Federal Court of Australia

Save Our Strathbogie Forest Inc v Secretary to the Department of Energy, Environment and Climate Action [2024] FCA 317

ORDERS

THE COURT ORDERS THAT:

1.    The applicant’s originating application dated 14 June 2023 be dismissed.

2.    Within 14 days of these orders, the parties file and serve any submissions on the question of costs of no more than five (5) pages in length.

3.    Subject to further order, the question of costs be determined on the papers.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

HORAN J:

Introduction

1    At some time during autumn this year, as part of the Victorian Government’s bushfire risk management program, the Secretary to the Department of Energy, Environment and Climate Action (DEECA) intends to conduct planned fuel management burns in four defined areas within the Strathbogie State Forest, which is located approximately 120 kilometres north-east of Melbourne, Victoria.

2    The applicant contends that the planned burns involve a controlled action under the Environment Protection and Biodiversity Conservation Act 1999 (Cth) (the EPBC Act) on the basis that they are likely to have a significant impact on the Southern Greater Glider, which is a listed threatened species included in the endangered category under the EPBC Act. Accordingly, the applicant contends that, in the absence of an applicable exemption, the action of the Secretary in carrying out the planned burns requires the approval of the Commonwealth Minister for the Environment (Commonwealth Minister) under Pt 9 of the EPBC Act. As no such approval has been given, the applicant seeks an injunction to prevent the Secretary from carrying out the planned burns without such an approval contrary to s 18(3) of the EPBC Act.

3    There is no dispute that the Strathbogie State Forest, and in particular the four planned burn areas, include habitat suitable for the Southern Greater Glider and that some gliders are likely to be present in the planned burn areas.

4    The ultimate question raised by this proceeding is whether the planned burns can proceed without being required to obtain the approval of the Commonwealth Minister under Pt 9 of the EPBC Act. That in turn raises three main issues.

(a)    whether it is likely that the planned burns will have a significant impact on the Southern Greater Glider as a listed threatened species included in the endangered category within the meaning of s 18(3) of the EPBC Act;

(b)    if so, whether s 43B of the EPBC Act operates to provide the Secretary with an exemption from the requirement to obtain an approval under Pt 9, on the basis that the planned burns amount to a lawful continuation of a use of land that was occurring immediately before the commencement of the EPBC Act on 16 July 2000; and

(c)    whether the application of s 18(3) of the EPBC Act to the functions and activities of the Secretary under s 62(2) of the Forests Act 1958 (Vic), or to the conduct of these four planned burns, offends the Constitutional principle recognised in the High Court’s decision in Melbourne Corporation v The Commonwealth (1947) 74 CLR 31, and is to that extent invalid.

5    Each of the parties adduced a significant amount of evidence directed to the manner in which the fire might behave if and when the planned burns are carried out, and the possible effects of the fire on Southern Greater Gliders present in the burn area and their habitat. The Secretary also adduced evidence addressing the broader context of the State’s bushfire risk management, including in particular the risk to lives and property, and the history of planned fuel reduction burns as one of the measures designed or intended to mitigate such risks.

6    There was a consensus between the parties that, in resolving the issues in dispute in this proceeding, this Court is not required to consider policy issues bearing upon whether or not the planned burns should be carried out, nor to attempt any balancing of the environmental values protected by the EPBC Act against the objectives of planned burning as a measure to address bushfire risk. The applicant submits that those issues are matters that are properly considered by the Commonwealth Minister in deciding whether or not to grant approval to the proposed action under the EPBC Act. The Secretary submits that no such approval is required.

7    In particular, neither the applicant nor the Secretary have invited the Court to consider or make any findings as to the effectiveness of planned burns as a bushfire risk management tool. There has been a long history of investigative commissions which have addressed many aspects of those matters, which lie well outside the scope of this proceeding: see e.g. 2009 Victorian Bushfires Royal Commission Final Report (2010), Fire Preparation, Response and Recovery: Land Management Vol II, Ch 7. The evidence and submissions in this proceeding focused on the impact of the planned burns on protected matters under the EPBC Act. The applicant’s case is concerned with the direct and indirect effects of the proposed planned burns on gliders in the Strathbogie State Forest, but it is no part of that case that the planned burns might have an effect of increasing the risk or severity of bushfires in the future. Conversely, the Secretary does not contend that any risk to the Southern Greater Glider arising from the conduct of the planned burns can or should be offset against a reduction in the risk to gliders in the Strathbogie State Forest from future bushfires as the result of the planned burns. While the Attorney-General of Victoria (Attorney-General), who intervened under s 78A of the Judiciary Act 1903 (Cth) in relation to the Melbourne Corporation issue, places some reliance on the objectives of planned burns in the context of the performance by the Secretary of the statutory duty under s 62(2) of the Forests Act to carry out proper and sufficient work on certain public land for the prevention and suppression of fire, as well as the broader public function or duty of the State to protect lives and property in Victoria, that submission does not turn on any factual contentions concerning the efficacy of planned burns in that context, which for present purposes may largely be assumed.

