Sanctuary and potential future reintroductions in the ACT and other jurisdictions. |\n| Aussie Ark (AA) | Maintains an insurance population. Potential development of a captive-breeding program for the Pookila and future reintroductions into cat and fox free sites in NSW. |\n| Corangamite Catchment Management Authority (CCMA) | Managers of habitat under threat from P. cinnamomi. |\n| Department of Climate Change, Energy, the Environment and Water (DCCEEW, Cwlth) | The Pookila is a priority species in the Threatened Species Action Plan 2022-2032, which provides guidance and support in the design and delivery of projects for priority species. Recovery plans inform decision making under the EPBC Act referral and assessment process. |\n| Department of Energy, Environment and Climate Action (DEECA) | Primary agency involved in threatened species management on public and private land in Vic. |\n| Department of Natural Resources and Environment (DNRE) | Primary agency involved in threatened species management on public and private land in Tas. |\n| Department of the Environment, Tourism, Science and Innovation (DETSI) | Primary agency involved in threatened species management on public and private land in Qld. |\n| Flinders Island Aboriginal Association Incorporated (FIAAI) | Traditional Owners of Flinders Island population Tas. |\n| Friends of the Prom (FOTP) | Group established to support the management of Wilsons Promontory National Park. |\n| Gippsland Water (GW) | Manages a significant Vic population. |\n| Gunaikurnai Land and Waters Aboriginal Corporation (GLaWAC) | Joint managers of Lakes National Park, and Gippsland Lakes Coastal Park, and have Registered Aboriginal Party status of Providence Ponds Flora and Fauna Reserve in Vic. |\n| Land holders/Landowners | Consultation and engagement with all types of landowners will play a role in the success in meeting the recovery plan’s actions and objectives. |\n| Local councils | Community involvement projects and cooperation with management actions on a local level. |\n| Monash University (MU) | Storage and research of Pookila cell lines and genetic tissue. Supporting research of the genetic sequencing of the Pookila and subsequent projects surrounding genetic information. |\n| Moonlit Sanctuary Wildlife Conservation Park (MSWCP) | Managing a captive population of Pookila as well as community conservation and engagement activities. |\n| Forestry agencies | Actions and management regarding forestry operations and Regional Forest Agreements. |\n| NSW Department of Climate Change, Energy, the Environment and Water (DCCEEW NSW) | Primary agency involved in threatened species management on public and private land in NSW. |\n| NSW National Parks and Wildlife Service (NPWS NSW) | Implementation of appropriate management actions within parks and reserves in NSW, including reintroduction of the Pookila to establish new populations in NPWS-managed feral predator-free areas. |\n| Parks Victoria (PV) | Implementation of appropriate management actions concerning habitat within Vic parks and reserves. |\n| National Threatened Species Institute (NTSI) Priam Psittaculture Centre (PPC) | Host of a captive breeding program founded from northern ESU individuals. |\n| Royal Botanical Gardens Victoria Cranbourne (RBGVC) | Location for reintroduction to establish a new population. |\n| Tasmanian Aboriginal Land and Sea Council Aboriginal Corporation (TALSCAC) | Traditional Owners of Northeast population Tas. |\n| Trust for Nature (TFN) | Managing properties adjacent to occupied habitat. |\n| Turner Family Foundation (TFF) | Manager of Spicers Peak Nature Refuge south-east Qld habitat. |\n| University of Canberra (UC) | Involved in captive husbandry and dietary analyses. |\n| University of Southern Queensland (USQ) | Collaborative research project with the DETSI and the Turner Family Foundation to refine knowledge of distribution and ecological requirements in south-east Queensland. |\n| University of Sydney (USyd) | Analysis and advice on captive-breeding genetics. |\n| Wadawurrung Traditional Owners Aboriginal Corporation (WTOAC) | Traditional Owners of Anglesea population Vic. |\n| Woodlands and Wetlands Trust (WWT) | Community engagement, education and volunteer programs at Mulligans Flat Woodland Sanctuary. |\n| World Wildlife Fund Australia (WWF) | Potential partner for island translocation. |\n| Zoos Victoria (ZV) | Programs underway for monitoring, management, captive breeding, genetic rescue, reintroductions, research, studbook management and species coordination, community conservation and engagement. |\n| Zoo and Aquarium Association Australasia (ZAA) | Coordination of the NPRT, initial engagement/consultation with First Nations people, and assistance with studbook management and coordination. |","sortOrder":10},{"sectionNumber":"2","sectionType":"part","heading":"Species information","content":"# 1 Background\n\n## 1.1 Introduction\n\nThe Pookila (New Holland mouse; Pseudomys novaehollandiae) is a small, burrowing, rodent native to south eastern Australia, that is patchily distributed from Tasmania to south-east Queensland. The species has had catastrophic declines that were recorded since the discovery of new populations in the 1970s. Isolated remnant populations are at high risk of extirpation and genetic inbreeding.\n\nDistribution is fragmented across southern Queensland (Qld), the northern and central coastal areas of New South Wales (NSW), central southern coastal Victoria (Vic), and north-eastern Tasmania (Tas) including Flinders Island (Burns et al. 2023a). While the species is typically restricted to coastal areas, it occupies areas up to 300 km inland in Qld and NSW.\n\nSince the species’ description in 1843, more than a dozen known populations have been lost and many more have declined. It has not been live trapped on mainland Tas since 2010 and in Vic, seven of twelve populations have become extinct since the 1970s (Burns et al. 2023b). Encouragingly, the Pookila was rediscovered on Flinders Island, Tasmania in 2021 and was last recorded in southern Qld in 2022. It has now been translocated into Mulligans Flat Woodland Sanctuary in the Australian Capital Territory (ACT), Yiraaldiya National Park feral predator-free area (NSW), reintroduced into the Cranbourne Royal Botanic Gardens (Vic) in 2024, and has numerous ongoing and successful captive-breeding programs.\n\nEuropean colonisation in Australia exposed the Pookila to compounding threats that caused population declines across its range. These threats include the direct destruction and fragmentation of habitat due to land clearing for agriculture, urbanisation, road construction, and forest plantations, as well as predation by introduced cats and European red foxes (DEWHA 2010; Burns 2019a). Introduced rodents may outcompete the Pookila under some conditions, and both introduced rodents and cats are vectors of novel diseases. Infection may also cause mortality and/or increase the Pookila’s susceptibility to predation (for example, toxoplasmosis spread by feral cats) (Lazenby et al. 2019; Burns 2019a).\n\n## 1.2 Conservation status\n\nThe Pookila is listed as Vulnerable under the Environment Protection, Biodiversity and Conservation Act 1999 (EPBC Act), Vulnerable (ACT, Qld, NSW, VIC) and Endangered (Tas) (Table 3).\n\nThe Pookila was first listed under the EPBC Act in the Vulnerable category on 11 August 2010. It was reassessed by the Threatened Species Scientific Committee in July 2024 retaining its Vulnerable category status under Criteria 1 and 2 (DCCEEW 2024b).\n\nCriterion 1 (A4b) was met due to the severe declines in population size estimated to have occurred in Tas and Vic and the significant declines suspected to have resulted from the impacts of the 2019 and 2020 bushfires in NSW, with the Committee concluding that the Pookila was subject to a population size reduction of at least 30% (DCCEEW 2024b).\n\nCriterion 2 B2ab(i,ii,iii,iv,v) was met due to the limited Area of Occupancy (AOO), low/very low number of locations (six), and estimated continuing decline in EOO, AOO, area of habitat, number of locations and populations, as well as the number of mature individuals (DCCEEW 2024b).\n\nTable 3: Conservation status of the Pookila\n\n| Listing | Conservation Status |\n| ---------------------------------------------------------------------------- | ------------------- |\n| Environment Protection and Biodiversity Conservation Act 1999 (Commonwealth) | Vulnerable |\n| Nature Conservation Act 2014 (Australian Capital Territory) | Vulnerable |\n| Nature Conservation Act 1992 (Queensland) | Vulnerable |\n| Biodiversity Conservation Act 2016 (New South Wales) | Vulnerable |\n| Flora and Fauna Guarantee Act 1988 (Victoria) | Vulnerable |\n| Threatened Species Protection Act 1995 (Tasmania) | Endangered |\n| IUCN Red List of Threatened Species | Vulnerable |\n\n## 1.3 Cultural and community significance\n\nThe Pookila currently occurs on the Country of the Awabakal, Banbai, Barunggam, Biripi, Bunurong, Boonwurrung, Bundjalung, Burrgati, Daingatti, Darkinung, Dharawal, Dharug, Eora, Geawegal, Gumbainggir, Gunaikurnai, Kuringgai, Gunaikurnai, Ngunnawal, Palawa, Wiradjuri, Wonnarua, Worimi and Yuggera Peoples.\n\nThe species has likely been known by numerous names in different Aboriginal languages across its geographic range. The name Pookila is derived from the Ngarigo word for ‘mouse’ – bugila (Braithwaite et al. 1995; Hercus n.d.), deviating somewhat from traditional Ngarigo pronunciation (Braithwaite et al. 1995). The assignation of that name was part of an initiative by the CSIRO to provide Indigenous Australian names for Australian rodents (Braithwaite et al. 1995). This was considered a way to avoid the undesirable associations with introduced vermin often attached to the words ‘rat’ and ‘mouse’, impeding conservation of native rodents due to negative public perception (Braithwaite et al. 1995). An alternative name could have been ‘tuwaranga’ which means ‘mouse’ in palawa kani, the language of the Tasmanian Aboriginal people (Tasmanian Aboriginal Centre, 2023), however, the name Pookila has been associated with this species since 1995 and is included on the Australian Government’s SPRAT database.\n\n## 1.4 National Pookila Recovery Team\n\nA National Pookila Recovery Team (the recovery team or NPRT) was established in 2021 with the aim of improving recovery efforts for the species at a national scale. A previous Pookila Recovery Team (1995 to 2002) was largely based in Vic. The current recovery team has representation across the species’ entire distribution. Beyond the core recovery team there are three sub-groups: captive breeding, habitat and fire management, and Vic management.\n\nThe role of the recovery team is to facilitate and encourage collaborations across jurisdictions, provide advice to partners and land managers, increase engagement, improve the public perception of the Pookila as well as the responsibility of advising and coordinating the implementation of recovery actions. The recovery team also seeks to facilitate opportunities for mutual learning and skill sharing with First Nations people. The membership of the recovery team includes species experts and land managers from relevant government agencies and non-government organisations and draws from a broader network of expertise from First Nations people, researchers, and community groups.\n\n## 1.5 Key partners\n\nImplementation of recovery planning is a collaborative effort involving a range of key partners and affected interests. The groups likely to have primary interest in recovery planning for the Pookila are listed in Table 4.\n\nTable 4: Groups with primary interest in Pookila recovery\n\n| Organisations | Involvement/ interest |\n| -------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| ACT Government | Management of the land of the first Pookila translocation site. Ongoing monitoring at the site and potential future releases in the area. |\n| Australian National University (ANU) | Captive-breeding program for NSW individuals. Ongoing monitoring and management in Mulligans Flat Woodland Sanctuary and potential future reintroductions in the ACT and other jurisdictions. |\n| Aussie Ark (AA) | Maintains an insurance population. Potential development of a captive-breeding program for the Pookila and future reintroductions into cat and fox free sites in NSW. |\n| Corangamite Catchment Management Authority (CCMA) | Managers of habitat under threat from P. cinnamomi. |\n| Department of Climate Change, Energy, the Environment and Water (DCCEEW, Cwlth) | The Pookila is a priority species in the Threatened Species Action Plan 2022-2032, which provides guidance and support in the design and delivery of projects for priority species. Recovery plans inform decision making under the EPBC Act referral and assessment process. |\n| Department of Energy, Environment and Climate Action (DEECA) | Primary agency involved in threatened species management on public and private land in Vic. |\n| Department of Natural Resources and Environment (DNRE) | Primary agency involved in threatened species management on public and private land in Tas. |\n| Department of the Environment, Tourism, Science and Innovation (DETSI) | Primary agency involved in threatened species management on public and private land in Qld. |\n| Flinders Island Aboriginal Association Incorporated (FIAAI) | Traditional Owners of Flinders Island population Tas. |\n| Friends of the Prom (FOTP) | Group established to support the management of Wilsons Promontory National Park. |\n| Gippsland Water (GW) | Manages a significant Vic population. |\n| Gunaikurnai Land and Waters Aboriginal Corporation (GLaWAC) | Joint managers of Lakes National Park, and Gippsland Lakes Coastal Park, and have Registered Aboriginal Party status of Providence Ponds Flora and Fauna Reserve in Vic. |\n| Land holders/Landowners | Consultation and engagement with all types of landowners will play a role in the success in meeting the recovery plan’s actions and objectives. |\n| Local councils | Community involvement projects and cooperation with management actions on a local level. |\n| Monash University (MU) | Storage and research of Pookila cell lines and genetic tissue. Supporting research of the genetic sequencing of the Pookila and subsequent projects surrounding genetic information. |\n| Moonlit Sanctuary Wildlife Conservation Park (MSWCP) | Managing a captive population of Pookila as well as community conservation and engagement activities. |\n| Forestry agencies | Actions and management regarding forestry operations and Regional Forest Agreements. |\n| NSW Department of Climate Change, Energy, the Environment and Water (DCCEEW NSW) | Primary agency involved in threatened species management on public and private land in NSW. |\n| NSW National Parks and Wildlife Service (NPWS NSW) | Implementation of appropriate management actions within parks and reserves in NSW, including reintroduction of the Pookila to establish new populations in NPWS-managed feral predator-free areas. |\n| Parks Victoria (PV) | Implementation of appropriate management actions concerning habitat within Vic parks and reserves. |\n| National Threatened Species Institute (NTSI) Priam Psittaculture Centre (PPC) | Host of a captive breeding program founded from northern ESU individuals. |\n| Royal Botanical Gardens Victoria Cranbourne (RBGVC) | Location for reintroduction to establish a new population. |\n| Tasmanian Aboriginal Land and Sea Council Aboriginal Corporation (TALSCAC) | Traditional Owners of Northeast population Tas. |\n| Trust for Nature (TFN) | Managing properties adjacent to occupied habitat. |\n| Turner Family Foundation (TFF) | Manager of Spicers Peak Nature Refuge south-east Qld habitat. |\n| University of Canberra (UC) | Involved in captive husbandry and dietary analyses. |\n| University of Southern Queensland (USQ) | Collaborative research project with the DETSI and the Turner Family Foundation to refine knowledge of distribution and ecological requirements in south-east Queensland. |\n| University of Sydney (USyd) | Analysis and advice on captive-breeding genetics. |\n| Wadawurrung Traditional Owners Aboriginal Corporation (WTOAC) | Traditional Owners of Anglesea population Vic. |\n| Woodlands and Wetlands Trust (WWT) | Community engagement, education and volunteer programs at Mulligans Flat Woodland Sanctuary. |\n| World Wildlife Fund Australia (WWF) | Potential partner for island translocation. |\n| Zoos Victoria (ZV) | Programs underway for monitoring, management, captive breeding, genetic rescue, reintroductions, research, studbook management and species coordination, community conservation and engagement. |\n| Zoo and Aquarium Association Australasia (ZAA) | Coordination of the NPRT, initial engagement/consultation with First Nations people, and assistance with studbook management and coordination. |\n\n# 2 Species information\n\n## 2.1 Taxonomy and description\n\nPseudomys novaehollandiae, family Muridae, is endemic to Australia and was first described in 1843 as Mus novaehollandiae (Waterhouse 1843). It was later transferred to the genus Pseudomys with the new name of Pseudomys (Gyomys) novaehollandiae (Thomas 1910). There is some geographic overlap and occasional evidence of hybridisation with its sister taxon, P. delicatulus (Delicate mouse), in the north of its range (Roycroft et al. 2024). However, the two taxa were found to be both genetically and morphologically distinct (Ford 2003; Roycroft et al. 2024).\n\nThe Pookila is a nocturnal, terrestrial, burrowing rodent. It is small, with a head-body length of approximately 65–90 mm and a tail length of 80–105 mm (Menkhorst & Knight 2011). The species’ dorsal coat is sandy brown with black flecks and the ventral coat is white (Figure 1). The tail appears bicoloured with a darker dorsal side and paler pink ventral side, and the feet are pink with fine white hairs (Figure 1).\n\nIndividuals in the south of the range are heavier than those further north. Tas individuals are the heaviest, weighing 19–28 g, followed by Vic at 18–‍25 g, then Qld and NSW at 12–20 g (Hocking 1980; Burns et al. 2023a). The species is most similar in appearance to the delicate mouse, which is generally smaller and more finely featured. It is superficially similar in appearance to the invasive house mouse (Mus musculus). It is distinguishable by its larger prominent eyes, larger rounded ears, lack of ‘mousy’ odour and absence of a notch on the upper incisors (note that examining dentition is not recommended for identifying live animals) (Seebeck et al. 2003; DEWHA 2010; Burns et al. 2023a; DCCEEW 2024b). The Pookila’s body shape is also thicker and more rounded compared to the house mouse, lacking an obvious neck (Burns 2019b).\n\nFigure 1: A Pookila (Pseudomys novaehollandiae) individual showing identifying characteristics: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.\n\n![Left: An image of a captive Pookila (New Holland Mouse) on a branch.\nRight: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.](image.004.png)Source: Left © Zoos Victoria (2023), Right and insets Kiarrah Smith (2025), Coexistence Conservation Laboratory, Australian National University.\n\n## 2.2 Distribution\n\nThe Pookila has a fragmented distribution across coastal areas of central-eastern and south‑eastern Australia (Figure 2). It occurs in southern Qld, the northern and central coastal areas of NSW, central-southern coastal Vic, and north-eastern Tas including Flinders Island (Burns et al. 2023b; DCCEEW 2024b). A large portion of its distribution occurs within areas with Regional Forest Agreements (RFAs), Indigenous Protected Areas and National Parks. In Qld and NSW, the Pookila occurs in the World and National Heritage Listed Greater Blue Mountains Area and the Gondwana Rainforests of Australia. It is known only from subfossil records in National Heritage Listed areas including the Australian Alps National Parks and Reserves, Ku-ring-gai Chase National Park, Lion, Long and Spectacle Island Nature Reserves and the Great Ocean Road region (Ford 2003; Smith et al. 2023c). While most modern detections of the species exist from coastal areas, it has also been located in areas up to 300 km inland in Qld and NSW. The species’ distribution is believed to have once been continuous, forming one large population throughout these areas.\n\nThe species’ historic distribution is inferred through the presence of subfossils as well as genetic information (Ford 2003; Breed & Ford 2007; Smith et al. 2023c). Its current distribution is attributable both to longer-term climatic and landscape changes, as well as a large population decline following European colonisation (Breed & Ford 2007; Theden-Ringl et al. 2018; Roycroft et al. 2021). Despite the ACT and surrounding region not being mapped within the current Pookila distribution (Figure 2), there is subfossil evidence of historical occurrence in nearby areas with similar habitat (Ford 2003; Smith et al. 2023c).\n\nThe extent of occurrence (EOO) for the species was estimated at 326,000 km2 and the area of occupancy (AOO) for the species was estimated at 694 km2 (DCCEEW 2024b). The AOO may have further contracted since this estimate, as nearly half of the species’ habitat was affected by the 2019 and 2020 bushfires. However, this includes approximately 30% of the species’ habitat that burned at low to medium severity (Legge et al. 2022). The species is likely to survive low to medium severity fires, but high severity fires could greatly reduce survivability (Wilson et al. 2018; Burns & Phillips 2020; DCCEEW 2024b).\n\nThe Pookila is associated with early to mid-stage vegetation succession following fire events (Woinarski & Burbidge 2016). Therefore, some recovery after the 2019 and 2020 bushfires is possible due to potential development of suitable habitat post-fire (DCCEEW 2024b).\n\nFigure 2: Modelled distribution of the Pookila\n\n![The modelled distribution of the Pookila or the Pookila habitat is known or likely to occur from south east Queensland along the northern and central east coast of New South Wales to costal regions in Victoria, predominantly to the east, all the way to Flinders Island and north and eastern Tasmania. The Pookila is also likely to occur 100-300 km inland in New South Wales. Pookila or Pookila habitat may occur in close proximity to these known or likely regions. Refer to subheading 'Species distribution' for more information. ](image.005.jpeg)\n\n> Note: Source: Base map Geoscience Australia; species distribution data Species of National Environmental Significance database.\n\n> Note: Caveat: The information presented in this map has been provided by a range of groups and agencies. While every effort has been made to ensure accuracy and completeness, no guarantee is given, nor responsibility taken by the Commonwealth for errors or omissions, and the Commonwealth does not accept responsibility in respect of any information or advice given in relation to, or as a consequence of, anything contained herein.\n\n> Note: Species distribution mapping: The species distribution mapping categories are indicative only and aim to capture (a) the habitat or geographic feature that represents recent observed locations of the species (known to occur) or habitat occurring in close proximity to these locations (likely to occur); and (b) the broad environmental envelope or geographic region that encompasses all areas that could provide habitat for the species (may occur). These presence categories are created using an extensive database of species observations records, national and regional-scale environmental data, environmental modelling techniques and documented scientific research.\n\n## 2.3 Population size and trends\n\nIn 2008, the national population size of the Pookila was estimated at fewer than 10,000 individuals (Menkhorst et al. 2008; DEWHA 2010) and was reported to be 8,000 individuals in the IUCN assessment by Woinarski and Burbidge (2016). In the assessment for listing under the EPBC Act, the maximum plausible population size prior to the 2019 and 2020 bushfires was estimated (using calculations of AOO and density) to be 34,700 individuals, with the post-fire plausible population range of 13,100–26,300 (DCCEEW 2024b). There was a reduction in the number of Pookila populations located in Vic (from 12 to 5; Burns 2019a), and in Tas (from 16 to 2; (Lazenby et al. 2019). However, there is limited information from which to infer similar trends across the NSW and Qld populations (Lazenby et al. 2019; Burns 2019a; DCCEEW 2024b). A study conducted over a 32-year period near the Tomago aluminium smelter in NSW indicated an overall declining population (SMEC 2015). Shorter-term studies conducted over 5 years recorded a possible decline at Wyrrabalong National Park but no evident trend over 7 years at Munmorah State Conservation Area (DCCEEW 2024b).\n\n## 2.4 Important populations\n\nThe Evolutionarily Significant Units (ESUs) concept is used by conservationists to strategically prioritise taxa or populations of a species for recovery purposes (for example, to target for captive-breeding), when resources are limited or when the granularity of genetic diversity is not sufficiently captured in the current overarching taxonomic status (Avise 1989; Moritz 1994). There are two recognised ESUs of Pookila, separated by more than 200,000 years of divergence (Burns et al. 2023b). The Southern ESU (Vic populations) is rarer, both genetically and geographically fragmented, and more threatened than the Northern ESU (NSW and Qld populations), which is comparatively genetically healthy and not at immediate risk of extinction (Burns et al. 2023b). Tas populations were not included in the Burns et al. (2023b) analysis due to the lack of data. However, these populations are expected to place either independently or within the Southern ESU and are considered to be at extremely high risk of extinction (Burns et al. 2023b). Given the Tas populations are likely closely related to the Vic populations, until further information is available these populations should be managed as part of the Southern ESU (Burns et al. 2024).\n\nPopulations in Vic and Tas are severely fragmented and are thought to make up less than 5% of Pookila numbers nationally (DCCEEW 2024b). In Tas, the species was known from 16 separate sites in 1998, but its status on the Tas mainland is in doubt with no recent records (Lazenby et al. 2019; DNRE Tas 2023). However, it was rediscovered on Flinders Island in 2021 (DNRE Tas 2023). In Vic, extensive surveys found the species in only 5 of the 12 parks and reserves in which it had previously been recorded (Burns 2019a; Burns et al. 2023b). The remaining populations are spatially isolated and at high risk of genetic issues without assisted migrations to support genetic rescue (Burns 2019a; Burns et al. 2023b).\n\nThe connectivity and extent of populations in NSW and south-east Qld is poorly understood. Genetic evidence indicates the isolation of these populations has occurred relatively recently compared to the isolation of populations in Vic (Burns et al. 2023a; Burns et al. 2023b; DCCEEW 2024b). The extant populations are in sites that are geographically distinct and unlikely to be connected by gene flow, as were the now locally extinct populations.\n\nThere is evidence that Pookila populations are strongly affected by rainfall. In NSW, the length of the breeding season, and thus reproductive output, has been shown to increase during times of above-average rainfall (Kemper 1976, 1980). Abundance in the Anglesea (Vic) population had a strong positive relationship with cumulative monthly residual rainfall (Lock 2005). Densities of this population peaked following 4 years of above-average rainfall and declined during below-average rainfall and drought conditions (Lock 2005; Wilson et al. 2017). There are also data to suggest that population fluctuations at Wilsons Promontory (Vic) were influenced by rainfall patterns (Wilson et al. 2005). All populations are considered important for the long-term recovery and survival of the species (Table 5).\n\n \n\nTable 5: Populations of Pookila\n\n| Region | Population | Status | Year of last detection | Year of last survey | Known or expected decline? | Genetic diversity (relative heterozygosity a) |\n| ------ | --------------------------------------------------- | --------------------------------------------------------------------- | ---------------------- | ------------------- | -------------------------- | -------------------------------------------------------------------------------------------------------------------------------- |\n| ACT | Mulligans Flat Woodlands Sanctuary | Not persisting, translocated | 2021 | 2023 | Known | Translocation trials with individuals from the Australian National University (ANU) captive-breeding colony (using NSW founders) |\n| NSW | Barcoongere State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Barrington Tops State Forest | Extant | 2001 | 2025 | Expected decline | Unknown |\n| | Booti Booti National Park | Extant | 2022 | 2022 | Unknown | Unknown |\n| | Broadwater National Park | Extant | 2024 | 2025 | Known | Low |\n| | Bunjalung National Park | Unknown | 2002 | 2024 | Known | Unknown |\n| | Bylong | Unknown | Unknown | Unknown | Unknown | Moderate |\n| | Carrai National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Chaelundi State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Dalmorton State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Doyles River State Forest | Unknown | 1984 | 1984 | Unknown | Unknown |\n| | Ellis State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Gibraltar Range State Forest | Unknown | 2015 | 2015 | Unknown | Unknown |\n| | Goobang National Park | Extant | 2025 | 2025 | Unknown | Moderate |\n| | Goulburn River National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Kiwarrak State Forest | Unknown | 2019 | 2019 | Unknown | Unknown |\n| | Malara State Forest | Unknown | 1999 | Unknown | Unknown | Moderate |\n| | Malara State Forest | Unknown | 1993 | 1993 | Unknown | Unknown |\n| | Manobalai Nature Reserve | Extant | 2024 | 2024 | Unknown | Unknown |\n| | Marara State Forest | Unknown | 2014 | 2014 | Unknown | Unknown |\n| | Marengo State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Mernot State Forest | Unknown | 2008 | 2008 | Unknown | Unknown |\n| | Mount Royal National Park | Extant | 2023 | 2023 | Unknown | Unknown |\n| | Mummel Gulf National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Munmorah/Chain Valley | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Myall Lakes | Extant | 2022 | 2022 | Known | Unknown |\n| | Nymboida National Park | Unknown | 2015 | Unknown | Unknown | Moderate |\n| | Olney State Forest | Unknown | 2004 | 2004 | Unknown | Unknown |\n| | Port Stephens | Extant | 2022 | 2022 | Unknown | Moderate |\n| | Ravensworth State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Royal National Park | Extant | 2022 | 2022 | Expected decline | Unknown |\n| | Styx River State Forest | Unknown | 2000 | 2000 | Unknown | Unknown |\n| | Watchimbark Reserve | Unknown | 2011 | Unknown | Unknown | High |\n| | Werrikimbe National Park | Unknown | 1984 | 2020 | Unknown | Unknown |\n| | Wollemi National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Wyrrabalong National Park | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Yiraaldiya National Park feral predator-free area | Translocated, persistence to be determined by post-release monitoring | 2025 | 2025 | Known | Translocation trials with individuals from the captive-breeding colonies at ANU and NTSI (PPC) using NSW founders |\n| | Yuraygir National Park | Extant | 2024 | 2025 | Expected decline | Unknown |\n| QLD | Crows Nest/ Ravensbourne/Helidon Hills | Unknown | 2022 | 2025 | Known | Moderate |\n| | Emu Creek (private property) | Unknown | 2022 | 2024 | Known | Unknown |\n| | Main Range National Park/Spicers Peak Nature Refuge | Unknown | 2022 | 2025 | Known | Unknown |\n| | Numinbah Valley | Unknown | 2005 | 2005 | Unknown | Unknown |\n| TAS | Beaconsfield | Unknown | 1978 | 2002 | Known | Unknown |\n| | Flinders Island | Extant in the north of the island, undetected in the south | 2025 | 2025 | Expected decline | Unknown |\n| | Freycinet | Unknown | 2003 | 2025 | Known | Unknown |\n| | St Helens | Unknown | 2009 | 2025 | Expected decline | Unknown |\n| | Waterhouse Conservation Area | Unknown | 2010 | 2025 | Known | Unknown |\n| | Wukalina/Mt William | Unknown | 2004 | 2025 | Known | Unknown |\n| VIC | Anglesea | Locally extinct | 1999 | 2021 | Known | Very low (from historic samples) |\n| | Cranbourne | Locally extinct | 1976 | Unknown | Known | Unknown |\n| | Dutson Downs | Extant | 2024 | 2025 | Known | Low |\n| | Gippsland Lakes | Extant | 2025 | 2025 | Known | Low |\n| | Hummock Island | Locally extinct | 1977 | Unknown | Known | Unknown |\n| | Langwarrin | Locally extinct | 1984 | Unknown | Known | Unknown |\n| | Mullungdung | Locally extinct | 1976 | 2017 | Known | Unknown |\n| | Providence Ponds | Extant | 2025 | 2025 | Known | Low |\n| | Reeves Beach | Locally extinct | 1978 | 2017 | Known | Unknown |\n| | The Lakes National Park | Extant | 2025 | 2025 | Known | Low |\n| | Tyabb | Locally extinct | 1975 | Unknown | Known | Low (from historic samples) |\n| | Wilsons Promontory | Extant | 2025 | 2025 | Known | Very low |\n\n> Note: a \\- Heterozygosity measures from Burns et al. (2023b).\n\n> Note: Source: Lazenby et al. (2019), Burns (2021a), DCCEEW (2024), NSW BioNet and unpublished data from members of the recovery team and Recovery Network.\n\n## 2.5 Habitat\n\nPookila habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heathy understorey, and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022) (see Figure 3). Historically, the species has been found across a broader array of habitat types (DCCEEW 2024b), and it does not occupy all plausibly suitable habitat at present (Burns et al. 2020; Smith et al. 2023a). Sandy soils suitable for burrowing are common across most occupied habitats and soft sandy substrates, while deep topsoils are thought to be optimal (Abicair et al. 2020). Within the northern parts of the species’ range, it also occupies sites up to 300 km from the coast and in subcoastal locations at higher elevation, where soil composition differs (Martin 1995; Van Dyck & Lawrie 1997; Vernes et al. 2006; Dixon 2014; Churchill et al. 2023).\n\nThe species is often, although not exclusively, found in areas with high floristic diversity with leguminous perennials (Wilson 1991; Seebeck et al. 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; DEWHA 2010). For example, the heathland element of the species’ habitat is represented by dry heathlands that are floristically diverse and even (a high number of plants that are all relatively abundant compared to each other) that grow on sandy soils with a high percentage of bare ground (Seebeck et al. 1996).\n\nThe species has been identified as preferring early successional habitat, at some sites with peak abundance approximately 3–7 years post-fire and 7–9 years post-sand mine rehabilitation (Braithwaite & Gullan 1978; Fox & Fox 1978; Fox & Fox 1984; Twigg et al. 1989; Wilson 1991; Wilson et al. 2018). However, research in Vic suggests that the relationship between the Pookila and fire is complex, and local densities may be high in areas that have not been affected by fires for over 60 years (Burns and Phillips 2020). Similarly, surveys conducted on the NSW central coast have found high densities in long-unburnt vegetation (F Ford 2023, personal communication 8 August). Season, fire severity and extent, predation pressures, and climatic conditions following fire, all affect habitat suitability more strongly than a linear effect of time-since-fire (Burns & Phillips 2020). Furthermore, it is likely that it shelters in burrows during times of fire. This behaviour may be one of the attributes that allows the Pookila to persist in situ through some fire events (Friend 1993; Burns & Phillips 2020).\n\nThe Pookila has been considered a ‘habitat specialist’ (Braithwaite & Gullan 1978), but fossil record observations, and a strategic translocation and assessment of microhabitat use, suggests that it had broad habitat preferences (Smith et al. 2023c). Remains of the species are common in most bone deposits in south-eastern Australia that has been examined, and these deposits occur across a broad range of historical vegetation types and altitudes (Ford 2003). The species’ perceived habitat specialisation may have been further challenged by the Pookila translocation into a predator exclusion zone in the ACT as part of the Mulligans Flat-Goorooyarroo Woodland Experiment. The box gum and grassy woodlands comprising the predator exclusion area are not the type of coastal habitat often associated with contemporary records of the species (Abicair et al. 2020). However, recent detections of the species in grassy open forests and woodlands in Qld means that these non-coastal habitat types are still represented in the Pookila distribution.\n\nThe currently skewed occurrence (and perhaps specificity) of the Pookila towards coastal areas may in part be due to the interaction among threats, such as habitat clearing, altered fire regimes and predation pressures by introduced species such as cats (Felis catus) and European red foxes (Vulpes vulpes). However, it is worth noting that individuals sourced from the ANU breeding colony in the Mulligans Flat population are significantly underweight compared to their founding populations (13.6 g vs. 17.5 g) (Abicair et al. 2020), and that the population may not be self-sustaining. Together, these attributes may indicate that the habitat there may now be suboptimal for the Pookila. However, it may be important to pursue translocations in a diversity of habitats (including those that may be suboptimal) to build the species' resilience to environmental change and to restore ecological functions (Smith et al. 2023a).\n\nFigure 3: Photographs of contemporary Pookila habitat\n\n![These grouped 6 images demonstrate the diversity in the Pookila's habitat throughout the species' range. Refer to the footnote of this collective image for descriptions on habitat per image in addition to subheading 'Habitat' for more information.](image.006.png)\n\n> Note: Contemporary habitat in which the Pookila has been detected. Depicted are: a) grassy woodlands at Cunninghams Gap in Qld (photo: Ian Gynther); b) woodland with a grass and sedge understorey at Mt Hyland, NSW (photo: Anna Lloyd); c) coastal heathland at Munmorah in NSW (photo: Barbara Wilson); d) woodland with a heath understorey at Anglesea in Vic (photo: Barbara Wilson); e) coastal heathland at Loch Sport in Vic (photo: Phoebe Burns); and f) coastal heathland with a woodland overstorey on Flinders Island in Tas (photo: Rosemary Gales) (Source: DNRE Tas 2022).\n\n### 2.5.1 Habitat critical to the survival of the Pookila\n\nAll potential habitat should be considered important to the Pookila’s persistence and recovery, this includes any:\n\n Areas within the species’ modelled distribution;\n Areas where the species previously occurred but has since become locally extinct;\n Patches of suitable habitat that may not be currently occupied by the species but have the potential to be occupied by natural dispersal; and\n Areas that could be restored for reintroduction.\n\nAll areas that contain habitat that serves one or more of the following ecological functions have the potential to be habitat critical to the survival of the Pookila:\n\n Foraging habitat: vegetation with appropriate density and diversity to be able to support a population, such as native vegetation with seasonal abundance of edible flowering, fruiting, and seeding plants. Vegetation cover is important for the species to be able to forage without increased risk of predation. This habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heath understorey and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022).\n\n Breeding habitat: substrate for the species to be able to create burrows, or alternatively, ‘surface nests’ in low-lying vegetation that is suitable for successful reproduction and rearing of young. Soft sandy substrates, with deep topsoils are thought to be optimal (Abicair et al. 2020) but not exclusive. The Pookila mostly occupies coastal areas that provide this substrate. However, it also occupies sites 100–300 km from the coast and areas with higher elevation where soil composition varies. For example, the presence of sandy soils is highly likely in Goobang National Park (NSW NPWS 2001), whilst in the NSW ranges and Northern Tableland areas, soil may vary with a more dominant loamy or quartz soil base where the Pookila occurs or may be present (NSW OEH 2023. pers comm 22 November).\n\n Habitat for the long-term maintenance of the species: including areas that are not subject to prolonged drought or at risk of extreme droughts in future (such as potential climate refugia), and habitat that is maintained through periodic small-scale burns to provide appropriate foraging habitat and connectivity for the species.\n\n Habitat for the maintenance of genetic diversity: includes all current habitats where the species is extant as they contain populations necessary to achieve maximum genetic diversity and increase resilience of the species against future threats. There are two recognised ESUs: the Southern ESU (Vic populations) and the Northern ESU (NSW and Qld populations). Tas may yet be identified as a third ESU or may cluster within the Southern ESU. Therefore, habitat across the species’ range is necessary to maintain the national genetic diversity of the species.\n\n Habitat for translocation: can be identified using habitat modelling followed by ground-truthing at the local-scale and can include locations that have accounted for future changes due to climate change regimes as well as the exclusion of invasive predators.\n\nRefining what constitutes habitat critical to the survival of the Pookila is a priority action in this recovery plan. Until those key characteristics are further clarified, all areas that contain habitat that serves any of the above five ecological functions have the potential to be habitat critical to the survival of the Pookila. Habitat that supports multiple ecological functions and/or a higher number of the ecological attributes identified within the descriptions provided above, increases the likelihood that the habitat may be critical to the survival of the Pookila.\n\nGiven that the species has declined markedly in parts of its range and populations may rapidly change in size, failure to detect the species during surveys does not indicate the absence of potential habitat critical to its survival.\n\n## 2.6 Ecology\n\nThe Pookila often nests underground in communal burrow systems in well-drained soils. Individuals have also been recorded sheltering under low growing shrubs in surface nests (Kemper 1980, Lazenby et al. 2007). Burrows vary in length, depth, and complexity, ranging from shallow surface nests to intricate tunnel systems (Kemper 1981). Of three burrows that have been excavated, the largest was 5 m in length with a maximum depth of 60 cm (Kemper 1981). The species is nocturnal, and with a significant proportion of its time spent foraging above ground, it is exposed to a high risk of predation by native predators including owls and introduced predators such as the cat and the European red fox.\n\nTen individuals were radio-collared in one area of Tas for approximately 2 weeks, yielding information on nightly movements (Lazenby 1999; Lazenby et al. 2007). Over a 2 ha grid, 9 individuals moved up to 206 m in one night, and up to 37 m in one hour (Lazenby 1999; Lazenby et al. 2007). Pookila home ranges have been known to overlap (Lazenby 1999; Lazenby et al. 2007). The species’ home range is approximately 0.5–1.8 ha, increasing as habitat quality decreases (Lock 2005; Burns et al. 2023a).\n\nThe generation length of the Pookila is between one and two years (Kemper 1980). Females reach sexual maturity at approximately 13 weeks and live 2–3 years (Kemper 1976; Kemper 1980). Males reach sexual maturity at approximately 20 weeks and longevity remains unclear (Kemper 1976; Kemper 1980). During good conditions in NSW, the average litter size for the species has been previously recorded as 4.6, with a range of 1–6 young, and with females usually producing 1–2 litters per year, or 3–4 in exceptional conditions (Kemper 1976; Kemper 1980). The gestation period is around one month (32 days) and young are weaned at 3–4 weeks of age (Kemper 1976; Kemper 1980).\n\nThe breeding season varies depending on location and rainfall, with breeding occurring in NSW from late winter to early summer, while in Vic and Tas the breeding season begins in mid-spring and ceases in late summer (Kemper 1980; Burns et al. 2023b). The impact of rainfall is likely related to its influence on resource availability, which influences variation in birth rates, recruitment, and adult survival (Woinarski et al. 2014). Energy requirements are likely to be high during breeding, particularly for females, and if dietary resources are inadequate during this time, females or their young may perish (Lock & Wilson 2017). Hence, the Pookila demonstrates fluctuating population dynamics where extended breeding seasons follow periods of higher rainfall (Lock & Wilson 2017; Burns 2021). Pookila abundance reduces during periods of drought due to resource shortages and likely failed litters (Lock & Wilson 2017; Burns 2021).\n\n### 2.6.1 Diet and foraging\n\nThe Pookila is an omnivore and feeds on a variety of seeds, leaves, fungi, and invertebrates, exhibiting seasonal and locality-dependent dietary variation (Cockburn 1980; Norton 1987; Wilson and Bradtke 1999; Braithwaite and Gullan 1978). The species plays an important ecological role in dispersing seed and fungi in scats (Wilson & Bradtke 1999; Seebeck et al. 2003; Burns et al. 2023b).\n\n \n\n# 3 Threats\n\n## 3.1 Historical causes of decline\n\nThe distribution of subfossils and the genetic structure of extant populations indicate that the historic distribution of the Pookila extended more continuously throughout parts of coastal eastern Australia, from south-east Qld extending down to the northern parts of Tas and up to 300 km inland on the Australian mainland (Ford 2003; Breed and Ford 2007; DEWHA 2010; Burns et al. 2024; Smith et al. 2023c). The species has since experienced significant declines in distribution and population size and is now distributed in fragmented patches across this area (Figure 2).