8    At the request of the parties, the trial of this matter was listed on an expedited basis and was heard over two weeks commencing on 29 January 2024.

9    For the reasons set out below, I have concluded that s 18(3) of the EPBC Act validly applies to the Secretary in the conduct of works for the prevention of fire, including planned fuel reduction burns. However, on the evidence before the Court, I am not satisfied that the four planned burns that the Secretary proposes to carry out in the Strathbogie State Forest will have, or are likely to have, a significant impact on the population of Southern Greater Gliders in the Strathbogie State Forest or on the species as a whole within the meaning of s 18(3).

10    In summary, I have found that the planned burns will be conducted in accordance with the prescriptions and objectives set out in their Delivery Plans, and will generally result in low intensity fire with limited impacts on the canopy in most of the planned burn areas. While there is a real chance the planned burns may kill or injure some individual gliders that are present in areas where the fire burns with higher severity, so as to result in a high degree of canopy scorch, such areas (if any) will constitute a relatively minor proportion (less than 2.5%) of the planned burn areas, and an even smaller proportion of the Strathbogie State Forest as a whole. There is also a real chance that the collapse of hollow-bearing trees during or following the planned burns may kill or injure any individual gliders that are sheltering in such trees when they collapse, but the evidence does not establish that such losses are likely to be common or widespread. More generally, there is a real chance that the planned burns will cause some reduction in the abundance of hollow-bearing trees in the planned burn areas, but the evidence does not establish that this is likely to lead to any significant reduction in the abundance of gliders in the planned burn areas, nor in the Strathbogie State Forest.

11    In the circumstances, any impacts of the planned burns on individual gliders in the areas affected by fire are not likely to have a significant impact on the population of Southern Greater Gliders in the Strathbogie State Forest, or on the species, for the purposes of s 18(3) of the EPBC Act.

12    Otherwise, I have found that the four planned burns cannot be characterised as a lawful continuation of a use of land that was occurring immediately before the commencement of the EPBC Act, and accordingly the exception in s 43B of the EPBC Act is not attracted. Further, in relation to the Melbourne Corporation principle, I do not accept that the application of s 18(3) of the EPBC Act to the conduct by the Secretary of works for the planned prevention of fire in State forests, or to the conduct of these four planned burns, curtails or impairs either the capacity of the State to function as a government or the exercise by the State of its constitutional powers. As a consequence, the Secretary is generally required to comply with Pt 3 of the EPBC Act when conducting any planned burn that has or will have, or is likely to have, a significant impact on any matters of national environmental significance within the meaning of that Part.

13    It follows that the applicant’s application for an injunction under s 475 of the EPBC Act in relation to the proposed conduct of the four planned burns must be dismissed.

Background factual matters

The planned burns

14    Planned burning involves the deliberate introduction of fire into the landscape in order to manage fuel and bushfire hazard and damage potential: see generally s 4 of the Code of Practice for Bushfire Management on Public Land 2012 (amended 2022) (the Code). It has been defined as “the controlled application of fire under specified environmental conditions to a predetermined area and at the time, intensity, and rate of spread required to attain planned resource management objectives”: Australian Fire Authorities Council, Bushfire Glossary (January 2012) (Australian Bushfire Glossary), p 24. The term “planned burning” is sometimes used interchangeably with terminology such as “prescribed burning” or “fuel reduction burning”. It is different from “backburning”, which refers to a fire suppression technique sometimes used when actively fighting bushfires.

15    As a part of the Victorian fuel management program to reduce bushfire risk in the Hume region, the Secretary proposes to conduct planned burns in four areas within the Strathbogie State Forest:

(a)    Lima East – Mt Albert (Lima East);

(b)    Barjarg / Harpers Road (Barjarg);

(c)    Strathbogie South / Ruoaks Road (Ruoaks Road); and

(d)    Tallangallook – Black Creek Track (Tallangallook).

16    I consider the evidence in relation to the planning and delivery of these four planned burns in more detail below. In order to give a general picture of the location of the planned burn areas in the context of the Strathbogie State Forest, the following map depicts the area of each of the planned burns (outlined in black) within the boundaries of the Forest, showing the distribution of ecological vegetation classes derived from the DEECA’s spatial data (predominantly herb-rich foothill forest, with some damp forest, grassy dry forest, healthy dry forest, shrubby dry forest, valley grassy forest and riparian forest).

The Strathbogie Ranges and the Strathbogie State Forest

17    The Strathbogie Ranges are situated approximately 120 kilometres north-east of Melbourne. As summarised in a report by Dr Charles Meredith, an expert witness called by the applicant, drawing on a publication by the Victorian Environmental Assessment Council (Assessment of the values of the Strathbogie Ranges Immediate Protection Area (March 2022)):

The Strathbogie Ranges cover an area of about 240,000 hectares north of the Great Dividing Range situated between the Goulburn River to the west, the Broken River to the east and the Hume Highway to the north (VEAC 2022). The Strathbogie Ranges are mostly within the Highlands - Northern Fall bioregion. The Ranges are effectively a promontory of granite extending northwest from the main part of the Great Dividing Range. Over much of the Ranges, the granite forms a relatively flat-topped ‘tableland’ between about 500 and 900 metres above sea level. This elevation draws rain from prevailing weather systems coming from the west and northeast, with less run off on the gentle topography of the tableland than on more dissected ranges in this part of Victoria (VEAC 2022).