\n\n## 3.2 Current known and suspected threatening processes\n\nCurrent threats and their prevalence and intensity are described for the national Pookila population below. It is important to note that different populations may be experiencing different combinations of threats in varying intensities. There is significant interaction between threats.\n\n### 3.2.1 Habitat loss, disturbance and modification\n\n#### Habitat clearing\n\nLand clearing for agriculture, infrastructure, road, and housing development continues to place Pookila habitat at risk of further loss and fragmentation. As very little targeted and systematic monitoring of Pookila populations has occurred across the range in NSW (DCCEEW 2024b), the species is at particular risk of losing habitat where presence has not yet been confirmed.\n\nIn 2019, 87% of the Australian population lived within 50 km of the coast (cited in Clark et al. 2021). Coastal development is a continuing, increasing pressure with very high impacts on coastal environments (Clark et al. 2021) that is sprawling beyond established urban areas to regional coastal areas (Infrastructure Australia 2020).\n\nWhen habitat is cleared, disturbed, or modified, the risk and magnitude of impact to adjacent extant habitat patches increases, along with the interactive capability of threats (Fischer & Lindenmayer 2007; DNRE Tas 2022). For example, land clearing for agriculture can increase the carrying capacity of native and introduced herbivores within the landscape, leading to increased grazing and soil impact pressures on adjacent habitat. This is exacerbated when fire results in a further increase in grazing pressure and movement as herbivores actively seek to consume newly emerging vegetative shoots (DAWE 2022). With an increasing trend of habitat loss due to land clearing, the likelihood that fragmentation has occurred and will continue to occur is very high.\n\n#### Habitat modification due to Phytophthora cinnamomi infestation\n\nPhytophthora dieback – caused by the root-rot pathogen Phytophthora cinnamomi – is listed as a Key Threatening Process under the EPBC Act (DOEE 2018) and has been observed across the national range of the Pookila (Liu et al. 2013; Wilson et al. 2015; Burns et al. 2020; DCCEEW 2024b). The pathogen is spread through the movement of soil – including on footwear, tools, vehicles and machinery – and in water (DOEE 2018). While the impacts of P. cinnamomi on Pookila populations and habitat have not been investigated directly, the species is likely to be at significant risk. A range of plant species within Pookila habitat have been identified as susceptible to infection by P. cinnamomi (see McDougall 2005).\n\nPhytophthora dieback causes the loss of plant species richness (including keystone species), reduction of structural complexity (Wilson et al. 1994; Garkaklis et al. 2004; Laidlaw & Wilson 2006; Annett 2008; Armistead 2008; Dundas et al. 2013, 2016; Davis et al. 2014; Wilson et al. 2020), loss of food resources (for example, seeds, leaves, nectar, pollen) and suitable refuges from weather and predation (Laidlaw & Wilson 2006). The impacts of Phytophthora dieback can also extend to invertebrate communities and thus impact insectivorous fauna (Cahill et al. 2008), reducing availability of this food resource.\n\nSignificant negative impacts from Phytophthora dieback on small mammals and their habitat has been previously documented (Newell & Wilson 1993; Lilith 2002; Laidlaw & Wilson 2006; Swinburn et al. 2007; Annett 2008; Armistead 2008; Dundas et al. 2013; 2016), and long-term studies have identified localised extinctions of species and mammal community declines (Zhuang-Griffin 2015; Wilson et al. 2017; 2020).\n\nPriority Protection Areas for Phytophthora dieback management have been identified for the Otway Ranges, which was informed by habitats occupied by EPBC Act listed species, including the Pookila (Walshe et al. 2021; 2022; Wilson & Garkaklis 2023a; 2023b).\n\n#### Habitat modification due to native and introduced weeds\n\nChanging grazing and fire regimes have led to habitat loss by encroachment of coast tea tree (Leptospermum laevigatum) in some Pookila habitat (Burns 2019a; Seebeck et al. 2003). This is particularly evident at Wilsons Promontory National Park and in other coastal Vic sites along Ninety Mile Beach where dune succession has given rise to dense thickets of coast tea tree, such as at Reeves Beech. Weed species are of growing concern in other Pookila habitat, for example, in Lakes National Park. Invasive weeds include Asparagus asparagoides (bridal creeper) and Lycium ferocissimum (boxthorn).\n\n### 3.2.2 Fire regimes that cause declines in biodiversity\n\n‘Fire regimes that cause declines in biodiversity’ is listed as a Key Threatening Process under the EPBC Act (DAWE 2022). Altered fire regimes, such as those resulting from changes to seasonality, frequency and severity of fire weather conditions, are causing declines and loss of biodiversity and ecosystem function (DAWE 2022). The Pookila and the habitat the species relies on have a complicated relationship with fire.\n\nUnder certain circumstances, the species can peak in abundance in the years following a fire as resources become more readily available due to regeneration of vegetation, with such areas described as ‘early successional’ (Fox & McKay 1981). Furthermore, the Pookila has been reported to have relatively high survival during and immediately following bushfires, which is also presumably attributed to its ability to shelter in burrows (Friend 1993; Burns & Phillips 2020). However, this response appears to be more common following small-scale, patchy, low-severity burns and is likely reliant on above-average rainfall in the months and years following the fire (Burns & Phillips 2020).\n\nLarge-scale, high-severity fires have the capacity to extirpate entire populations, as occurred in 1983 on Ash Wednesday at Anglesea (Wilson et al. 2018). Fires being too frequent, too large, too severe, or poorly timed may lead to direct loss of individuals and/or sudden or gradual loss of habitat through reduction of habitat patches of suitable successional age, size, distribution, and plant diversity that the Pookila requires for availability of food and shelter from predators.\n\nSimilarly, prolonged exclusion or poorly timed application of fire may encourage coast tea tree encroachment and loss of favourable food and habitat species. Also, introduced predators may target recently burnt areas, compounding the increased risk of predation from loss of vegetative cover (Hradsky et al. 2017). Planned burning may also threaten the species if conducted during or immediately prior to drought, at too high an intensity, or across a widespread area (Wilson et al. 2018).\n\nThe increasing scale, severity, and frequency of fires under a changing climate pose a risk to the Pookila and its habitat. In 2019 and 2020, catastrophic bushfires impacted a large area of eastern Australia with varying severity (Legge et al. 2021). It was estimated that over 45% of the Pookila distribution was burnt, with a third of this area burning with high severity (Legge et al. 2021). The overall population of Pookila was predicted to have declined by 24% by 2021 (one year after fire), with the species suffering an estimated 75% decrease in abundance in severely burnt areas in that same period (Legge et al. 2021).\n\n### 3.2.3 Climate change: Increasing number of hot days, prolonged drought and changes to precipitation patterns\n\nIncreasing frequency, intensity, and duration of drought across much of the species’ range (CSIRO & BoM 2020) may lead to further catastrophic declines and local extinctions. Extreme heat events may exacerbate the effects of drought and could lead to mass mortality events as individuals dehydrate and fail to thermoregulate but this would require further research to confirm whether behaviours would mitigate this threat. The projections for the NSW climate by 2050 include an increase in minimum and maximum temperatures, an increase (double) in the number of hot days per year (days greater than 35°C), and longer fire seasons, with around a 40% increase in ‘very high’ fire danger days (CSIRO & BoM 2021).\n\nExtended periods of low rainfall reduce food availability for Pookila. Recent periods of drought have caused demonstrable crashes in Pookila populations. The local extinction of the Anglesea population is in part attributed to the Millennium Drought (Lock & Wilson 2017), and the species is yet to recover from the 2019 Gippsland drought (Burns 2021).\n\nFire–drought interactions may have a significant impact on the Pookila with potential compounding effects. This is evident from the 2019 and 2020 bushfires, which followed a period of significant drought in 2017 to 2019 (BoM 2020; DPIE 2020).\n\n### 3.2.4 Invasive fauna species\n\n#### Predation by cats and European red foxes\n\nFeral predators can have rapid and catastrophic impacts on populations of threatened rodents (Tuft et al. 2021). While there is limited direct evidence of the consumption of Pookila by feral predators (records in dog scats; Hoskins 1997; Davis et al. 2015; Lock & Wilson 2017; MidCoast Council NSW 2024. pers comm 13 February), it is recognised in the Threat Abatement Plan for Predation by Feral Cats 2024 that the Pookila is likely highly susceptible to predation by cats (and/or in some cases by foxes) (DCCEEW 2024a) and that cats and foxes can significantly impact population numbers (Seebeck et al. 2003; Burns 2019a).\n\nLike many other Pseudomys species, Pookila are docile and display poor awareness of exotic predators (P. Burns, pers. obs.). In such species, selective predation can occur (as reported, for example, on the broad-toothed rat, Mastacomys fuscus; Green 2002), leading to a disproportionate impact of predation on the vulnerable species even when co-occurring species (for example, the house mouse) are at greater abundance. Furthermore, communal burrowing and small local population densities mean that predation events at burrow entrances can cause rapid local declines (Burns 2019a).\n\nThe relative impact of predation by feral cats and foxes on the Pookila is not yet understood. Based on factors such as small body size, lower breeding output, and occupation of open habitat, some researchers have identified the Pookila as having a high threat rating from feral cat predation (Dickman 1996; Stobo-Wilson et al. 2021). Radford et al. (2018) reported the cat and fox threat to Pookila to be approximately the same based on expert opinion, although with weak evidence. While foxes are currently absent from the Tas mainland and Flinders Island (DNRE Tas 2022), and fox control is in place in other parts of the species’ range, effective cat control in open, non-insular systems beyond the local scale is not yet possible and potentially counterproductive (Lazenby et al. 2015; Lazenby et al. 2021; see Predator management).\n\nThe density and predatory impact of cats and foxes can be altered by competitive interactions with other predators. For example, reduced numbers of foxes via fox control can increase cat density, alter cat behaviour, and ultimately increase the impact of predation by feral cats (Risbey et al. 2000; Rees et al. 2023). Tasmanian devils have been shown to alter the behaviour and abundance of feral cats (Hollings et al. 2014; Lazenby & Dickman 2013), and cats have a much higher predatory impact on small mammals compared to devils (Hollings et al. 2014). The decline of devils caused by Devil Facial Tumour Disease (DFTD) is likely to have increased the predatory impacts of feral cats in Pookila habitat on the Tas mainland.\n\n#### Impact of hooved feral animals including deer, goats, horses, and pigs\n\nThe negative impacts of hooved feral animals on vegetation health and soil erosion are well documented (Keith & Pellow 2005; Bengsen et al. 2017; Wills et al. 2023; Burns et al. 2021). This is also recognised within the Draft threat abatement plan for competition and land degradation by unmanaged goats (DCCEEW 2024c), the Threat abatement plan for predation, habitat degradation, competition and disease transmission by feral pigs (Sus scrofa) (DOEE 2017), and the National feral deer action plan (Government of South Australia 2023). Where these species co-occur with the Pookila, there are risks of habitat degradation through trampling, wallowing, and overgrazing (Burns 2021). Overgrazing pressures may increase in post-fire environments, particularly in heathy vegetation types preferred by the Pookila (Heaton et al. 2022), compounding post-fire food shortages and predation risk, while impeding regrowth of suitable habitat and food species (DAWE 2022). The effects of hooved feral animal control programs on Pookila occurrence and abundance have not been directly studied, but based on vegetation response studies, control programs are expected to benefit the Pookila and its habitat, particularly around fire events.\n\n#### Competition/predation from introduced rodents\n\nWhile experimental trials have shown the Pookila to be competitively superior to the invasive house mouse (Fox & Twigg 1991; Haering & Fox 1997), there is growing evidence that house mouse plagues may supress Pookila recovery following periods of drought where resources are subsequently scarce (Burns 2021). Rattus rattus and R. norvegicus (black and brown rats, respectively) may also out-compete the Pookila for resources and prey on adults and juveniles, although this requires further research.\n\n#### Baiting of rabbits and invasive rodents\n\nThe susceptibility of the Pookila to 1080, Pindone, and rodenticide has not been tested. However, such poisons are lethal at low doses to some other Pseudomys species and are expected to be lethal to the Pookila (McIlroy 1982; Eisler 1995; Fisher 1999). Running a lethal dose trial on the Pookila is not appropriate given the species’ threatened status and ethical considerations, so it is safest to apply the Precautionary Principle and assume the species is likely susceptible to 1080, Pindone and rodenticide. Rabbit poison baiting is usually conducted with a carrot or oat bait, both foods readily consumed by the Pookila (Prosser et al. 2007), and it is possible that rabbit-baiting programs have contributed to the decline of the species in some areas (Fisher 1999; Burns 2019a). When finding the balance between protecting Pookila habitat by controlling rabbits and potentially poisoning individual Pookila, it is recommended to entirely avoid poison baiting in areas where Pookila presence is known. This is particularly important for at-risk populations; however, it may also apply to locations where Pookila presence is uncertain but these areas are adjacent to suitable habitat and/or a known population (within a 5 km radius). Rabbit and invasive rodent population control measures should use alternative methods. Control measures should be carefully considered depending on the area of control and the potential side effects for native species. Currently, this threat is considered to be high in Wilsons Promontory in Vic, not applicable in Qld or Tas, and unknown in NSW (DNRE Tas 2023).\n\n### 3.2.5 Small and fragmented populations: Low genetic diversity and inbreeding\n\nHistorical and contemporary habitat degradation and destruction, and the species’ very low dispersal ability have left the Pookila highly fragmented across its distribution, with some populations genetically isolated. This severe fragmentation leaves populations at higher risk of extinction, as a single catastrophic event can more easily impact the total area occupied by a population (Garner et al. 2020).\n\nFurthermore, genetically isolated populations are at higher risk of inbreeding depression. Inbreeding reduces genetic diversity, which reduces adaptability to changing environments and anthropogenic impacts. Inbreeding depression also increases species’ vulnerability to disease while also allowing potentially detrimental traits to surface in the population (Keller 2002). Furthermore, inbreeding depression can cause reduced fecundity and further genetic health issues (for example, chromosomal abnormalities) (Keller 2002).\n\nVic populations (Southern ESU) of the Pookila have relatively low heterozygosity (genetic diversity) and are at risk of inbreeding, with no natural gene flow possible between localities given lack of dispersal (Burns et al. 2023b). Therefore, the Southern ESU is considered an ‘at-risk ESU’. Populations across NSW retain greater heterozygosity and are genetically distinct from those in Vic, although animals in Broadwater National Park appear to have undergone a genetic bottleneck (Burns et al. 2023b). At the time of publishing this recovery plan, two historic mainland Tas samples and a small number of contemporary samples from Flinders Island have been sequenced and are being analysed. Additional Tas samples are being sought through survey work. Genetic mixing within Vic populations has been recommended and is being pursued, whereas no immediate action is required in the Northern ESU (NSW and Qld) (Burns et al. 2023b; Burns et al. 2024) since it is more genetically healthy than the Southern ESU. The recovery team has developed a National Pookila genetic management strategy (Zoos Victoria) and will continue to update it when new information is forthcoming.\n\n#### Disease\n\nThere is limited understanding of the potential prevalence or impact of disease on the Pookila. In general, screening of native rodents for pathogens has been very limited and little information is available on symptomology (Thompson et al. 2010). However, feral cats and invasive rodents carry diseases that could impact the Pookila. In particular, Toxoplasma gondii, a virulent parasite spread by cats, is likely present through much of the species’ former range (Dickson 2018). Previous experimental trials in other Pseudomys species demonstrated rapid, high mortality rates after infection with T. gondii, hence this parasite is likely to also increase the mortality of the Pookila (Burns 2019a).\n\n## 3.3 Threat prioritisation\n\nEach of the threats outlined in this recovery plan has been assessed to determine the risk posed across the national distribution of the Pookila using a risk matrix (see Table 6). The risk matrix considers the likelihood of an incident occurring and the consequences of that incident and uses a qualitative assessment drawing on both peer-reviewed literature and expert opinion.\n\nTable 6: Risk matrix\n\n| Consequence (x) / Likelihood (y) | Not significant | Minor | Moderate | Major | Catastrophic |\n| -------------------------------- | --------------- | ------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------- |\n| Almost certain | Low risk | Moderate risk | Very high riskPredation by European red foxes | Very high riskHabitat clearing (past/present)Fire regimes that cause declines in biodiversityIncreasing number of hot days, prolonged drought and changes to precipitation patternsPredation by cats | Very high risk |\n| Likely | Low risk | Moderate riskImpact of hooved feral animals including deer, goats, horses, and pigs | High riskHabitat modification due to Phytophthora cinnamomi infestationHabitat modification due to native and introduced weeds | Very high riskHabitat clearing (future) | Very high risk |\n| Possible | Low risk | Moderate riskCompetition/predation from introduced rodentsBaiting of rabbits and invasive rodents | High riskLow genetic diversity and inbreedingDisease | Very high risk | Very high risk |\n| Unlikely | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n| Unknown | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n\nRisk matrix key\n\n| Low risk | Moderate risk | High risk | Very-high risk |\n| -------- | ------------- | --------- | -------------- |\n\nDefinition of likelihood and consequence categories\n\nConsequence is a measure of the severity of the impact of the threat on the survival of the species.\n\n- Catastrophic – affecting survival, reproduction or essential movement of individuals in >80% of the population or across >80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Major – affecting survival, reproduction or essential movement of individuals in 50-80 % of the population or across 50-80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Moderate – affecting survival, reproduction or essential movement of individuals in 20-50% of the population or across 20-50% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Minor – affecting survival, reproduction or essential movement of individuals in 5-30% of the population or across 5-30% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Not significant – affecting survival, reproduction or essential movement of individuals in <5% of the population or across <5% of the distribution within the next 3 generations or 10 years, whichever is longer.\n\nLikelihood is a measure of how regularly the threat impacts the species.\n\n- Almost certain – >90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Likely – 66-90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Possible – 33-66% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unlikely – <33% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unknown – probability that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer, could be 0-100%.","sortOrder":11},{"sectionNumber":"2.1","sectionType":"part","heading":"Taxonomy and description","content":"## 2.1 Taxonomy and description\n\nPseudomys novaehollandiae, family Muridae, is endemic to Australia and was first described in 1843 as Mus novaehollandiae (Waterhouse 1843). It was later transferred to the genus Pseudomys with the new name of Pseudomys (Gyomys) novaehollandiae (Thomas 1910). There is some geographic overlap and occasional evidence of hybridisation with its sister taxon, P. delicatulus (Delicate mouse), in the north of its range (Roycroft et al. 2024). However, the two taxa were found to be both genetically and morphologically distinct (Ford 2003; Roycroft et al. 2024).\n\nThe Pookila is a nocturnal, terrestrial, burrowing rodent. It is small, with a head-body length of approximately 65–90 mm and a tail length of 80–105 mm (Menkhorst & Knight 2011). The species’ dorsal coat is sandy brown with black flecks and the ventral coat is white (Figure 1). The tail appears bicoloured with a darker dorsal side and paler pink ventral side, and the feet are pink with fine white hairs (Figure 1).\n\nIndividuals in the south of the range are heavier than those further north. Tas individuals are the heaviest, weighing 19–28 g, followed by Vic at 18–‍25 g, then Qld and NSW at 12–20 g (Hocking 1980; Burns et al. 2023a). The species is most similar in appearance to the delicate mouse, which is generally smaller and more finely featured. It is superficially similar in appearance to the invasive house mouse (Mus musculus). It is distinguishable by its larger prominent eyes, larger rounded ears, lack of ‘mousy’ odour and absence of a notch on the upper incisors (note that examining dentition is not recommended for identifying live animals) (Seebeck et al. 2003; DEWHA 2010; Burns et al. 2023a; DCCEEW 2024b). The Pookila’s body shape is also thicker and more rounded compared to the house mouse, lacking an obvious neck (Burns 2019b).\n\nFigure 1: A Pookila (Pseudomys novaehollandiae) individual showing identifying characteristics: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.\n\n![Left: An image of a captive Pookila (New Holland Mouse) on a branch.\nRight: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.](image.004.png)Source: Left © Zoos Victoria (2023), Right and insets Kiarrah Smith (2025), Coexistence Conservation Laboratory, Australian National University.","sortOrder":12},{"sectionNumber":"2.2","sectionType":"part","heading":"Distribution","content":"## 2.2 Distribution\n\nThe Pookila has a fragmented distribution across coastal areas of central-eastern and south‑eastern Australia (Figure 2). It occurs in southern Qld, the northern and central coastal areas of NSW, central-southern coastal Vic, and north-eastern Tas including Flinders Island (Burns et al. 2023b; DCCEEW 2024b). A large portion of its distribution occurs within areas with Regional Forest Agreements (RFAs), Indigenous Protected Areas and National Parks. In Qld and NSW, the Pookila occurs in the World and National Heritage Listed Greater Blue Mountains Area and the Gondwana Rainforests of Australia. It is known only from subfossil records in National Heritage Listed areas including the Australian Alps National Parks and Reserves, Ku-ring-gai Chase National Park, Lion, Long and Spectacle Island Nature Reserves and the Great Ocean Road region (Ford 2003; Smith et al. 2023c). While most modern detections of the species exist from coastal areas, it has also been located in areas up to 300 km inland in Qld and NSW. The species’ distribution is believed to have once been continuous, forming one large population throughout these areas.\n\nThe species’ historic distribution is inferred through the presence of subfossils as well as genetic information (Ford 2003; Breed & Ford 2007; Smith et al. 2023c). Its current distribution is attributable both to longer-term climatic and landscape changes, as well as a large population decline following European colonisation (Breed & Ford 2007; Theden-Ringl et al. 2018; Roycroft et al. 2021). Despite the ACT and surrounding region not being mapped within the current Pookila distribution (Figure 2), there is subfossil evidence of historical occurrence in nearby areas with similar habitat (Ford 2003; Smith et al. 2023c).\n\nThe extent of occurrence (EOO) for the species was estimated at 326,000 km2 and the area of occupancy (AOO) for the species was estimated at 694 km2 (DCCEEW 2024b). The AOO may have further contracted since this estimate, as nearly half of the species’ habitat was affected by the 2019 and 2020 bushfires. However, this includes approximately 30% of the species’ habitat that burned at low to medium severity (Legge et al. 2022). The species is likely to survive low to medium severity fires, but high severity fires could greatly reduce survivability (Wilson et al. 2018; Burns & Phillips 2020; DCCEEW 2024b).\n\nThe Pookila is associated with early to mid-stage vegetation succession following fire events (Woinarski & Burbidge 2016). Therefore, some recovery after the 2019 and 2020 bushfires is possible due to potential development of suitable habitat post-fire (DCCEEW 2024b).\n\nFigure 2: Modelled distribution of the Pookila\n\n![The modelled distribution of the Pookila or the Pookila habitat is known or likely to occur from south east Queensland along the northern and central east coast of New South Wales to costal regions in Victoria, predominantly to the east, all the way to Flinders Island and north and eastern Tasmania. The Pookila is also likely to occur 100-300 km inland in New South Wales. Pookila or Pookila habitat may occur in close proximity to these known or likely regions. Refer to subheading 'Species distribution' for more information. ](image.005.jpeg)\n\n> Note: Source: Base map Geoscience Australia; species distribution data Species of National Environmental Significance database.\n\n> Note: Caveat: The information presented in this map has been provided by a range of groups and agencies. While every effort has been made to ensure accuracy and completeness, no guarantee is given, nor responsibility taken by the Commonwealth for errors or omissions, and the Commonwealth does not accept responsibility in respect of any information or advice given in relation to, or as a consequence of, anything contained herein.\n\n> Note: Species distribution mapping: The species distribution mapping categories are indicative only and aim to capture (a) the habitat or geographic feature that represents recent observed locations of the species (known to occur) or habitat occurring in close proximity to these locations (likely to occur); and (b) the broad environmental envelope or geographic region that encompasses all areas that could provide habitat for the species (may occur). These presence categories are created using an extensive database of species observations records, national and regional-scale environmental data, environmental modelling techniques and documented scientific research.","sortOrder":13},{"sectionNumber":"2.3","sectionType":"part","heading":"Population size and trends","content":"## 2.3 Population size and trends\n\nIn 2008, the national population size of the Pookila was estimated at fewer than 10,000 individuals (Menkhorst et al. 2008; DEWHA 2010) and was reported to be 8,000 individuals in the IUCN assessment by Woinarski and Burbidge (2016). In the assessment for listing under the EPBC Act, the maximum plausible population size prior to the 2019 and 2020 bushfires was estimated (using calculations of AOO and density) to be 34,700 individuals, with the post-fire plausible population range of 13,100–26,300 (DCCEEW 2024b). There was a reduction in the number of Pookila populations located in Vic (from 12 to 5; Burns 2019a), and in Tas (from 16 to 2; (Lazenby et al. 2019). However, there is limited information from which to infer similar trends across the NSW and Qld populations (Lazenby et al. 2019; Burns 2019a; DCCEEW 2024b). A study conducted over a 32-year period near the Tomago aluminium smelter in NSW indicated an overall declining population (SMEC 2015). Shorter-term studies conducted over 5 years recorded a possible decline at Wyrrabalong National Park but no evident trend over 7 years at Munmorah State Conservation Area (DCCEEW 2024b).","sortOrder":14},{"sectionNumber":"2.4","sectionType":"part","heading":"Important populations","content":"## 2.4 Important populations\n\nThe Evolutionarily Significant Units (ESUs) concept is used by conservationists to strategically prioritise taxa or populations of a species for recovery purposes (for example, to target for captive-breeding), when resources are limited or when the granularity of genetic diversity is not sufficiently captured in the current overarching taxonomic status (Avise 1989; Moritz 1994). There are two recognised ESUs of Pookila, separated by more than 200,000 years of divergence (Burns et al. 2023b). The Southern ESU (Vic populations) is rarer, both genetically and geographically fragmented, and more threatened than the Northern ESU (NSW and Qld populations), which is comparatively genetically healthy and not at immediate risk of extinction (Burns et al. 2023b). Tas populations were not included in the Burns et al. (2023b) analysis due to the lack of data. However, these populations are expected to place either independently or within the Southern ESU and are considered to be at extremely high risk of extinction (Burns et al. 2023b). Given the Tas populations are likely closely related to the Vic populations, until further information is available these populations should be managed as part of the Southern ESU (Burns et al. 2024).\n\nPopulations in Vic and Tas are severely fragmented and are thought to make up less than 5% of Pookila numbers nationally (DCCEEW 2024b). In Tas, the species was known from 16 separate sites in 1998, but its status on the Tas mainland is in doubt with no recent records (Lazenby et al. 2019; DNRE Tas 2023). However, it was rediscovered on Flinders Island in 2021 (DNRE Tas 2023). In Vic, extensive surveys found the species in only 5 of the 12 parks and reserves in which it had previously been recorded (Burns 2019a; Burns et al. 2023b). The remaining populations are spatially isolated and at high risk of genetic issues without assisted migrations to support genetic rescue (Burns 2019a; Burns et al. 2023b).\n\nThe connectivity and extent of populations in NSW and south-east Qld is poorly understood. Genetic evidence indicates the isolation of these populations has occurred relatively recently compared to the isolation of populations in Vic (Burns et al. 2023a; Burns et al. 2023b; DCCEEW 2024b). The extant populations are in sites that are geographically distinct and unlikely to be connected by gene flow, as were the now locally extinct populations.\n\nThere is evidence that Pookila populations are strongly affected by rainfall. In NSW, the length of the breeding season, and thus reproductive output, has been shown to increase during times of above-average rainfall (Kemper 1976, 1980). Abundance in the Anglesea (Vic) population had a strong positive relationship with cumulative monthly residual rainfall (Lock 2005). Densities of this population peaked following 4 years of above-average rainfall and declined during below-average rainfall and drought conditions (Lock 2005; Wilson et al. 2017). There are also data to suggest that population fluctuations at Wilsons Promontory (Vic) were influenced by rainfall patterns (Wilson et al. 2005). All populations are considered important for the long-term recovery and survival of the species (Table 5).\n\n \n\nTable 5: Populations of Pookila\n\n| Region | Population | Status | Year of last detection | Year of last survey | Known or expected decline? | Genetic diversity (relative heterozygosity a) |\n| ------ | --------------------------------------------------- | --------------------------------------------------------------------- | ---------------------- | ------------------- | -------------------------- | -------------------------------------------------------------------------------------------------------------------------------- |\n| ACT | Mulligans Flat Woodlands Sanctuary | Not persisting, translocated | 2021 | 2023 | Known | Translocation trials with individuals from the Australian National University (ANU) captive-breeding colony (using NSW founders) |\n| NSW | Barcoongere State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Barrington Tops State Forest | Extant | 2001 | 2025 | Expected decline | Unknown |\n| | Booti Booti National Park | Extant | 2022 | 2022 | Unknown | Unknown |\n| | Broadwater National Park | Extant | 2024 | 2025 | Known | Low |\n| | Bunjalung National Park | Unknown | 2002 | 2024 | Known | Unknown |\n| | Bylong | Unknown | Unknown | Unknown | Unknown | Moderate |\n| | Carrai National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Chaelundi State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Dalmorton State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Doyles River State Forest | Unknown | 1984 | 1984 | Unknown | Unknown |\n| | Ellis State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Gibraltar Range State Forest | Unknown | 2015 | 2015 | Unknown | Unknown |\n| | Goobang National Park | Extant | 2025 | 2025 | Unknown | Moderate |\n| | Goulburn River National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Kiwarrak State Forest | Unknown | 2019 | 2019 | Unknown | Unknown |\n| | Malara State Forest | Unknown | 1999 | Unknown | Unknown | Moderate |\n| | Malara State Forest | Unknown | 1993 | 1993 | Unknown | Unknown |\n| | Manobalai Nature Reserve | Extant | 2024 | 2024 | Unknown | Unknown |\n| | Marara State Forest | Unknown | 2014 | 2014 | Unknown | Unknown |\n| | Marengo State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Mernot State Forest | Unknown | 2008 | 2008 | Unknown | Unknown |\n| | Mount Royal National Park | Extant | 2023 | 2023 | Unknown | Unknown |\n| | Mummel Gulf National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Munmorah/Chain Valley | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Myall Lakes | Extant | 2022 | 2022 | Known | Unknown |\n| | Nymboida National Park | Unknown | 2015 | Unknown | Unknown | Moderate |\n| | Olney State Forest | Unknown | 2004 | 2004 | Unknown | Unknown |\n| | Port Stephens | Extant | 2022 | 2022 | Unknown | Moderate |\n| | Ravensworth State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Royal National Park | Extant | 2022 | 2022 | Expected decline | Unknown |\n| | Styx River State Forest | Unknown | 2000 | 2000 | Unknown | Unknown |\n| | Watchimbark Reserve | Unknown | 2011 | Unknown | Unknown | High |\n| | Werrikimbe National Park | Unknown | 1984 | 2020 | Unknown | Unknown |\n| | Wollemi National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Wyrrabalong National Park | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Yiraaldiya National Park feral predator-free area | Translocated, persistence to be determined by post-release monitoring | 2025 | 2025 | Known | Translocation trials with individuals from the captive-breeding colonies at ANU and NTSI (PPC) using NSW founders |\n| | Yuraygir National Park | Extant | 2024 | 2025 | Expected decline | Unknown |\n| QLD | Crows Nest/ Ravensbourne/Helidon Hills | Unknown | 2022 | 2025 | Known | Moderate |\n| | Emu Creek (private property) | Unknown | 2022 | 2024 | Known | Unknown |\n| | Main Range National Park/Spicers Peak Nature Refuge | Unknown | 2022 | 2025 | Known | Unknown |\n| | Numinbah Valley | Unknown | 2005 | 2005 | Unknown | Unknown |\n| TAS | Beaconsfield | Unknown | 1978 | 2002 | Known | Unknown |\n| | Flinders Island | Extant in the north of the island, undetected in the south | 2025 | 2025 | Expected decline | Unknown |\n| | Freycinet | Unknown | 2003 | 2025 | Known | Unknown |\n| | St Helens | Unknown | 2009 | 2025 | Expected decline | Unknown |\n| | Waterhouse Conservation Area | Unknown | 2010 | 2025 | Known | Unknown |\n| | Wukalina/Mt William | Unknown | 2004 | 2025 | Known | Unknown |\n| VIC | Anglesea | Locally extinct | 1999 | 2021 | Known | Very low (from historic samples) |\n| | Cranbourne | Locally extinct | 1976 | Unknown | Known | Unknown |\n| | Dutson Downs | Extant | 2024 | 2025 | Known | Low |\n| | Gippsland Lakes | Extant | 2025 | 2025 | Known | Low |\n| | Hummock Island | Locally extinct | 1977 | Unknown | Known | Unknown |\n| | Langwarrin | Locally extinct | 1984 | Unknown | Known | Unknown |\n| | Mullungdung | Locally extinct | 1976 | 2017 | Known | Unknown |\n| | Providence Ponds | Extant | 2025 | 2025 | Known | Low |\n| | Reeves Beach | Locally extinct | 1978 | 2017 | Known | Unknown |\n| | The Lakes National Park | Extant | 2025 | 2025 | Known | Low |\n| | Tyabb | Locally extinct | 1975 | Unknown | Known | Low (from historic samples) |\n| | Wilsons Promontory | Extant | 2025 | 2025 | Known | Very low |\n\n> Note: a \\- Heterozygosity measures from Burns et al. (2023b).\n\n> Note: Source: Lazenby et al. (2019), Burns (2021a), DCCEEW (2024), NSW BioNet and unpublished data from members of the recovery team and Recovery Network.","sortOrder":15},{"sectionNumber":"2.5","sectionType":"part","heading":"Habitat","content":"## 2.5 Habitat\n\nPookila habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heathy understorey, and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022) (see Figure 3). Historically, the species has been found across a broader array of habitat types (DCCEEW 2024b), and it does not occupy all plausibly suitable habitat at present (Burns et al. 2020; Smith et al. 2023a). Sandy soils suitable for burrowing are common across most occupied habitats and soft sandy substrates, while deep topsoils are thought to be optimal (Abicair et al. 2020). Within the northern parts of the species’ range, it also occupies sites up to 300 km from the coast and in subcoastal locations at higher elevation, where soil composition differs (Martin 1995; Van Dyck & Lawrie 1997; Vernes et al. 2006; Dixon 2014; Churchill et al. 2023).\n\nThe species is often, although not exclusively, found in areas with high floristic diversity with leguminous perennials (Wilson 1991; Seebeck et al. 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; DEWHA 2010). For example, the heathland element of the species’ habitat is represented by dry heathlands that are floristically diverse and even (a high number of plants that are all relatively abundant compared to each other) that grow on sandy soils with a high percentage of bare ground (Seebeck et al. 1996).\n\nThe species has been identified as preferring early successional habitat, at some sites with peak abundance approximately 3–7 years post-fire and 7–9 years post-sand mine rehabilitation (Braithwaite & Gullan 1978; Fox & Fox 1978; Fox & Fox 1984; Twigg et al. 1989; Wilson 1991; Wilson et al. 2018). However, research in Vic suggests that the relationship between the Pookila and fire is complex, and local densities may be high in areas that have not been affected by fires for over 60 years (Burns and Phillips 2020). Similarly, surveys conducted on the NSW central coast have found high densities in long-unburnt vegetation (F Ford 2023, personal communication 8 August). Season, fire severity and extent, predation pressures, and climatic conditions following fire, all affect habitat suitability more strongly than a linear effect of time-since-fire (Burns & Phillips 2020). Furthermore, it is likely that it shelters in burrows during times of fire. This behaviour may be one of the attributes that allows the Pookila to persist in situ through some fire events (Friend 1993; Burns & Phillips 2020).\n\nThe Pookila has been considered a ‘habitat specialist’ (Braithwaite & Gullan 1978), but fossil record observations, and a strategic translocation and assessment of microhabitat use, suggests that it had broad habitat preferences (Smith et al. 2023c). Remains of the species are common in most bone deposits in south-eastern Australia that has been examined, and these deposits occur across a broad range of historical vegetation types and altitudes (Ford 2003). The species’ perceived habitat specialisation may have been further challenged by the Pookila translocation into a predator exclusion zone in the ACT as part of the Mulligans Flat-Goorooyarroo Woodland Experiment. The box gum and grassy woodlands comprising the predator exclusion area are not the type of coastal habitat often associated with contemporary records of the species (Abicair et al. 2020). However, recent detections of the species in grassy open forests and woodlands in Qld means that these non-coastal habitat types are still represented in the Pookila distribution.\n\nThe currently skewed occurrence (and perhaps specificity) of the Pookila towards coastal areas may in part be due to the interaction among threats, such as habitat clearing, altered fire regimes and predation pressures by introduced species such as cats (Felis catus) and European red foxes (Vulpes vulpes). However, it is worth noting that individuals sourced from the ANU breeding colony in the Mulligans Flat population are significantly underweight compared to their founding populations (13.6 g vs. 17.5 g) (Abicair et al. 2020), and that the population may not be self-sustaining. Together, these attributes may indicate that the habitat there may now be suboptimal for the Pookila. However, it may be important to pursue translocations in a diversity of habitats (including those that may be suboptimal) to build the species' resilience to environmental change and to restore ecological functions (Smith et al. 2023a).\n\nFigure 3: Photographs of contemporary Pookila habitat\n\n![These grouped 6 images demonstrate the diversity in the Pookila's habitat throughout the species' range. Refer to the footnote of this collective image for descriptions on habitat per image in addition to subheading 'Habitat' for more information.](image.006.png)\n\n> Note: Contemporary habitat in which the Pookila has been detected. Depicted are: a) grassy woodlands at Cunninghams Gap in Qld (photo: Ian Gynther); b) woodland with a grass and sedge understorey at Mt Hyland, NSW (photo: Anna Lloyd); c) coastal heathland at Munmorah in NSW (photo: Barbara Wilson); d) woodland with a heath understorey at Anglesea in Vic (photo: Barbara Wilson); e) coastal heathland at Loch Sport in Vic (photo: Phoebe Burns); and f) coastal heathland with a woodland overstorey on Flinders Island in Tas (photo: Rosemary Gales) (Source: DNRE Tas 2022).","sortOrder":16},{"sectionNumber":"2.5.1 Habitat critical to the survival o","sectionType":"division","heading":"2.5.1 Habitat critical to the survival of the Pookila","content":"### 2.5.1 Habitat critical to the survival of the Pookila\n\nAll potential habitat should be considered important to the Pookila’s persistence and recovery, this includes any:\n\n Areas within the species’ modelled distribution;\n Areas where the species previously occurred but has since become locally extinct;\n Patches of suitable habitat that may not be currently occupied by the species but have the potential to be occupied by natural dispersal; and\n Areas that could be restored for reintroduction.\n\nAll areas that contain habitat that serves one or more of the following ecological functions have the potential to be habitat critical to the survival of the Pookila:\n\n Foraging habitat: vegetation with appropriate density and diversity to be able to support a population, such as native vegetation with seasonal abundance of edible flowering, fruiting, and seeding plants. Vegetation cover is important for the species to be able to forage without increased risk of predation. This habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heath understorey and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022).\n\n Breeding habitat: substrate for the species to be able to create burrows, or alternatively, ‘surface nests’ in low-lying vegetation that is suitable for successful reproduction and rearing of young. Soft sandy substrates, with deep topsoils are thought to be optimal (Abicair et al. 2020) but not exclusive. The Pookila mostly occupies coastal areas that provide this substrate. However, it also occupies sites 100–300 km from the coast and areas with higher elevation where soil composition varies. For example, the presence of sandy soils is highly likely in Goobang National Park (NSW NPWS 2001), whilst in the NSW ranges and Northern Tableland areas, soil may vary with a more dominant loamy or quartz soil base where the Pookila occurs or may be present (NSW OEH 2023. pers comm 22 November).\n\n Habitat for the long-term maintenance of the species: including areas that are not subject to prolonged drought or at risk of extreme droughts in future (such as potential climate refugia), and habitat that is maintained through periodic small-scale burns to provide appropriate foraging habitat and connectivity for the species.\n\n Habitat for the maintenance of genetic diversity: includes all current habitats where the species is extant as they contain populations necessary to achieve maximum genetic diversity and increase resilience of the species against future threats. There are two recognised ESUs: the Southern ESU (Vic populations) and the Northern ESU (NSW and Qld populations). Tas may yet be identified as a third ESU or may cluster within the Southern ESU. Therefore, habitat across the species’ range is necessary to maintain the national genetic diversity of the species.\n\n Habitat for translocation: can be identified using habitat modelling followed by ground-truthing at the local-scale and can include locations that have accounted for future changes due to climate change regimes as well as the exclusion of invasive predators.\n\nRefining what constitutes habitat critical to the survival of the Pookila is a priority action in this recovery plan. Until those key characteristics are further clarified, all areas that contain habitat that serves any of the above five ecological functions have the potential to be habitat critical to the survival of the Pookila. Habitat that supports multiple ecological functions and/or a higher number of the ecological attributes identified within the descriptions provided above, increases the likelihood that the habitat may be critical to the survival of the Pookila.\n\nGiven that the species has declined markedly in parts of its range and populations may rapidly change in size, failure to detect the species during surveys does not indicate the absence of potential habitat critical to its survival.","sortOrder":17},{"sectionNumber":"2.6","sectionType":"part","heading":"Ecology","content":"## 2.6 Ecology\n\nThe Pookila often nests underground in communal burrow systems in well-drained soils. Individuals have also been recorded sheltering under low growing shrubs in surface nests (Kemper 1980, Lazenby et al. 2007). Burrows vary in length, depth, and complexity, ranging from shallow surface nests to intricate tunnel systems (Kemper 1981). Of three burrows that have been excavated, the largest was 5 m in length with a maximum depth of 60 cm (Kemper 1981). The species is nocturnal, and with a significant proportion of its time spent foraging above ground, it is exposed to a high risk of predation by native predators including owls and introduced predators such as the cat and the European red fox.\n\nTen individuals were radio-collared in one area of Tas for approximately 2 weeks, yielding information on nightly movements (Lazenby 1999; Lazenby et al. 2007). Over a 2 ha grid, 9 individuals moved up to 206 m in one night, and up to 37 m in one hour (Lazenby 1999; Lazenby et al. 2007). Pookila home ranges have been known to overlap (Lazenby 1999; Lazenby et al. 2007). The species’ home range is approximately 0.5–1.8 ha, increasing as habitat quality decreases (Lock 2005; Burns et al. 2023a).\n\nThe generation length of the Pookila is between one and two years (Kemper 1980). Females reach sexual maturity at approximately 13 weeks and live 2–3 years (Kemper 1976; Kemper 1980). Males reach sexual maturity at approximately 20 weeks and longevity remains unclear (Kemper 1976; Kemper 1980). During good conditions in NSW, the average litter size for the species has been previously recorded as 4.6, with a range of 1–6 young, and with females usually producing 1–2 litters per year, or 3–4 in exceptional conditions (Kemper 1976; Kemper 1980). The gestation period is around one month (32 days) and young are weaned at 3–4 weeks of age (Kemper 1976; Kemper 1980).\n\nThe breeding season varies depending on location and rainfall, with breeding occurring in NSW from late winter to early summer, while in Vic and Tas the breeding season begins in mid-spring and ceases in late summer (Kemper 1980; Burns et al. 2023b). The impact of rainfall is likely related to its influence on resource availability, which influences variation in birth rates, recruitment, and adult survival (Woinarski et al. 2014). Energy requirements are likely to be high during breeding, particularly for females, and if dietary resources are inadequate during this time, females or their young may perish (Lock & Wilson 2017). Hence, the Pookila demonstrates fluctuating population dynamics where extended breeding seasons follow periods of higher rainfall (Lock & Wilson 2017; Burns 2021). Pookila abundance reduces during periods of drought due to resource shortages and likely failed litters (Lock & Wilson 2017; Burns 2021).","sortOrder":18},{"sectionNumber":"2.6.1 Diet and foraging","sectionType":"division","heading":"2.6.1 Diet and foraging","content":"### 2.6.1 Diet and foraging\n\nThe Pookila is an omnivore and feeds on a variety of seeds, leaves, fungi, and invertebrates, exhibiting seasonal and locality-dependent dietary variation (Cockburn 1980; Norton 1987; Wilson and Bradtke 1999; Braithwaite and Gullan 1978). The species plays an important ecological role in dispersing seed and fungi in scats (Wilson & Bradtke 1999; Seebeck et al. 2003; Burns et al. 2023b).","sortOrder":19},{"sectionNumber":"3","sectionType":"part","heading":"Threats","content":"# 1 Background\n\n## 1.1 Introduction\n\nThe Pookila (New Holland mouse; Pseudomys novaehollandiae) is a small, burrowing, rodent native to south eastern Australia, that is patchily distributed from Tasmania to south-east Queensland. The species has had catastrophic declines that were recorded since the discovery of new populations in the 1970s. Isolated remnant populations are at high risk of extirpation and genetic inbreeding.\n\nDistribution is fragmented across southern Queensland (Qld), the northern and central coastal areas of New South Wales (NSW), central southern coastal Victoria (Vic), and north-eastern Tasmania (Tas) including Flinders Island (Burns et al. 2023a). While the species is typically restricted to coastal areas, it occupies areas up to 300 km inland in Qld and NSW.\n\nSince the species’ description in 1843, more than a dozen known populations have been lost and many more have declined. It has not been live trapped on mainland Tas since 2010 and in Vic, seven of twelve populations have become extinct since the 1970s (Burns et al. 2023b). Encouragingly, the Pookila was rediscovered on Flinders Island, Tasmania in 2021 and was last recorded in southern Qld in 2022. It has now been translocated into Mulligans Flat Woodland Sanctuary in the Australian Capital Territory (ACT), Yiraaldiya National Park feral predator-free area (NSW), reintroduced into the Cranbourne Royal Botanic Gardens (Vic) in 2024, and has numerous ongoing and successful captive-breeding programs.\n\nEuropean colonisation in Australia exposed the Pookila to compounding threats that caused population declines across its range. These threats include the direct destruction and fragmentation of habitat due to land clearing for agriculture, urbanisation, road construction, and forest plantations, as well as predation by introduced cats and European red foxes (DEWHA 2010; Burns 2019a). Introduced rodents may outcompete the Pookila under some conditions, and both introduced rodents and cats are vectors of novel diseases. Infection may also cause mortality and/or increase the Pookila’s susceptibility to predation (for example, toxoplasmosis spread by feral cats) (Lazenby et al. 2019; Burns 2019a).\n\n## 1.2 Conservation status\n\nThe Pookila is listed as Vulnerable under the Environment Protection, Biodiversity and Conservation Act 1999 (EPBC Act), Vulnerable (ACT, Qld, NSW, VIC) and Endangered (Tas) (Table 3).\n\nThe Pookila was first listed under the EPBC Act in the Vulnerable category on 11 August 2010. It was reassessed by the Threatened Species Scientific Committee in July 2024 retaining its Vulnerable category status under Criteria 1 and 2 (DCCEEW 2024b).\n\nCriterion 1 (A4b) was met due to the severe declines in population size estimated to have occurred in Tas and Vic and the significant declines suspected to have resulted from the impacts of the 2019 and 2020 bushfires in NSW, with the Committee concluding that the Pookila was subject to a population size reduction of at least 30% (DCCEEW 2024b).\n\nCriterion 2 B2ab(i,ii,iii,iv,v) was met due to the limited Area of Occupancy (AOO), low/very low number of locations (six), and estimated continuing decline in EOO, AOO, area of habitat, number of locations and populations, as well as the number of mature individuals (DCCEEW 2024b).\n\nTable 3: Conservation status of the Pookila\n\n| Listing | Conservation Status |\n| ---------------------------------------------------------------------------- | ------------------- |\n| Environment Protection and Biodiversity Conservation Act 1999 (Commonwealth) | Vulnerable |\n| Nature Conservation Act 2014 (Australian Capital Territory) | Vulnerable |\n| Nature Conservation Act 1992 (Queensland) | Vulnerable |\n| Biodiversity Conservation Act 2016 (New South Wales) | Vulnerable |\n| Flora and Fauna Guarantee Act 1988 (Victoria) | Vulnerable |\n| Threatened Species Protection Act 1995 (Tasmania) | Endangered |\n| IUCN Red List of Threatened Species | Vulnerable |\n\n## 1.3 Cultural and community significance\n\nThe Pookila currently occurs on the Country of the Awabakal, Banbai, Barunggam, Biripi, Bunurong, Boonwurrung, Bundjalung, Burrgati, Daingatti, Darkinung, Dharawal, Dharug, Eora, Geawegal, Gumbainggir, Gunaikurnai, Kuringgai, Gunaikurnai, Ngunnawal, Palawa, Wiradjuri, Wonnarua, Worimi and Yuggera Peoples.\n\nThe species has likely been known by numerous names in different Aboriginal languages across its geographic range. The name Pookila is derived from the Ngarigo word for ‘mouse’ – bugila (Braithwaite et al. 1995; Hercus n.d.), deviating somewhat from traditional Ngarigo pronunciation (Braithwaite et al. 1995). The assignation of that name was part of an initiative by the CSIRO to provide Indigenous Australian names for Australian rodents (Braithwaite et al. 1995). This was considered a way to avoid the undesirable associations with introduced vermin often attached to the words ‘rat’ and ‘mouse’, impeding conservation of native rodents due to negative public perception (Braithwaite et al. 1995). An alternative name could have been ‘tuwaranga’ which means ‘mouse’ in palawa kani, the language of the Tasmanian Aboriginal people (Tasmanian Aboriginal Centre, 2023), however, the name Pookila has been associated with this species since 1995 and is included on the Australian Government’s SPRAT database.\n\n## 1.4 National Pookila Recovery Team\n\nA National Pookila Recovery Team (the recovery team or NPRT) was established in 2021 with the aim of improving recovery efforts for the species at a national scale. A previous Pookila Recovery Team (1995 to 2002) was largely based in Vic. The current recovery team has representation across the species’ entire distribution. Beyond the core recovery team there are three sub-groups: captive breeding, habitat and fire management, and Vic management.\n\nThe role of the recovery team is to facilitate and encourage collaborations across jurisdictions, provide advice to partners and land managers, increase engagement, improve the public perception of the Pookila as well as the responsibility of advising and coordinating the implementation of recovery actions. The recovery team also seeks to facilitate opportunities for mutual learning and skill sharing with First Nations people. The membership of the recovery team includes species experts and land managers from relevant government agencies and non-government organisations and draws from a broader network of expertise from First Nations people, researchers, and community groups.\n\n## 1.5 Key partners\n\nImplementation of recovery planning is a collaborative effort involving a range of key partners and affected interests. The groups likely to have primary interest in recovery planning for the Pookila are listed in Table 4.\n\nTable 4: Groups with primary interest in Pookila recovery\n\n| Organisations | Involvement/ interest |\n| -------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| ACT Government | Management of the land of the first Pookila translocation site. Ongoing monitoring at the site and potential future releases in the area. |\n| Australian National University (ANU) | Captive-breeding program for NSW individuals. Ongoing monitoring and management in Mulligans Flat Woodland Sanctuary and potential future reintroductions in the ACT and other jurisdictions. |\n| Aussie Ark (AA) | Maintains an insurance population. Potential development of a captive-breeding program for the Pookila and future reintroductions into cat and fox free sites in NSW. |\n| Corangamite Catchment Management Authority (CCMA) | Managers of habitat under threat from P. cinnamomi. |\n| Department of Climate Change, Energy, the Environment and Water (DCCEEW, Cwlth) | The Pookila is a priority species in the Threatened Species Action Plan 2022-2032, which provides guidance and support in the design and delivery of projects for priority species. Recovery plans inform decision making under the EPBC Act referral and assessment process. |\n| Department of Energy, Environment and Climate Action (DEECA) | Primary agency involved in threatened species management on public and private land in Vic. |\n| Department of Natural Resources and Environment (DNRE) | Primary agency involved in threatened species management on public and private land in Tas. |\n| Department of the Environment, Tourism, Science and Innovation (DETSI) | Primary agency involved in threatened species management on public and private land in Qld. |\n| Flinders Island Aboriginal Association Incorporated (FIAAI) | Traditional Owners of Flinders Island population Tas. |\n| Friends of the Prom (FOTP) | Group established to support the management of Wilsons Promontory National Park. |\n| Gippsland Water (GW) | Manages a significant Vic population. |\n| Gunaikurnai Land and Waters Aboriginal Corporation (GLaWAC) | Joint managers of Lakes National Park, and Gippsland Lakes Coastal Park, and have Registered Aboriginal Party status of Providence Ponds Flora and Fauna Reserve in Vic. |\n| Land holders/Landowners | Consultation and engagement with all types of landowners will play a role in the success in meeting the recovery plan’s actions and objectives. |\n| Local councils | Community involvement projects and cooperation with management actions on a local level. |\n| Monash University (MU) | Storage and research of Pookila cell lines and genetic tissue. Supporting research of the genetic sequencing of the Pookila and subsequent projects surrounding genetic information. |\n| Moonlit Sanctuary Wildlife Conservation Park (MSWCP) | Managing a captive population of Pookila as well as community conservation and engagement activities. |\n| Forestry agencies | Actions and management regarding forestry operations and Regional Forest Agreements. |\n| NSW Department of Climate Change, Energy, the Environment and Water (DCCEEW NSW) | Primary agency involved in threatened species management on public and private land in NSW. |\n| NSW National Parks and Wildlife Service (NPWS NSW) | Implementation of appropriate management actions within parks and reserves in NSW, including reintroduction of the Pookila to establish new populations in NPWS-managed feral predator-free areas. |\n| Parks Victoria (PV) | Implementation of appropriate management actions concerning habitat within Vic parks and reserves. |\n| National Threatened Species Institute (NTSI) Priam Psittaculture Centre (PPC) | Host of a captive breeding program founded from northern ESU individuals. |\n| Royal Botanical Gardens Victoria Cranbourne (RBGVC) | Location for reintroduction to establish a new population. |\n| Tasmanian Aboriginal Land and Sea Council Aboriginal Corporation (TALSCAC) | Traditional Owners of Northeast population Tas. |\n| Trust for Nature (TFN) | Managing properties adjacent to occupied habitat. |\n| Turner Family Foundation (TFF) | Manager of Spicers Peak Nature Refuge south-east Qld habitat. |\n| University of Canberra (UC) | Involved in captive husbandry and dietary analyses. |\n| University of Southern Queensland (USQ) | Collaborative research project with the DETSI and the Turner Family Foundation to refine knowledge of distribution and ecological requirements in south-east Queensland. |\n| University of Sydney (USyd) | Analysis and advice on captive-breeding genetics. |\n| Wadawurrung Traditional Owners Aboriginal Corporation (WTOAC) | Traditional Owners of Anglesea population Vic. |\n| Woodlands and Wetlands Trust (WWT) | Community engagement, education and volunteer programs at Mulligans Flat Woodland Sanctuary. |\n| World Wildlife Fund Australia (WWF) | Potential partner for island translocation. |\n| Zoos Victoria (ZV) | Programs underway for monitoring, management, captive breeding, genetic rescue, reintroductions, research, studbook management and species coordination, community conservation and engagement. |\n| Zoo and Aquarium Association Australasia (ZAA) | Coordination of the NPRT, initial engagement/consultation with First Nations people, and assistance with studbook management and coordination. |\n\n# 2 Species information\n\n## 2.1 Taxonomy and description\n\nPseudomys novaehollandiae, family Muridae, is endemic to Australia and was first described in 1843 as Mus novaehollandiae (Waterhouse 1843). It was later transferred to the genus Pseudomys with the new name of Pseudomys (Gyomys) novaehollandiae (Thomas 1910). There is some geographic overlap and occasional evidence of hybridisation with its sister taxon, P. delicatulus (Delicate mouse), in the north of its range (Roycroft et al. 2024). However, the two taxa were found to be both genetically and morphologically distinct (Ford 2003; Roycroft et al. 2024).\n\nThe Pookila is a nocturnal, terrestrial, burrowing rodent. It is small, with a head-body length of approximately 65–90 mm and a tail length of 80–105 mm (Menkhorst & Knight 2011). The species’ dorsal coat is sandy brown with black flecks and the ventral coat is white (Figure 1). The tail appears bicoloured with a darker dorsal side and paler pink ventral side, and the feet are pink with fine white hairs (Figure 1).\n\nIndividuals in the south of the range are heavier than those further north. Tas individuals are the heaviest, weighing 19–28 g, followed by Vic at 18–‍25 g, then Qld and NSW at 12–20 g (Hocking 1980; Burns et al. 2023a). The species is most similar in appearance to the delicate mouse, which is generally smaller and more finely featured. It is superficially similar in appearance to the invasive house mouse (Mus musculus). It is distinguishable by its larger prominent eyes, larger rounded ears, lack of ‘mousy’ odour and absence of a notch on the upper incisors (note that examining dentition is not recommended for identifying live animals) (Seebeck et al. 2003; DEWHA 2010; Burns et al. 2023a; DCCEEW 2024b). The Pookila’s body shape is also thicker and more rounded compared to the house mouse, lacking an obvious neck (Burns 2019b).\n\nFigure 1: A Pookila (Pseudomys novaehollandiae) individual showing identifying characteristics: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.\n\n![Left: An image of a captive Pookila (New Holland Mouse) on a branch.\nRight: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.](image.004.png)Source: Left © Zoos Victoria (2023), Right and insets Kiarrah Smith (2025), Coexistence Conservation Laboratory, Australian National University.\n\n## 2.2 Distribution\n\nThe Pookila has a fragmented distribution across coastal areas of central-eastern and south‑eastern Australia (Figure 2). It occurs in southern Qld, the northern and central coastal areas of NSW, central-southern coastal Vic, and north-eastern Tas including Flinders Island (Burns et al. 2023b; DCCEEW 2024b). A large portion of its distribution occurs within areas with Regional Forest Agreements (RFAs), Indigenous Protected Areas and National Parks. In Qld and NSW, the Pookila occurs in the World and National Heritage Listed Greater Blue Mountains Area and the Gondwana Rainforests of Australia. It is known only from subfossil records in National Heritage Listed areas including the Australian Alps National Parks and Reserves, Ku-ring-gai Chase National Park, Lion, Long and Spectacle Island Nature Reserves and the Great Ocean Road region (Ford 2003; Smith et al. 2023c). While most modern detections of the species exist from coastal areas, it has also been located in areas up to 300 km inland in Qld and NSW. The species’ distribution is believed to have once been continuous, forming one large population throughout these areas.\n\nThe species’ historic distribution is inferred through the presence of subfossils as well as genetic information (Ford 2003; Breed & Ford 2007; Smith et al. 2023c). Its current distribution is attributable both to longer-term climatic and landscape changes, as well as a large population decline following European colonisation (Breed & Ford 2007; Theden-Ringl et al. 2018; Roycroft et al. 2021). Despite the ACT and surrounding region not being mapped within the current Pookila distribution (Figure 2), there is subfossil evidence of historical occurrence in nearby areas with similar habitat (Ford 2003; Smith et al. 2023c).\n\nThe extent of occurrence (EOO) for the species was estimated at 326,000 km2 and the area of occupancy (AOO) for the species was estimated at 694 km2 (DCCEEW 2024b). The AOO may have further contracted since this estimate, as nearly half of the species’ habitat was affected by the 2019 and 2020 bushfires. However, this includes approximately 30% of the species’ habitat that burned at low to medium severity (Legge et al. 2022). The species is likely to survive low to medium severity fires, but high severity fires could greatly reduce survivability (Wilson et al. 2018; Burns & Phillips 2020; DCCEEW 2024b).\n\nThe Pookila is associated with early to mid-stage vegetation succession following fire events (Woinarski & Burbidge 2016). Therefore, some recovery after the 2019 and 2020 bushfires is possible due to potential development of suitable habitat post-fire (DCCEEW 2024b).\n\nFigure 2: Modelled distribution of the Pookila\n\n![The modelled distribution of the Pookila or the Pookila habitat is known or likely to occur from south east Queensland along the northern and central east coast of New South Wales to costal regions in Victoria, predominantly to the east, all the way to Flinders Island and north and eastern Tasmania. The Pookila is also likely to occur 100-300 km inland in New South Wales. Pookila or Pookila habitat may occur in close proximity to these known or likely regions. Refer to subheading 'Species distribution' for more information. ](image.005.jpeg)\n\n> Note: Source: Base map Geoscience Australia; species distribution data Species of National Environmental Significance database.\n\n> Note: Caveat: The information presented in this map has been provided by a range of groups and agencies. While every effort has been made to ensure accuracy and completeness, no guarantee is given, nor responsibility taken by the Commonwealth for errors or omissions, and the Commonwealth does not accept responsibility in respect of any information or advice given in relation to, or as a consequence of, anything contained herein.\n\n> Note: Species distribution mapping: The species distribution mapping categories are indicative only and aim to capture (a) the habitat or geographic feature that represents recent observed locations of the species (known to occur) or habitat occurring in close proximity to these locations (likely to occur); and (b) the broad environmental envelope or geographic region that encompasses all areas that could provide habitat for the species (may occur). These presence categories are created using an extensive database of species observations records, national and regional-scale environmental data, environmental modelling techniques and documented scientific research.\n\n## 2.3 Population size and trends\n\nIn 2008, the national population size of the Pookila was estimated at fewer than 10,000 individuals (Menkhorst et al. 2008; DEWHA 2010) and was reported to be 8,000 individuals in the IUCN assessment by Woinarski and Burbidge (2016). In the assessment for listing under the EPBC Act, the maximum plausible population size prior to the 2019 and 2020 bushfires was estimated (using calculations of AOO and density) to be 34,700 individuals, with the post-fire plausible population range of 13,100–26,300 (DCCEEW 2024b). There was a reduction in the number of Pookila populations located in Vic (from 12 to 5; Burns 2019a), and in Tas (from 16 to 2; (Lazenby et al. 2019). However, there is limited information from which to infer similar trends across the NSW and Qld populations (Lazenby et al. 2019; Burns 2019a; DCCEEW 2024b). A study conducted over a 32-year period near the Tomago aluminium smelter in NSW indicated an overall declining population (SMEC 2015). Shorter-term studies conducted over 5 years recorded a possible decline at Wyrrabalong National Park but no evident trend over 7 years at Munmorah State Conservation Area (DCCEEW 2024b).\n\n## 2.4 Important populations\n\nThe Evolutionarily Significant Units (ESUs) concept is used by conservationists to strategically prioritise taxa or populations of a species for recovery purposes (for example, to target for captive-breeding), when resources are limited or when the granularity of genetic diversity is not sufficiently captured in the current overarching taxonomic status (Avise 1989; Moritz 1994). There are two recognised ESUs of Pookila, separated by more than 200,000 years of divergence (Burns et al. 2023b). The Southern ESU (Vic populations) is rarer, both genetically and geographically fragmented, and more threatened than the Northern ESU (NSW and Qld populations), which is comparatively genetically healthy and not at immediate risk of extinction (Burns et al. 2023b). Tas populations were not included in the Burns et al. (2023b) analysis due to the lack of data. However, these populations are expected to place either independently or within the Southern ESU and are considered to be at extremely high risk of extinction (Burns et al. 2023b). Given the Tas populations are likely closely related to the Vic populations, until further information is available these populations should be managed as part of the Southern ESU (Burns et al. 2024).\n\nPopulations in Vic and Tas are severely fragmented and are thought to make up less than 5% of Pookila numbers nationally (DCCEEW 2024b). In Tas, the species was known from 16 separate sites in 1998, but its status on the Tas mainland is in doubt with no recent records (Lazenby et al. 2019; DNRE Tas 2023). However, it was rediscovered on Flinders Island in 2021 (DNRE Tas 2023). In Vic, extensive surveys found the species in only 5 of the 12 parks and reserves in which it had previously been recorded (Burns 2019a; Burns et al. 2023b). The remaining populations are spatially isolated and at high risk of genetic issues without assisted migrations to support genetic rescue (Burns 2019a; Burns et al. 2023b).\n\nThe connectivity and extent of populations in NSW and south-east Qld is poorly understood. Genetic evidence indicates the isolation of these populations has occurred relatively recently compared to the isolation of populations in Vic (Burns et al. 2023a; Burns et al. 2023b; DCCEEW 2024b). The extant populations are in sites that are geographically distinct and unlikely to be connected by gene flow, as were the now locally extinct populations.\n\nThere is evidence that Pookila populations are strongly affected by rainfall. In NSW, the length of the breeding season, and thus reproductive output, has been shown to increase during times of above-average rainfall (Kemper 1976, 1980). Abundance in the Anglesea (Vic) population had a strong positive relationship with cumulative monthly residual rainfall (Lock 2005). Densities of this population peaked following 4 years of above-average rainfall and declined during below-average rainfall and drought conditions (Lock 2005; Wilson et al. 2017). There are also data to suggest that population fluctuations at Wilsons Promontory (Vic) were influenced by rainfall patterns (Wilson et al. 2005). All populations are considered important for the long-term recovery and survival of the species (Table 5).\n\n \n\nTable 5: Populations of Pookila\n\n| Region | Population | Status | Year of last detection | Year of last survey | Known or expected decline? | Genetic diversity (relative heterozygosity a) |\n| ------ | --------------------------------------------------- | --------------------------------------------------------------------- | ---------------------- | ------------------- | -------------------------- | -------------------------------------------------------------------------------------------------------------------------------- |\n| ACT | Mulligans Flat Woodlands Sanctuary | Not persisting, translocated | 2021 | 2023 | Known | Translocation trials with individuals from the Australian National University (ANU) captive-breeding colony (using NSW founders) |\n| NSW | Barcoongere State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Barrington Tops State Forest | Extant | 2001 | 2025 | Expected decline | Unknown |\n| | Booti Booti National Park | Extant | 2022 | 2022 | Unknown | Unknown |\n| | Broadwater National Park | Extant | 2024 | 2025 | Known | Low |\n| | Bunjalung National Park | Unknown | 2002 | 2024 | Known | Unknown |\n| | Bylong | Unknown | Unknown | Unknown | Unknown | Moderate |\n| | Carrai National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Chaelundi State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Dalmorton State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Doyles River State Forest | Unknown | 1984 | 1984 | Unknown | Unknown |\n| | Ellis State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Gibraltar Range State Forest | Unknown | 2015 | 2015 | Unknown | Unknown |\n| | Goobang National Park | Extant | 2025 | 2025 | Unknown | Moderate |\n| | Goulburn River National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Kiwarrak State Forest | Unknown | 2019 | 2019 | Unknown | Unknown |\n| | Malara State Forest | Unknown | 1999 | Unknown | Unknown | Moderate |\n| | Malara State Forest | Unknown | 1993 | 1993 | Unknown | Unknown |\n| | Manobalai Nature Reserve | Extant | 2024 | 2024 | Unknown | Unknown |\n| | Marara State Forest | Unknown | 2014 | 2014 | Unknown | Unknown |\n| | Marengo State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Mernot State Forest | Unknown | 2008 | 2008 | Unknown | Unknown |\n| | Mount Royal National Park | Extant | 2023 | 2023 | Unknown | Unknown |\n| | Mummel Gulf National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Munmorah/Chain Valley | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Myall Lakes | Extant | 2022 | 2022 | Known | Unknown |\n| | Nymboida National Park | Unknown | 2015 | Unknown | Unknown | Moderate |\n| | Olney State Forest | Unknown | 2004 | 2004 | Unknown | Unknown |\n| | Port Stephens | Extant | 2022 | 2022 | Unknown | Moderate |\n| | Ravensworth State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Royal National Park | Extant | 2022 | 2022 | Expected decline | Unknown |\n| | Styx River State Forest | Unknown | 2000 | 2000 | Unknown | Unknown |\n| | Watchimbark Reserve | Unknown | 2011 | Unknown | Unknown | High |\n| | Werrikimbe National Park | Unknown | 1984 | 2020 | Unknown | Unknown |\n| | Wollemi National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Wyrrabalong National Park | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Yiraaldiya National Park feral predator-free area | Translocated, persistence to be determined by post-release monitoring | 2025 | 2025 | Known | Translocation trials with individuals from the captive-breeding colonies at ANU and NTSI (PPC) using NSW founders |\n| | Yuraygir National Park | Extant | 2024 | 2025 | Expected decline | Unknown |\n| QLD | Crows Nest/ Ravensbourne/Helidon Hills | Unknown | 2022 | 2025 | Known | Moderate |\n| | Emu Creek (private property) | Unknown | 2022 | 2024 | Known | Unknown |\n| | Main Range National Park/Spicers Peak Nature Refuge | Unknown | 2022 | 2025 | Known | Unknown |\n| | Numinbah Valley | Unknown | 2005 | 2005 | Unknown | Unknown |\n| TAS | Beaconsfield | Unknown | 1978 | 2002 | Known | Unknown |\n| | Flinders Island | Extant in the north of the island, undetected in the south | 2025 | 2025 | Expected decline | Unknown |\n| | Freycinet | Unknown | 2003 | 2025 | Known | Unknown |\n| | St Helens | Unknown | 2009 | 2025 | Expected decline | Unknown |\n| | Waterhouse Conservation Area | Unknown | 2010 | 2025 | Known | Unknown |\n| | Wukalina/Mt William | Unknown | 2004 | 2025 | Known | Unknown |\n| VIC | Anglesea | Locally extinct | 1999 | 2021 | Known | Very low (from historic samples) |\n| | Cranbourne | Locally extinct | 1976 | Unknown | Known | Unknown |\n| | Dutson Downs | Extant | 2024 | 2025 | Known | Low |\n| | Gippsland Lakes | Extant | 2025 | 2025 | Known | Low |\n| | Hummock Island | Locally extinct | 1977 | Unknown | Known | Unknown |\n| | Langwarrin | Locally extinct | 1984 | Unknown | Known | Unknown |\n| | Mullungdung | Locally extinct | 1976 | 2017 | Known | Unknown |\n| | Providence Ponds | Extant | 2025 | 2025 | Known | Low |\n| | Reeves Beach | Locally extinct | 1978 | 2017 | Known | Unknown |\n| | The Lakes National Park | Extant | 2025 | 2025 | Known | Low |\n| | Tyabb | Locally extinct | 1975 | Unknown | Known | Low (from historic samples) |\n| | Wilsons Promontory | Extant | 2025 | 2025 | Known | Very low |\n\n> Note: a \\- Heterozygosity measures from Burns et al. (2023b).\n\n> Note: Source: Lazenby et al. (2019), Burns (2021a), DCCEEW (2024), NSW BioNet and unpublished data from members of the recovery team and Recovery Network.\n\n## 2.5 Habitat\n\nPookila habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heathy understorey, and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022) (see Figure 3). Historically, the species has been found across a broader array of habitat types (DCCEEW 2024b), and it does not occupy all plausibly suitable habitat at present (Burns et al. 2020; Smith et al. 2023a). Sandy soils suitable for burrowing are common across most occupied habitats and soft sandy substrates, while deep topsoils are thought to be optimal (Abicair et al. 2020). Within the northern parts of the species’ range, it also occupies sites up to 300 km from the coast and in subcoastal locations at higher elevation, where soil composition differs (Martin 1995; Van Dyck & Lawrie 1997; Vernes et al. 2006; Dixon 2014; Churchill et al. 2023).\n\nThe species is often, although not exclusively, found in areas with high floristic diversity with leguminous perennials (Wilson 1991; Seebeck et al. 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; DEWHA 2010). For example, the heathland element of the species’ habitat is represented by dry heathlands that are floristically diverse and even (a high number of plants that are all relatively abundant compared to each other) that grow on sandy soils with a high percentage of bare ground (Seebeck et al. 1996).\n\nThe species has been identified as preferring early successional habitat, at some sites with peak abundance approximately 3–7 years post-fire and 7–9 years post-sand mine rehabilitation (Braithwaite & Gullan 1978; Fox & Fox 1978; Fox & Fox 1984; Twigg et al. 1989; Wilson 1991; Wilson et al. 2018). However, research in Vic suggests that the relationship between the Pookila and fire is complex, and local densities may be high in areas that have not been affected by fires for over 60 years (Burns and Phillips 2020). Similarly, surveys conducted on the NSW central coast have found high densities in long-unburnt vegetation (F Ford 2023, personal communication 8 August). Season, fire severity and extent, predation pressures, and climatic conditions following fire, all affect habitat suitability more strongly than a linear effect of time-since-fire (Burns & Phillips 2020). Furthermore, it is likely that it shelters in burrows during times of fire. This behaviour may be one of the attributes that allows the Pookila to persist in situ through some fire events (Friend 1993; Burns & Phillips 2020).\n\nThe Pookila has been considered a ‘habitat specialist’ (Braithwaite & Gullan 1978), but fossil record observations, and a strategic translocation and assessment of microhabitat use, suggests that it had broad habitat preferences (Smith et al. 2023c). Remains of the species are common in most bone deposits in south-eastern Australia that has been examined, and these deposits occur across a broad range of historical vegetation types and altitudes (Ford 2003). The species’ perceived habitat specialisation may have been further challenged by the Pookila translocation into a predator exclusion zone in the ACT as part of the Mulligans Flat-Goorooyarroo Woodland Experiment. The box gum and grassy woodlands comprising the predator exclusion area are not the type of coastal habitat often associated with contemporary records of the species (Abicair et al. 2020). However, recent detections of the species in grassy open forests and woodlands in Qld means that these non-coastal habitat types are still represented in the Pookila distribution.\n\nThe currently skewed occurrence (and perhaps specificity) of the Pookila towards coastal areas may in part be due to the interaction among threats, such as habitat clearing, altered fire regimes and predation pressures by introduced species such as cats (Felis catus) and European red foxes (Vulpes vulpes). However, it is worth noting that individuals sourced from the ANU breeding colony in the Mulligans Flat population are significantly underweight compared to their founding populations (13.6 g vs. 17.5 g) (Abicair et al. 2020), and that the population may not be self-sustaining. Together, these attributes may indicate that the habitat there may now be suboptimal for the Pookila. However, it may be important to pursue translocations in a diversity of habitats (including those that may be suboptimal) to build the species' resilience to environmental change and to restore ecological functions (Smith et al. 2023a).\n\nFigure 3: Photographs of contemporary Pookila habitat\n\n![These grouped 6 images demonstrate the diversity in the Pookila's habitat throughout the species' range. Refer to the footnote of this collective image for descriptions on habitat per image in addition to subheading 'Habitat' for more information.](image.006.png)\n\n> Note: Contemporary habitat in which the Pookila has been detected. Depicted are: a) grassy woodlands at Cunninghams Gap in Qld (photo: Ian Gynther); b) woodland with a grass and sedge understorey at Mt Hyland, NSW (photo: Anna Lloyd); c) coastal heathland at Munmorah in NSW (photo: Barbara Wilson); d) woodland with a heath understorey at Anglesea in Vic (photo: Barbara Wilson); e) coastal heathland at Loch Sport in Vic (photo: Phoebe Burns); and f) coastal heathland with a woodland overstorey on Flinders Island in Tas (photo: Rosemary Gales) (Source: DNRE Tas 2022).\n\n### 2.5.1 Habitat critical to the survival of the Pookila\n\nAll potential habitat should be considered important to the Pookila’s persistence and recovery, this includes any:\n\n Areas within the species’ modelled distribution;\n Areas where the species previously occurred but has since become locally extinct;\n Patches of suitable habitat that may not be currently occupied by the species but have the potential to be occupied by natural dispersal; and\n Areas that could be restored for reintroduction.\n\nAll areas that contain habitat that serves one or more of the following ecological functions have the potential to be habitat critical to the survival of the Pookila:\n\n Foraging habitat: vegetation with appropriate density and diversity to be able to support a population, such as native vegetation with seasonal abundance of edible flowering, fruiting, and seeding plants. Vegetation cover is important for the species to be able to forage without increased risk of predation. This habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heath understorey and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022).\n\n Breeding habitat: substrate for the species to be able to create burrows, or alternatively, ‘surface nests’ in low-lying vegetation that is suitable for successful reproduction and rearing of young. Soft sandy substrates, with deep topsoils are thought to be optimal (Abicair et al. 2020) but not exclusive. The Pookila mostly occupies coastal areas that provide this substrate. However, it also occupies sites 100–300 km from the coast and areas with higher elevation where soil composition varies. For example, the presence of sandy soils is highly likely in Goobang National Park (NSW NPWS 2001), whilst in the NSW ranges and Northern Tableland areas, soil may vary with a more dominant loamy or quartz soil base where the Pookila occurs or may be present (NSW OEH 2023. pers comm 22 November).\n\n Habitat for the long-term maintenance of the species: including areas that are not subject to prolonged drought or at risk of extreme droughts in future (such as potential climate refugia), and habitat that is maintained through periodic small-scale burns to provide appropriate foraging habitat and connectivity for the species.\n\n Habitat for the maintenance of genetic diversity: includes all current habitats where the species is extant as they contain populations necessary to achieve maximum genetic diversity and increase resilience of the species against future threats. There are two recognised ESUs: the Southern ESU (Vic populations) and the Northern ESU (NSW and Qld populations). Tas may yet be identified as a third ESU or may cluster within the Southern ESU. Therefore, habitat across the species’ range is necessary to maintain the national genetic diversity of the species.\n\n Habitat for translocation: can be identified using habitat modelling followed by ground-truthing at the local-scale and can include locations that have accounted for future changes due to climate change regimes as well as the exclusion of invasive predators.\n\nRefining what constitutes habitat critical to the survival of the Pookila is a priority action in this recovery plan. Until those key characteristics are further clarified, all areas that contain habitat that serves any of the above five ecological functions have the potential to be habitat critical to the survival of the Pookila. Habitat that supports multiple ecological functions and/or a higher number of the ecological attributes identified within the descriptions provided above, increases the likelihood that the habitat may be critical to the survival of the Pookila.\n\nGiven that the species has declined markedly in parts of its range and populations may rapidly change in size, failure to detect the species during surveys does not indicate the absence of potential habitat critical to its survival.\n\n## 2.6 Ecology\n\nThe Pookila often nests underground in communal burrow systems in well-drained soils. Individuals have also been recorded sheltering under low growing shrubs in surface nests (Kemper 1980, Lazenby et al. 2007). Burrows vary in length, depth, and complexity, ranging from shallow surface nests to intricate tunnel systems (Kemper 1981). Of three burrows that have been excavated, the largest was 5 m in length with a maximum depth of 60 cm (Kemper 1981). The species is nocturnal, and with a significant proportion of its time spent foraging above ground, it is exposed to a high risk of predation by native predators including owls and introduced predators such as the cat and the European red fox.\n\nTen individuals were radio-collared in one area of Tas for approximately 2 weeks, yielding information on nightly movements (Lazenby 1999; Lazenby et al. 2007). Over a 2 ha grid, 9 individuals moved up to 206 m in one night, and up to 37 m in one hour (Lazenby 1999; Lazenby et al. 2007). Pookila home ranges have been known to overlap (Lazenby 1999; Lazenby et al. 2007). The species’ home range is approximately 0.5–1.8 ha, increasing as habitat quality decreases (Lock 2005; Burns et al. 2023a).\n\nThe generation length of the Pookila is between one and two years (Kemper 1980). Females reach sexual maturity at approximately 13 weeks and live 2–3 years (Kemper 1976; Kemper 1980). Males reach sexual maturity at approximately 20 weeks and longevity remains unclear (Kemper 1976; Kemper 1980). During good conditions in NSW, the average litter size for the species has been previously recorded as 4.6, with a range of 1–6 young, and with females usually producing 1–2 litters per year, or 3–4 in exceptional conditions (Kemper 1976; Kemper 1980). The gestation period is around one month (32 days) and young are weaned at 3–4 weeks of age (Kemper 1976; Kemper 1980).\n\nThe breeding season varies depending on location and rainfall, with breeding occurring in NSW from late winter to early summer, while in Vic and Tas the breeding season begins in mid-spring and ceases in late summer (Kemper 1980; Burns et al. 2023b). The impact of rainfall is likely related to its influence on resource availability, which influences variation in birth rates, recruitment, and adult survival (Woinarski et al. 2014). Energy requirements are likely to be high during breeding, particularly for females, and if dietary resources are inadequate during this time, females or their young may perish (Lock & Wilson 2017). Hence, the Pookila demonstrates fluctuating population dynamics where extended breeding seasons follow periods of higher rainfall (Lock & Wilson 2017; Burns 2021). Pookila abundance reduces during periods of drought due to resource shortages and likely failed litters (Lock & Wilson 2017; Burns 2021).\n\n### 2.6.1 Diet and foraging\n\nThe Pookila is an omnivore and feeds on a variety of seeds, leaves, fungi, and invertebrates, exhibiting seasonal and locality-dependent dietary variation (Cockburn 1980; Norton 1987; Wilson and Bradtke 1999; Braithwaite and Gullan 1978). The species plays an important ecological role in dispersing seed and fungi in scats (Wilson & Bradtke 1999; Seebeck et al. 2003; Burns et al. 2023b).\n\n \n\n# 3 Threats\n\n## 3.1 Historical causes of decline\n\nThe distribution of subfossils and the genetic structure of extant populations indicate that the historic distribution of the Pookila extended more continuously throughout parts of coastal eastern Australia, from south-east Qld extending down to the northern parts of Tas and up to 300 km inland on the Australian mainland (Ford 2003; Breed and Ford 2007; DEWHA 2010; Burns et al. 2024; Smith et al. 2023c). The species has since experienced significant declines in distribution and population size and is now distributed in fragmented patches across this area (Figure 2).\n\n## 3.2 Current known and suspected threatening processes\n\nCurrent threats and their prevalence and intensity are described for the national Pookila population below. It is important to note that different populations may be experiencing different combinations of threats in varying intensities. There is significant interaction between threats.\n\n### 3.2.1 Habitat loss, disturbance and modification\n\n#### Habitat clearing\n\nLand clearing for agriculture, infrastructure, road, and housing development continues to place Pookila habitat at risk of further loss and fragmentation. As very little targeted and systematic monitoring of Pookila populations has occurred across the range in NSW (DCCEEW 2024b), the species is at particular risk of losing habitat where presence has not yet been confirmed.\n\nIn 2019, 87% of the Australian population lived within 50 km of the coast (cited in Clark et al. 2021). Coastal development is a continuing, increasing pressure with very high impacts on coastal environments (Clark et al. 2021) that is sprawling beyond established urban areas to regional coastal areas (Infrastructure Australia 2020).\n\nWhen habitat is cleared, disturbed, or modified, the risk and magnitude of impact to adjacent extant habitat patches increases, along with the interactive capability of threats (Fischer & Lindenmayer 2007; DNRE Tas 2022). For example, land clearing for agriculture can increase the carrying capacity of native and introduced herbivores within the landscape, leading to increased grazing and soil impact pressures on adjacent habitat. This is exacerbated when fire results in a further increase in grazing pressure and movement as herbivores actively seek to consume newly emerging vegetative shoots (DAWE 2022). With an increasing trend of habitat loss due to land clearing, the likelihood that fragmentation has occurred and will continue to occur is very high.\n\n#### Habitat modification due to Phytophthora cinnamomi infestation\n\nPhytophthora dieback – caused by the root-rot pathogen Phytophthora cinnamomi – is listed as a Key Threatening Process under the EPBC Act (DOEE 2018) and has been observed across the national range of the Pookila (Liu et al. 2013; Wilson et al. 2015; Burns et al. 2020; DCCEEW 2024b). The pathogen is spread through the movement of soil – including on footwear, tools, vehicles and machinery – and in water (DOEE 2018). While the impacts of P. cinnamomi on Pookila populations and habitat have not been investigated directly, the species is likely to be at significant risk. A range of plant species within Pookila habitat have been identified as susceptible to infection by P. cinnamomi (see McDougall 2005).\n\nPhytophthora dieback causes the loss of plant species richness (including keystone species), reduction of structural complexity (Wilson et al. 1994; Garkaklis et al. 2004; Laidlaw & Wilson 2006; Annett 2008; Armistead 2008; Dundas et al. 2013, 2016; Davis et al. 2014; Wilson et al. 2020), loss of food resources (for example, seeds, leaves, nectar, pollen) and suitable refuges from weather and predation (Laidlaw & Wilson 2006). The impacts of Phytophthora dieback can also extend to invertebrate communities and thus impact insectivorous fauna (Cahill et al. 2008), reducing availability of this food resource.\n\nSignificant negative impacts from Phytophthora dieback on small mammals and their habitat has been previously documented (Newell & Wilson 1993; Lilith 2002; Laidlaw & Wilson 2006; Swinburn et al. 2007; Annett 2008; Armistead 2008; Dundas et al. 2013; 2016), and long-term studies have identified localised extinctions of species and mammal community declines (Zhuang-Griffin 2015; Wilson et al. 2017; 2020).\n\nPriority Protection Areas for Phytophthora dieback management have been identified for the Otway Ranges, which was informed by habitats occupied by EPBC Act listed species, including the Pookila (Walshe et al. 2021; 2022; Wilson & Garkaklis 2023a; 2023b).\n\n#### Habitat modification due to native and introduced weeds\n\nChanging grazing and fire regimes have led to habitat loss by encroachment of coast tea tree (Leptospermum laevigatum) in some Pookila habitat (Burns 2019a; Seebeck et al. 2003). This is particularly evident at Wilsons Promontory National Park and in other coastal Vic sites along Ninety Mile Beach where dune succession has given rise to dense thickets of coast tea tree, such as at Reeves Beech. Weed species are of growing concern in other Pookila habitat, for example, in Lakes National Park. Invasive weeds include Asparagus asparagoides (bridal creeper) and Lycium ferocissimum (boxthorn).\n\n### 3.2.2 Fire regimes that cause declines in biodiversity\n\n‘Fire regimes that cause declines in biodiversity’ is listed as a Key Threatening Process under the EPBC Act (DAWE 2022). Altered fire regimes, such as those resulting from changes to seasonality, frequency and severity of fire weather conditions, are causing declines and loss of biodiversity and ecosystem function (DAWE 2022). The Pookila and the habitat the species relies on have a complicated relationship with fire.\n\nUnder certain circumstances, the species can peak in abundance in the years following a fire as resources become more readily available due to regeneration of vegetation, with such areas described as ‘early successional’ (Fox & McKay 1981). Furthermore, the Pookila has been reported to have relatively high survival during and immediately following bushfires, which is also presumably attributed to its ability to shelter in burrows (Friend 1993; Burns & Phillips 2020). However, this response appears to be more common following small-scale, patchy, low-severity burns and is likely reliant on above-average rainfall in the months and years following the fire (Burns & Phillips 2020).\n\nLarge-scale, high-severity fires have the capacity to extirpate entire populations, as occurred in 1983 on Ash Wednesday at Anglesea (Wilson et al. 2018). Fires being too frequent, too large, too severe, or poorly timed may lead to direct loss of individuals and/or sudden or gradual loss of habitat through reduction of habitat patches of suitable successional age, size, distribution, and plant diversity that the Pookila requires for availability of food and shelter from predators.\n\nSimilarly, prolonged exclusion or poorly timed application of fire may encourage coast tea tree encroachment and loss of favourable food and habitat species. Also, introduced predators may target recently burnt areas, compounding the increased risk of predation from loss of vegetative cover (Hradsky et al. 2017). Planned burning may also threaten the species if conducted during or immediately prior to drought, at too high an intensity, or across a widespread area (Wilson et al. 2018).\n\nThe increasing scale, severity, and frequency of fires under a changing climate pose a risk to the Pookila and its habitat. In 2019 and 2020, catastrophic bushfires impacted a large area of eastern Australia with varying severity (Legge et al. 2021). It was estimated that over 45% of the Pookila distribution was burnt, with a third of this area burning with high severity (Legge et al. 2021). The overall population of Pookila was predicted to have declined by 24% by 2021 (one year after fire), with the species suffering an estimated 75% decrease in abundance in severely burnt areas in that same period (Legge et al. 2021).\n\n### 3.2.3 Climate change: Increasing number of hot days, prolonged drought and changes to precipitation patterns\n\nIncreasing frequency, intensity, and duration of drought across much of the species’ range (CSIRO & BoM 2020) may lead to further catastrophic declines and local extinctions. Extreme heat events may exacerbate the effects of drought and could lead to mass mortality events as individuals dehydrate and fail to thermoregulate but this would require further research to confirm whether behaviours would mitigate this threat. The projections for the NSW climate by 2050 include an increase in minimum and maximum temperatures, an increase (double) in the number of hot days per year (days greater than 35°C), and longer fire seasons, with around a 40% increase in ‘very high’ fire danger days (CSIRO & BoM 2021).\n\nExtended periods of low rainfall reduce food availability for Pookila. Recent periods of drought have caused demonstrable crashes in Pookila populations. The local extinction of the Anglesea population is in part attributed to the Millennium Drought (Lock & Wilson 2017), and the species is yet to recover from the 2019 Gippsland drought (Burns 2021).\n\nFire–drought interactions may have a significant impact on the Pookila with potential compounding effects. This is evident from the 2019 and 2020 bushfires, which followed a period of significant drought in 2017 to 2019 (BoM 2020; DPIE 2020).\n\n### 3.2.4 Invasive fauna species\n\n#### Predation by cats and European red foxes\n\nFeral predators can have rapid and catastrophic impacts on populations of threatened rodents (Tuft et al. 2021). While there is limited direct evidence of the consumption of Pookila by feral predators (records in dog scats; Hoskins 1997; Davis et al. 2015; Lock & Wilson 2017; MidCoast Council NSW 2024. pers comm 13 February), it is recognised in the Threat Abatement Plan for Predation by Feral Cats 2024 that the Pookila is likely highly susceptible to predation by cats (and/or in some cases by foxes) (DCCEEW 2024a) and that cats and foxes can significantly impact population numbers (Seebeck et al. 2003; Burns 2019a).\n\nLike many other Pseudomys species, Pookila are docile and display poor awareness of exotic predators (P. Burns, pers. obs.). In such species, selective predation can occur (as reported, for example, on the broad-toothed rat, Mastacomys fuscus; Green 2002), leading to a disproportionate impact of predation on the vulnerable species even when co-occurring species (for example, the house mouse) are at greater abundance. Furthermore, communal burrowing and small local population densities mean that predation events at burrow entrances can cause rapid local declines (Burns 2019a).\n\nThe relative impact of predation by feral cats and foxes on the Pookila is not yet understood. Based on factors such as small body size, lower breeding output, and occupation of open habitat, some researchers have identified the Pookila as having a high threat rating from feral cat predation (Dickman 1996; Stobo-Wilson et al. 2021). Radford et al. (2018) reported the cat and fox threat to Pookila to be approximately the same based on expert opinion, although with weak evidence. While foxes are currently absent from the Tas mainland and Flinders Island (DNRE Tas 2022), and fox control is in place in other parts of the species’ range, effective cat control in open, non-insular systems beyond the local scale is not yet possible and potentially counterproductive (Lazenby et al. 2015; Lazenby et al. 2021; see Predator management).\n\nThe density and predatory impact of cats and foxes can be altered by competitive interactions with other predators. For example, reduced numbers of foxes via fox control can increase cat density, alter cat behaviour, and ultimately increase the impact of predation by feral cats (Risbey et al. 2000; Rees et al. 2023). Tasmanian devils have been shown to alter the behaviour and abundance of feral cats (Hollings et al. 2014; Lazenby & Dickman 2013), and cats have a much higher predatory impact on small mammals compared to devils (Hollings et al. 2014). The decline of devils caused by Devil Facial Tumour Disease (DFTD) is likely to have increased the predatory impacts of feral cats in Pookila habitat on the Tas mainland.\n\n#### Impact of hooved feral animals including deer, goats, horses, and pigs\n\nThe negative impacts of hooved feral animals on vegetation health and soil erosion are well documented (Keith & Pellow 2005; Bengsen et al. 2017; Wills et al. 2023; Burns et al. 2021). This is also recognised within the Draft threat abatement plan for competition and land degradation by unmanaged goats (DCCEEW 2024c), the Threat abatement plan for predation, habitat degradation, competition and disease transmission by feral pigs (Sus scrofa) (DOEE 2017), and the National feral deer action plan (Government of South Australia 2023). Where these species co-occur with the Pookila, there are risks of habitat degradation through trampling, wallowing, and overgrazing (Burns 2021). Overgrazing pressures may increase in post-fire environments, particularly in heathy vegetation types preferred by the Pookila (Heaton et al. 2022), compounding post-fire food shortages and predation risk, while impeding regrowth of suitable habitat and food species (DAWE 2022). The effects of hooved feral animal control programs on Pookila occurrence and abundance have not been directly studied, but based on vegetation response studies, control programs are expected to benefit the Pookila and its habitat, particularly around fire events.\n\n#### Competition/predation from introduced rodents\n\nWhile experimental trials have shown the Pookila to be competitively superior to the invasive house mouse (Fox & Twigg 1991; Haering & Fox 1997), there is growing evidence that house mouse plagues may supress Pookila recovery following periods of drought where resources are subsequently scarce (Burns 2021). Rattus rattus and R. norvegicus (black and brown rats, respectively) may also out-compete the Pookila for resources and prey on adults and juveniles, although this requires further research.\n\n#### Baiting of rabbits and invasive rodents\n\nThe susceptibility of the Pookila to 1080, Pindone, and rodenticide has not been tested. However, such poisons are lethal at low doses to some other Pseudomys species and are expected to be lethal to the Pookila (McIlroy 1982; Eisler 1995; Fisher 1999). Running a lethal dose trial on the Pookila is not appropriate given the species’ threatened status and ethical considerations, so it is safest to apply the Precautionary Principle and assume the species is likely susceptible to 1080, Pindone and rodenticide. Rabbit poison baiting is usually conducted with a carrot or oat bait, both foods readily consumed by the Pookila (Prosser et al. 2007), and it is possible that rabbit-baiting programs have contributed to the decline of the species in some areas (Fisher 1999; Burns 2019a). When finding the balance between protecting Pookila habitat by controlling rabbits and potentially poisoning individual Pookila, it is recommended to entirely avoid poison baiting in areas where Pookila presence is known. This is particularly important for at-risk populations; however, it may also apply to locations where Pookila presence is uncertain but these areas are adjacent to suitable habitat and/or a known population (within a 5 km radius). Rabbit and invasive rodent population control measures should use alternative methods. Control measures should be carefully considered depending on the area of control and the potential side effects for native species. Currently, this threat is considered to be high in Wilsons Promontory in Vic, not applicable in Qld or Tas, and unknown in NSW (DNRE Tas 2023).\n\n### 3.2.5 Small and fragmented populations: Low genetic diversity and inbreeding\n\nHistorical and contemporary habitat degradation and destruction, and the species’ very low dispersal ability have left the Pookila highly fragmented across its distribution, with some populations genetically isolated. This severe fragmentation leaves populations at higher risk of extinction, as a single catastrophic event can more easily impact the total area occupied by a population (Garner et al. 2020).\n\nFurthermore, genetically isolated populations are at higher risk of inbreeding depression. Inbreeding reduces genetic diversity, which reduces adaptability to changing environments and anthropogenic impacts. Inbreeding depression also increases species’ vulnerability to disease while also allowing potentially detrimental traits to surface in the population (Keller 2002). Furthermore, inbreeding depression can cause reduced fecundity and further genetic health issues (for example, chromosomal abnormalities) (Keller 2002).\n\nVic populations (Southern ESU) of the Pookila have relatively low heterozygosity (genetic diversity) and are at risk of inbreeding, with no natural gene flow possible between localities given lack of dispersal (Burns et al. 2023b). Therefore, the Southern ESU is considered an ‘at-risk ESU’. Populations across NSW retain greater heterozygosity and are genetically distinct from those in Vic, although animals in Broadwater National Park appear to have undergone a genetic bottleneck (Burns et al. 2023b). At the time of publishing this recovery plan, two historic mainland Tas samples and a small number of contemporary samples from Flinders Island have been sequenced and are being analysed. Additional Tas samples are being sought through survey work. Genetic mixing within Vic populations has been recommended and is being pursued, whereas no immediate action is required in the Northern ESU (NSW and Qld) (Burns et al. 2023b; Burns et al. 2024) since it is more genetically healthy than the Southern ESU. The recovery team has developed a National Pookila genetic management strategy (Zoos Victoria) and will continue to update it when new information is forthcoming.\n\n#### Disease\n\nThere is limited understanding of the potential prevalence or impact of disease on the Pookila. In general, screening of native rodents for pathogens has been very limited and little information is available on symptomology (Thompson et al. 2010). However, feral cats and invasive rodents carry diseases that could impact the Pookila. In particular, Toxoplasma gondii, a virulent parasite spread by cats, is likely present through much of the species’ former range (Dickson 2018). Previous experimental trials in other Pseudomys species demonstrated rapid, high mortality rates after infection with T. gondii, hence this parasite is likely to also increase the mortality of the Pookila (Burns 2019a).\n\n## 3.3 Threat prioritisation\n\nEach of the threats outlined in this recovery plan has been assessed to determine the risk posed across the national distribution of the Pookila using a risk matrix (see Table 6). The risk matrix considers the likelihood of an incident occurring and the consequences of that incident and uses a qualitative assessment drawing on both peer-reviewed literature and expert opinion.\n\nTable 6: Risk matrix\n\n| Consequence (x) / Likelihood (y) | Not significant | Minor | Moderate | Major | Catastrophic |\n| -------------------------------- | --------------- | ------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------- |\n| Almost certain | Low risk | Moderate risk | Very high riskPredation by European red foxes | Very high riskHabitat clearing (past/present)Fire regimes that cause declines in biodiversityIncreasing number of hot days, prolonged drought and changes to precipitation patternsPredation by cats | Very high risk |\n| Likely | Low risk | Moderate riskImpact of hooved feral animals including deer, goats, horses, and pigs | High riskHabitat modification due to Phytophthora cinnamomi infestationHabitat modification due to native and introduced weeds | Very high riskHabitat clearing (future) | Very high risk |\n| Possible | Low risk | Moderate riskCompetition/predation from introduced rodentsBaiting of rabbits and invasive rodents | High riskLow genetic diversity and inbreedingDisease | Very high risk | Very high risk |\n| Unlikely | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n| Unknown | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n\nRisk matrix key\n\n| Low risk | Moderate risk | High risk | Very-high risk |\n| -------- | ------------- | --------- | -------------- |\n\nDefinition of likelihood and consequence categories\n\nConsequence is a measure of the severity of the impact of the threat on the survival of the species.\n\n- Catastrophic – affecting survival, reproduction or essential movement of individuals in >80% of the population or across >80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Major – affecting survival, reproduction or essential movement of individuals in 50-80 % of the population or across 50-80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Moderate – affecting survival, reproduction or essential movement of individuals in 20-50% of the population or across 20-50% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Minor – affecting survival, reproduction or essential movement of individuals in 5-30% of the population or across 5-30% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Not significant – affecting survival, reproduction or essential movement of individuals in <5% of the population or across <5% of the distribution within the next 3 generations or 10 years, whichever is longer.\n\nLikelihood is a measure of how regularly the threat impacts the species.\n\n- Almost certain – >90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Likely – 66-90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Possible – 33-66% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unlikely – <33% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unknown – probability that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer, could be 0-100%.","sortOrder":20},{"sectionNumber":"3.1","sectionType":"part","heading":"Historical causes of decline","content":"## 3.1 Historical causes of decline\n\nThe distribution of subfossils and the genetic structure of extant populations indicate that the historic distribution of the Pookila extended more continuously throughout parts of coastal eastern Australia, from south-east Qld extending down to the northern parts of Tas and up to 300 km inland on the Australian mainland (Ford 2003; Breed and Ford 2007; DEWHA 2010; Burns et al. 2024; Smith et al. 2023c). The species has since experienced significant declines in distribution and population size and is now distributed in fragmented patches across this area (Figure 2).","sortOrder":21},{"sectionNumber":"3.2","sectionType":"part","heading":"Current known and suspected threatening processes","content":"## 3.2 Current known and suspected threatening processes\n\nCurrent threats and their prevalence and intensity are described for the national Pookila population below. It is important to note that different populations may be experiencing different combinations of threats in varying intensities. There is significant interaction between threats.","sortOrder":22},{"sectionNumber":"3.2.1 Habitat loss, disturbance and modi","sectionType":"division","heading":"3.2.1 Habitat loss, disturbance and modification","content":"# 1 Background\n\n## 1.1 Introduction\n\nThe Pookila (New Holland mouse; Pseudomys novaehollandiae) is a small, burrowing, rodent native to south eastern Australia, that is patchily distributed from Tasmania to south-east Queensland. The species has had catastrophic declines that were recorded since the discovery of new populations in the 1970s. Isolated remnant populations are at high risk of extirpation and genetic inbreeding.\n\nDistribution is fragmented across southern Queensland (Qld), the northern and central coastal areas of New South Wales (NSW), central southern coastal Victoria (Vic), and north-eastern Tasmania (Tas) including Flinders Island (Burns et al. 2023a). While the species is typically restricted to coastal areas, it occupies areas up to 300 km inland in Qld and NSW.\n\nSince the species’ description in 1843, more than a dozen known populations have been lost and many more have declined. It has not been live trapped on mainland Tas since 2010 and in Vic, seven of twelve populations have become extinct since the 1970s (Burns et al. 2023b). Encouragingly, the Pookila was rediscovered on Flinders Island, Tasmania in 2021 and was last recorded in southern Qld in 2022. It has now been translocated into Mulligans Flat Woodland Sanctuary in the Australian Capital Territory (ACT), Yiraaldiya National Park feral predator-free area (NSW), reintroduced into the Cranbourne Royal Botanic Gardens (Vic) in 2024, and has numerous ongoing and successful captive-breeding programs.\n\nEuropean colonisation in Australia exposed the Pookila to compounding threats that caused population declines across its range. These threats include the direct destruction and fragmentation of habitat due to land clearing for agriculture, urbanisation, road construction, and forest plantations, as well as predation by introduced cats and European red foxes (DEWHA 2010; Burns 2019a). Introduced rodents may outcompete the Pookila under some conditions, and both introduced rodents and cats are vectors of novel diseases. Infection may also cause mortality and/or increase the Pookila’s susceptibility to predation (for example, toxoplasmosis spread by feral cats) (Lazenby et al. 2019; Burns 2019a).\n\n## 1.2 Conservation status\n\nThe Pookila is listed as Vulnerable under the Environment Protection, Biodiversity and Conservation Act 1999 (EPBC Act), Vulnerable (ACT, Qld, NSW, VIC) and Endangered (Tas) (Table 3).\n\nThe Pookila was first listed under the EPBC Act in the Vulnerable category on 11 August 2010. It was reassessed by the Threatened Species Scientific Committee in July 2024 retaining its Vulnerable category status under Criteria 1 and 2 (DCCEEW 2024b).\n\nCriterion 1 (A4b) was met due to the severe declines in population size estimated to have occurred in Tas and Vic and the significant declines suspected to have resulted from the impacts of the 2019 and 2020 bushfires in NSW, with the Committee concluding that the Pookila was subject to a population size reduction of at least 30% (DCCEEW 2024b).\n\nCriterion 2 B2ab(i,ii,iii,iv,v) was met due to the limited Area of Occupancy (AOO), low/very low number of locations (six), and estimated continuing decline in EOO, AOO, area of habitat, number of locations and populations, as well as the number of mature individuals (DCCEEW 2024b).\n\nTable 3: Conservation status of the Pookila\n\n| Listing | Conservation Status |\n| ---------------------------------------------------------------------------- | ------------------- |\n| Environment Protection and Biodiversity Conservation Act 1999 (Commonwealth) | Vulnerable |\n| Nature Conservation Act 2014 (Australian Capital Territory) | Vulnerable |\n| Nature Conservation Act 1992 (Queensland) | Vulnerable |\n| Biodiversity Conservation Act 2016 (New South Wales) | Vulnerable |\n| Flora and Fauna Guarantee Act 1988 (Victoria) | Vulnerable |\n| Threatened Species Protection Act 1995 (Tasmania) | Endangered |\n| IUCN Red List of Threatened Species | Vulnerable |\n\n## 1.3 Cultural and community significance\n\nThe Pookila currently occurs on the Country of the Awabakal, Banbai, Barunggam, Biripi, Bunurong, Boonwurrung, Bundjalung, Burrgati, Daingatti, Darkinung, Dharawal, Dharug, Eora, Geawegal, Gumbainggir, Gunaikurnai, Kuringgai, Gunaikurnai, Ngunnawal, Palawa, Wiradjuri, Wonnarua, Worimi and Yuggera Peoples.\n\nThe species has likely been known by numerous names in different Aboriginal languages across its geographic range. The name Pookila is derived from the Ngarigo word for ‘mouse’ – bugila (Braithwaite et al. 1995; Hercus n.d.), deviating somewhat from traditional Ngarigo pronunciation (Braithwaite et al. 1995). The assignation of that name was part of an initiative by the CSIRO to provide Indigenous Australian names for Australian rodents (Braithwaite et al. 1995). This was considered a way to avoid the undesirable associations with introduced vermin often attached to the words ‘rat’ and ‘mouse’, impeding conservation of native rodents due to negative public perception (Braithwaite et al. 1995). An alternative name could have been ‘tuwaranga’ which means ‘mouse’ in palawa kani, the language of the Tasmanian Aboriginal people (Tasmanian Aboriginal Centre, 2023), however, the name Pookila has been associated with this species since 1995 and is included on the Australian Government’s SPRAT database.\n\n## 1.4 National Pookila Recovery Team\n\nA National Pookila Recovery Team (the recovery team or NPRT) was established in 2021 with the aim of improving recovery efforts for the species at a national scale. A previous Pookila Recovery Team (1995 to 2002) was largely based in Vic. The current recovery team has representation across the species’ entire distribution. Beyond the core recovery team there are three sub-groups: captive breeding, habitat and fire management, and Vic management.\n\nThe role of the recovery team is to facilitate and encourage collaborations across jurisdictions, provide advice to partners and land managers, increase engagement, improve the public perception of the Pookila as well as the responsibility of advising and coordinating the implementation of recovery actions. The recovery team also seeks to facilitate opportunities for mutual learning and skill sharing with First Nations people. The membership of the recovery team includes species experts and land managers from relevant government agencies and non-government organisations and draws from a broader network of expertise from First Nations people, researchers, and community groups.\n\n## 1.5 Key partners\n\nImplementation of recovery planning is a collaborative effort involving a range of key partners and affected interests. The groups likely to have primary interest in recovery planning for the Pookila are listed in Table 4.\n\nTable 4: Groups with primary interest in Pookila recovery\n\n| Organisations | Involvement/ interest |\n| -------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| ACT Government | Management of the land of the first Pookila translocation site. Ongoing monitoring at the site and potential future releases in the area. |\n| Australian National University (ANU) | Captive-breeding program for NSW individuals. Ongoing monitoring and management in Mulligans Flat Woodland Sanctuary and potential future reintroductions in the ACT and other jurisdictions. |\n| Aussie Ark (AA) | Maintains an insurance population. Potential development of a captive-breeding program for the Pookila and future reintroductions into cat and fox free sites in NSW. |\n| Corangamite Catchment Management Authority (CCMA) | Managers of habitat under threat from P. cinnamomi. |\n| Department of Climate Change, Energy, the Environment and Water (DCCEEW, Cwlth) | The Pookila is a priority species in the Threatened Species Action Plan 2022-2032, which provides guidance and support in the design and delivery of projects for priority species. Recovery plans inform decision making under the EPBC Act referral and assessment process. |\n| Department of Energy, Environment and Climate Action (DEECA) | Primary agency involved in threatened species management on public and private land in Vic. |\n| Department of Natural Resources and Environment (DNRE) | Primary agency involved in threatened species management on public and private land in Tas. |\n| Department of the Environment, Tourism, Science and Innovation (DETSI) | Primary agency involved in threatened species management on public and private land in Qld. |\n| Flinders Island Aboriginal Association Incorporated (FIAAI) | Traditional Owners of Flinders Island population Tas. |\n| Friends of the Prom (FOTP) | Group established to support the management of Wilsons Promontory National Park. |\n| Gippsland Water (GW) | Manages a significant Vic population. |\n| Gunaikurnai Land and Waters Aboriginal Corporation (GLaWAC) | Joint managers of Lakes National Park, and Gippsland Lakes Coastal Park, and have Registered Aboriginal Party status of Providence Ponds Flora and Fauna Reserve in Vic. |\n| Land holders/Landowners | Consultation and engagement with all types of landowners will play a role in the success in meeting the recovery plan’s actions and objectives. |\n| Local councils | Community involvement projects and cooperation with management actions on a local level. |\n| Monash University (MU) | Storage and research of Pookila cell lines and genetic tissue. Supporting research of the genetic sequencing of the Pookila and subsequent projects surrounding genetic information. |\n| Moonlit Sanctuary Wildlife Conservation Park (MSWCP) | Managing a captive population of Pookila as well as community conservation and engagement activities. |\n| Forestry agencies | Actions and management regarding forestry operations and Regional Forest Agreements. |\n| NSW Department of Climate Change, Energy, the Environment and Water (DCCEEW NSW) | Primary agency involved in threatened species management on public and private land in NSW. |\n| NSW National Parks and Wildlife Service (NPWS NSW) | Implementation of appropriate management actions within parks and reserves in NSW, including reintroduction of the Pookila to establish new populations in NPWS-managed feral predator-free areas. |\n| Parks Victoria (PV) | Implementation of appropriate management actions concerning habitat within Vic parks and reserves. |\n| National Threatened Species Institute (NTSI) Priam Psittaculture Centre (PPC) | Host of a captive breeding program founded from northern ESU individuals. |\n| Royal Botanical Gardens Victoria Cranbourne (RBGVC) | Location for reintroduction to establish a new population. |\n| Tasmanian Aboriginal Land and Sea Council Aboriginal Corporation (TALSCAC) | Traditional Owners of Northeast population Tas. |\n| Trust for Nature (TFN) | Managing properties adjacent to occupied habitat. |\n| Turner Family Foundation (TFF) | Manager of Spicers Peak Nature Refuge south-east Qld habitat. |\n| University of Canberra (UC) | Involved in captive husbandry and dietary analyses. |\n| University of Southern Queensland (USQ) | Collaborative research project with the DETSI and the Turner Family Foundation to refine knowledge of distribution and ecological requirements in south-east Queensland. |\n| University of Sydney (USyd) | Analysis and advice on captive-breeding genetics. |\n| Wadawurrung Traditional Owners Aboriginal Corporation (WTOAC) | Traditional Owners of Anglesea population Vic. |\n| Woodlands and Wetlands Trust (WWT) | Community engagement, education and volunteer programs at Mulligans Flat Woodland Sanctuary. |\n| World Wildlife Fund Australia (WWF) | Potential partner for island translocation. |\n| Zoos Victoria (ZV) | Programs underway for monitoring, management, captive breeding, genetic rescue, reintroductions, research, studbook management and species coordination, community conservation and engagement. |\n| Zoo and Aquarium Association Australasia (ZAA) | Coordination of the NPRT, initial engagement/consultation with First Nations people, and assistance with studbook management and coordination. |\n\n# 2 Species information\n\n## 2.1 Taxonomy and description\n\nPseudomys novaehollandiae, family Muridae, is endemic to Australia and was first described in 1843 as Mus novaehollandiae (Waterhouse 1843). It was later transferred to the genus Pseudomys with the new name of Pseudomys (Gyomys) novaehollandiae (Thomas 1910). There is some geographic overlap and occasional evidence of hybridisation with its sister taxon, P. delicatulus (Delicate mouse), in the north of its range (Roycroft et al. 2024). However, the two taxa were found to be both genetically and morphologically distinct (Ford 2003; Roycroft et al. 2024).\n\nThe Pookila is a nocturnal, terrestrial, burrowing rodent. It is small, with a head-body length of approximately 65–90 mm and a tail length of 80–105 mm (Menkhorst & Knight 2011). The species’ dorsal coat is sandy brown with black flecks and the ventral coat is white (Figure 1). The tail appears bicoloured with a darker dorsal side and paler pink ventral side, and the feet are pink with fine white hairs (Figure 1).\n\nIndividuals in the south of the range are heavier than those further north. Tas individuals are the heaviest, weighing 19–28 g, followed by Vic at 18–‍25 g, then Qld and NSW at 12–20 g (Hocking 1980; Burns et al. 2023a). The species is most similar in appearance to the delicate mouse, which is generally smaller and more finely featured. It is superficially similar in appearance to the invasive house mouse (Mus musculus). It is distinguishable by its larger prominent eyes, larger rounded ears, lack of ‘mousy’ odour and absence of a notch on the upper incisors (note that examining dentition is not recommended for identifying live animals) (Seebeck et al. 2003; DEWHA 2010; Burns et al. 2023a; DCCEEW 2024b). The Pookila’s body shape is also thicker and more rounded compared to the house mouse, lacking an obvious neck (Burns 2019b).\n\nFigure 1: A Pookila (Pseudomys novaehollandiae) individual showing identifying characteristics: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.\n\n![Left: An image of a captive Pookila (New Holland Mouse) on a branch.\nRight: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.](image.004.png)Source: Left © Zoos Victoria (2023), Right and insets Kiarrah Smith (2025), Coexistence Conservation Laboratory, Australian National University.\n\n## 2.2 Distribution\n\nThe Pookila has a fragmented distribution across coastal areas of central-eastern and south‑eastern Australia (Figure 2). It occurs in southern Qld, the northern and central coastal areas of NSW, central-southern coastal Vic, and north-eastern Tas including Flinders Island (Burns et al. 2023b; DCCEEW 2024b). A large portion of its distribution occurs within areas with Regional Forest Agreements (RFAs), Indigenous Protected Areas and National Parks. In Qld and NSW, the Pookila occurs in the World and National Heritage Listed Greater Blue Mountains Area and the Gondwana Rainforests of Australia. It is known only from subfossil records in National Heritage Listed areas including the Australian Alps National Parks and Reserves, Ku-ring-gai Chase National Park, Lion, Long and Spectacle Island Nature Reserves and the Great Ocean Road region (Ford 2003; Smith et al. 2023c). While most modern detections of the species exist from coastal areas, it has also been located in areas up to 300 km inland in Qld and NSW. The species’ distribution is believed to have once been continuous, forming one large population throughout these areas.\n\nThe species’ historic distribution is inferred through the presence of subfossils as well as genetic information (Ford 2003; Breed & Ford 2007; Smith et al. 2023c). Its current distribution is attributable both to longer-term climatic and landscape changes, as well as a large population decline following European colonisation (Breed & Ford 2007; Theden-Ringl et al. 2018; Roycroft et al. 2021). Despite the ACT and surrounding region not being mapped within the current Pookila distribution (Figure 2), there is subfossil evidence of historical occurrence in nearby areas with similar habitat (Ford 2003; Smith et al. 2023c).\n\nThe extent of occurrence (EOO) for the species was estimated at 326,000 km2 and the area of occupancy (AOO) for the species was estimated at 694 km2 (DCCEEW 2024b). The AOO may have further contracted since this estimate, as nearly half of the species’ habitat was affected by the 2019 and 2020 bushfires. However, this includes approximately 30% of the species’ habitat that burned at low to medium severity (Legge et al. 2022). The species is likely to survive low to medium severity fires, but high severity fires could greatly reduce survivability (Wilson et al. 2018; Burns & Phillips 2020; DCCEEW 2024b).\n\nThe Pookila is associated with early to mid-stage vegetation succession following fire events (Woinarski & Burbidge 2016). Therefore, some recovery after the 2019 and 2020 bushfires is possible due to potential development of suitable habitat post-fire (DCCEEW 2024b).\n\nFigure 2: Modelled distribution of the Pookila\n\n![The modelled distribution of the Pookila or the Pookila habitat is known or likely to occur from south east Queensland along the northern and central east coast of New South Wales to costal regions in Victoria, predominantly to the east, all the way to Flinders Island and north and eastern Tasmania. The Pookila is also likely to occur 100-300 km inland in New South Wales. Pookila or Pookila habitat may occur in close proximity to these known or likely regions. Refer to subheading 'Species distribution' for more information. ](image.005.jpeg)\n\n> Note: Source: Base map Geoscience Australia; species distribution data Species of National Environmental Significance database.\n\n> Note: Caveat: The information presented in this map has been provided by a range of groups and agencies. While every effort has been made to ensure accuracy and completeness, no guarantee is given, nor responsibility taken by the Commonwealth for errors or omissions, and the Commonwealth does not accept responsibility in respect of any information or advice given in relation to, or as a consequence of, anything contained herein.\n\n> Note: Species distribution mapping: The species distribution mapping categories are indicative only and aim to capture (a) the habitat or geographic feature that represents recent observed locations of the species (known to occur) or habitat occurring in close proximity to these locations (likely to occur); and (b) the broad environmental envelope or geographic region that encompasses all areas that could provide habitat for the species (may occur). These presence categories are created using an extensive database of species observations records, national and regional-scale environmental data, environmental modelling techniques and documented scientific research.\n\n## 2.3 Population size and trends\n\nIn 2008, the national population size of the Pookila was estimated at fewer than 10,000 individuals (Menkhorst et al. 2008; DEWHA 2010) and was reported to be 8,000 individuals in the IUCN assessment by Woinarski and Burbidge (2016). In the assessment for listing under the EPBC Act, the maximum plausible population size prior to the 2019 and 2020 bushfires was estimated (using calculations of AOO and density) to be 34,700 individuals, with the post-fire plausible population range of 13,100–26,300 (DCCEEW 2024b). There was a reduction in the number of Pookila populations located in Vic (from 12 to 5; Burns 2019a), and in Tas (from 16 to 2; (Lazenby et al. 2019). However, there is limited information from which to infer similar trends across the NSW and Qld populations (Lazenby et al. 2019; Burns 2019a; DCCEEW 2024b). A study conducted over a 32-year period near the Tomago aluminium smelter in NSW indicated an overall declining population (SMEC 2015). Shorter-term studies conducted over 5 years recorded a possible decline at Wyrrabalong National Park but no evident trend over 7 years at Munmorah State Conservation Area (DCCEEW 2024b).\n\n## 2.4 Important populations\n\nThe Evolutionarily Significant Units (ESUs) concept is used by conservationists to strategically prioritise taxa or populations of a species for recovery purposes (for example, to target for captive-breeding), when resources are limited or when the granularity of genetic diversity is not sufficiently captured in the current overarching taxonomic status (Avise 1989; Moritz 1994). There are two recognised ESUs of Pookila, separated by more than 200,000 years of divergence (Burns et al. 2023b). The Southern ESU (Vic populations) is rarer, both genetically and geographically fragmented, and more threatened than the Northern ESU (NSW and Qld populations), which is comparatively genetically healthy and not at immediate risk of extinction (Burns et al. 2023b). Tas populations were not included in the Burns et al. (2023b) analysis due to the lack of data. However, these populations are expected to place either independently or within the Southern ESU and are considered to be at extremely high risk of extinction (Burns et al. 2023b). Given the Tas populations are likely closely related to the Vic populations, until further information is available these populations should be managed as part of the Southern ESU (Burns et al. 2024).\n\nPopulations in Vic and Tas are severely fragmented and are thought to make up less than 5% of Pookila numbers nationally (DCCEEW 2024b). In Tas, the species was known from 16 separate sites in 1998, but its status on the Tas mainland is in doubt with no recent records (Lazenby et al. 2019; DNRE Tas 2023). However, it was rediscovered on Flinders Island in 2021 (DNRE Tas 2023). In Vic, extensive surveys found the species in only 5 of the 12 parks and reserves in which it had previously been recorded (Burns 2019a; Burns et al. 2023b). The remaining populations are spatially isolated and at high risk of genetic issues without assisted migrations to support genetic rescue (Burns 2019a; Burns et al. 2023b).\n\nThe connectivity and extent of populations in NSW and south-east Qld is poorly understood. Genetic evidence indicates the isolation of these populations has occurred relatively recently compared to the isolation of populations in Vic (Burns et al. 2023a; Burns et al. 2023b; DCCEEW 2024b). The extant populations are in sites that are geographically distinct and unlikely to be connected by gene flow, as were the now locally extinct populations.\n\nThere is evidence that Pookila populations are strongly affected by rainfall. In NSW, the length of the breeding season, and thus reproductive output, has been shown to increase during times of above-average rainfall (Kemper 1976, 1980). Abundance in the Anglesea (Vic) population had a strong positive relationship with cumulative monthly residual rainfall (Lock 2005). Densities of this population peaked following 4 years of above-average rainfall and declined during below-average rainfall and drought conditions (Lock 2005; Wilson et al. 2017). There are also data to suggest that population fluctuations at Wilsons Promontory (Vic) were influenced by rainfall patterns (Wilson et al. 2005). All populations are considered important for the long-term recovery and survival of the species (Table 5).\n\n \n\nTable 5: Populations of Pookila\n\n| Region | Population | Status | Year of last detection | Year of last survey | Known or expected decline? | Genetic diversity (relative heterozygosity a) |\n| ------ | --------------------------------------------------- | --------------------------------------------------------------------- | ---------------------- | ------------------- | -------------------------- | -------------------------------------------------------------------------------------------------------------------------------- |\n| ACT | Mulligans Flat Woodlands Sanctuary | Not persisting, translocated | 2021 | 2023 | Known | Translocation trials with individuals from the Australian National University (ANU) captive-breeding colony (using NSW founders) |\n| NSW | Barcoongere State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Barrington Tops State Forest | Extant | 2001 | 2025 | Expected decline | Unknown |\n| | Booti Booti National Park | Extant | 2022 | 2022 | Unknown | Unknown |\n| | Broadwater National Park | Extant | 2024 | 2025 | Known | Low |\n| | Bunjalung National Park | Unknown | 2002 | 2024 | Known | Unknown |\n| | Bylong | Unknown | Unknown | Unknown | Unknown | Moderate |\n| | Carrai National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Chaelundi State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Dalmorton State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Doyles River State Forest | Unknown | 1984 | 1984 | Unknown | Unknown |\n| | Ellis State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Gibraltar Range State Forest | Unknown | 2015 | 2015 | Unknown | Unknown |\n| | Goobang National Park | Extant | 2025 | 2025 | Unknown | Moderate |\n| | Goulburn River National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Kiwarrak State Forest | Unknown | 2019 | 2019 | Unknown | Unknown |\n| | Malara State Forest | Unknown | 1999 | Unknown | Unknown | Moderate |\n| | Malara State Forest | Unknown | 1993 | 1993 | Unknown | Unknown |\n| | Manobalai Nature Reserve | Extant | 2024 | 2024 | Unknown | Unknown |\n| | Marara State Forest | Unknown | 2014 | 2014 | Unknown | Unknown |\n| | Marengo State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Mernot State Forest | Unknown | 2008 | 2008 | Unknown | Unknown |\n| | Mount Royal National Park | Extant | 2023 | 2023 | Unknown | Unknown |\n| | Mummel Gulf National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Munmorah/Chain Valley | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Myall Lakes | Extant | 2022 | 2022 | Known | Unknown |\n| | Nymboida National Park | Unknown | 2015 | Unknown | Unknown | Moderate |\n| | Olney State Forest | Unknown | 2004 | 2004 | Unknown | Unknown |\n| | Port Stephens | Extant | 2022 | 2022 | Unknown | Moderate |\n| | Ravensworth State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Royal National Park | Extant | 2022 | 2022 | Expected decline | Unknown |\n| | Styx River State Forest | Unknown | 2000 | 2000 | Unknown | Unknown |\n| | Watchimbark Reserve | Unknown | 2011 | Unknown | Unknown | High |\n| | Werrikimbe National Park | Unknown | 1984 | 2020 | Unknown | Unknown |\n| | Wollemi National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Wyrrabalong National Park | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Yiraaldiya National Park feral predator-free area | Translocated, persistence to be determined by post-release monitoring | 2025 | 2025 | Known | Translocation trials with individuals from the captive-breeding colonies at ANU and NTSI (PPC) using NSW founders |\n| | Yuraygir National Park | Extant | 2024 | 2025 | Expected decline | Unknown |\n| QLD | Crows Nest/ Ravensbourne/Helidon Hills | Unknown | 2022 | 2025 | Known | Moderate |\n| | Emu Creek (private property) | Unknown | 2022 | 2024 | Known | Unknown |\n| | Main Range National Park/Spicers Peak Nature Refuge | Unknown | 2022 | 2025 | Known | Unknown |\n| | Numinbah Valley | Unknown | 2005 | 2005 | Unknown | Unknown |\n| TAS | Beaconsfield | Unknown | 1978 | 2002 | Known | Unknown |\n| | Flinders Island | Extant in the north of the island, undetected in the south | 2025 | 2025 | Expected decline | Unknown |\n| | Freycinet | Unknown | 2003 | 2025 | Known | Unknown |\n| | St Helens | Unknown | 2009 | 2025 | Expected decline | Unknown |\n| | Waterhouse Conservation Area | Unknown | 2010 | 2025 | Known | Unknown |\n| | Wukalina/Mt William | Unknown | 2004 | 2025 | Known | Unknown |\n| VIC | Anglesea | Locally extinct | 1999 | 2021 | Known | Very low (from historic samples) |\n| | Cranbourne | Locally extinct | 1976 | Unknown | Known | Unknown |\n| | Dutson Downs | Extant | 2024 | 2025 | Known | Low |\n| | Gippsland Lakes | Extant | 2025 | 2025 | Known | Low |\n| | Hummock Island | Locally extinct | 1977 | Unknown | Known | Unknown |\n| | Langwarrin | Locally extinct | 1984 | Unknown | Known | Unknown |\n| | Mullungdung | Locally extinct | 1976 | 2017 | Known | Unknown |\n| | Providence Ponds | Extant | 2025 | 2025 | Known | Low |\n| | Reeves Beach | Locally extinct | 1978 | 2017 | Known | Unknown |\n| | The Lakes National Park | Extant | 2025 | 2025 | Known | Low |\n| | Tyabb | Locally extinct | 1975 | Unknown | Known | Low (from historic samples) |\n| | Wilsons Promontory | Extant | 2025 | 2025 | Known | Very low |\n\n> Note: a \\- Heterozygosity measures from Burns et al. (2023b).\n\n> Note: Source: Lazenby et al. (2019), Burns (2021a), DCCEEW (2024), NSW BioNet and unpublished data from members of the recovery team and Recovery Network.\n\n## 2.5 Habitat\n\nPookila habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heathy understorey, and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022) (see Figure 3). Historically, the species has been found across a broader array of habitat types (DCCEEW 2024b), and it does not occupy all plausibly suitable habitat at present (Burns et al. 2020; Smith et al. 2023a). Sandy soils suitable for burrowing are common across most occupied habitats and soft sandy substrates, while deep topsoils are thought to be optimal (Abicair et al. 2020). Within the northern parts of the species’ range, it also occupies sites up to 300 km from the coast and in subcoastal locations at higher elevation, where soil composition differs (Martin 1995; Van Dyck & Lawrie 1997; Vernes et al. 2006; Dixon 2014; Churchill et al. 2023).\n\nThe species is often, although not exclusively, found in areas with high floristic diversity with leguminous perennials (Wilson 1991; Seebeck et al. 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; DEWHA 2010). For example, the heathland element of the species’ habitat is represented by dry heathlands that are floristically diverse and even (a high number of plants that are all relatively abundant compared to each other) that grow on sandy soils with a high percentage of bare ground (Seebeck et al. 1996).\n\nThe species has been identified as preferring early successional habitat, at some sites with peak abundance approximately 3–7 years post-fire and 7–9 years post-sand mine rehabilitation (Braithwaite & Gullan 1978; Fox & Fox 1978; Fox & Fox 1984; Twigg et al. 1989; Wilson 1991; Wilson et al. 2018). However, research in Vic suggests that the relationship between the Pookila and fire is complex, and local densities may be high in areas that have not been affected by fires for over 60 years (Burns and Phillips 2020). Similarly, surveys conducted on the NSW central coast have found high densities in long-unburnt vegetation (F Ford 2023, personal communication 8 August). Season, fire severity and extent, predation pressures, and climatic conditions following fire, all affect habitat suitability more strongly than a linear effect of time-since-fire (Burns & Phillips 2020). Furthermore, it is likely that it shelters in burrows during times of fire. This behaviour may be one of the attributes that allows the Pookila to persist in situ through some fire events (Friend 1993; Burns & Phillips 2020).\n\nThe Pookila has been considered a ‘habitat specialist’ (Braithwaite & Gullan 1978), but fossil record observations, and a strategic translocation and assessment of microhabitat use, suggests that it had broad habitat preferences (Smith et al. 2023c). Remains of the species are common in most bone deposits in south-eastern Australia that has been examined, and these deposits occur across a broad range of historical vegetation types and altitudes (Ford 2003). The species’ perceived habitat specialisation may have been further challenged by the Pookila translocation into a predator exclusion zone in the ACT as part of the Mulligans Flat-Goorooyarroo Woodland Experiment. The box gum and grassy woodlands comprising the predator exclusion area are not the type of coastal habitat often associated with contemporary records of the species (Abicair et al. 2020). However, recent detections of the species in grassy open forests and woodlands in Qld means that these non-coastal habitat types are still represented in the Pookila distribution.\n\nThe currently skewed occurrence (and perhaps specificity) of the Pookila towards coastal areas may in part be due to the interaction among threats, such as habitat clearing, altered fire regimes and predation pressures by introduced species such as cats (Felis catus) and European red foxes (Vulpes vulpes). However, it is worth noting that individuals sourced from the ANU breeding colony in the Mulligans Flat population are significantly underweight compared to their founding populations (13.6 g vs. 17.5 g) (Abicair et al. 2020), and that the population may not be self-sustaining. Together, these attributes may indicate that the habitat there may now be suboptimal for the Pookila. However, it may be important to pursue translocations in a diversity of habitats (including those that may be suboptimal) to build the species' resilience to environmental change and to restore ecological functions (Smith et al. 2023a).\n\nFigure 3: Photographs of contemporary Pookila habitat\n\n![These grouped 6 images demonstrate the diversity in the Pookila's habitat throughout the species' range. Refer to the footnote of this collective image for descriptions on habitat per image in addition to subheading 'Habitat' for more information.](image.006.png)\n\n> Note: Contemporary habitat in which the Pookila has been detected. Depicted are: a) grassy woodlands at Cunninghams Gap in Qld (photo: Ian Gynther); b) woodland with a grass and sedge understorey at Mt Hyland, NSW (photo: Anna Lloyd); c) coastal heathland at Munmorah in NSW (photo: Barbara Wilson); d) woodland with a heath understorey at Anglesea in Vic (photo: Barbara Wilson); e) coastal heathland at Loch Sport in Vic (photo: Phoebe Burns); and f) coastal heathland with a woodland overstorey on Flinders Island in Tas (photo: Rosemary Gales) (Source: DNRE Tas 2022).\n\n### 2.5.1 Habitat critical to the survival of the Pookila\n\nAll potential habitat should be considered important to the Pookila’s persistence and recovery, this includes any:\n\n Areas within the species’ modelled distribution;\n Areas where the species previously occurred but has since become locally extinct;\n Patches of suitable habitat that may not be currently occupied by the species but have the potential to be occupied by natural dispersal; and\n Areas that could be restored for reintroduction.\n\nAll areas that contain habitat that serves one or more of the following ecological functions have the potential to be habitat critical to the survival of the Pookila:\n\n Foraging habitat: vegetation with appropriate density and diversity to be able to support a population, such as native vegetation with seasonal abundance of edible flowering, fruiting, and seeding plants. Vegetation cover is important for the species to be able to forage without increased risk of predation. This habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heath understorey and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022).\n\n Breeding habitat: substrate for the species to be able to create burrows, or alternatively, ‘surface nests’ in low-lying vegetation that is suitable for successful reproduction and rearing of young. Soft sandy substrates, with deep topsoils are thought to be optimal (Abicair et al. 2020) but not exclusive. The Pookila mostly occupies coastal areas that provide this substrate. However, it also occupies sites 100–300 km from the coast and areas with higher elevation where soil composition varies. For example, the presence of sandy soils is highly likely in Goobang National Park (NSW NPWS 2001), whilst in the NSW ranges and Northern Tableland areas, soil may vary with a more dominant loamy or quartz soil base where the Pookila occurs or may be present (NSW OEH 2023. pers comm 22 November).\n\n Habitat for the long-term maintenance of the species: including areas that are not subject to prolonged drought or at risk of extreme droughts in future (such as potential climate refugia), and habitat that is maintained through periodic small-scale burns to provide appropriate foraging habitat and connectivity for the species.\n\n Habitat for the maintenance of genetic diversity: includes all current habitats where the species is extant as they contain populations necessary to achieve maximum genetic diversity and increase resilience of the species against future threats. There are two recognised ESUs: the Southern ESU (Vic populations) and the Northern ESU (NSW and Qld populations). Tas may yet be identified as a third ESU or may cluster within the Southern ESU. Therefore, habitat across the species’ range is necessary to maintain the national genetic diversity of the species.\n\n Habitat for translocation: can be identified using habitat modelling followed by ground-truthing at the local-scale and can include locations that have accounted for future changes due to climate change regimes as well as the exclusion of invasive predators.\n\nRefining what constitutes habitat critical to the survival of the Pookila is a priority action in this recovery plan. Until those key characteristics are further clarified, all areas that contain habitat that serves any of the above five ecological functions have the potential to be habitat critical to the survival of the Pookila. Habitat that supports multiple ecological functions and/or a higher number of the ecological attributes identified within the descriptions provided above, increases the likelihood that the habitat may be critical to the survival of the Pookila.\n\nGiven that the species has declined markedly in parts of its range and populations may rapidly change in size, failure to detect the species during surveys does not indicate the absence of potential habitat critical to its survival.\n\n## 2.6 Ecology\n\nThe Pookila often nests underground in communal burrow systems in well-drained soils. Individuals have also been recorded sheltering under low growing shrubs in surface nests (Kemper 1980, Lazenby et al. 2007). Burrows vary in length, depth, and complexity, ranging from shallow surface nests to intricate tunnel systems (Kemper 1981). Of three burrows that have been excavated, the largest was 5 m in length with a maximum depth of 60 cm (Kemper 1981). The species is nocturnal, and with a significant proportion of its time spent foraging above ground, it is exposed to a high risk of predation by native predators including owls and introduced predators such as the cat and the European red fox.\n\nTen individuals were radio-collared in one area of Tas for approximately 2 weeks, yielding information on nightly movements (Lazenby 1999; Lazenby et al. 2007). Over a 2 ha grid, 9 individuals moved up to 206 m in one night, and up to 37 m in one hour (Lazenby 1999; Lazenby et al. 2007). Pookila home ranges have been known to overlap (Lazenby 1999; Lazenby et al. 2007). The species’ home range is approximately 0.5–1.8 ha, increasing as habitat quality decreases (Lock 2005; Burns et al. 2023a).\n\nThe generation length of the Pookila is between one and two years (Kemper 1980). Females reach sexual maturity at approximately 13 weeks and live 2–3 years (Kemper 1976; Kemper 1980). Males reach sexual maturity at approximately 20 weeks and longevity remains unclear (Kemper 1976; Kemper 1980). During good conditions in NSW, the average litter size for the species has been previously recorded as 4.6, with a range of 1–6 young, and with females usually producing 1–2 litters per year, or 3–4 in exceptional conditions (Kemper 1976; Kemper 1980). The gestation period is around one month (32 days) and young are weaned at 3–4 weeks of age (Kemper 1976; Kemper 1980).\n\nThe breeding season varies depending on location and rainfall, with breeding occurring in NSW from late winter to early summer, while in Vic and Tas the breeding season begins in mid-spring and ceases in late summer (Kemper 1980; Burns et al. 2023b). The impact of rainfall is likely related to its influence on resource availability, which influences variation in birth rates, recruitment, and adult survival (Woinarski et al. 2014). Energy requirements are likely to be high during breeding, particularly for females, and if dietary resources are inadequate during this time, females or their young may perish (Lock & Wilson 2017). Hence, the Pookila demonstrates fluctuating population dynamics where extended breeding seasons follow periods of higher rainfall (Lock & Wilson 2017; Burns 2021). Pookila abundance reduces during periods of drought due to resource shortages and likely failed litters (Lock & Wilson 2017; Burns 2021).\n\n### 2.6.1 Diet and foraging\n\nThe Pookila is an omnivore and feeds on a variety of seeds, leaves, fungi, and invertebrates, exhibiting seasonal and locality-dependent dietary variation (Cockburn 1980; Norton 1987; Wilson and Bradtke 1999; Braithwaite and Gullan 1978). The species plays an important ecological role in dispersing seed and fungi in scats (Wilson & Bradtke 1999; Seebeck et al. 2003; Burns et al. 2023b).\n\n \n\n# 3 Threats\n\n## 3.1 Historical causes of decline\n\nThe distribution of subfossils and the genetic structure of extant populations indicate that the historic distribution of the Pookila extended more continuously throughout parts of coastal eastern Australia, from south-east Qld extending down to the northern parts of Tas and up to 300 km inland on the Australian mainland (Ford 2003; Breed and Ford 2007; DEWHA 2010; Burns et al. 2024; Smith et al. 2023c). The species has since experienced significant declines in distribution and population size and is now distributed in fragmented patches across this area (Figure 2).\n\n## 3.2 Current known and suspected threatening processes\n\nCurrent threats and their prevalence and intensity are described for the national Pookila population below. It is important to note that different populations may be experiencing different combinations of threats in varying intensities. There is significant interaction between threats.\n\n### 3.2.1 Habitat loss, disturbance and modification\n\n#### Habitat clearing\n\nLand clearing for agriculture, infrastructure, road, and housing development continues to place Pookila habitat at risk of further loss and fragmentation. As very little targeted and systematic monitoring of Pookila populations has occurred across the range in NSW (DCCEEW 2024b), the species is at particular risk of losing habitat where presence has not yet been confirmed.\n\nIn 2019, 87% of the Australian population lived within 50 km of the coast (cited in Clark et al. 2021). Coastal development is a continuing, increasing pressure with very high impacts on coastal environments (Clark et al. 2021) that is sprawling beyond established urban areas to regional coastal areas (Infrastructure Australia 2020).\n\nWhen habitat is cleared, disturbed, or modified, the risk and magnitude of impact to adjacent extant habitat patches increases, along with the interactive capability of threats (Fischer & Lindenmayer 2007; DNRE Tas 2022). For example, land clearing for agriculture can increase the carrying capacity of native and introduced herbivores within the landscape, leading to increased grazing and soil impact pressures on adjacent habitat. This is exacerbated when fire results in a further increase in grazing pressure and movement as herbivores actively seek to consume newly emerging vegetative shoots (DAWE 2022). With an increasing trend of habitat loss due to land clearing, the likelihood that fragmentation has occurred and will continue to occur is very high.\n\n#### Habitat modification due to Phytophthora cinnamomi infestation\n\nPhytophthora dieback – caused by the root-rot pathogen Phytophthora cinnamomi – is listed as a Key Threatening Process under the EPBC Act (DOEE 2018) and has been observed across the national range of the Pookila (Liu et al. 2013; Wilson et al. 2015; Burns et al. 2020; DCCEEW 2024b). The pathogen is spread through the movement of soil – including on footwear, tools, vehicles and machinery – and in water (DOEE 2018). While the impacts of P. cinnamomi on Pookila populations and habitat have not been investigated directly, the species is likely to be at significant risk. A range of plant species within Pookila habitat have been identified as susceptible to infection by P. cinnamomi (see McDougall 2005).\n\nPhytophthora dieback causes the loss of plant species richness (including keystone species), reduction of structural complexity (Wilson et al. 1994; Garkaklis et al. 2004; Laidlaw & Wilson 2006; Annett 2008; Armistead 2008; Dundas et al. 2013, 2016; Davis et al. 2014; Wilson et al. 2020), loss of food resources (for example, seeds, leaves, nectar, pollen) and suitable refuges from weather and predation (Laidlaw & Wilson 2006). The impacts of Phytophthora dieback can also extend to invertebrate communities and thus impact insectivorous fauna (Cahill et al. 2008), reducing availability of this food resource.\n\nSignificant negative impacts from Phytophthora dieback on small mammals and their habitat has been previously documented (Newell & Wilson 1993; Lilith 2002; Laidlaw & Wilson 2006; Swinburn et al. 2007; Annett 2008; Armistead 2008; Dundas et al. 2013; 2016), and long-term studies have identified localised extinctions of species and mammal community declines (Zhuang-Griffin 2015; Wilson et al. 2017; 2020).\n\nPriority Protection Areas for Phytophthora dieback management have been identified for the Otway Ranges, which was informed by habitats occupied by EPBC Act listed species, including the Pookila (Walshe et al. 2021; 2022; Wilson & Garkaklis 2023a; 2023b).\n\n#### Habitat modification due to native and introduced weeds\n\nChanging grazing and fire regimes have led to habitat loss by encroachment of coast tea tree (Leptospermum laevigatum) in some Pookila habitat (Burns 2019a; Seebeck et al. 2003). This is particularly evident at Wilsons Promontory National Park and in other coastal Vic sites along Ninety Mile Beach where dune succession has given rise to dense thickets of coast tea tree, such as at Reeves Beech. Weed species are of growing concern in other Pookila habitat, for example, in Lakes National Park. Invasive weeds include Asparagus asparagoides (bridal creeper) and Lycium ferocissimum (boxthorn).\n\n### 3.2.2 Fire regimes that cause declines in biodiversity\n\n‘Fire regimes that cause declines in biodiversity’ is listed as a Key Threatening Process under the EPBC Act (DAWE 2022). Altered fire regimes, such as those resulting from changes to seasonality, frequency and severity of fire weather conditions, are causing declines and loss of biodiversity and ecosystem function (DAWE 2022). The Pookila and the habitat the species relies on have a complicated relationship with fire.\n\nUnder certain circumstances, the species can peak in abundance in the years following a fire as resources become more readily available due to regeneration of vegetation, with such areas described as ‘early successional’ (Fox & McKay 1981). Furthermore, the Pookila has been reported to have relatively high survival during and immediately following bushfires, which is also presumably attributed to its ability to shelter in burrows (Friend 1993; Burns & Phillips 2020). However, this response appears to be more common following small-scale, patchy, low-severity burns and is likely reliant on above-average rainfall in the months and years following the fire (Burns & Phillips 2020).\n\nLarge-scale, high-severity fires have the capacity to extirpate entire populations, as occurred in 1983 on Ash Wednesday at Anglesea (Wilson et al. 2018). Fires being too frequent, too large, too severe, or poorly timed may lead to direct loss of individuals and/or sudden or gradual loss of habitat through reduction of habitat patches of suitable successional age, size, distribution, and plant diversity that the Pookila requires for availability of food and shelter from predators.\n\nSimilarly, prolonged exclusion or poorly timed application of fire may encourage coast tea tree encroachment and loss of favourable food and habitat species. Also, introduced predators may target recently burnt areas, compounding the increased risk of predation from loss of vegetative cover (Hradsky et al. 2017). Planned burning may also threaten the species if conducted during or immediately prior to drought, at too high an intensity, or across a widespread area (Wilson et al. 2018).\n\nThe increasing scale, severity, and frequency of fires under a changing climate pose a risk to the Pookila and its habitat. In 2019 and 2020, catastrophic bushfires impacted a large area of eastern Australia with varying severity (Legge et al. 2021). It was estimated that over 45% of the Pookila distribution was burnt, with a third of this area burning with high severity (Legge et al. 2021). The overall population of Pookila was predicted to have declined by 24% by 2021 (one year after fire), with the species suffering an estimated 75% decrease in abundance in severely burnt areas in that same period (Legge et al. 2021).\n\n### 3.2.3 Climate change: Increasing number of hot days, prolonged drought and changes to precipitation patterns\n\nIncreasing frequency, intensity, and duration of drought across much of the species’ range (CSIRO & BoM 2020) may lead to further catastrophic declines and local extinctions. Extreme heat events may exacerbate the effects of drought and could lead to mass mortality events as individuals dehydrate and fail to thermoregulate but this would require further research to confirm whether behaviours would mitigate this threat. The projections for the NSW climate by 2050 include an increase in minimum and maximum temperatures, an increase (double) in the number of hot days per year (days greater than 35°C), and longer fire seasons, with around a 40% increase in ‘very high’ fire danger days (CSIRO & BoM 2021).\n\nExtended periods of low rainfall reduce food availability for Pookila. Recent periods of drought have caused demonstrable crashes in Pookila populations. The local extinction of the Anglesea population is in part attributed to the Millennium Drought (Lock & Wilson 2017), and the species is yet to recover from the 2019 Gippsland drought (Burns 2021).\n\nFire–drought interactions may have a significant impact on the Pookila with potential compounding effects. This is evident from the 2019 and 2020 bushfires, which followed a period of significant drought in 2017 to 2019 (BoM 2020; DPIE 2020).\n\n### 3.2.4 Invasive fauna species\n\n#### Predation by cats and European red foxes\n\nFeral predators can have rapid and catastrophic impacts on populations of threatened rodents (Tuft et al. 2021). While there is limited direct evidence of the consumption of Pookila by feral predators (records in dog scats; Hoskins 1997; Davis et al. 2015; Lock & Wilson 2017; MidCoast Council NSW 2024. pers comm 13 February), it is recognised in the Threat Abatement Plan for Predation by Feral Cats 2024 that the Pookila is likely highly susceptible to predation by cats (and/or in some cases by foxes) (DCCEEW 2024a) and that cats and foxes can significantly impact population numbers (Seebeck et al. 2003; Burns 2019a).\n\nLike many other Pseudomys species, Pookila are docile and display poor awareness of exotic predators (P. Burns, pers. obs.). In such species, selective predation can occur (as reported, for example, on the broad-toothed rat, Mastacomys fuscus; Green 2002), leading to a disproportionate impact of predation on the vulnerable species even when co-occurring species (for example, the house mouse) are at greater abundance. Furthermore, communal burrowing and small local population densities mean that predation events at burrow entrances can cause rapid local declines (Burns 2019a).\n\nThe relative impact of predation by feral cats and foxes on the Pookila is not yet understood. Based on factors such as small body size, lower breeding output, and occupation of open habitat, some researchers have identified the Pookila as having a high threat rating from feral cat predation (Dickman 1996; Stobo-Wilson et al. 2021). Radford et al. (2018) reported the cat and fox threat to Pookila to be approximately the same based on expert opinion, although with weak evidence. While foxes are currently absent from the Tas mainland and Flinders Island (DNRE Tas 2022), and fox control is in place in other parts of the species’ range, effective cat control in open, non-insular systems beyond the local scale is not yet possible and potentially counterproductive (Lazenby et al. 2015; Lazenby et al. 2021; see Predator management).\n\nThe density and predatory impact of cats and foxes can be altered by competitive interactions with other predators. For example, reduced numbers of foxes via fox control can increase cat density, alter cat behaviour, and ultimately increase the impact of predation by feral cats (Risbey et al. 2000; Rees et al. 2023). Tasmanian devils have been shown to alter the behaviour and abundance of feral cats (Hollings et al. 2014; Lazenby & Dickman 2013), and cats have a much higher predatory impact on small mammals compared to devils (Hollings et al. 2014). The decline of devils caused by Devil Facial Tumour Disease (DFTD) is likely to have increased the predatory impacts of feral cats in Pookila habitat on the Tas mainland.\n\n#### Impact of hooved feral animals including deer, goats, horses, and pigs\n\nThe negative impacts of hooved feral animals on vegetation health and soil erosion are well documented (Keith & Pellow 2005; Bengsen et al. 2017; Wills et al. 2023; Burns et al. 2021). This is also recognised within the Draft threat abatement plan for competition and land degradation by unmanaged goats (DCCEEW 2024c), the Threat abatement plan for predation, habitat degradation, competition and disease transmission by feral pigs (Sus scrofa) (DOEE 2017), and the National feral deer action plan (Government of South Australia 2023). Where these species co-occur with the Pookila, there are risks of habitat degradation through trampling, wallowing, and overgrazing (Burns 2021). Overgrazing pressures may increase in post-fire environments, particularly in heathy vegetation types preferred by the Pookila (Heaton et al. 2022), compounding post-fire food shortages and predation risk, while impeding regrowth of suitable habitat and food species (DAWE 2022). The effects of hooved feral animal control programs on Pookila occurrence and abundance have not been directly studied, but based on vegetation response studies, control programs are expected to benefit the Pookila and its habitat, particularly around fire events.\n\n#### Competition/predation from introduced rodents\n\nWhile experimental trials have shown the Pookila to be competitively superior to the invasive house mouse (Fox & Twigg 1991; Haering & Fox 1997), there is growing evidence that house mouse plagues may supress Pookila recovery following periods of drought where resources are subsequently scarce (Burns 2021). Rattus rattus and R. norvegicus (black and brown rats, respectively) may also out-compete the Pookila for resources and prey on adults and juveniles, although this requires further research.\n\n#### Baiting of rabbits and invasive rodents\n\nThe susceptibility of the Pookila to 1080, Pindone, and rodenticide has not been tested. However, such poisons are lethal at low doses to some other Pseudomys species and are expected to be lethal to the Pookila (McIlroy 1982; Eisler 1995; Fisher 1999). Running a lethal dose trial on the Pookila is not appropriate given the species’ threatened status and ethical considerations, so it is safest to apply the Precautionary Principle and assume the species is likely susceptible to 1080, Pindone and rodenticide. Rabbit poison baiting is usually conducted with a carrot or oat bait, both foods readily consumed by the Pookila (Prosser et al. 2007), and it is possible that rabbit-baiting programs have contributed to the decline of the species in some areas (Fisher 1999; Burns 2019a). When finding the balance between protecting Pookila habitat by controlling rabbits and potentially poisoning individual Pookila, it is recommended to entirely avoid poison baiting in areas where Pookila presence is known. This is particularly important for at-risk populations; however, it may also apply to locations where Pookila presence is uncertain but these areas are adjacent to suitable habitat and/or a known population (within a 5 km radius). Rabbit and invasive rodent population control measures should use alternative methods. Control measures should be carefully considered depending on the area of control and the potential side effects for native species. Currently, this threat is considered to be high in Wilsons Promontory in Vic, not applicable in Qld or Tas, and unknown in NSW (DNRE Tas 2023).\n\n### 3.2.5 Small and fragmented populations: Low genetic diversity and inbreeding\n\nHistorical and contemporary habitat degradation and destruction, and the species’ very low dispersal ability have left the Pookila highly fragmented across its distribution, with some populations genetically isolated. This severe fragmentation leaves populations at higher risk of extinction, as a single catastrophic event can more easily impact the total area occupied by a population (Garner et al. 2020).\n\nFurthermore, genetically isolated populations are at higher risk of inbreeding depression. Inbreeding reduces genetic diversity, which reduces adaptability to changing environments and anthropogenic impacts. Inbreeding depression also increases species’ vulnerability to disease while also allowing potentially detrimental traits to surface in the population (Keller 2002). Furthermore, inbreeding depression can cause reduced fecundity and further genetic health issues (for example, chromosomal abnormalities) (Keller 2002).\n\nVic populations (Southern ESU) of the Pookila have relatively low heterozygosity (genetic diversity) and are at risk of inbreeding, with no natural gene flow possible between localities given lack of dispersal (Burns et al. 2023b). Therefore, the Southern ESU is considered an ‘at-risk ESU’. Populations across NSW retain greater heterozygosity and are genetically distinct from those in Vic, although animals in Broadwater National Park appear to have undergone a genetic bottleneck (Burns et al. 2023b). At the time of publishing this recovery plan, two historic mainland Tas samples and a small number of contemporary samples from Flinders Island have been sequenced and are being analysed. Additional Tas samples are being sought through survey work. Genetic mixing within Vic populations has been recommended and is being pursued, whereas no immediate action is required in the Northern ESU (NSW and Qld) (Burns et al. 2023b; Burns et al. 2024) since it is more genetically healthy than the Southern ESU. The recovery team has developed a National Pookila genetic management strategy (Zoos Victoria) and will continue to update it when new information is forthcoming.\n\n#### Disease\n\nThere is limited understanding of the potential prevalence or impact of disease on the Pookila. In general, screening of native rodents for pathogens has been very limited and little information is available on symptomology (Thompson et al. 2010). However, feral cats and invasive rodents carry diseases that could impact the Pookila. In particular, Toxoplasma gondii, a virulent parasite spread by cats, is likely present through much of the species’ former range (Dickson 2018). Previous experimental trials in other Pseudomys species demonstrated rapid, high mortality rates after infection with T. gondii, hence this parasite is likely to also increase the mortality of the Pookila (Burns 2019a).\n\n## 3.3 Threat prioritisation\n\nEach of the threats outlined in this recovery plan has been assessed to determine the risk posed across the national distribution of the Pookila using a risk matrix (see Table 6). The risk matrix considers the likelihood of an incident occurring and the consequences of that incident and uses a qualitative assessment drawing on both peer-reviewed literature and expert opinion.\n\nTable 6: Risk matrix\n\n| Consequence (x) / Likelihood (y) | Not significant | Minor | Moderate | Major | Catastrophic |\n| -------------------------------- | --------------- | ------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------- |\n| Almost certain | Low risk | Moderate risk | Very high riskPredation by European red foxes | Very high riskHabitat clearing (past/present)Fire regimes that cause declines in biodiversityIncreasing number of hot days, prolonged drought and changes to precipitation patternsPredation by cats | Very high risk |\n| Likely | Low risk | Moderate riskImpact of hooved feral animals including deer, goats, horses, and pigs | High riskHabitat modification due to Phytophthora cinnamomi infestationHabitat modification due to native and introduced weeds | Very high riskHabitat clearing (future) | Very high risk |\n| Possible | Low risk | Moderate riskCompetition/predation from introduced rodentsBaiting of rabbits and invasive rodents | High riskLow genetic diversity and inbreedingDisease | Very high risk | Very high risk |\n| Unlikely | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n| Unknown | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n\nRisk matrix key\n\n| Low risk | Moderate risk | High risk | Very-high risk |\n| -------- | ------------- | --------- | -------------- |\n\nDefinition of likelihood and consequence categories\n\nConsequence is a measure of the severity of the impact of the threat on the survival of the species.\n\n- Catastrophic – affecting survival, reproduction or essential movement of individuals in >80% of the population or across >80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Major – affecting survival, reproduction or essential movement of individuals in 50-80 % of the population or across 50-80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Moderate – affecting survival, reproduction or essential movement of individuals in 20-50% of the population or across 20-50% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Minor – affecting survival, reproduction or essential movement of individuals in 5-30% of the population or across 5-30% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Not significant – affecting survival, reproduction or essential movement of individuals in <5% of the population or across <5% of the distribution within the next 3 generations or 10 years, whichever is longer.\n\nLikelihood is a measure of how regularly the threat impacts the species.\n\n- Almost certain – >90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Likely – 66-90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Possible – 33-66% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unlikely – <33% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unknown – probability that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer, could be 0-100%.","sortOrder":23},{"sectionNumber":"Habitat clearing","sectionType":"subdivision","heading":"Habitat clearing","content":"#### Habitat clearing\n\nLand clearing for agriculture, infrastructure, road, and housing development continues to place Pookila habitat at risk of further loss and fragmentation. As very little targeted and systematic monitoring of Pookila populations has occurred across the range in NSW (DCCEEW 2024b), the species is at particular risk of losing habitat where presence has not yet been confirmed.\n\nIn 2019, 87% of the Australian population lived within 50 km of the coast (cited in Clark et al. 2021). Coastal development is a continuing, increasing pressure with very high impacts on coastal environments (Clark et al. 2021) that is sprawling beyond established urban areas to regional coastal areas (Infrastructure Australia 2020).\n\nWhen habitat is cleared, disturbed, or modified, the risk and magnitude of impact to adjacent extant habitat patches increases, along with the interactive capability of threats (Fischer & Lindenmayer 2007; DNRE Tas 2022). For example, land clearing for agriculture can increase the carrying capacity of native and introduced herbivores within the landscape, leading to increased grazing and soil impact pressures on adjacent habitat. This is exacerbated when fire results in a further increase in grazing pressure and movement as herbivores actively seek to consume newly emerging vegetative shoots (DAWE 2022). With an increasing trend of habitat loss due to land clearing, the likelihood that fragmentation has occurred and will continue to occur is very high.","sortOrder":24},{"sectionNumber":"Habitat modification due to Phytophthora","sectionType":"subdivision","heading":"Habitat modification due to Phytophthora cinnamomi infestation","content":"#### Habitat modification due to Phytophthora cinnamomi infestation\n\nPhytophthora dieback – caused by the root-rot pathogen Phytophthora cinnamomi – is listed as a Key Threatening Process under the EPBC Act (DOEE 2018) and has been observed across the national range of the Pookila (Liu et al. 2013; Wilson et al. 2015; Burns et al. 2020; DCCEEW 2024b). The pathogen is spread through the movement of soil – including on footwear, tools, vehicles and machinery – and in water (DOEE 2018). While the impacts of P. cinnamomi on Pookila populations and habitat have not been investigated directly, the species is likely to be at significant risk. A range of plant species within Pookila habitat have been identified as susceptible to infection by P. cinnamomi (see McDougall 2005).\n\nPhytophthora dieback causes the loss of plant species richness (including keystone species), reduction of structural complexity (Wilson et al. 1994; Garkaklis et al. 2004; Laidlaw & Wilson 2006; Annett 2008; Armistead 2008; Dundas et al. 2013, 2016; Davis et al. 2014; Wilson et al. 2020), loss of food resources (for example, seeds, leaves, nectar, pollen) and suitable refuges from weather and predation (Laidlaw & Wilson 2006). The impacts of Phytophthora dieback can also extend to invertebrate communities and thus impact insectivorous fauna (Cahill et al. 2008), reducing availability of this food resource.\n\nSignificant negative impacts from Phytophthora dieback on small mammals and their habitat has been previously documented (Newell & Wilson 1993; Lilith 2002; Laidlaw & Wilson 2006; Swinburn et al. 2007; Annett 2008; Armistead 2008; Dundas et al. 2013; 2016), and long-term studies have identified localised extinctions of species and mammal community declines (Zhuang-Griffin 2015; Wilson et al. 2017; 2020).\n\nPriority Protection Areas for Phytophthora dieback management have been identified for the Otway Ranges, which was informed by habitats occupied by EPBC Act listed species, including the Pookila (Walshe et al. 2021; 2022; Wilson & Garkaklis 2023a; 2023b).","sortOrder":25},{"sectionNumber":"Habitat modification due to native and i","sectionType":"subdivision","heading":"Habitat modification due to native and introduced weeds","content":"#### Habitat modification due to native and introduced weeds\n\nChanging grazing and fire regimes have led to habitat loss by encroachment of coast tea tree (Leptospermum laevigatum) in some Pookila habitat (Burns 2019a; Seebeck et al. 2003). This is particularly evident at Wilsons Promontory National Park and in other coastal Vic sites along Ninety Mile Beach where dune succession has given rise to dense thickets of coast tea tree, such as at Reeves Beech. Weed species are of growing concern in other Pookila habitat, for example, in Lakes National Park. Invasive weeds include Asparagus asparagoides (bridal creeper) and Lycium ferocissimum (boxthorn).","sortOrder":26},{"sectionNumber":"3.2.2 Fire regimes that cause declines i","sectionType":"division","heading":"3.2.2 Fire regimes that cause declines in biodiversity","content":"### 3.2.2 Fire regimes that cause declines in biodiversity\n\n‘Fire regimes that cause declines in biodiversity’ is listed as a Key Threatening Process under the EPBC Act (DAWE 2022). Altered fire regimes, such as those resulting from changes to seasonality, frequency and severity of fire weather conditions, are causing declines and loss of biodiversity and ecosystem function (DAWE 2022). The Pookila and the habitat the species relies on have a complicated relationship with fire.\n\nUnder certain circumstances, the species can peak in abundance in the years following a fire as resources become more readily available due to regeneration of vegetation, with such areas described as ‘early successional’ (Fox & McKay 1981). Furthermore, the Pookila has been reported to have relatively high survival during and immediately following bushfires, which is also presumably attributed to its ability to shelter in burrows (Friend 1993; Burns & Phillips 2020). However, this response appears to be more common following small-scale, patchy, low-severity burns and is likely reliant on above-average rainfall in the months and years following the fire (Burns & Phillips 2020).\n\nLarge-scale, high-severity fires have the capacity to extirpate entire populations, as occurred in 1983 on Ash Wednesday at Anglesea (Wilson et al. 2018). Fires being too frequent, too large, too severe, or poorly timed may lead to direct loss of individuals and/or sudden or gradual loss of habitat through reduction of habitat patches of suitable successional age, size, distribution, and plant diversity that the Pookila requires for availability of food and shelter from predators.\n\nSimilarly, prolonged exclusion or poorly timed application of fire may encourage coast tea tree encroachment and loss of favourable food and habitat species. Also, introduced predators may target recently burnt areas, compounding the increased risk of predation from loss of vegetative cover (Hradsky et al. 2017). Planned burning may also threaten the species if conducted during or immediately prior to drought, at too high an intensity, or across a widespread area (Wilson et al. 2018).\n\nThe increasing scale, severity, and frequency of fires under a changing climate pose a risk to the Pookila and its habitat. In 2019 and 2020, catastrophic bushfires impacted a large area of eastern Australia with varying severity (Legge et al. 2021). It was estimated that over 45% of the Pookila distribution was burnt, with a third of this area burning with high severity (Legge et al. 2021). The overall population of Pookila was predicted to have declined by 24% by 2021 (one year after fire), with the species suffering an estimated 75% decrease in abundance in severely burnt areas in that same period (Legge et al. 2021).","sortOrder":27},{"sectionNumber":"3.2.3 Climate change: Increasing number ","sectionType":"division","heading":"3.2.3 Climate change: Increasing number of hot days, prolonged drought and changes to precipitation patterns","content":"### 3.2.3 Climate change: Increasing number of hot days, prolonged drought and changes to precipitation patterns\n\nIncreasing frequency, intensity, and duration of drought across much of the species’ range (CSIRO & BoM 2020) may lead to further catastrophic declines and local extinctions. Extreme heat events may exacerbate the effects of drought and could lead to mass mortality events as individuals dehydrate and fail to thermoregulate but this would require further research to confirm whether behaviours would mitigate this threat. The projections for the NSW climate by 2050 include an increase in minimum and maximum temperatures, an increase (double) in the number of hot days per year (days greater than 35°C), and longer fire seasons, with around a 40% increase in ‘very high’ fire danger days (CSIRO & BoM 2021).\n\nExtended periods of low rainfall reduce food availability for Pookila. Recent periods of drought have caused demonstrable crashes in Pookila populations. The local extinction of the Anglesea population is in part attributed to the Millennium Drought (Lock & Wilson 2017), and the species is yet to recover from the 2019 Gippsland drought (Burns 2021).\n\nFire–drought interactions may have a significant impact on the Pookila with potential compounding effects. This is evident from the 2019 and 2020 bushfires, which followed a period of significant drought in 2017 to 2019 (BoM 2020; DPIE 2020).","sortOrder":28},{"sectionNumber":"3.2.4 Invasive fauna species","sectionType":"division","heading":"3.2.4 Invasive fauna species","content":"# 1 Background\n\n## 1.1 Introduction\n\nThe Pookila (New Holland mouse; Pseudomys novaehollandiae) is a small, burrowing, rodent native to south eastern Australia, that is patchily distributed from Tasmania to south-east Queensland. The species has had catastrophic declines that were recorded since the discovery of new populations in the 1970s. Isolated remnant populations are at high risk of extirpation and genetic inbreeding.\n\nDistribution is fragmented across southern Queensland (Qld), the northern and central coastal areas of New South Wales (NSW), central southern coastal Victoria (Vic), and north-eastern Tasmania (Tas) including Flinders Island (Burns et al. 2023a). While the species is typically restricted to coastal areas, it occupies areas up to 300 km inland in Qld and NSW.\n\nSince the species’ description in 1843, more than a dozen known populations have been lost and many more have declined. It has not been live trapped on mainland Tas since 2010 and in Vic, seven of twelve populations have become extinct since the 1970s (Burns et al. 2023b). Encouragingly, the Pookila was rediscovered on Flinders Island, Tasmania in 2021 and was last recorded in southern Qld in 2022. It has now been translocated into Mulligans Flat Woodland Sanctuary in the Australian Capital Territory (ACT), Yiraaldiya National Park feral predator-free area (NSW), reintroduced into the Cranbourne Royal Botanic Gardens (Vic) in 2024, and has numerous ongoing and successful captive-breeding programs.\n\nEuropean colonisation in Australia exposed the Pookila to compounding threats that caused population declines across its range. These threats include the direct destruction and fragmentation of habitat due to land clearing for agriculture, urbanisation, road construction, and forest plantations, as well as predation by introduced cats and European red foxes (DEWHA 2010; Burns 2019a). Introduced rodents may outcompete the Pookila under some conditions, and both introduced rodents and cats are vectors of novel diseases. Infection may also cause mortality and/or increase the Pookila’s susceptibility to predation (for example, toxoplasmosis spread by feral cats) (Lazenby et al. 2019; Burns 2019a).\n\n## 1.2 Conservation status\n\nThe Pookila is listed as Vulnerable under the Environment Protection, Biodiversity and Conservation Act 1999 (EPBC Act), Vulnerable (ACT, Qld, NSW, VIC) and Endangered (Tas) (Table 3).\n\nThe Pookila was first listed under the EPBC Act in the Vulnerable category on 11 August 2010. It was reassessed by the Threatened Species Scientific Committee in July 2024 retaining its Vulnerable category status under Criteria 1 and 2 (DCCEEW 2024b).\n\nCriterion 1 (A4b) was met due to the severe declines in population size estimated to have occurred in Tas and Vic and the significant declines suspected to have resulted from the impacts of the 2019 and 2020 bushfires in NSW, with the Committee concluding that the Pookila was subject to a population size reduction of at least 30% (DCCEEW 2024b).\n\nCriterion 2 B2ab(i,ii,iii,iv,v) was met due to the limited Area of Occupancy (AOO), low/very low number of locations (six), and estimated continuing decline in EOO, AOO, area of habitat, number of locations and populations, as well as the number of mature individuals (DCCEEW 2024b).\n\nTable 3: Conservation status of the Pookila\n\n| Listing | Conservation Status |\n| ---------------------------------------------------------------------------- | ------------------- |\n| Environment Protection and Biodiversity Conservation Act 1999 (Commonwealth) | Vulnerable |\n| Nature Conservation Act 2014 (Australian Capital Territory) | Vulnerable |\n| Nature Conservation Act 1992 (Queensland) | Vulnerable |\n| Biodiversity Conservation Act 2016 (New South Wales) | Vulnerable |\n| Flora and Fauna Guarantee Act 1988 (Victoria) | Vulnerable |\n| Threatened Species Protection Act 1995 (Tasmania) | Endangered |\n| IUCN Red List of Threatened Species | Vulnerable |\n\n## 1.3 Cultural and community significance\n\nThe Pookila currently occurs on the Country of the Awabakal, Banbai, Barunggam, Biripi, Bunurong, Boonwurrung, Bundjalung, Burrgati, Daingatti, Darkinung, Dharawal, Dharug, Eora, Geawegal, Gumbainggir, Gunaikurnai, Kuringgai, Gunaikurnai, Ngunnawal, Palawa, Wiradjuri, Wonnarua, Worimi and Yuggera Peoples.\n\nThe species has likely been known by numerous names in different Aboriginal languages across its geographic range. The name Pookila is derived from the Ngarigo word for ‘mouse’ – bugila (Braithwaite et al. 1995; Hercus n.d.), deviating somewhat from traditional Ngarigo pronunciation (Braithwaite et al. 1995). The assignation of that name was part of an initiative by the CSIRO to provide Indigenous Australian names for Australian rodents (Braithwaite et al. 1995). This was considered a way to avoid the undesirable associations with introduced vermin often attached to the words ‘rat’ and ‘mouse’, impeding conservation of native rodents due to negative public perception (Braithwaite et al. 1995). An alternative name could have been ‘tuwaranga’ which means ‘mouse’ in palawa kani, the language of the Tasmanian Aboriginal people (Tasmanian Aboriginal Centre, 2023), however, the name Pookila has been associated with this species since 1995 and is included on the Australian Government’s SPRAT database.\n\n## 1.4 National Pookila Recovery Team\n\nA National Pookila Recovery Team (the recovery team or NPRT) was established in 2021 with the aim of improving recovery efforts for the species at a national scale. A previous Pookila Recovery Team (1995 to 2002) was largely based in Vic. The current recovery team has representation across the species’ entire distribution. Beyond the core recovery team there are three sub-groups: captive breeding, habitat and fire management, and Vic management.\n\nThe role of the recovery team is to facilitate and encourage collaborations across jurisdictions, provide advice to partners and land managers, increase engagement, improve the public perception of the Pookila as well as the responsibility of advising and coordinating the implementation of recovery actions. The recovery team also seeks to facilitate opportunities for mutual learning and skill sharing with First Nations people. The membership of the recovery team includes species experts and land managers from relevant government agencies and non-government organisations and draws from a broader network of expertise from First Nations people, researchers, and community groups.\n\n## 1.5 Key partners\n\nImplementation of recovery planning is a collaborative effort involving a range of key partners and affected interests. The groups likely to have primary interest in recovery planning for the Pookila are listed in Table 4.\n\nTable 4: Groups with primary interest in Pookila recovery\n\n| Organisations | Involvement/ interest |\n| -------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| ACT Government | Management of the land of the first Pookila translocation site. Ongoing monitoring at the site and potential future releases in the area. |\n| Australian National University (ANU) | Captive-breeding program for NSW individuals. Ongoing monitoring and management in Mulligans Flat Woodland Sanctuary and potential future reintroductions in the ACT and other jurisdictions. |\n| Aussie Ark (AA) | Maintains an insurance population. Potential development of a captive-breeding program for the Pookila and future reintroductions into cat and fox free sites in NSW. |\n| Corangamite Catchment Management Authority (CCMA) | Managers of habitat under threat from P. cinnamomi. |\n| Department of Climate Change, Energy, the Environment and Water (DCCEEW, Cwlth) | The Pookila is a priority species in the Threatened Species Action Plan 2022-2032, which provides guidance and support in the design and delivery of projects for priority species. Recovery plans inform decision making under the EPBC Act referral and assessment process. |\n| Department of Energy, Environment and Climate Action (DEECA) | Primary agency involved in threatened species management on public and private land in Vic. |\n| Department of Natural Resources and Environment (DNRE) | Primary agency involved in threatened species management on public and private land in Tas. |\n| Department of the Environment, Tourism, Science and Innovation (DETSI) | Primary agency involved in threatened species management on public and private land in Qld. |\n| Flinders Island Aboriginal Association Incorporated (FIAAI) | Traditional Owners of Flinders Island population Tas. |\n| Friends of the Prom (FOTP) | Group established to support the management of Wilsons Promontory National Park. |\n| Gippsland Water (GW) | Manages a significant Vic population. |\n| Gunaikurnai Land and Waters Aboriginal Corporation (GLaWAC) | Joint managers of Lakes National Park, and Gippsland Lakes Coastal Park, and have Registered Aboriginal Party status of Providence Ponds Flora and Fauna Reserve in Vic. |\n| Land holders/Landowners | Consultation and engagement with all types of landowners will play a role in the success in meeting the recovery plan’s actions and objectives. |\n| Local councils | Community involvement projects and cooperation with management actions on a local level. |\n| Monash University (MU) | Storage and research of Pookila cell lines and genetic tissue. Supporting research of the genetic sequencing of the Pookila and subsequent projects surrounding genetic information. |\n| Moonlit Sanctuary Wildlife Conservation Park (MSWCP) | Managing a captive population of Pookila as well as community conservation and engagement activities. |\n| Forestry agencies | Actions and management regarding forestry operations and Regional Forest Agreements. |\n| NSW Department of Climate Change, Energy, the Environment and Water (DCCEEW NSW) | Primary agency involved in threatened species management on public and private land in NSW. |\n| NSW National Parks and Wildlife Service (NPWS NSW) | Implementation of appropriate management actions within parks and reserves in NSW, including reintroduction of the Pookila to establish new populations in NPWS-managed feral predator-free areas. |\n| Parks Victoria (PV) | Implementation of appropriate management actions concerning habitat within Vic parks and reserves. |\n| National Threatened Species Institute (NTSI) Priam Psittaculture Centre (PPC) | Host of a captive breeding program founded from northern ESU individuals. |\n| Royal Botanical Gardens Victoria Cranbourne (RBGVC) | Location for reintroduction to establish a new population. |\n| Tasmanian Aboriginal Land and Sea Council Aboriginal Corporation (TALSCAC) | Traditional Owners of Northeast population Tas. |\n| Trust for Nature (TFN) | Managing properties adjacent to occupied habitat. |\n| Turner Family Foundation (TFF) | Manager of Spicers Peak Nature Refuge south-east Qld habitat. |\n| University of Canberra (UC) | Involved in captive husbandry and dietary analyses. |\n| University of Southern Queensland (USQ) | Collaborative research project with the DETSI and the Turner Family Foundation to refine knowledge of distribution and ecological requirements in south-east Queensland. |\n| University of Sydney (USyd) | Analysis and advice on captive-breeding genetics. |\n| Wadawurrung Traditional Owners Aboriginal Corporation (WTOAC) | Traditional Owners of Anglesea population Vic. |\n| Woodlands and Wetlands Trust (WWT) | Community engagement, education and volunteer programs at Mulligans Flat Woodland Sanctuary. |\n| World Wildlife Fund Australia (WWF) | Potential partner for island translocation. |\n| Zoos Victoria (ZV) | Programs underway for monitoring, management, captive breeding, genetic rescue, reintroductions, research, studbook management and species coordination, community conservation and engagement. |\n| Zoo and Aquarium Association Australasia (ZAA) | Coordination of the NPRT, initial engagement/consultation with First Nations people, and assistance with studbook management and coordination. |\n\n# 2 Species information\n\n## 2.1 Taxonomy and description\n\nPseudomys novaehollandiae, family Muridae, is endemic to Australia and was first described in 1843 as Mus novaehollandiae (Waterhouse 1843). It was later transferred to the genus Pseudomys with the new name of Pseudomys (Gyomys) novaehollandiae (Thomas 1910). There is some geographic overlap and occasional evidence of hybridisation with its sister taxon, P. delicatulus (Delicate mouse), in the north of its range (Roycroft et al. 2024). However, the two taxa were found to be both genetically and morphologically distinct (Ford 2003; Roycroft et al. 2024).\n\nThe Pookila is a nocturnal, terrestrial, burrowing rodent. It is small, with a head-body length of approximately 65–90 mm and a tail length of 80–105 mm (Menkhorst & Knight 2011). The species’ dorsal coat is sandy brown with black flecks and the ventral coat is white (Figure 1). The tail appears bicoloured with a darker dorsal side and paler pink ventral side, and the feet are pink with fine white hairs (Figure 1).\n\nIndividuals in the south of the range are heavier than those further north. Tas individuals are the heaviest, weighing 19–28 g, followed by Vic at 18–‍25 g, then Qld and NSW at 12–20 g (Hocking 1980; Burns et al. 2023a). The species is most similar in appearance to the delicate mouse, which is generally smaller and more finely featured. It is superficially similar in appearance to the invasive house mouse (Mus musculus). It is distinguishable by its larger prominent eyes, larger rounded ears, lack of ‘mousy’ odour and absence of a notch on the upper incisors (note that examining dentition is not recommended for identifying live animals) (Seebeck et al. 2003; DEWHA 2010; Burns et al. 2023a; DCCEEW 2024b). The Pookila’s body shape is also thicker and more rounded compared to the house mouse, lacking an obvious neck (Burns 2019b).\n\nFigure 1: A Pookila (Pseudomys novaehollandiae) individual showing identifying characteristics: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.\n\n![Left: An image of a captive Pookila (New Holland Mouse) on a branch.\nRight: Underbelly of a male Pookila showing the white fur, paler surface of the ventral tail and pink feet with white hairs. The inset top picture shows the darker dorsal side of the tail and the bottom picture showing the white belly with black underfur.](image.004.png)Source: Left © Zoos Victoria (2023), Right and insets Kiarrah Smith (2025), Coexistence Conservation Laboratory, Australian National University.\n\n## 2.2 Distribution\n\nThe Pookila has a fragmented distribution across coastal areas of central-eastern and south‑eastern Australia (Figure 2). It occurs in southern Qld, the northern and central coastal areas of NSW, central-southern coastal Vic, and north-eastern Tas including Flinders Island (Burns et al. 2023b; DCCEEW 2024b). A large portion of its distribution occurs within areas with Regional Forest Agreements (RFAs), Indigenous Protected Areas and National Parks. In Qld and NSW, the Pookila occurs in the World and National Heritage Listed Greater Blue Mountains Area and the Gondwana Rainforests of Australia. It is known only from subfossil records in National Heritage Listed areas including the Australian Alps National Parks and Reserves, Ku-ring-gai Chase National Park, Lion, Long and Spectacle Island Nature Reserves and the Great Ocean Road region (Ford 2003; Smith et al. 2023c). While most modern detections of the species exist from coastal areas, it has also been located in areas up to 300 km inland in Qld and NSW. The species’ distribution is believed to have once been continuous, forming one large population throughout these areas.\n\nThe species’ historic distribution is inferred through the presence of subfossils as well as genetic information (Ford 2003; Breed & Ford 2007; Smith et al. 2023c). Its current distribution is attributable both to longer-term climatic and landscape changes, as well as a large population decline following European colonisation (Breed & Ford 2007; Theden-Ringl et al. 2018; Roycroft et al. 2021). Despite the ACT and surrounding region not being mapped within the current Pookila distribution (Figure 2), there is subfossil evidence of historical occurrence in nearby areas with similar habitat (Ford 2003; Smith et al. 2023c).\n\nThe extent of occurrence (EOO) for the species was estimated at 326,000 km2 and the area of occupancy (AOO) for the species was estimated at 694 km2 (DCCEEW 2024b). The AOO may have further contracted since this estimate, as nearly half of the species’ habitat was affected by the 2019 and 2020 bushfires. However, this includes approximately 30% of the species’ habitat that burned at low to medium severity (Legge et al. 2022). The species is likely to survive low to medium severity fires, but high severity fires could greatly reduce survivability (Wilson et al. 2018; Burns & Phillips 2020; DCCEEW 2024b).\n\nThe Pookila is associated with early to mid-stage vegetation succession following fire events (Woinarski & Burbidge 2016). Therefore, some recovery after the 2019 and 2020 bushfires is possible due to potential development of suitable habitat post-fire (DCCEEW 2024b).\n\nFigure 2: Modelled distribution of the Pookila\n\n![The modelled distribution of the Pookila or the Pookila habitat is known or likely to occur from south east Queensland along the northern and central east coast of New South Wales to costal regions in Victoria, predominantly to the east, all the way to Flinders Island and north and eastern Tasmania. The Pookila is also likely to occur 100-300 km inland in New South Wales. Pookila or Pookila habitat may occur in close proximity to these known or likely regions. Refer to subheading 'Species distribution' for more information. ](image.005.jpeg)\n\n> Note: Source: Base map Geoscience Australia; species distribution data Species of National Environmental Significance database.\n\n> Note: Caveat: The information presented in this map has been provided by a range of groups and agencies. While every effort has been made to ensure accuracy and completeness, no guarantee is given, nor responsibility taken by the Commonwealth for errors or omissions, and the Commonwealth does not accept responsibility in respect of any information or advice given in relation to, or as a consequence of, anything contained herein.\n\n> Note: Species distribution mapping: The species distribution mapping categories are indicative only and aim to capture (a) the habitat or geographic feature that represents recent observed locations of the species (known to occur) or habitat occurring in close proximity to these locations (likely to occur); and (b) the broad environmental envelope or geographic region that encompasses all areas that could provide habitat for the species (may occur). These presence categories are created using an extensive database of species observations records, national and regional-scale environmental data, environmental modelling techniques and documented scientific research.\n\n## 2.3 Population size and trends\n\nIn 2008, the national population size of the Pookila was estimated at fewer than 10,000 individuals (Menkhorst et al. 2008; DEWHA 2010) and was reported to be 8,000 individuals in the IUCN assessment by Woinarski and Burbidge (2016). In the assessment for listing under the EPBC Act, the maximum plausible population size prior to the 2019 and 2020 bushfires was estimated (using calculations of AOO and density) to be 34,700 individuals, with the post-fire plausible population range of 13,100–26,300 (DCCEEW 2024b). There was a reduction in the number of Pookila populations located in Vic (from 12 to 5; Burns 2019a), and in Tas (from 16 to 2; (Lazenby et al. 2019). However, there is limited information from which to infer similar trends across the NSW and Qld populations (Lazenby et al. 2019; Burns 2019a; DCCEEW 2024b). A study conducted over a 32-year period near the Tomago aluminium smelter in NSW indicated an overall declining population (SMEC 2015). Shorter-term studies conducted over 5 years recorded a possible decline at Wyrrabalong National Park but no evident trend over 7 years at Munmorah State Conservation Area (DCCEEW 2024b).\n\n## 2.4 Important populations\n\nThe Evolutionarily Significant Units (ESUs) concept is used by conservationists to strategically prioritise taxa or populations of a species for recovery purposes (for example, to target for captive-breeding), when resources are limited or when the granularity of genetic diversity is not sufficiently captured in the current overarching taxonomic status (Avise 1989; Moritz 1994). There are two recognised ESUs of Pookila, separated by more than 200,000 years of divergence (Burns et al. 2023b). The Southern ESU (Vic populations) is rarer, both genetically and geographically fragmented, and more threatened than the Northern ESU (NSW and Qld populations), which is comparatively genetically healthy and not at immediate risk of extinction (Burns et al. 2023b). Tas populations were not included in the Burns et al. (2023b) analysis due to the lack of data. However, these populations are expected to place either independently or within the Southern ESU and are considered to be at extremely high risk of extinction (Burns et al. 2023b). Given the Tas populations are likely closely related to the Vic populations, until further information is available these populations should be managed as part of the Southern ESU (Burns et al. 2024).\n\nPopulations in Vic and Tas are severely fragmented and are thought to make up less than 5% of Pookila numbers nationally (DCCEEW 2024b). In Tas, the species was known from 16 separate sites in 1998, but its status on the Tas mainland is in doubt with no recent records (Lazenby et al. 2019; DNRE Tas 2023). However, it was rediscovered on Flinders Island in 2021 (DNRE Tas 2023). In Vic, extensive surveys found the species in only 5 of the 12 parks and reserves in which it had previously been recorded (Burns 2019a; Burns et al. 2023b). The remaining populations are spatially isolated and at high risk of genetic issues without assisted migrations to support genetic rescue (Burns 2019a; Burns et al. 2023b).\n\nThe connectivity and extent of populations in NSW and south-east Qld is poorly understood. Genetic evidence indicates the isolation of these populations has occurred relatively recently compared to the isolation of populations in Vic (Burns et al. 2023a; Burns et al. 2023b; DCCEEW 2024b). The extant populations are in sites that are geographically distinct and unlikely to be connected by gene flow, as were the now locally extinct populations.\n\nThere is evidence that Pookila populations are strongly affected by rainfall. In NSW, the length of the breeding season, and thus reproductive output, has been shown to increase during times of above-average rainfall (Kemper 1976, 1980). Abundance in the Anglesea (Vic) population had a strong positive relationship with cumulative monthly residual rainfall (Lock 2005). Densities of this population peaked following 4 years of above-average rainfall and declined during below-average rainfall and drought conditions (Lock 2005; Wilson et al. 2017). There are also data to suggest that population fluctuations at Wilsons Promontory (Vic) were influenced by rainfall patterns (Wilson et al. 2005). All populations are considered important for the long-term recovery and survival of the species (Table 5).\n\n \n\nTable 5: Populations of Pookila\n\n| Region | Population | Status | Year of last detection | Year of last survey | Known or expected decline? | Genetic diversity (relative heterozygosity a) |\n| ------ | --------------------------------------------------- | --------------------------------------------------------------------- | ---------------------- | ------------------- | -------------------------- | -------------------------------------------------------------------------------------------------------------------------------- |\n| ACT | Mulligans Flat Woodlands Sanctuary | Not persisting, translocated | 2021 | 2023 | Known | Translocation trials with individuals from the Australian National University (ANU) captive-breeding colony (using NSW founders) |\n| NSW | Barcoongere State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Barrington Tops State Forest | Extant | 2001 | 2025 | Expected decline | Unknown |\n| | Booti Booti National Park | Extant | 2022 | 2022 | Unknown | Unknown |\n| | Broadwater National Park | Extant | 2024 | 2025 | Known | Low |\n| | Bunjalung National Park | Unknown | 2002 | 2024 | Known | Unknown |\n| | Bylong | Unknown | Unknown | Unknown | Unknown | Moderate |\n| | Carrai National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Chaelundi State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Dalmorton State Forest | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Doyles River State Forest | Unknown | 1984 | 1984 | Unknown | Unknown |\n| | Ellis State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Gibraltar Range State Forest | Unknown | 2015 | 2015 | Unknown | Unknown |\n| | Goobang National Park | Extant | 2025 | 2025 | Unknown | Moderate |\n| | Goulburn River National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Kiwarrak State Forest | Unknown | 2019 | 2019 | Unknown | Unknown |\n| | Malara State Forest | Unknown | 1999 | Unknown | Unknown | Moderate |\n| | Malara State Forest | Unknown | 1993 | 1993 | Unknown | Unknown |\n| | Manobalai Nature Reserve | Extant | 2024 | 2024 | Unknown | Unknown |\n| | Marara State Forest | Unknown | 2014 | 2014 | Unknown | Unknown |\n| | Marengo State Forest | Unknown | 2022 | 2022 | Unknown | Unknown |\n| | Mernot State Forest | Unknown | 2008 | 2008 | Unknown | Unknown |\n| | Mount Royal National Park | Extant | 2023 | 2023 | Unknown | Unknown |\n| | Mummel Gulf National Park | Unknown | 2020 | 2023 | Unknown | Unknown |\n| | Munmorah/Chain Valley | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Myall Lakes | Extant | 2022 | 2022 | Known | Unknown |\n| | Nymboida National Park | Unknown | 2015 | Unknown | Unknown | Moderate |\n| | Olney State Forest | Unknown | 2004 | 2004 | Unknown | Unknown |\n| | Port Stephens | Extant | 2022 | 2022 | Unknown | Moderate |\n| | Ravensworth State Forest | Unknown | 2016 | 2016 | Unknown | Unknown |\n| | Royal National Park | Extant | 2022 | 2022 | Expected decline | Unknown |\n| | Styx River State Forest | Unknown | 2000 | 2000 | Unknown | Unknown |\n| | Watchimbark Reserve | Unknown | 2011 | Unknown | Unknown | High |\n| | Werrikimbe National Park | Unknown | 1984 | 2020 | Unknown | Unknown |\n| | Wollemi National Park | Extant | 2025 | 2025 | Unknown | Unknown |\n| | Wyrrabalong National Park | Unknown | 2013 | 2013 | Unknown | Unknown |\n| | Yiraaldiya National Park feral predator-free area | Translocated, persistence to be determined by post-release monitoring | 2025 | 2025 | Known | Translocation trials with individuals from the captive-breeding colonies at ANU and NTSI (PPC) using NSW founders |\n| | Yuraygir National Park | Extant | 2024 | 2025 | Expected decline | Unknown |\n| QLD | Crows Nest/ Ravensbourne/Helidon Hills | Unknown | 2022 | 2025 | Known | Moderate |\n| | Emu Creek (private property) | Unknown | 2022 | 2024 | Known | Unknown |\n| | Main Range National Park/Spicers Peak Nature Refuge | Unknown | 2022 | 2025 | Known | Unknown |\n| | Numinbah Valley | Unknown | 2005 | 2005 | Unknown | Unknown |\n| TAS | Beaconsfield | Unknown | 1978 | 2002 | Known | Unknown |\n| | Flinders Island | Extant in the north of the island, undetected in the south | 2025 | 2025 | Expected decline | Unknown |\n| | Freycinet | Unknown | 2003 | 2025 | Known | Unknown |\n| | St Helens | Unknown | 2009 | 2025 | Expected decline | Unknown |\n| | Waterhouse Conservation Area | Unknown | 2010 | 2025 | Known | Unknown |\n| | Wukalina/Mt William | Unknown | 2004 | 2025 | Known | Unknown |\n| VIC | Anglesea | Locally extinct | 1999 | 2021 | Known | Very low (from historic samples) |\n| | Cranbourne | Locally extinct | 1976 | Unknown | Known | Unknown |\n| | Dutson Downs | Extant | 2024 | 2025 | Known | Low |\n| | Gippsland Lakes | Extant | 2025 | 2025 | Known | Low |\n| | Hummock Island | Locally extinct | 1977 | Unknown | Known | Unknown |\n| | Langwarrin | Locally extinct | 1984 | Unknown | Known | Unknown |\n| | Mullungdung | Locally extinct | 1976 | 2017 | Known | Unknown |\n| | Providence Ponds | Extant | 2025 | 2025 | Known | Low |\n| | Reeves Beach | Locally extinct | 1978 | 2017 | Known | Unknown |\n| | The Lakes National Park | Extant | 2025 | 2025 | Known | Low |\n| | Tyabb | Locally extinct | 1975 | Unknown | Known | Low (from historic samples) |\n| | Wilsons Promontory | Extant | 2025 | 2025 | Known | Very low |\n\n> Note: a \\- Heterozygosity measures from Burns et al. (2023b).\n\n> Note: Source: Lazenby et al. (2019), Burns (2021a), DCCEEW (2024), NSW BioNet and unpublished data from members of the recovery team and Recovery Network.\n\n## 2.5 Habitat\n\nPookila habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heathy understorey, and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022) (see Figure 3). Historically, the species has been found across a broader array of habitat types (DCCEEW 2024b), and it does not occupy all plausibly suitable habitat at present (Burns et al. 2020; Smith et al. 2023a). Sandy soils suitable for burrowing are common across most occupied habitats and soft sandy substrates, while deep topsoils are thought to be optimal (Abicair et al. 2020). Within the northern parts of the species’ range, it also occupies sites up to 300 km from the coast and in subcoastal locations at higher elevation, where soil composition differs (Martin 1995; Van Dyck & Lawrie 1997; Vernes et al. 2006; Dixon 2014; Churchill et al. 2023).\n\nThe species is often, although not exclusively, found in areas with high floristic diversity with leguminous perennials (Wilson 1991; Seebeck et al. 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; DEWHA 2010). For example, the heathland element of the species’ habitat is represented by dry heathlands that are floristically diverse and even (a high number of plants that are all relatively abundant compared to each other) that grow on sandy soils with a high percentage of bare ground (Seebeck et al. 1996).\n\nThe species has been identified as preferring early successional habitat, at some sites with peak abundance approximately 3–7 years post-fire and 7–9 years post-sand mine rehabilitation (Braithwaite & Gullan 1978; Fox & Fox 1978; Fox & Fox 1984; Twigg et al. 1989; Wilson 1991; Wilson et al. 2018). However, research in Vic suggests that the relationship between the Pookila and fire is complex, and local densities may be high in areas that have not been affected by fires for over 60 years (Burns and Phillips 2020). Similarly, surveys conducted on the NSW central coast have found high densities in long-unburnt vegetation (F Ford 2023, personal communication 8 August). Season, fire severity and extent, predation pressures, and climatic conditions following fire, all affect habitat suitability more strongly than a linear effect of time-since-fire (Burns & Phillips 2020). Furthermore, it is likely that it shelters in burrows during times of fire. This behaviour may be one of the attributes that allows the Pookila to persist in situ through some fire events (Friend 1993; Burns & Phillips 2020).\n\nThe Pookila has been considered a ‘habitat specialist’ (Braithwaite & Gullan 1978), but fossil record observations, and a strategic translocation and assessment of microhabitat use, suggests that it had broad habitat preferences (Smith et al. 2023c). Remains of the species are common in most bone deposits in south-eastern Australia that has been examined, and these deposits occur across a broad range of historical vegetation types and altitudes (Ford 2003). The species’ perceived habitat specialisation may have been further challenged by the Pookila translocation into a predator exclusion zone in the ACT as part of the Mulligans Flat-Goorooyarroo Woodland Experiment. The box gum and grassy woodlands comprising the predator exclusion area are not the type of coastal habitat often associated with contemporary records of the species (Abicair et al. 2020). However, recent detections of the species in grassy open forests and woodlands in Qld means that these non-coastal habitat types are still represented in the Pookila distribution.\n\nThe currently skewed occurrence (and perhaps specificity) of the Pookila towards coastal areas may in part be due to the interaction among threats, such as habitat clearing, altered fire regimes and predation pressures by introduced species such as cats (Felis catus) and European red foxes (Vulpes vulpes). However, it is worth noting that individuals sourced from the ANU breeding colony in the Mulligans Flat population are significantly underweight compared to their founding populations (13.6 g vs. 17.5 g) (Abicair et al. 2020), and that the population may not be self-sustaining. Together, these attributes may indicate that the habitat there may now be suboptimal for the Pookila. However, it may be important to pursue translocations in a diversity of habitats (including those that may be suboptimal) to build the species' resilience to environmental change and to restore ecological functions (Smith et al. 2023a).\n\nFigure 3: Photographs of contemporary Pookila habitat\n\n![These grouped 6 images demonstrate the diversity in the Pookila's habitat throughout the species' range. Refer to the footnote of this collective image for descriptions on habitat per image in addition to subheading 'Habitat' for more information.](image.006.png)\n\n> Note: Contemporary habitat in which the Pookila has been detected. Depicted are: a) grassy woodlands at Cunninghams Gap in Qld (photo: Ian Gynther); b) woodland with a grass and sedge understorey at Mt Hyland, NSW (photo: Anna Lloyd); c) coastal heathland at Munmorah in NSW (photo: Barbara Wilson); d) woodland with a heath understorey at Anglesea in Vic (photo: Barbara Wilson); e) coastal heathland at Loch Sport in Vic (photo: Phoebe Burns); and f) coastal heathland with a woodland overstorey on Flinders Island in Tas (photo: Rosemary Gales) (Source: DNRE Tas 2022).\n\n### 2.5.1 Habitat critical to the survival of the Pookila\n\nAll potential habitat should be considered important to the Pookila’s persistence and recovery, this includes any:\n\n Areas within the species’ modelled distribution;\n Areas where the species previously occurred but has since become locally extinct;\n Patches of suitable habitat that may not be currently occupied by the species but have the potential to be occupied by natural dispersal; and\n Areas that could be restored for reintroduction.\n\nAll areas that contain habitat that serves one or more of the following ecological functions have the potential to be habitat critical to the survival of the Pookila:\n\n Foraging habitat: vegetation with appropriate density and diversity to be able to support a population, such as native vegetation with seasonal abundance of edible flowering, fruiting, and seeding plants. Vegetation cover is important for the species to be able to forage without increased risk of predation. This habitat includes open heathland, tall open forest with a grassy understorey, open woodland with a grassy or heath understorey and vegetated sand dunes (Keith & Calaby 1968; Posamentier & Recher 1974; Fox & Fox 1978; Hocking 1980; Fox & McKay 1981; Pye 1991; Wilson 1991; Van Dyck & Lawrie 1997; Lock & Wilson 1999; Sparshott & Sparshott 2003; Wilson & Laidlaw 2003; Lazenby et al. 2007; Dixon 2014; DNRE Tas 2022).\n\n Breeding habitat: substrate for the species to be able to create burrows, or alternatively, ‘surface nests’ in low-lying vegetation that is suitable for successful reproduction and rearing of young. Soft sandy substrates, with deep topsoils are thought to be optimal (Abicair et al. 2020) but not exclusive. The Pookila mostly occupies coastal areas that provide this substrate. However, it also occupies sites 100–300 km from the coast and areas with higher elevation where soil composition varies. For example, the presence of sandy soils is highly likely in Goobang National Park (NSW NPWS 2001), whilst in the NSW ranges and Northern Tableland areas, soil may vary with a more dominant loamy or quartz soil base where the Pookila occurs or may be present (NSW OEH 2023. pers comm 22 November).\n\n Habitat for the long-term maintenance of the species: including areas that are not subject to prolonged drought or at risk of extreme droughts in future (such as potential climate refugia), and habitat that is maintained through periodic small-scale burns to provide appropriate foraging habitat and connectivity for the species.\n\n Habitat for the maintenance of genetic diversity: includes all current habitats where the species is extant as they contain populations necessary to achieve maximum genetic diversity and increase resilience of the species against future threats. There are two recognised ESUs: the Southern ESU (Vic populations) and the Northern ESU (NSW and Qld populations). Tas may yet be identified as a third ESU or may cluster within the Southern ESU. Therefore, habitat across the species’ range is necessary to maintain the national genetic diversity of the species.\n\n Habitat for translocation: can be identified using habitat modelling followed by ground-truthing at the local-scale and can include locations that have accounted for future changes due to climate change regimes as well as the exclusion of invasive predators.\n\nRefining what constitutes habitat critical to the survival of the Pookila is a priority action in this recovery plan. Until those key characteristics are further clarified, all areas that contain habitat that serves any of the above five ecological functions have the potential to be habitat critical to the survival of the Pookila. Habitat that supports multiple ecological functions and/or a higher number of the ecological attributes identified within the descriptions provided above, increases the likelihood that the habitat may be critical to the survival of the Pookila.\n\nGiven that the species has declined markedly in parts of its range and populations may rapidly change in size, failure to detect the species during surveys does not indicate the absence of potential habitat critical to its survival.\n\n## 2.6 Ecology\n\nThe Pookila often nests underground in communal burrow systems in well-drained soils. Individuals have also been recorded sheltering under low growing shrubs in surface nests (Kemper 1980, Lazenby et al. 2007). Burrows vary in length, depth, and complexity, ranging from shallow surface nests to intricate tunnel systems (Kemper 1981). Of three burrows that have been excavated, the largest was 5 m in length with a maximum depth of 60 cm (Kemper 1981). The species is nocturnal, and with a significant proportion of its time spent foraging above ground, it is exposed to a high risk of predation by native predators including owls and introduced predators such as the cat and the European red fox.\n\nTen individuals were radio-collared in one area of Tas for approximately 2 weeks, yielding information on nightly movements (Lazenby 1999; Lazenby et al. 2007). Over a 2 ha grid, 9 individuals moved up to 206 m in one night, and up to 37 m in one hour (Lazenby 1999; Lazenby et al. 2007). Pookila home ranges have been known to overlap (Lazenby 1999; Lazenby et al. 2007). The species’ home range is approximately 0.5–1.8 ha, increasing as habitat quality decreases (Lock 2005; Burns et al. 2023a).\n\nThe generation length of the Pookila is between one and two years (Kemper 1980). Females reach sexual maturity at approximately 13 weeks and live 2–3 years (Kemper 1976; Kemper 1980). Males reach sexual maturity at approximately 20 weeks and longevity remains unclear (Kemper 1976; Kemper 1980). During good conditions in NSW, the average litter size for the species has been previously recorded as 4.6, with a range of 1–6 young, and with females usually producing 1–2 litters per year, or 3–4 in exceptional conditions (Kemper 1976; Kemper 1980). The gestation period is around one month (32 days) and young are weaned at 3–4 weeks of age (Kemper 1976; Kemper 1980).\n\nThe breeding season varies depending on location and rainfall, with breeding occurring in NSW from late winter to early summer, while in Vic and Tas the breeding season begins in mid-spring and ceases in late summer (Kemper 1980; Burns et al. 2023b). The impact of rainfall is likely related to its influence on resource availability, which influences variation in birth rates, recruitment, and adult survival (Woinarski et al. 2014). Energy requirements are likely to be high during breeding, particularly for females, and if dietary resources are inadequate during this time, females or their young may perish (Lock & Wilson 2017). Hence, the Pookila demonstrates fluctuating population dynamics where extended breeding seasons follow periods of higher rainfall (Lock & Wilson 2017; Burns 2021). Pookila abundance reduces during periods of drought due to resource shortages and likely failed litters (Lock & Wilson 2017; Burns 2021).\n\n### 2.6.1 Diet and foraging\n\nThe Pookila is an omnivore and feeds on a variety of seeds, leaves, fungi, and invertebrates, exhibiting seasonal and locality-dependent dietary variation (Cockburn 1980; Norton 1987; Wilson and Bradtke 1999; Braithwaite and Gullan 1978). The species plays an important ecological role in dispersing seed and fungi in scats (Wilson & Bradtke 1999; Seebeck et al. 2003; Burns et al. 2023b).\n\n \n\n# 3 Threats\n\n## 3.1 Historical causes of decline\n\nThe distribution of subfossils and the genetic structure of extant populations indicate that the historic distribution of the Pookila extended more continuously throughout parts of coastal eastern Australia, from south-east Qld extending down to the northern parts of Tas and up to 300 km inland on the Australian mainland (Ford 2003; Breed and Ford 2007; DEWHA 2010; Burns et al. 2024; Smith et al. 2023c). The species has since experienced significant declines in distribution and population size and is now distributed in fragmented patches across this area (Figure 2).\n\n## 3.2 Current known and suspected threatening processes\n\nCurrent threats and their prevalence and intensity are described for the national Pookila population below. It is important to note that different populations may be experiencing different combinations of threats in varying intensities. There is significant interaction between threats.\n\n### 3.2.1 Habitat loss, disturbance and modification\n\n#### Habitat clearing\n\nLand clearing for agriculture, infrastructure, road, and housing development continues to place Pookila habitat at risk of further loss and fragmentation. As very little targeted and systematic monitoring of Pookila populations has occurred across the range in NSW (DCCEEW 2024b), the species is at particular risk of losing habitat where presence has not yet been confirmed.\n\nIn 2019, 87% of the Australian population lived within 50 km of the coast (cited in Clark et al. 2021). Coastal development is a continuing, increasing pressure with very high impacts on coastal environments (Clark et al. 2021) that is sprawling beyond established urban areas to regional coastal areas (Infrastructure Australia 2020).\n\nWhen habitat is cleared, disturbed, or modified, the risk and magnitude of impact to adjacent extant habitat patches increases, along with the interactive capability of threats (Fischer & Lindenmayer 2007; DNRE Tas 2022). For example, land clearing for agriculture can increase the carrying capacity of native and introduced herbivores within the landscape, leading to increased grazing and soil impact pressures on adjacent habitat. This is exacerbated when fire results in a further increase in grazing pressure and movement as herbivores actively seek to consume newly emerging vegetative shoots (DAWE 2022). With an increasing trend of habitat loss due to land clearing, the likelihood that fragmentation has occurred and will continue to occur is very high.\n\n#### Habitat modification due to Phytophthora cinnamomi infestation\n\nPhytophthora dieback – caused by the root-rot pathogen Phytophthora cinnamomi – is listed as a Key Threatening Process under the EPBC Act (DOEE 2018) and has been observed across the national range of the Pookila (Liu et al. 2013; Wilson et al. 2015; Burns et al. 2020; DCCEEW 2024b). The pathogen is spread through the movement of soil – including on footwear, tools, vehicles and machinery – and in water (DOEE 2018). While the impacts of P. cinnamomi on Pookila populations and habitat have not been investigated directly, the species is likely to be at significant risk. A range of plant species within Pookila habitat have been identified as susceptible to infection by P. cinnamomi (see McDougall 2005).\n\nPhytophthora dieback causes the loss of plant species richness (including keystone species), reduction of structural complexity (Wilson et al. 1994; Garkaklis et al. 2004; Laidlaw & Wilson 2006; Annett 2008; Armistead 2008; Dundas et al. 2013, 2016; Davis et al. 2014; Wilson et al. 2020), loss of food resources (for example, seeds, leaves, nectar, pollen) and suitable refuges from weather and predation (Laidlaw & Wilson 2006). The impacts of Phytophthora dieback can also extend to invertebrate communities and thus impact insectivorous fauna (Cahill et al. 2008), reducing availability of this food resource.\n\nSignificant negative impacts from Phytophthora dieback on small mammals and their habitat has been previously documented (Newell & Wilson 1993; Lilith 2002; Laidlaw & Wilson 2006; Swinburn et al. 2007; Annett 2008; Armistead 2008; Dundas et al. 2013; 2016), and long-term studies have identified localised extinctions of species and mammal community declines (Zhuang-Griffin 2015; Wilson et al. 2017; 2020).\n\nPriority Protection Areas for Phytophthora dieback management have been identified for the Otway Ranges, which was informed by habitats occupied by EPBC Act listed species, including the Pookila (Walshe et al. 2021; 2022; Wilson & Garkaklis 2023a; 2023b).\n\n#### Habitat modification due to native and introduced weeds\n\nChanging grazing and fire regimes have led to habitat loss by encroachment of coast tea tree (Leptospermum laevigatum) in some Pookila habitat (Burns 2019a; Seebeck et al. 2003). This is particularly evident at Wilsons Promontory National Park and in other coastal Vic sites along Ninety Mile Beach where dune succession has given rise to dense thickets of coast tea tree, such as at Reeves Beech. Weed species are of growing concern in other Pookila habitat, for example, in Lakes National Park. Invasive weeds include Asparagus asparagoides (bridal creeper) and Lycium ferocissimum (boxthorn).\n\n### 3.2.2 Fire regimes that cause declines in biodiversity\n\n‘Fire regimes that cause declines in biodiversity’ is listed as a Key Threatening Process under the EPBC Act (DAWE 2022). Altered fire regimes, such as those resulting from changes to seasonality, frequency and severity of fire weather conditions, are causing declines and loss of biodiversity and ecosystem function (DAWE 2022). The Pookila and the habitat the species relies on have a complicated relationship with fire.\n\nUnder certain circumstances, the species can peak in abundance in the years following a fire as resources become more readily available due to regeneration of vegetation, with such areas described as ‘early successional’ (Fox & McKay 1981). Furthermore, the Pookila has been reported to have relatively high survival during and immediately following bushfires, which is also presumably attributed to its ability to shelter in burrows (Friend 1993; Burns & Phillips 2020). However, this response appears to be more common following small-scale, patchy, low-severity burns and is likely reliant on above-average rainfall in the months and years following the fire (Burns & Phillips 2020).\n\nLarge-scale, high-severity fires have the capacity to extirpate entire populations, as occurred in 1983 on Ash Wednesday at Anglesea (Wilson et al. 2018). Fires being too frequent, too large, too severe, or poorly timed may lead to direct loss of individuals and/or sudden or gradual loss of habitat through reduction of habitat patches of suitable successional age, size, distribution, and plant diversity that the Pookila requires for availability of food and shelter from predators.\n\nSimilarly, prolonged exclusion or poorly timed application of fire may encourage coast tea tree encroachment and loss of favourable food and habitat species. Also, introduced predators may target recently burnt areas, compounding the increased risk of predation from loss of vegetative cover (Hradsky et al. 2017). Planned burning may also threaten the species if conducted during or immediately prior to drought, at too high an intensity, or across a widespread area (Wilson et al. 2018).\n\nThe increasing scale, severity, and frequency of fires under a changing climate pose a risk to the Pookila and its habitat. In 2019 and 2020, catastrophic bushfires impacted a large area of eastern Australia with varying severity (Legge et al. 2021). It was estimated that over 45% of the Pookila distribution was burnt, with a third of this area burning with high severity (Legge et al. 2021). The overall population of Pookila was predicted to have declined by 24% by 2021 (one year after fire), with the species suffering an estimated 75% decrease in abundance in severely burnt areas in that same period (Legge et al. 2021).\n\n### 3.2.3 Climate change: Increasing number of hot days, prolonged drought and changes to precipitation patterns\n\nIncreasing frequency, intensity, and duration of drought across much of the species’ range (CSIRO & BoM 2020) may lead to further catastrophic declines and local extinctions. Extreme heat events may exacerbate the effects of drought and could lead to mass mortality events as individuals dehydrate and fail to thermoregulate but this would require further research to confirm whether behaviours would mitigate this threat. The projections for the NSW climate by 2050 include an increase in minimum and maximum temperatures, an increase (double) in the number of hot days per year (days greater than 35°C), and longer fire seasons, with around a 40% increase in ‘very high’ fire danger days (CSIRO & BoM 2021).\n\nExtended periods of low rainfall reduce food availability for Pookila. Recent periods of drought have caused demonstrable crashes in Pookila populations. The local extinction of the Anglesea population is in part attributed to the Millennium Drought (Lock & Wilson 2017), and the species is yet to recover from the 2019 Gippsland drought (Burns 2021).\n\nFire–drought interactions may have a significant impact on the Pookila with potential compounding effects. This is evident from the 2019 and 2020 bushfires, which followed a period of significant drought in 2017 to 2019 (BoM 2020; DPIE 2020).\n\n### 3.2.4 Invasive fauna species\n\n#### Predation by cats and European red foxes\n\nFeral predators can have rapid and catastrophic impacts on populations of threatened rodents (Tuft et al. 2021). While there is limited direct evidence of the consumption of Pookila by feral predators (records in dog scats; Hoskins 1997; Davis et al. 2015; Lock & Wilson 2017; MidCoast Council NSW 2024. pers comm 13 February), it is recognised in the Threat Abatement Plan for Predation by Feral Cats 2024 that the Pookila is likely highly susceptible to predation by cats (and/or in some cases by foxes) (DCCEEW 2024a) and that cats and foxes can significantly impact population numbers (Seebeck et al. 2003; Burns 2019a).\n\nLike many other Pseudomys species, Pookila are docile and display poor awareness of exotic predators (P. Burns, pers. obs.). In such species, selective predation can occur (as reported, for example, on the broad-toothed rat, Mastacomys fuscus; Green 2002), leading to a disproportionate impact of predation on the vulnerable species even when co-occurring species (for example, the house mouse) are at greater abundance. Furthermore, communal burrowing and small local population densities mean that predation events at burrow entrances can cause rapid local declines (Burns 2019a).\n\nThe relative impact of predation by feral cats and foxes on the Pookila is not yet understood. Based on factors such as small body size, lower breeding output, and occupation of open habitat, some researchers have identified the Pookila as having a high threat rating from feral cat predation (Dickman 1996; Stobo-Wilson et al. 2021). Radford et al. (2018) reported the cat and fox threat to Pookila to be approximately the same based on expert opinion, although with weak evidence. While foxes are currently absent from the Tas mainland and Flinders Island (DNRE Tas 2022), and fox control is in place in other parts of the species’ range, effective cat control in open, non-insular systems beyond the local scale is not yet possible and potentially counterproductive (Lazenby et al. 2015; Lazenby et al. 2021; see Predator management).\n\nThe density and predatory impact of cats and foxes can be altered by competitive interactions with other predators. For example, reduced numbers of foxes via fox control can increase cat density, alter cat behaviour, and ultimately increase the impact of predation by feral cats (Risbey et al. 2000; Rees et al. 2023). Tasmanian devils have been shown to alter the behaviour and abundance of feral cats (Hollings et al. 2014; Lazenby & Dickman 2013), and cats have a much higher predatory impact on small mammals compared to devils (Hollings et al. 2014). The decline of devils caused by Devil Facial Tumour Disease (DFTD) is likely to have increased the predatory impacts of feral cats in Pookila habitat on the Tas mainland.\n\n#### Impact of hooved feral animals including deer, goats, horses, and pigs\n\nThe negative impacts of hooved feral animals on vegetation health and soil erosion are well documented (Keith & Pellow 2005; Bengsen et al. 2017; Wills et al. 2023; Burns et al. 2021). This is also recognised within the Draft threat abatement plan for competition and land degradation by unmanaged goats (DCCEEW 2024c), the Threat abatement plan for predation, habitat degradation, competition and disease transmission by feral pigs (Sus scrofa) (DOEE 2017), and the National feral deer action plan (Government of South Australia 2023). Where these species co-occur with the Pookila, there are risks of habitat degradation through trampling, wallowing, and overgrazing (Burns 2021). Overgrazing pressures may increase in post-fire environments, particularly in heathy vegetation types preferred by the Pookila (Heaton et al. 2022), compounding post-fire food shortages and predation risk, while impeding regrowth of suitable habitat and food species (DAWE 2022). The effects of hooved feral animal control programs on Pookila occurrence and abundance have not been directly studied, but based on vegetation response studies, control programs are expected to benefit the Pookila and its habitat, particularly around fire events.\n\n#### Competition/predation from introduced rodents\n\nWhile experimental trials have shown the Pookila to be competitively superior to the invasive house mouse (Fox & Twigg 1991; Haering & Fox 1997), there is growing evidence that house mouse plagues may supress Pookila recovery following periods of drought where resources are subsequently scarce (Burns 2021). Rattus rattus and R. norvegicus (black and brown rats, respectively) may also out-compete the Pookila for resources and prey on adults and juveniles, although this requires further research.\n\n#### Baiting of rabbits and invasive rodents\n\nThe susceptibility of the Pookila to 1080, Pindone, and rodenticide has not been tested. However, such poisons are lethal at low doses to some other Pseudomys species and are expected to be lethal to the Pookila (McIlroy 1982; Eisler 1995; Fisher 1999). Running a lethal dose trial on the Pookila is not appropriate given the species’ threatened status and ethical considerations, so it is safest to apply the Precautionary Principle and assume the species is likely susceptible to 1080, Pindone and rodenticide. Rabbit poison baiting is usually conducted with a carrot or oat bait, both foods readily consumed by the Pookila (Prosser et al. 2007), and it is possible that rabbit-baiting programs have contributed to the decline of the species in some areas (Fisher 1999; Burns 2019a). When finding the balance between protecting Pookila habitat by controlling rabbits and potentially poisoning individual Pookila, it is recommended to entirely avoid poison baiting in areas where Pookila presence is known. This is particularly important for at-risk populations; however, it may also apply to locations where Pookila presence is uncertain but these areas are adjacent to suitable habitat and/or a known population (within a 5 km radius). Rabbit and invasive rodent population control measures should use alternative methods. Control measures should be carefully considered depending on the area of control and the potential side effects for native species. Currently, this threat is considered to be high in Wilsons Promontory in Vic, not applicable in Qld or Tas, and unknown in NSW (DNRE Tas 2023).\n\n### 3.2.5 Small and fragmented populations: Low genetic diversity and inbreeding\n\nHistorical and contemporary habitat degradation and destruction, and the species’ very low dispersal ability have left the Pookila highly fragmented across its distribution, with some populations genetically isolated. This severe fragmentation leaves populations at higher risk of extinction, as a single catastrophic event can more easily impact the total area occupied by a population (Garner et al. 2020).\n\nFurthermore, genetically isolated populations are at higher risk of inbreeding depression. Inbreeding reduces genetic diversity, which reduces adaptability to changing environments and anthropogenic impacts. Inbreeding depression also increases species’ vulnerability to disease while also allowing potentially detrimental traits to surface in the population (Keller 2002). Furthermore, inbreeding depression can cause reduced fecundity and further genetic health issues (for example, chromosomal abnormalities) (Keller 2002).\n\nVic populations (Southern ESU) of the Pookila have relatively low heterozygosity (genetic diversity) and are at risk of inbreeding, with no natural gene flow possible between localities given lack of dispersal (Burns et al. 2023b). Therefore, the Southern ESU is considered an ‘at-risk ESU’. Populations across NSW retain greater heterozygosity and are genetically distinct from those in Vic, although animals in Broadwater National Park appear to have undergone a genetic bottleneck (Burns et al. 2023b). At the time of publishing this recovery plan, two historic mainland Tas samples and a small number of contemporary samples from Flinders Island have been sequenced and are being analysed. Additional Tas samples are being sought through survey work. Genetic mixing within Vic populations has been recommended and is being pursued, whereas no immediate action is required in the Northern ESU (NSW and Qld) (Burns et al. 2023b; Burns et al. 2024) since it is more genetically healthy than the Southern ESU. The recovery team has developed a National Pookila genetic management strategy (Zoos Victoria) and will continue to update it when new information is forthcoming.\n\n#### Disease\n\nThere is limited understanding of the potential prevalence or impact of disease on the Pookila. In general, screening of native rodents for pathogens has been very limited and little information is available on symptomology (Thompson et al. 2010). However, feral cats and invasive rodents carry diseases that could impact the Pookila. In particular, Toxoplasma gondii, a virulent parasite spread by cats, is likely present through much of the species’ former range (Dickson 2018). Previous experimental trials in other Pseudomys species demonstrated rapid, high mortality rates after infection with T. gondii, hence this parasite is likely to also increase the mortality of the Pookila (Burns 2019a).\n\n## 3.3 Threat prioritisation\n\nEach of the threats outlined in this recovery plan has been assessed to determine the risk posed across the national distribution of the Pookila using a risk matrix (see Table 6). The risk matrix considers the likelihood of an incident occurring and the consequences of that incident and uses a qualitative assessment drawing on both peer-reviewed literature and expert opinion.\n\nTable 6: Risk matrix\n\n| Consequence (x) / Likelihood (y) | Not significant | Minor | Moderate | Major | Catastrophic |\n| -------------------------------- | --------------- | ------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------- |\n| Almost certain | Low risk | Moderate risk | Very high riskPredation by European red foxes | Very high riskHabitat clearing (past/present)Fire regimes that cause declines in biodiversityIncreasing number of hot days, prolonged drought and changes to precipitation patternsPredation by cats | Very high risk |\n| Likely | Low risk | Moderate riskImpact of hooved feral animals including deer, goats, horses, and pigs | High riskHabitat modification due to Phytophthora cinnamomi infestationHabitat modification due to native and introduced weeds | Very high riskHabitat clearing (future) | Very high risk |\n| Possible | Low risk | Moderate riskCompetition/predation from introduced rodentsBaiting of rabbits and invasive rodents | High riskLow genetic diversity and inbreedingDisease | Very high risk | Very high risk |\n| Unlikely | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n| Unknown | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n\nRisk matrix key\n\n| Low risk | Moderate risk | High risk | Very-high risk |\n| -------- | ------------- | --------- | -------------- |\n\nDefinition of likelihood and consequence categories\n\nConsequence is a measure of the severity of the impact of the threat on the survival of the species.\n\n- Catastrophic – affecting survival, reproduction or essential movement of individuals in >80% of the population or across >80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Major – affecting survival, reproduction or essential movement of individuals in 50-80 % of the population or across 50-80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Moderate – affecting survival, reproduction or essential movement of individuals in 20-50% of the population or across 20-50% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Minor – affecting survival, reproduction or essential movement of individuals in 5-30% of the population or across 5-30% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Not significant – affecting survival, reproduction or essential movement of individuals in <5% of the population or across <5% of the distribution within the next 3 generations or 10 years, whichever is longer.\n\nLikelihood is a measure of how regularly the threat impacts the species.\n\n- Almost certain – >90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Likely – 66-90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Possible – 33-66% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unlikely – <33% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unknown – probability that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer, could be 0-100%.","sortOrder":29},{"sectionNumber":"Predation by cats and European red foxes","sectionType":"subdivision","heading":"Predation by cats and European red foxes","content":"#### Predation by cats and European red foxes\n\nFeral predators can have rapid and catastrophic impacts on populations of threatened rodents (Tuft et al. 2021). While there is limited direct evidence of the consumption of Pookila by feral predators (records in dog scats; Hoskins 1997; Davis et al. 2015; Lock & Wilson 2017; MidCoast Council NSW 2024. pers comm 13 February), it is recognised in the Threat Abatement Plan for Predation by Feral Cats 2024 that the Pookila is likely highly susceptible to predation by cats (and/or in some cases by foxes) (DCCEEW 2024a) and that cats and foxes can significantly impact population numbers (Seebeck et al. 2003; Burns 2019a).\n\nLike many other Pseudomys species, Pookila are docile and display poor awareness of exotic predators (P. Burns, pers. obs.). In such species, selective predation can occur (as reported, for example, on the broad-toothed rat, Mastacomys fuscus; Green 2002), leading to a disproportionate impact of predation on the vulnerable species even when co-occurring species (for example, the house mouse) are at greater abundance. Furthermore, communal burrowing and small local population densities mean that predation events at burrow entrances can cause rapid local declines (Burns 2019a).\n\nThe relative impact of predation by feral cats and foxes on the Pookila is not yet understood. Based on factors such as small body size, lower breeding output, and occupation of open habitat, some researchers have identified the Pookila as having a high threat rating from feral cat predation (Dickman 1996; Stobo-Wilson et al. 2021). Radford et al. (2018) reported the cat and fox threat to Pookila to be approximately the same based on expert opinion, although with weak evidence. While foxes are currently absent from the Tas mainland and Flinders Island (DNRE Tas 2022), and fox control is in place in other parts of the species’ range, effective cat control in open, non-insular systems beyond the local scale is not yet possible and potentially counterproductive (Lazenby et al. 2015; Lazenby et al. 2021; see Predator management).\n\nThe density and predatory impact of cats and foxes can be altered by competitive interactions with other predators. For example, reduced numbers of foxes via fox control can increase cat density, alter cat behaviour, and ultimately increase the impact of predation by feral cats (Risbey et al. 2000; Rees et al. 2023). Tasmanian devils have been shown to alter the behaviour and abundance of feral cats (Hollings et al. 2014; Lazenby & Dickman 2013), and cats have a much higher predatory impact on small mammals compared to devils (Hollings et al. 2014). The decline of devils caused by Devil Facial Tumour Disease (DFTD) is likely to have increased the predatory impacts of feral cats in Pookila habitat on the Tas mainland.","sortOrder":30},{"sectionNumber":"Impact of hooved feral animals including","sectionType":"subdivision","heading":"Impact of hooved feral animals including deer, goats, horses, and pigs","content":"#### Impact of hooved feral animals including deer, goats, horses, and pigs\n\nThe negative impacts of hooved feral animals on vegetation health and soil erosion are well documented (Keith & Pellow 2005; Bengsen et al. 2017; Wills et al. 2023; Burns et al. 2021). This is also recognised within the Draft threat abatement plan for competition and land degradation by unmanaged goats (DCCEEW 2024c), the Threat abatement plan for predation, habitat degradation, competition and disease transmission by feral pigs (Sus scrofa) (DOEE 2017), and the National feral deer action plan (Government of South Australia 2023). Where these species co-occur with the Pookila, there are risks of habitat degradation through trampling, wallowing, and overgrazing (Burns 2021). Overgrazing pressures may increase in post-fire environments, particularly in heathy vegetation types preferred by the Pookila (Heaton et al. 2022), compounding post-fire food shortages and predation risk, while impeding regrowth of suitable habitat and food species (DAWE 2022). The effects of hooved feral animal control programs on Pookila occurrence and abundance have not been directly studied, but based on vegetation response studies, control programs are expected to benefit the Pookila and its habitat, particularly around fire events.","sortOrder":31},{"sectionNumber":"Competition/predation from introduced ro","sectionType":"subdivision","heading":"Competition/predation from introduced rodents","content":"#### Competition/predation from introduced rodents\n\nWhile experimental trials have shown the Pookila to be competitively superior to the invasive house mouse (Fox & Twigg 1991; Haering & Fox 1997), there is growing evidence that house mouse plagues may supress Pookila recovery following periods of drought where resources are subsequently scarce (Burns 2021). Rattus rattus and R. norvegicus (black and brown rats, respectively) may also out-compete the Pookila for resources and prey on adults and juveniles, although this requires further research.","sortOrder":32},{"sectionNumber":"Baiting of rabbits and invasive rodents","sectionType":"subdivision","heading":"Baiting of rabbits and invasive rodents","content":"#### Baiting of rabbits and invasive rodents\n\nThe susceptibility of the Pookila to 1080, Pindone, and rodenticide has not been tested. However, such poisons are lethal at low doses to some other Pseudomys species and are expected to be lethal to the Pookila (McIlroy 1982; Eisler 1995; Fisher 1999). Running a lethal dose trial on the Pookila is not appropriate given the species’ threatened status and ethical considerations, so it is safest to apply the Precautionary Principle and assume the species is likely susceptible to 1080, Pindone and rodenticide. Rabbit poison baiting is usually conducted with a carrot or oat bait, both foods readily consumed by the Pookila (Prosser et al. 2007), and it is possible that rabbit-baiting programs have contributed to the decline of the species in some areas (Fisher 1999; Burns 2019a). When finding the balance between protecting Pookila habitat by controlling rabbits and potentially poisoning individual Pookila, it is recommended to entirely avoid poison baiting in areas where Pookila presence is known. This is particularly important for at-risk populations; however, it may also apply to locations where Pookila presence is uncertain but these areas are adjacent to suitable habitat and/or a known population (within a 5 km radius). Rabbit and invasive rodent population control measures should use alternative methods. Control measures should be carefully considered depending on the area of control and the potential side effects for native species. Currently, this threat is considered to be high in Wilsons Promontory in Vic, not applicable in Qld or Tas, and unknown in NSW (DNRE Tas 2023).","sortOrder":33},{"sectionNumber":"3.2.5 Small and fragmented populations: ","sectionType":"division","heading":"3.2.5 Small and fragmented populations: Low genetic diversity and inbreeding","content":"### 3.2.5 Small and fragmented populations: Low genetic diversity and inbreeding\n\nHistorical and contemporary habitat degradation and destruction, and the species’ very low dispersal ability have left the Pookila highly fragmented across its distribution, with some populations genetically isolated. This severe fragmentation leaves populations at higher risk of extinction, as a single catastrophic event can more easily impact the total area occupied by a population (Garner et al. 2020).\n\nFurthermore, genetically isolated populations are at higher risk of inbreeding depression. Inbreeding reduces genetic diversity, which reduces adaptability to changing environments and anthropogenic impacts. Inbreeding depression also increases species’ vulnerability to disease while also allowing potentially detrimental traits to surface in the population (Keller 2002). Furthermore, inbreeding depression can cause reduced fecundity and further genetic health issues (for example, chromosomal abnormalities) (Keller 2002).\n\nVic populations (Southern ESU) of the Pookila have relatively low heterozygosity (genetic diversity) and are at risk of inbreeding, with no natural gene flow possible between localities given lack of dispersal (Burns et al. 2023b). Therefore, the Southern ESU is considered an ‘at-risk ESU’. Populations across NSW retain greater heterozygosity and are genetically distinct from those in Vic, although animals in Broadwater National Park appear to have undergone a genetic bottleneck (Burns et al. 2023b). At the time of publishing this recovery plan, two historic mainland Tas samples and a small number of contemporary samples from Flinders Island have been sequenced and are being analysed. Additional Tas samples are being sought through survey work. Genetic mixing within Vic populations has been recommended and is being pursued, whereas no immediate action is required in the Northern ESU (NSW and Qld) (Burns et al. 2023b; Burns et al. 2024) since it is more genetically healthy than the Southern ESU. The recovery team has developed a National Pookila genetic management strategy (Zoos Victoria) and will continue to update it when new information is forthcoming.","sortOrder":34},{"sectionNumber":"Disease","sectionType":"subdivision","heading":"Disease","content":"#### Disease\n\nThere is limited understanding of the potential prevalence or impact of disease on the Pookila. In general, screening of native rodents for pathogens has been very limited and little information is available on symptomology (Thompson et al. 2010). However, feral cats and invasive rodents carry diseases that could impact the Pookila. In particular, Toxoplasma gondii, a virulent parasite spread by cats, is likely present through much of the species’ former range (Dickson 2018). Previous experimental trials in other Pseudomys species demonstrated rapid, high mortality rates after infection with T. gondii, hence this parasite is likely to also increase the mortality of the Pookila (Burns 2019a).","sortOrder":35},{"sectionNumber":"3.3","sectionType":"part","heading":"Threat prioritisation","content":"## 3.3 Threat prioritisation\n\nEach of the threats outlined in this recovery plan has been assessed to determine the risk posed across the national distribution of the Pookila using a risk matrix (see Table 6). The risk matrix considers the likelihood of an incident occurring and the consequences of that incident and uses a qualitative assessment drawing on both peer-reviewed literature and expert opinion.\n\nTable 6: Risk matrix\n\n| Consequence (x) / Likelihood (y) | Not significant | Minor | Moderate | Major | Catastrophic |\n| -------------------------------- | --------------- | ------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------- |\n| Almost certain | Low risk | Moderate risk | Very high riskPredation by European red foxes | Very high riskHabitat clearing (past/present)Fire regimes that cause declines in biodiversityIncreasing number of hot days, prolonged drought and changes to precipitation patternsPredation by cats | Very high risk |\n| Likely | Low risk | Moderate riskImpact of hooved feral animals including deer, goats, horses, and pigs | High riskHabitat modification due to Phytophthora cinnamomi infestationHabitat modification due to native and introduced weeds | Very high riskHabitat clearing (future) | Very high risk |\n| Possible | Low risk | Moderate riskCompetition/predation from introduced rodentsBaiting of rabbits and invasive rodents | High riskLow genetic diversity and inbreedingDisease | Very high risk | Very high risk |\n| Unlikely | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n| Unknown | Low risk | Low risk | Moderate risk | High risk | Very high risk |\n\nRisk matrix key\n\n| Low risk | Moderate risk | High risk | Very-high risk |\n| -------- | ------------- | --------- | -------------- |\n\nDefinition of likelihood and consequence categories\n\nConsequence is a measure of the severity of the impact of the threat on the survival of the species.\n\n- Catastrophic – affecting survival, reproduction or essential movement of individuals in >80% of the population or across >80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Major – affecting survival, reproduction or essential movement of individuals in 50-80 % of the population or across 50-80% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Moderate – affecting survival, reproduction or essential movement of individuals in 20-50% of the population or across 20-50% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Minor – affecting survival, reproduction or essential movement of individuals in 5-30% of the population or across 5-30% of the distribution within the next 3 generations or 10 years, whichever is longer.\n- Not significant – affecting survival, reproduction or essential movement of individuals in <5% of the population or across <5% of the distribution within the next 3 generations or 10 years, whichever is longer.\n\nLikelihood is a measure of how regularly the threat impacts the species.\n\n- Almost certain – >90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Likely – 66-90% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Possible – 33-66% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unlikely – <33% chance that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer.\n- Unknown – probability that threat will have an impact on the species within the next 3 generations or 10 years, whichever is longer, could be 0-100%.","sortOrder":36},{"sectionNumber":"4","sectionType":"part","heading":"Recovery Planning","content":"# 4 Recovery Planning\n\n## 4.1 Vision\n\nIn 2044, wild self-sustaining populations of the Pookila persist in a greater area of occupancy than at present across Qld, NSW, Vic, and Tas. Known threatening processes are well understood and successfully managed in collaboration with First Nations people, land management agencies, conservation organisations and private landowners. The public embrace the Pookila as a native rodent species worth protecting.\n\n## 4.2 Objectives and performance criteria\n\nRecovery of the Pookila is a continuing commitment. The focus of recovery actions is to prevent local extinctions, improve genetic diversity, increase the population size, and expand the area of occupancy by gaining a greater understanding of the ecological requirements of the Pookila, threatening processes and addressing other knowledge gaps.\n\n### Objective 1: The Pookila’s area of occupancy has increased\n\nPerformance Criteria:\n\n- Within 10 years, the area of occupancy has increased within known existing areas with reintroductions into historically occupied areas in Qld, NSW, Vic, and Tas (see Table 7).\n\n### Objective 2: The Pookila’s national population size is stable or increasing\n\nPerformance Criteria:\n\n- Within 10 years, the sizes of identified important populations are maintained or increased from the baseline (see Table 8).\n\n### Objective 3: The genetic diversity of the Pookila is improved\n\nPerformance Criteria:\n\n- Within 10 years, there is increased gene flow within Evolutionarily Significant Units (ESUs). Genetic health of important populations is stable or improving (see Table 9).\n\n### Objective 4: First Nations people are collaborators in the management and recovery of the Pookila\n\nPerformance Criteria:\n\n- Within 10 years, long-term partnerships have been established with First Nations people and representative organisations across the species’ range are collaborators in Pookila recovery efforts. Where they choose to be, First Nations people are leading or actively engaged through partnerships in recovery actions on Country (see Table 10).\n- Where First Nations people have chosen to share knowledge and perspectives on Pookila, these have been respectfully acknowledged, incorporated, and managed as desired by those First Nations people and organisations (see Table 10).\n\n### Objective 5: Community awareness of the Pookila and engagement with recovery efforts have increased\n\nPerformance Criteria:\n\n- Within 10 years, habitat protection has increased by collaborating with landowners and land managers (see Table 11).\n- Surveys demonstrate greater community awareness of the Pookila, the threats to its persistence and the priority actions required to support its recovery (see Table 11).\n- The number of local communities and people involved in Pookila recovery efforts, including campaign attendance, has increased over the life of the recovery plan (see Table 11).\n- Surveys demonstrate greater community awareness of P. cinnamomi and its impacts which contributes to slowing the human-facilitated spread of P. cinnamomi around key locations (see Table 11).\n\n## 4.3 Strategies and recovery actions\n\nActions to achieve the objectives are grouped into strategies are included in the Tables 7–11. Outcome measures are identified against actions to guide their successful implementation. Actions may not specifically meet only one objective, for example, actions under Objective 1 are also relevant to Objective 2.","sortOrder":37},{"sectionNumber":"4.1","sectionType":"part","heading":"Vision","content":"## 4.1 Vision\n\nIn 2044, wild self-sustaining populations of the Pookila persist in a greater area of occupancy than at present across Qld, NSW, Vic, and Tas. Known threatening processes are well understood and successfully managed in collaboration with First Nations people, land management agencies, conservation organisations and private landowners. The public embrace the Pookila as a native rodent species worth protecting.","sortOrder":38},{"sectionNumber":"4.2","sectionType":"part","heading":"Objectives and performance criteria","content":"## 4.2 Objectives and performance criteria\n\nRecovery of the Pookila is a continuing commitment. The focus of recovery actions is to prevent local extinctions, improve genetic diversity, increase the population size, and expand the area of occupancy by gaining a greater understanding of the ecological requirements of the Pookila, threatening processes and addressing other knowledge gaps.","sortOrder":39},{"sectionNumber":"Objective 1: The Pookila’s area of occup","sectionType":"division","heading":"Objective 1: The Pookila’s area of occupancy has increased","content":"### Objective 1: The Pookila’s area of occupancy has increased\n\nPerformance Criteria:\n\n- Within 10 years, the area of occupancy has increased within known existing areas with reintroductions into historically occupied areas in Qld, NSW, Vic, and Tas (see Table 7).","sortOrder":40},{"sectionNumber":"Objective 2: The Pookila’s national popu","sectionType":"division","heading":"Objective 2: The Pookila’s national population size is stable or increasing","content":"### Objective 2: The Pookila’s national population size is stable or increasing\n\nPerformance Criteria:\n\n- Within 10 years, the sizes of identified important populations are maintained or increased from the baseline (see Table 8).","sortOrder":41},{"sectionNumber":"Objective 3: The genetic diversity of th","sectionType":"division","heading":"Objective 3: The genetic diversity of the Pookila is improved","content":"### Objective 3: The genetic diversity of the Pookila is improved\n\nPerformance Criteria:\n\n- Within 10 years, there is increased gene flow within Evolutionarily Significant Units (ESUs). Genetic health of important populations is stable or improving (see Table 9).","sortOrder":42},{"sectionNumber":"Objective 4: First Nations people are co","sectionType":"division","heading":"Objective 4: First Nations people are collaborators in the management and recovery of the Pookila","content":"### Objective 4: First Nations people are collaborators in the management and recovery of the Pookila\n\nPerformance Criteria:\n\n- Within 10 years, long-term partnerships have been established with First Nations people and representative organisations across the species’ range are collaborators in Pookila recovery efforts. Where they choose to be, First Nations people are leading or actively engaged through partnerships in recovery actions on Country (see Table 10).\n- Where First Nations people have chosen to share knowledge and perspectives on Pookila, these have been respectfully acknowledged, incorporated, and managed as desired by those First Nations people and organisations (see Table 10).","sortOrder":43},{"sectionNumber":"Objective 5: Community awareness of the ","sectionType":"division","heading":"Objective 5: Community awareness of the Pookila and engagement with recovery efforts have increased","content":"### Objective 5: Community awareness of the Pookila and engagement with recovery efforts have increased\n\nPerformance Criteria:\n\n- Within 10 years, habitat protection has increased by collaborating with landowners and land managers (see Table 11).\n- Surveys demonstrate greater community awareness of the Pookila, the threats to its persistence and the priority actions required to support its recovery (see Table 11).\n- The number of local communities and people involved in Pookila recovery efforts, including campaign attendance, has increased over the life of the recovery plan (see Table 11).\n- Surveys demonstrate greater community awareness of P. cinnamomi and its impacts which contributes to slowing the human-facilitated spread of P. cinnamomi around key locations (see Table 11).","sortOrder":44},{"sectionNumber":"4.3","sectionType":"part","heading":"Strategies and recovery actions","content":"## 4.3 Strategies and recovery actions\n\nActions to achieve the objectives are grouped into strategies are included in the Tables 7–11. Outcome measures are identified against actions to guide their successful implementation. Actions may not specifically meet only one objective, for example, actions under Objective 1 are also relevant to Objective 2.","sortOrder":45},{"sectionNumber":"5","sectionType":"part","heading":"Conservation management","content":"# 5 Conservation management\n\nThe Pookila has been the subject of much conservation management activity for more than 20 years. Previous monitoring, surveying and management actions have largely been implemented with state-specific focus until the recovery team was formed in 2021. Previous conservation effort has built a substantial legacy and foundation for current conservation activity and has allowed for the development of future recovery action priorities.","sortOrder":46},{"sectionNumber":"5.1","sectionType":"part","heading":"Surveying and monitoring","content":"## 5.1 Surveying and monitoring\n\nConducting regular detection surveys in Pookila habitat and regular monitoring to determine a baseline of population numbers are priorities for Pookila recovery. Further detection will inform management practices in newly identified Pookila populations and may increase protection of Pookila habitat from land clearing and degradation.\n\nThe Australian Government in partnership with the DNRE Tas, funded a national online workshop in November 2021 to consider potential causes for decline of the species, compare management practices across the species’ national range and identify key knowledge gaps. An output of this workshop was the development of the ‘Management and Survey Guidelines’ (DNRE Tas 2022) which will assist in standardising survey efforts across the national range.\n\nIn Vic, a comprehensive survey of historic Pookila sites was performed from 2015 to 2018, with only 5 of the historic 12 parks and reserves detecting current occupancy (Burns 2019a). The current focus for Vic is ongoing monitoring of the extant populations, habitat surveys to understand floristic and structural habitat features, fire management where species-focused mosaic burns are being implemented, genetic maintenance of extant populations, establishment and maintenance of the captive-breeding program and a reintroduction to Cranbourne Botanic Gardens.\n\nConsistent surveying methods across the national range of the species will allow for collection of comparable data which can provide an overall insight into population trends. A camera-trapping protocol is under development to assist with future surveying and monitoring across the species’ range (see Figure 4). This protocol will support the ‘Management and Survey Guidelines’. \n\nFigure 4 Camera trap and bait canister set-up to improve detection\n\n![A collective image of four images that demonstrates the ideal camera trap and bait canister set-up for detecting Pookila. Refer to the footnote of this collective image for a description of each individual image, in addition to referring to subheading 'Surveying and monitoring'. ](image.007.png)\n\n> Note: A camera and bait canister set-up aimed at maximising the identification rate of small mammals and resulting photographs of the Pookila using a) an elevated bait cannister and Reconyx Hyperfire 2 white flash camera showing i. the camera and bait station set-up and ii. resulting image of a Pookila, and b) a ground mounted cannister and Bolyguard SG 2060-D white flash camera showing i. the camera and bait station set-up, and ii. resulting image of a Pookila (Source: DNRE Tas 2022).","sortOrder":47},{"sectionNumber":"5.2","sectionType":"part","heading":"Captive-breeding and translocations","content":"## 5.2 Captive-breeding and translocations\n\nCurrent captive-breeding efforts separately manage Pookila of Vic and NSW provenance to account for the two genetically distinct ESUs.\n\nAs of 2023, a translocated population made up of individuals from the ANU breeding colony existed at Mulligans Flat Woodland Sanctuary (MFWS), (Abicair et al. 2020; Smith et al. 2023a). Monitoring programs have been implemented to track the population’s progress. Learnings from MFWS translocations will inform future conservation decisions.\n\nIn Victoria, eleven breeding pairs were collected and used to establish a Vic-ESU conservation breeding program in early 2022 (Victoria State Government 2022) with further collections of seven animals in 2023, and two animals in 2024 (P Burns unpub. data). This population provides some insurance against loss of Pookila in the wild, although is not large enough to constitute an ‘insurance population,’ and will provide individuals to use for supplementation of extant populations and reintroductions. Mate choice pairings, informed by genetic analyses, are used to ensure genetically healthy offspring are produced at breeding facilities located at Melbourne Zoo (ZV) and Moonlit Sanctuary Wildlife Conservation Park. Initial activities are focused on genetic rescue for the population at Wilsons Promontory, as it has shown to have very low relative heterozygosity (Table 3), and reintroduction into the fenced RBGV Cranbourne Annex. The Vic-sourced captive population is managed as a ZAA Species Management Program, with ZV as program hosts and fulfilling Species Coordinator and Studbook Keeper roles.\n\nAt the RBGV, Pookila were last recorded in 1976. The site has been prepared to ensure habitat is suitable, and a long-term fire regime will be designed to maximise suitability over time. In Autumn 2024, the first animals were released into the existing predator-proof fenced area at the RBGV Cranbourne Annex to establish a semi-wild insurance population. The release, and ongoing work with a second release in June 2025, is being conducted in accordance with appropriate permits and Animal Ethics, and agreement between TEP and the recovery team. Supplemental food and artificial nesting structures have been provided until monitoring indicates they are no longer required. Additional releases are planned to bolster numbers and maintain suitable genetic diversity.\n\nReintroduction of Pookila individuals, sourced from the captive-breeding colonies at ANU and Priam Psittaculture Centre, is underway to a feral predator-free area in NSW, with the aim to expand the species’ area of occupancy and facilitate research that will benefit the Pookila and similar species.\n\nFuture translocations, whether new programs or the continuation of existing programs (for example, RBGV Cranbourne Annex; Mulligans Flat Woodland Sanctuary, NPWS NSW Feral Predator-free Area Program), should be subject to approval on the grounds that there is demonstrable conservation benefit, permit and ethics approval, relevant federal and state and territory government policy is adhered to, and that the action is informed by the IUCN guidelines.\n\nThe Southern and Northern ESUs are to be managed as separate genetic units. Translocations from Northern ESU states (NSW & Qld) to the Southern ESU (Vic) and Tas, and vice versa, are not recommended, at this stage, due to significant genetic and morphological differences (Burns et al. 2024).","sortOrder":48},{"sectionNumber":"5.3","sectionType":"part","heading":"Fire management","content":"## 5.3 Fire management\n\nThe Pookila has previously been described to prefer early successional habitat. Burn severity, timing and size, as well as predation pressures and climatic conditions following fire may all impact habitat suitability (Burns & Phillips 2020). Fire management for the species and its associated local ecological communities, needs to be undertaken in a way that will maintain pockets of suitable habitat, meet fuel reduction needs and mitigate risk of high intensity bushfires. Species-appropriate fire management practices have been trialled and implemented in the Gippsland Lakes Coastal Park. The planned burn provided unburnt refuge areas for the Pookila through the application of slow burns with controlled ignition which targeted selected blocks of habitat. Pre- and post-burn surveys of Pookila sites indicated that the fire management approach applied was successful, with detection of individuals post-fire at all sites that were previously occupied, essentially meaning that the species persisted across the burn area. This planned burn research has allowed for further understanding of fire management practices that may be beneficial for the species (Forest Fire Management Victoria 2022).","sortOrder":49},{"sectionNumber":"5.4","sectionType":"part","heading":"Predator management","content":"## 5.4 Predator management\n\nGeneral fox management programs are implemented in national parks and managed areas across the species’ range, supported by the Australian Pest Animal Strategy and the Threat Abatement Plan for Feral Cats. These control measures may contribute to protection of the Pookila by reducing fox predation pressure. However, further research is underway to understand and improve the efficiency of feral cat control. Current techniques for cat control are often labour intensive, expensive and have limited success in eradicating or controlling cat populations (NSW OEH 2012). Importantly, they can have unintended consequences and result in temporary increases in cat activity and abundance (Lazenby et al. 2015). Cats are cautious of traps and are also difficult to shoot, and where baits are utilised, there are potential non-target species effects (NSW OEH 2012; Sharp & Saunders 2012). Therefore, advanced control methods are currently being explored, such as the Felixer, an automated device that uses sensors to detect target animals and dispenses a toxic gel on the animal’s fur, to be consumed when the animal subsequently grooms itself (Thylation 2020). Other alternative methods focus on coexistence conservation using Mini Safe Havens, which may facilitate the learning and adaptation of prey species, like the Pookila, to the presence of invasive predators (see Evans et al. 2022; Smith et al. 2023c).","sortOrder":50},{"sectionNumber":"5.5","sectionType":"part","heading":"Invasive plant and pathogen management","content":"## 5.5 Invasive plant and pathogen management\n\nThe removal of invasive plant species that may impact Pookila habitat, and further limiting the spread of weeds, are identified as priorities for the species’ recovery. However, any potential threats and impacts from weeds remains poorly understood, are a significant research gap and require evaluation to support management decisions. The management of P. cinnamomi currently includes the application of phosphite and prevention of the introduction of this pathogen into uninfected areas (DOEE 2018). Under the Australian Government’s Wild Otway initiative, the Phytophthora management program is implemented for long-term protection of diversity in the region, which includes areas where the Pookila historically inhabited. Current practices include aerial spraying with phosphite, barriers to protect uninfected areas, an educational campaign for visitors to spray shoes and treatment gates for bike riders (Corangamite Catchment Management Authority 2020). Further collaboration within the recovery team will facilitate learnings from this program that may be applied to other areas of Pookila habitat that require Phytophthora management.","sortOrder":51},{"sectionNumber":"References","sectionType":"part","heading":"References","content":"# References\n\nAbicair K, Manning AD, Ford F, Newport J & Banks SC (2020) Habitat selection and genetic diversity of a reintroduced ‘refugee species’. Animal Conservation 23, 330–341.\n\nAnnett K (2008) Effects of Phytophthora cinnamomi on Heathland Fauna and Ecosystem Function. PhD Thesis, Deakin University, Geelong, Victoria.\n\nArmistead R (2008) The impact of Phytophthora cinnamomi on yellow footed antechinus (mardo). PhD Thesis, Murdoch University, Perth, Western Australia.\n\nAvise JC (1989) A role for molecular genetics in the recognition and conservation of endangered species. Trends in Ecology & Evolution 4, 279–281.\n\nBengsen AJ, West P & Krull CR (2017) Feral pigs in Australia and New Zealand: range, trend, management and impacts of an invasive species. Ecology, conservation and management of wild pigs and peccaries, 325–338.\n\nBraithwaite RW & Gullan PK (1978) Habitat selection by small mammals in a Victorian heathland. Austral Ecology 3, 109–127.\n\nBraithwaite RW, Morton SR, Burbidge AA & Calaby JH (1995) Australian names for Australian rodents, editor R Longmore. Australian Nature Conservation Agency & CSIRO Division of Wildlife and Ecology, Canberra.\n\nBreed B & Ford F (2007) Native Mice and Rats. CSIRO Publishing, Collingwood, Victoria.\n\nBrown A, Cox T, Wishart J & Harris C (ed) (2020) Glovebox Guide for Managing Rabbits. PestSmart Toolkit publication. The Centre for Invasive Species Solutions, Canberra, ACT.\n\nBurns PA (2019a) Testing the decline of the New Holland mouse (Pseudomys novaehollandiae) in Victoria. Australian Mammalogy 42, 185–193.\n\nBurns PA (2019b) Small mammal camera trap image identification guide - Gippsland, Grampians, Otways. University of Melbourne. Accessed 24 October 2024.\n\nBurns PA (2021) Status of the New Holland Mouse in Gippsland and the role of drought in population fluctuations. Report to East Gippsland Catchment Management Authority. Native Mouse Ecological Consulting, Melbourne.\n\nBurns PA, Ford F, Gynther IC, Hill R, Lazenby B, Lloyd A, Parrott ML, Smith L, Tolhurst B, Wilson BA & Roycroft E (2024) National Pookila genetic management strategy. Pookila Recovery Team.\n\nBurns H, Gibbons P & Claridge A (2021) Quantifying variations in browsing pressure caused by feral deer for a range of threatened ecological communities and plant growth forms. Austral Ecology 46(7), 1156–1169.\n\nBurns PA, Kemper CM & Wilson BA (2023a) New Holland Mouse Pseudomys novaehollandiae. In AM Baker & IC Gynther (eds) Strahan’s Mammals of Australia, 4th ed. New Holland Publishers: Sydney, Australia.\n\nBurns PA & Phillips BL (2020) Time since fire is an over-simplified measure of habitat suitability for the New Holland mouse. Journal of Mammalogy 101, 476–486.\n\nBurns PA, Rowe KC, Parrott ML & Roycroft EJ (2023b) Population genomics of decline and local extinction in the endangered Australian Pookila. Biological Conservation 284, 110183.\n\nCahill DM, Rookes JE, Wilson BA, Gibson L & McDougall KL (2008) Phytophthora cinnamomi and Australia’s biodiversity: impacts, predictions and progress towards control. Australian Journal of Botany 56, 279–310.\n\nChurchill TB, Laidlaw MJ, McCall AH, Hines HB, Gynther IC, Molyneux J, Teixeira D, Stewart D, Rix MG, Burwell CJ, Lambkin CL, Venz MF, Zadkovich J, Cooper S, Buch W, O’Connell J, Hughes D, Finlayson S, Gurra A, & Job C (2023) Bushfire Recovery Program 2020–2022: Priority actions for threatened species in the Gondwana Rainforests of Australia World Heritage Area, South East Queensland. Queensland Department of Environment and Science, Brisbane.\n\nClark G, Fischer M & Hunter C (2021) Australia state of the environment 2021: coasts. Independent report to the Australian Government Minister for the Environment, Commonwealth of Australia, Canberra.\n\nCockburn A (1980) The diet of the New Holland mouse (Pseudomys novaehollandiae) and the house mouse (Mus musculus) in a Victorian coastal heathland. Australian Mammalogy 3, 31–34.\n\nCorangamite Catchment Management Authority (2020) Australian Government’s Wild Otway initiative: Phytophthora Management. Accessed 24 October 2024.\n\nCSIRO & BoM (Commonwealth Scientific and Industrial Research Organisation & Bureau of Meteorology) (2020) State of the Climate 2020. Australian Government.\n\nCSIRO & BoM (Commonwealth Scientific and Industrial Research Organisation & Bureau of Meteorology) (2021) New South Wales' and ACT's Changing Climate. Accessed 24 October 2024.\n\nDAFF (2023). Regional Forest Agreements. Department of Agriculture, Fisheries and Forestry, Canberra. Available here: Regional Forest Agreements - DAFF (agriculture.gov.au).\n\nDavis NE, Forsyth DM, Triggs B, Pascoe C, Benshemesh J, Robley A, Lawrence J, Ritchie EG, Nimmo DG & Lumsden LF (2015) Interspecific and geographic variation in the diets of sympatric carnivores: Dingoes/Wild Dogs and Red Foxes in South-Eastern Australia. PLOS ONE 10, e0120975.\n\nDavis RA, Valentine LE, Craig MD, Wilson B, Bancroft WJ & Mallie M (2014) Impact of Phytophthora-dieback on birds in Banksia woodlands in south west Western Australia. Biological Conservation 171, 136–144.\n\nDAWE (Department of Agriculture, Water and the Environment) (2022) Fire regimes that cause declines in biodiversity. Department of Agriculture, Water and the Environment, Canberra, April. CC BY 4.0.\n\nDCCEEW (Department of Climate Change, Energy, the Environment and Water) (2024a) Threat Abatement Plan for Predation by Feral Cats. Department of Climate Change, Energy, the Environment and Water, Canberra.\n\nDCCEEW (Department of Climate Change, Energy, the Environment and Water) (2024b) Conservation Advice for Pseudomys novaehollandiae (New Holland mouse, Pookila). Department of Climate Change, Energy, the Environment and Water, Canberra.\n\nDCCEEW (Department of Climate Change, Energy, the Environment and Water) (2024c) Draft threat abatement plan for competition and land degradation by unmanaged goats. Department of Climate Change, Energy, the Environment and Water, Canberra.\n\nDEWHA (Department of the Environment, Water, Heritage and the Arts) (2010) Approved Conservation Advice for Pseudomys novaehollandiae (New Holland Mouse). Department of the Environment, Water, Heritage and the Arts, Canberra.\n\nDickman CR (1996) Overview of the impacts of feral cats on Australian native fauna. Australian Nature Conservation Agency, Canberra.\n\nDickson JA (2018) The distribution of Toxoplasma gondii in Australia. Honours Thesis, University of Melbourne, Parkville, Australia.\n\nDixon K (2014) Habitat use and the effects of fire and grazing on the Hastings River mouse, brown antechinus, bush rat and swamp rat. BappSc Honours Thesis, The University of Queensland, Gatton.\n\nDOEE (Department of the Environment and Energy) (2017) Background document for the Threat abatement plan for predation, habitat degradation, competition and disease transmission by feral pigs (Sus scrofa). Department of the Environment and Energy, Canberra.\n\nDOEE (Department of the Environment and Energy) (2018) Threat abatement plan for disease in natural ecosystems caused by Phytophthora cinnamomi. Department of the Environment and Energy, Canberra.\n\nDNRE Tas (Department of Natural Resources and Environment, Tasmania) (2022) Management and survey guidelines for wild populations of New Holland Mouse, Pookila – Pseudomys novaehollandiae. DNRE Tas, Hobart.\n\nDNRE Tas (Department of Natural Resources and Environment, Tasmania) (2023) New Holland Mouse, Pookila, National Conservation Assessment Workshop 24th and 25th November 2021, Workshop Proceedings. Nature Conservation Report 2023/02. DNRE Tas, Hobart.\n\nDPIE (Department of Planning, Industry and Environment) (2020) Drought in NSW. Accessed 14 September 2020.\n\nDundas SJ, Fleming PA & Hardy GESJ (2013) Flower visitation by honey possums (Tarsipes rostratus) in a coastal Banksia heathland infested with the plant pathogen Phytophthora cinnamomi. Australian Mammalogy 35, 166–174.\n\nDundas SJ, Fleming PA & Hardy GESJ (2016) The plant pathogen Phytophthora cinnamomi influences habitat use by the obligate nectarivore honey possum (Tarsipes rostratus). Australian Journal of Zoology 64, 122–131.\n\nEisler R (1995) Sodium monofluoroacetate (1080) hazards to fish, wildlife, and invertebrates: a synoptic review. No. 30. US Department of the Interior, National Biological Service, United States.\n\nEvans MJ, Weeks AR, Scheele BC, Gordon IJ, Neaves LE, Andrewartha TA, Brockett B, Rapley S, Smith KJ, Wilson BA & Manning AD (2022) Coexistence conservation: reconciling threatened species and invasive predators through adaptive ecological and evolutionary approaches. Conservation Science and Practice 4(7), e12742.\n\nFischer J & Lindenmayer DB (2007) Landscape modification and habitat fragmentation: a synthesis. Global ecology and biogeography 16(3), 265–280.\n\nFisher P (1999) Factors affecting the risk of poisoning to New Holland Mouse (Pseudomys novaehollandiae) from rabbit oat baits. Victorian Institute of Animal Science, Victoria.\n\nFord F (2003) Conilurine Rodent Evolution: The role of ecology in modifying the evolutionary consequences of environmental change. PhD Thesis, James Cook University, Townsville.\n\nFord F (2023) Personal communication by email, 8 August 2023. NSW Department of Planning and Environment.\n\nForest Fire Management Victoria (2022) Pookila: masterful mice worth protecting. Media release, 19 October 2022.\n\nFox BJ & Fox MD (1978) Recolonization of coastal heath by Pseudomys novaehollandiae (Muridae) following sand mining. Austral Ecology 3, 447–465.\n\nFox BJ & Fox MD (1984) Small-mammal recolonization of open-forest following sand mining. Austral Ecology 9, 241–252.\n\nFox BJ & McKay GM (1981) Small mammal responses to pyric successional changes in eucalypt forest. Australian Journal of Ecology 6, 29–41.\n\nFox BJ & Twigg LE (1991) Experimental transplants of mice (Pseudomys and Mus) on to early stages of post-mining regeneration in open forest. Austral Ecology 16, 281–287.\n\nFriend GR (1993) Impact of fire on small vertebrates in mallee woodlands and heathlands of temperate Australia: A review. Biological Conservation 65, 99–114.\n\nGarkaklis MJ, Calver MC, Wilson BA, & Hardy GESJ (2004) Habitat alteration caused by introduced plant disease: a significant threat to the conservation of Australian forest fauna. In D Lunney (ed) Management of Australia’s Forest Fauna. Royal Zoological Society of New South Wales, Mosman, NSW.\n\nGarner BA, Hoban S & Luikart G (2020) IUCN Red List and the value of integrating genetics. Conservation Genetics 21, 795–801.\n\nGovernment of South Australia (2023) National Feral Deer Action Plan 2023–2028. Government of South Australia.\n\nGreen K (2002) Selective predation on the broad-toothed rat, Mastacomys fuscus (Rodentia: Muridae), by the introduced red fox, Vulpes vulpes (Carnivora: Canidae), in the Snowy Mountains, Australia. Austral Ecology 27, 353–359.\n\nHaering R & Fox BJ (1997) Habitat use by sympatric populations of Pseudomys novaehollandiae and Mus domesticus in coastal heathland. Austral Ecology 22, 69–80.\n\nHeaton DJ, McHenry MT & Kirkpatrick JB (2022) The fire and fodder reversal phenomenon: vertebrate herbivore activity in burned and unburned Tasmanian ecosystems. Fire 5(4), 111.\n\nHercus LA (n.d.) Southern Ngarigu vocabulary. AIATSIS Australian Indigenous Languages Collection. Accessed 24 October 2024.\n\nHocking GJ (1980) The occurrence of the New Holland Mouse, Pseudomys novaehollandiae (Waterhouse), in Tasmania. Wildlife Research 7, 71–77.\n\nHollings T, Jones M, Mooney N & McCallum H (2014) Trophic cascades following the disease‐induced decline of an apex predator, the Tasmanian devil. Conservation biology 28(1), 63–‍75.\n\nHoskins M (1997) Fox baiting for the protection of the New Holland Mouse. Report to the New Holland Mouse Recovery Team. Parks Victoria, Foster, Victoria.\n\nHradsky BA, Mildwaters C, Ritchie EG, Christie F & Di Stefano J (2017) Responses of invasive predators and native prey to a prescribed forest fire. Journal of Mammalogy 98, 835–847.\n\nInfrastructure Australia (2020) Infrastructure priority list, project and initiative summaries. Infrastructure Australia, Canberra. Accessed 24 October 2024.\n\nIUCN/SSC (2013) Guidelines for Reintroductions and Other Conservation Translocations. Version 1.0. Gland, Switzerland: IUCN Species Survival Commission, viiii & pp. 57.\n\nKeith K & Calaby JH (1968) The New Holland Mouse, Pseudomys novaehollandiae (Waterhouse), in the Port Stephens district, New South Wales. CSIRO Wildlife Research 13, 45–58.\n\nKeith D & Pellow B (2005) Effects of Javan rusa deer (Cervus timorensis) on native plant species in the Jibbon-Bundeena area, Royal National Park, New South Wales. In Proceedings of the Linnean Society of New South Wales 126, 99–110.\n\nKeller L (2002) Inbreeding effects in wild populations. Trends in Ecology & Evolution 17, 230–‍241.\n\nKemper CM (1976) Reproduction of Pseudomys novaehollandiae (Muridae) in the laboratory. Australian Journal of Zoology 24, 159–167.\n\nKemper CM (1980) Reproduction of Pseudomys novaehollandiae (Muridae) in the wild. Wildlife Research 7, 385–402.\n\nKemper C (1981) Description of Pseudomys novaehollandiae burrows located with radioisotopes. Australian Mammalogy 4, 141–143. doi:10.1071/AM81008.\n\nLaidlaw WS & Wilson BA (2006) Habitat utilisation by small mammals in a coastal heathland exhibiting symptoms of Phytophthora cinnamomi infestation in the eastern Otway Ranges, Victoria. Wildlife Research 33, 639–649.\n\nLazenby BT (1999) Vegetation associations and spatial relations in the New Holland Mouse, Pseudomys novaehollandiae (Rodentia: Muridae) in Tasmania. Honours Thesis, University of Tasmania.\n\nLazenby BT, Bell P, Driessen MM, Pemberton D & Dickman CR (2019) Evidence for a recent decline in the distribution and abundance of the New Holland mouse (Pseudomys novaehollandiae) in Tasmania, Australia. Australian Mammalogy 41, 179–185.\n\nLazenby BT & Dickman CR (2013) Patterns of detection and capture are associated with cohabiting predators and prey. PLoS One, 8(4), p.e59846.\n\nLazenby BT, Mooney NJ & Dickman CR (2015) Effects of low-level culling of feral cats in open populations: a case study from the forests of southern Tasmania. Wildlife Research 41(5), 407–420.\n\nLazenby BT, Mooney NJ & Dickman CR (2021) Raiders of the last ark: the impacts of feral cats on small mammals in Tasmanian forest ecosystems. Ecological Applications 31(6), p.e02362.\n\nLazenby BT, Pye T, Richardson A & Bryant SA (2007) Towards a habitat model for the New Holland Mouse Pseudomys novaehollandiae in Tasmania - population vegetation associations and an investigation into individual habitat use. Australian Mammalogy 29, 137–148.\n\nLegge S, Rumpff L, Woinarski JCZ, Whiterod NS, Ward M, Southwell DG, Scheele BC, Nimmo DG, Lintermans M, Geyle H, Garnett ST, Hayward-Brown B, Ensbey M, Ehmke G, Ahyong ST, Blackmore CJ, Bower DS, Brizuela-Torres D, Burbidge AH, Burns PA, Butler G, Catullo R, Chapple DG, Dickman CR, Doyle K, Ferris J, Fisher D, Gallagher R, Gillespie GR, Greenlees MJ, Hohnen R, Hoskin CJ, Hunter D, Jolly C, Kennard M, King A, Kuchinke D, Law B, Lawler I, Lawler S, Loyn R, Lunney D, Lyon J, MacHunter J, Mahony M, Mahony S, McCormack RB, Melville J, Menkhorst P, Michael D, Mitchell N, Mulder E, Newell D, Pearce L, Raadik TA, Rowley J, Sitters H, Spencer R, Valavi R, West M, Wilkinson DP & Zukowski S (2022) The conservation impacts of ecological disturbance: time-bound estimates of population loss and recovery for fauna affected by the 2019-20 Australian megafires. Global Ecology Biogeography 31(10), 2085–2104.\n\nLegge S, Woinarski JCZ, Garnett ST, Geyle H, Lintermans, M, Nimmo DG, Rumpff L, Scheele BC, Southwell DG, Ward M, Whiterod NS, Ahyong ST, Blackmore CJ, Bower DS, Brizuela-Torres D, Burbidge AH, Burns PA, Butler G, Catullo R, Dickman CR, Doyle K, Ehmke G, Ensbey M, Ferris J, Fisher D, Gallagher R, Gillespie GR, Greenlees MJ, Hayward-Brown B, Hohnen R, Hoskin CJ, Hunter D, Jolly C, Kennard M, King A, Kuchinke D, Law B, Lawler I, Lawler S, Loyn R, Lunney D, Lyon J, MacHunter J, Mahony M, Mahony S, McCormack RB, Melville J, Menkhorst P, Michael D, Mitchell N, Mulder E, Newell D, Pearce L, Raadik TA, Rowley J, Sitters H, Spencer R, Valavi R, West M, Wilkinson DP & Zukowski S (2021) Estimates of the impacts of the 2019–20 fires on populations of native animal species. Project 8.3.2 report. NESP Threatened Species Recovery Hub, Brisbane, Australia.\n\nLilith M (2002) The effect of plant disease on fauna and validation of trapping methodology in the Darling range forest ecosystem. BSc Hons, Murdoch University, Western Australia.\n\nLiu C, White M, Newell G & Griffioen P (2013) Species distribution modelling for conservation planning in Victoria, Australia. Ecological Modelling 249, 68–74.\n\nLock ML (2005) Population ecology and captive breeding of Pseudomys novaehollandiae at Anglesea. PhD Thesis, Deakin University Geelong, Victoria.\n\nLock ML & Wilson BA (1999) The distribution of the New Holland Mouse (Pseudomys novaehollandiae) with respect to vegetation near Anglesea, Victoria. Australian Wildlife Research 26, 565–577.\n\nLock M & Wilson BA (2017) Influence of rainfall on population dynamics and survival of a threatened rodent (Pseudomys novaehollandiae) under a drying climate in coastal woodlands of south-eastern Australia. Australian Journal of Zoology 65, 60–70.\n\nManning A (2023) Personal communication by email, 4 August 2023. Australian National University.\n\nMartin A (1995) Flora and Fauna Impact Statement, Timbarra Gold Project via Tenterfield. Report prepared for Capricornia Prospecting, Pty Ltd.\n\nMcDougall KL (2005) Appendix 4: The responses of native Australian plant species to Phytophthora cinnamomi. In O’Gara E, Howard K, Wilson B, Hardy GESJ (eds) Management of Phytophthora cinnamomi for Biodiversity Conservation in Australia: Part 2. National Best Practice. Department of the Environment and Heritage, Canberra.\n\nMcIlroy JC (1982) The sensitivity of Australian animals to 1080 poison IV. Native and introduced rodents. Wildlife Research 9, 505–517.\n\nMenkhorst PW, Dickman CR, Denny MW, Aplin K, Lunney D & Ellis M (2008) Pseudomys novaehollandiae: The IUCN Red List of Threatened Species 2010. Version 2010.4. IUCN Red List of Threatened Species.\n\nMenkhorst P & Knight F (2011) Field Guide to the Mammals of Australia 3rd ed. Oxford University Press, South Melbourne.\n\nMidCoast Council NSW (2024) Personal communication by email, 13 February 2024. MidCoast Council NSW.\n\nMoritz C (1994) Defining ‘evolutionarily significant units’ for conservation. Trends in Ecology & Evolution 9, 373–375.\n\nNewell GR & Wilson BA (1993) The relationship between cinnamon fungus (Phytophthora cinnamomi) and the abundance of Antechinus stuartii (Dasyuridae: Marsupialia) in the Brisbane Ranges, Victoria. Wildlife Research 20, 251–59.\n\nNorton TW (1987) The Ecology of Small Mammals in Northeastern Tasmania. II. Pseudomys novaehollandiae and the introduced Mus musculus. Wildlife Research 14, 435–441.\n\nNSW NPWS (National Parks and Wildlife Service) (2001) Goobang National Park Plan of Management. NSW National Parks and Wildlife Service. Department of Planning and Environment, Parramatta.\n\nNSW OEH (Office of Environment & Heritage) (2012) Regional Pest Management Strategy 2012–17, North Coast Region: a new approach for reducing impacts on native species and park neighbours. NSW Office of Environment and Heritage, Sydney.\n\nNSW OEH (Office of Environment & Heritage) (2023) Personal communication by email, 22 November 2023.\n\nPosamentier HG & Recher HF (1974) The status of Pseudomys novaehollandiae (the New Holland mouse). Australian Journal of Zoology 18, 66–71.\n\nProsser S, Lock M, Bodley K, Groat J, Myronuik P & Courtney P (2007) Husbandry manual for the New Holland Mouse Pseudomys novaehollandiae. Zoos Victoria, Parkville, Melbourne.\n\nPye T (1991) The New Holland mouse (Pseudomys novaehollandiae) (Rodentia: Muridae) in Tasmania: a field study. Wildlife Research 18, 521–531.\n\nRadford JQ, Woinarski JCZ, Legge S, Baseler M, Bentley J, Burbidge AA, Bode M, Copley P, Dexter N, Dickman CR, Gillespie G, Hill B, Johnson CN, Kanowski J, Latch P, Letnic M, Manning A, Menkhorst PW, Mitchell N, Morris K, Moseby KE, Page M & Ringma J (2018) Degrees of population-level susceptibility of Australian mammal species to predation by the introduced red fox Vulpes vulpes and feral cat Felis catus. Wildlife Research 45, 645–657.\n\nRees MW, Pascoe JH, Le Pla M, Robley A, Birnbaum EK, Wintle BA & Hradsky BA (2023) Mesopredator release among invasive predators: Controlling red foxes can increase feral cat density and alter their behaviour. Journal of Applied Ecology 60, 1100–1114.\n\nRisbey DA, Calver MC, Short J, Bradley JS & Wright IW (2000) The impact of cats and foxes on the small vertebrate fauna of Heirisson Prong, Western Australia. II. A field experiment. Wildlife Research 27, 223–235.\n\nRoycroft EJ, MacDonald AJ, Moritz C, Moussalli A, Portela Miguez R & Rowe KC (2021) Museum genomics reveals the rapid decline and extinction of Australian rodents since European settlement. Proceedings of the National Academy of Sciences 118, e2021390118.\n\nRoycroft E, Ford F, Ramm T, Schembri R, Breed WG, Burns PA, Rowe KC & Moritz C (2024) Speciation across biomes: Rapid diversification with reproductive isolation in the Australian delicate mice. Molecular Ecology, e17301.\n\nSeebeck J, Menkhorst P, Wilson B & Lowe K (1996) Flora and Fauna Guarantee Action Statement No. 74, New Holland mouse Pseudomys novaehollandiae. Department of Natural Resources and Environment, Victoria.\n\nSeebeck JH, Menkhorst PW, Wilson BA & Lowe KW (2003) New Holland Mouse (Pseudomys novaehollandiae) - Flora and Fauna Guarantee Action Statement No. 74. Department of Sustainability and Environment, East Melbourne.\n\nSharp T & Saunders G (2012) Code of Practice for the humane control of cats. Department of Sustainability, Environment, Water, Population and Communities, Canberra.\n\nSMEC (Snowy Mountains Engineering Corporation) (2015) Tomago Aluminium Company Small Mammal Trapping Program Review. Unpublished report to the Tomago Aluminium Company. SMEC Australia Pty Ltd.\n\nSmith KJ, Evans MJ, Gordon IJ, Pierson JC, McIntyre S & Manning AD (2023a) Countering ecological misconceptions with strategic translocation and assessment of microhabitat use. Biological Conservation, 284, 110143.\n\nSmith KJ, Evans MJ, Gordon IJ, Pierson JC, Newport J & Manning AD (2023b) Analysing captive breeding outcomes to inform reintroduction practice: lessons from the pookila (Pseudomys novaehollandiae). Journal of Mammalogy 104, 1–15.\n\nSmith KJ, Evans MJ, Gordon IJ, Pierson JC, Stratford S & Manning AD (2023c) Mini safe havens for population recovery and reintroductions ‘beyond-the-fence’. Biodiversity and Conservation 32, 203–225.\n\nSparshott P & Sparshott K (2003) Notes on New Holland Mouse (Pseudomys novaehollandiae) (Rodentia: Muridae) at “Tyirrima”, Ravensbourne, South-east Queensland. Queensland Naturalist 41(1–3), 42–47.\n\nStobo-Wilson AM, Murphy BP, Crawford HM, Dawson SJ, Dickman CR, Doherty TS, Fleming PA, Gentle MN, Legge SM, Newsome TM, Palmer R, Rees MW, Ritchie EG, Speed J, Stuart J-M, Thompson E, Turpin J & Woinarski JCZ (2021) Sharing meals: Predation on Australian mammals by the introduced European red fox compounds and complements predation by feral cats. Biological Conservation 261, 109284.\n\nSwinburn ML, Fleming PA, Craig MD, Grigg AH, Garkaklis MJ, Hobbs RJ & Hardy GESJ (2007) The importance of grasstrees (Xanthorrhoea preissii) as habitat for mardo (Antechinus flavipes leucogaster) during post-fire recovery. Wildlife Research 34, 640−651.\n\nTasmanian Aboriginal Centre (2023) Personal communication by email 11 December 2023. Daisy Allan, Language and Research Project Officer, palawa kani Language Program, Tasmanian Aboriginal Centre.\n\nTheden-Ringl F, Hislop KP, Aplin K, Grün R & Schurr MR (2018) The chronology and environmental context of a cave deposit and associated faunal assemblage including megafauna teeth near Wee Jasper, southeastern Australia. The Holocene 28, 1467–1482.\n\nThomas O (1910) The generic arrangement of the Australian Murines hitherto referred to “Mus.” Journal of Natural History Series 8 (6), 603–607.\n\nThompson RCA, Lymbery AJ & Smith A (2010) Parasites, emerging disease and wildlife conservation. International Journal for Parasitology 40, 1163–1170.\n\nThylation (2020) Felixer feral cat grooming traps – under development. PetSmart. Accessed 4 August 2023.\n\nTuft K, Legge S, Frank ASK, James AI, May T, Page E, Radford IJ, Woinarski JCZ, Fisher A, Lawes MJ, Gordon IJ & Johnson CN (2021) Cats are a key threatening factor to the survival of local populations of native small mammals in Australia. Wildlife Research 48 (7), 654–662.\n\nTwigg LE, Fox BJ & Jia L (1989) The modified primary succession following sand mining: A validation of the use of chronosequence analysis. Austral Ecology 14, 441–447.\n\nVan Dyck S & Lawrie B (1997) The New Holland Mouse Pseudomys novaehollandiae (Rodentia: Muridae), an addition to the mammal fauna of Queensland. Memoirs of the Queensland Museum 42, 367–376.\n\nVernes K, Green S, Howes A & Dunn L (2006) Species Richness and Habitat Associations of Non-flying Mammals in Gibraltar Range National Park. Proceedings of the Linnean Society of New South Wales 127, 93–105.\n\nVictoria State Government (2022) Mouse matchmakers bring native back from the brink. Media Release, 17 June 2022.\n\nWalshe T, Wilson BA, Bridges D & Garkaklis MJ (2022) Decision support for management of Phytophthora in Victoria’s Otway Ranges - Final Report. Unpublished Report to the Corangamite Catchment Management Authority, Victoria.\n\nWalshe T, Wilson BA, Garkaklis MJ & Bridges D (2021) Wild Otways Initiative – Project 3: Protecting threatened species habitat in the Otway Ranges, Bells Beach (Ironbark Basin) and Great Ocean Road hinterland from the impacts of Phytophthora dieback. Unpublished Report to the Corangamite Catchment Management Authority, Victoria.\n\nWaterhouse GR (1843) Descriptions of two new species of the genus. Museum Proceedings of the Zoological Society of London 1842, 145–146.\n\nWhiteley AR, Fitzpatrick SW, Funk WC & Tallmon DA (2015) Genetic rescue to the rescue. Trends in Ecology & Evolution 30, 42–49.\n\nWills TJ, Retallick RW, Greet J & Bennett A (2023) Browsing by non-native invasive sambar deer dramatically impacts forest structure. Forest Ecology and Management 543, 21153.\n\nWilson BA (1991) The ecology of Pseudomys novaehollandiae (Waterhouse, 1843) in the Eastern Otway Ranges, Victoria. Wildlife Research 18, 233–247.\n\nWilson BA, Annett K, Laidlaw WS, Cahill DM, Garkaklis MJ & Zhuang-Griffin L (2020) Long term impacts of Phytophthora cinnamomi infestation on heathy woodland in the Great Otway National Park in south-eastern Australia. Australian Journal of Botany 68, 542–556.\n\nWilson BA & Bradtke E (1999) The diet of the New Holland mouse, Pseudomys novaehollandiae (Waterhouse) in Victoria. Wildlife Research 26, 439–451.\n\nWilson BA & Garkaklis MJ (2023a) Wild Otways Initiative – Project 3: Protecting Plant and Animal biodiversity in the Otway Ranges, Bells Beach (Ironbark Basin) and Great Ocean Road hinterland from Phytophthora dieback - Threat Abatement Plan (TAP) for Phytophthora dieback in the Otway Ranges, Victoria. Unpublished TAP to the Corangamite Catchment Management Authority, Victoria.\n\nWilson BA & Garkaklis MJ (2023b) Two-year Report – Protecting plant and animal diversity in the Otway Ranges, Bells Beach (Ironbark Basin) and Great Ocean Road hinterland from Phytophthora cinnamomi dieback; Phosphite treatment of high-risk Phytophthora infestation points. Unpublished Report to the Corangamite Catchment Management Authority, Victoria.\n\nWilson BA & Laidlaw WS (2003) Habitat characteristics for New Holland mouse Pseudomys novaehollandiae in Victoria. Australian Mammalogy 25, 1–11.\n\nWilson BA, Lock M & Garkaklis MJ (2018) Long-term fluctuations in distribution and populations of a threatened rodent (Pseudomys novaehollandiae) in coastal woodlands of the Otway Ranges, Victoria: a regional decline or extinction? Australian Mammalogy 40, 281–293.\n\nWilson BA, Millie S & Aberton J (2015) Identification of optimal habitat for the recovery and management of New Holland mouse (Pseudomys novaehollandiae) in the eastern Otways, southern Victoria. Report to Parks Victoria. School of Life & Environmental Sciences, Deakin University, Victoria.\n\nWilson BA, Newell G, Laidlaw WS & Friend G (1994) Impact of plant diseases on faunal communities. Journal of the Royal Society Interface 77, 139–43.\n\nWilson BA, White NM, Hanley A & Tidey DL (2005) Population fluctuations of the New Holland mouse Pseudomys novaehollandiae at Wilson’s Promontory National Park, Victoria. Australian Mammalogy 27, 49–60.\n\nWilson BA, Zhuang-Griffin L & Garkaklis MJ (2017) Decline of the dasyurid marsupial Antechinus minimus maritimus in south-east Australia: implications for recovery and management under a drying climate. Australian Journal of Zoology 65, 203–216.\n\nWoinarski J, Burbidge A & Harrison P (2014) The Action Plan for Australian Mammals 2012. CSIRO Publishing.\n\nWoinarski JCZ & Burbidge AA (2016) Pseudomys novaehollandiae: The IUCN Red List of Threatened Species 2016. Accessed 24 October 2024.\n\nZhuang-Griffin L (2015) Assessment of the status of the endangered swamp antechinus (Antechinus minimus maritimus) in the Eastern Otways: factors influencing decline. Honours Thesis, Deakin University, Melbourne.","sortOrder":52}],"analysis":{"flash_summary":{"complexity_score":5,"scope_assessment":{"changed":true,"description":"Compared to older, narrower recovery documents for the Pookila (previously largely Victoria‑focused), this national recovery plan expands scope in several concrete ways: (1) it sets a national coordination role and formalises a cross‑jurisdictional recovery team with named partners across Commonwealth, state agencies, Zoos and universities (1.4; Table 4); (2) it adds explicit genetic management (ESU approach), captive‑breeding and coordinated translocation protocols as central recovery tools (2.4; 5.2); (3) it prescribes operational guidance that influences land management practices (e.g., recommended avoidance of poison baiting near known sites, Phytophthora management and fire regime guidance) rather than only high‑level goals (3.2.4; 5.5; 5.3); and (4) it embeds social goals (First Nations partnerships and community engagement) as core objectives with performance criteria (Objective 4). These additions move the instrument beyond a descriptive conservation statement into an operational national programme that will materially affect permitting, management choices and funding priorities."},"complexity_factors":["Extensive length and repetition across sections (multiple repeated headings and duplicated content) increasing reading time and risk of versioning confusion","Glossary with ~11 defined technical terms (Table 1) that are used to interpret metrics such as AOO and ESU","Eleven tables (Tables 1–11) plus four figures that cross‑reference species status, populations, risk matrix and action lists","Five recovery objectives with nested strategies and dozens of specific actions (Tables 7–11) producing conditional and multi‑actor implementation pathways","Risk matrix (Table 6) combining qualitative likelihood and consequence categories that feed prioritisation decisions (requires judgement to apply)","Multiple cross‑jurisdictional references (Commonwealth EPBC Act; state/Territory Acts; Regional Forest Agreements; Threat Abatement Plans) increasing legal and administrative cross‑references","Technical genetic distinctions (two ESUs, potential third for Tasmania) that constrain translocation and breeding options (adds procedural complexity)","Multiple conditional management prescriptions (e.g., avoid baiting within 5 km of known presence; Phytophthora treatment zones; predator control caveats) — 'exceptions to exceptions' style rules","Requirements for ethical permits and IUCN‑aligned translocation approval steps (5.2) that add procedural checkpoints","Significant data gaps flagged (many populations with unknown status) meaning the plan relies on iterative data collection and adaptive revision"],"plain_english_summary":"**What this instrument does (mechanics)\n\n- Makes the \"National Recovery Plan for the Pookila (New Holland mouse) Pseudomys novaehollandiae\" the official recovery plan under the Environment Protection and Biodiversity Conservation Act framework. In practice it: (a) records the species' status, distribution and threats; (b) sets five recovery objectives with 10‑year performance criteria; and (c) lists specific management strategies and actions (surveying, monitoring, captive‑breeding, translocations, fire, predator, weed and pathogen management, genetic management and community/First Nations engagement) to meet those objectives (Executive Summary; 4.2 Objectives; Tables 7–11).\n\n**Who the plan affects**\n\n- Commonwealth and state/territory environment agencies and park managers (they are named partners and implement many actions) (Table 4; 5.1–5.5).\n- Conservation organisations, zoos and university research groups active in captive breeding, genetics and reintroductions (ANU, Zoos Victoria, NTSI, RBGV, etc.) that will receive, hold or use animals, samples and grants (Table 4; 5.2).\n- First Nations groups — the plan requires collaboration and provides for partnerships and recognition of Traditional Ecological Knowledge where those groups choose to engage (1.3; Objective 4; Table 10).\n- Private landholders, local councils, forestry and infrastructure interests whose land management or development decisions intersect with modelled distribution, habitat critical to survival or sites targeted for surveys and actions — the plan both advises and is intended to inform EPBC referral/assessment decisions (Table 4; 2.5.1; Executive Summary).\n- Pest‑control operators and land managers: the plan recommends precautionary changes to some control practices near known populations (for example, avoid poison baiting within ~5 km of known presence) (3.2.4 Baiting).\n\n**Why it matters (purpose claims and what it delivers)**\n\n- Purpose as stated: slow the species’ decline and maximise long‑term survival by increasing area of occupancy, stabilising or increasing population size, improving genetic diversity, involving First Nations people, and boosting community awareness (Executive Summary; 4.2 Objectives).\n- What it does operationally: provides a national, standardised set of objectives and priority actions; identifies populations and habitats of concern (Table 5; 2.5.1); establishes genetic management guidance (two ESUs and a genetic strategy) and protocols for captive breeding and translocations (2.4; 5.2); sets monitoring/survey protocols and a risk matrix to prioritise threats (5.1; Table 6).\n\n**Tests of the plan’s claims against costs, trade‑offs and implementation risk**\n\n- Direct and recurring costs: implementing actions (surveys, long‑term monitoring, captive‑breeding, translocations, predator control, phosphite spraying for Phytophthora) requires funding and skilled staff. Costs are borne across Commonwealth, state agencies, NGOs and partners named in the plan (Table 4; 5.1–5.5). The plan does not itself appropriate funds; it frames priorities that will compete with other budget items.\n\n- Who benefits vs who pays: benefits (recovery, insurance populations, scientific knowledge) are concentrated among species managers, zoos, research institutions, and local conservation groups that run programs (5.2; Table 4). Costs are diffuse across government budgets, and potentially borne in practice by private landholders who must alter management or face stricter EPBC assessment outcomes where habitat is present (2.5.1; Table 4).\n\n- Trade‑offs and opportunity costs: recommended actions (e.g., restricting poison baiting near known sites) impose operational trade‑offs for pest control and land managers and may require substitution to more expensive methods (3.2.4 Baiting). Habitat protection and fire regimes designed for Pookila may limit some forestry or development practices where habitat critical to survival is identified (2.5.1; 5.3).\n\n- Implementation risk and biological uncertainty: the plan repeatedly notes data gaps (many NSW populations have unknown status; Tas mainland status is uncertain), detection difficulties and uncertainty about threats (e.g., relative impacts of cats vs foxes). These gaps increase the risk that actions will be mis‑targeted or require revision (2.3; 2.4; 3.2; Risk matrix Table 6).\n\n- Management complexity and unintended consequences: predator control is identified as complex — for example, fox control can alter cat dynamics and worsen cat predation (mesopredator interactions are discussed) (3.2.4; 5.4). The plan recognises these interactions and recommends research and adaptive approaches rather than one‑size‑fits‑all control.\n\n- Bureaucratic discretion and coordination requirement: the plan relies on multijurisdictional coordination (Commonwealth, state agencies, local councils, Registered Aboriginal Parties and many NGOs). That creates implementation dependence on interagency cooperation, discretionary permitting and shared funding decisions (1.4; Table 4; 4.3). Recovery plans under the EPBC framework inform, but do not automatically impose, statutory land‑use bans; they guide referral and assessment decisions (Table 4).\n\n- Compliance burden and regulatory effect: the instrument is a recovery plan — it primarily provides policy and technical guidance. However, because decision‑makers under the EPBC Act must consider recovery plans when assessing referrals, the plan will influence regulatory outcomes (e.g., permit conditions, offsets, or refusal) for projects affecting mapped habitat or identified important populations (Executive Summary; Table 4). The plan itself does not create criminal penalties or new private liabilities; it increases the evidentiary and policy basis regulators use.\n\n**Market and private‑choice effects (brief)**\n\n- Private enterprise (forestry, development, agriculture) may face higher approval costs, mitigation requirements or altered pest‑control prescriptions near identified habitat; such effects are indirect (the plan informs EPBC assessments rather than directly changing property rights) (2.5.1; 3.2.1; Table 4).\n- The plan incentivises conservation partnerships and creates opportunities for specialist service providers (surveys, captive husbandry, genetic labs), concentrating revenue/skills in particular organisations (Table 4; 5.2).\n\n**Implementation risks and monitoring**\n\n- The plan embeds monitoring and performance criteria to manage risk (five objectives with 10‑year criteria, and standardised survey methods/camera protocols), enabling adaptive management if initial actions fail (4.2; 5.1).\n\n**Net effect summary (mechanical, not normative)**\n\n- Mechanically, the instrument converts a detailed multi‑jurisdictional conservation strategy into the official national recovery plan that will shape Commonwealth and state decision‑making on Pookila‑related matters, prioritise spending and coordinate partners. It does not itself appropriate funds or create new criminal liabilities, but it does raise the policy and evidentiary bar for activities that may affect identified habitat or populations (Executive Summary; Table 4; 2.5.1)."},"flash_summary_failed":{"failed":true,"reason":"A positive credit balance is required for all requests, including BYOK, so fallback providers remain available. Add credits at https://vercel.com/d?to=%2F%5Bteam%5D%2F%7E%2Fai%3Fmodal%3Dtop-up to continue.","source":"analysis-cron"},"kimi_summary":{"_metrics":{"completionTokens":710},"content_quality":"ok","complexity_score":6,"scope_assessment":{"changed":false,"description":"This is the first recovery plan developed for the Pookila under the EPBC Act. The scope aligns with the standard template for threatened species recovery plans under Commonwealth legislation, covering species biology, threats, population status, and recovery objectives. No evidence of scope creep beyond the original purpose of species recovery planning."},"complexity_factors":["Multiple overlapping jurisdictional frameworks (Commonwealth EPBC Act plus five state/territory conservation laws)","Extensive scientific terminology requiring glossary definitions (11 defined terms including ESU, AOO, PVA, SNA)","Detailed threat assessment matrix with 5×5 qualitative risk ratings and custom consequence/likelihood definitions","Population tracking across 60+ discrete locations with varying data quality (extant, unknown, locally extinct, translocated)","Genetic management strategy involving two Evolutionarily Significant Units (ESUs) with different conservation priorities","Cross-referencing to multiple other statutory instruments (Threat Abatement Plans, Regional Forest Agreements, Key Threatening Processes)","Conditional management recommendations based on regional threat variation (e.g., baiting risks differ Qld/Tas/Vic/NSW)","Recovery tables with 5 objectives and multiple strategies/actions per objective requiring coordinated implementation"],"plain_english_summary":"The national 10-year Pookila (New Holland mouse) recovery plan sets a coordinated strategy to halt population decline and restore this Vulnerable/Endangered native rodent across its Australian range. It guides governments, conservation agencies, zoos, universities, First Nations and communities to address habitat loss, fire, feral predators, climate impacts and genetic isolation through captive breeding, translocations, genetic rescue and Indigenous collaboration. Saving the Pookila helps protect coastal and woodland ecosystem health."}},"importantCases":[],"_links":{"self":"/api/acts/environment-protection-and-biodiversity-conservation-national-recovery-plan-for-the-pookila-new-holl","history":"/api/acts/environment-protection-and-biodiversity-conservation-national-recovery-plan-for-the-pookila-new-holl/history","analysis":"/api/acts/environment-protection-and-biodiversity-conservation-national-recovery-plan-for-the-pookila-new-holl/analysis","conflicts":"/api/acts/environment-protection-and-biodiversity-conservation-national-recovery-plan-for-the-pookila-new-holl/conflicts","importantCases":"/api/acts/environment-protection-and-biodiversity-conservation-national-recovery-plan-for-the-pookila-new-holl/important-cases","documents":"/api/acts/environment-protection-and-biodiversity-conservation-national-recovery-plan-for-the-pookila-new-holl/documents"}}