The Crown land within the Strathbogie Ranges falls into several land-use designations. The largest in area of these is State forest and the second largest is land leased for pine plantations. Small areas of other land-use designations include a reference area, several historic areas and an education area. By far the bulk of the native vegetation in the Ranges occurs in the State forest.

18    The Strathbogie State Forest comprises an area of approximately 23,000 hectares within the Strathbogie Ranges. State forests are Crown lands that are reserved or dedicated under the Forests Act and are managed by DEECA for a number of different uses, including timber production, recreation and the conservation of flora and fauna. State forests are distinct from parks and conservation reserves managed by Parks Victoria, committees of management or local government: see generally DEECA, Overview of Victoria’s Forest Management System, December 2019, pp 7-8.

The Southern Greater Glider

19    The Greater Glider is a large gliding possum, which has a head and body length of 35-46 cm with a long, furry, non-prehensile tail measuring 45-60 cm, and an adult weight range from 900-1,700 grams. There are currently two recognised species: Petauroides minor (Greater Glider (northern)) found in north-eastern Queensland, and Petauroides volans (Greater Glider (southern and central)) found in south-eastern Australia. This proceeding is concerned with the latter species, which is generally referred to in Victoria as the Southern Greater Glider. I will adopt that terminology in these reasons for judgment, so as to distinguish the Southern Greater Glider from the Greater Glider (northern). I note in passing that the latter species is now separately listed under the EPBC Act and is the subject of a separate conservation advice.

20    The Southern Greater Glider is a listed threatened species included in the endangered category under the EPBC Act (with effect from 5 July 2022), and is included in the Threatened List with a category of threat of endangered under the Flora and Fauna Guarantee Act 1988 (Vic) (FFG Act). Under the applicable statutory definitions, this means that the species is one that is “facing a very high risk of extinction in the wild in the near future”, but which is not critically endangered (i.e. facing an extremely high risk of extinction in the wild in the immediate future): see s 179(4) of the EPBC Act, s 3(1) of the FFG Act.

Conservation Advice under the EPBC Act

21    The EPBC Act requires an approved conservation advice to be prepared for each listed threatened species: s 266B(1). An approved conservation advice must contain a statement setting out the grounds on which the species is eligible to be included in the category in which it is listed and the main factors that are the cause of it being so eligible: s 266B(2)(a). It must also set out information about what could appropriately be done to stop the decline of, or support the recovery of, the species: s 266B(2)(b)(i).

22    Conservation advices under the EPBC Act are provided to the Commonwealth Minister by the Threatened Species Scientific Committee established under s 502 of the EPBC Act, whose expert members are appointed by the Commonwealth Minister. The functions of the Scientific Committee include advising the Minister on approved conservation advice and the amendment and updating of the lists established under Pt 13 of the EPBC Act (including the lists of threatened species for which ss 178 and 179 provide): s 503. A conservation advice must be approved in writing by the Commonwealth Minister, following consultation with the Scientific Committee, before its publication on the internet: s 266B(2)-(7).

23    In Friends of Leadbeater’s Possum Inc v VicForests (No 4) [2020] FCA 704; 244 LGERA 92 at [25]-[26], Mortimer J (as her Honour then was) described such conservation advices as “the mandatory and foundational documents describing each threatened species, its characteristics and habitat, and the threats posed to it”, and as containing “the formal recognition, for the purposes of the EPBC Act, of why the listed threatened species has been determined to need protection and what measures need to be taken to ensure its conservation and recovery”.

24    In the present case, the applicant places considerable reliance on the approved conservation advice published under the EPBC Act for the Southern Greater Glider: Conservation Advice for Petauroides volans (greater glider (southern and central)) (as in effect from 5 July 2022) (the Conservation Advice). Although counsel for the Secretary accepted that the Conservation Advice is an “authoritative document”, they submitted that it should not be “used as a proxy or surrogate for expert evidence in the case”. While that may be so, the Conservation Advice is nevertheless itself in evidence and was referred to and relied on by several expert witnesses.

25    There does not appear to be any dispute about the correctness of the contents of the Conservation Advice in so far as it has any bearing on the issues for determination in this case (although counsel for the Secretary noted that the contents of the Conservation Advice had not been specifically pleaded). I note that Mortimer J, in her fact-finding in Friends of Leadbeater’s Possum, considered that it was appropriate to place “significant weight” on the conservation advice for each of the species addressed in that case (which included an earlier iteration of the Conservation Advice for the Greater Glider, prior to the elevation of the Southern Greater Glider from the vulnerable category to the endangered category). As an instrument that has been approved by the Commonwealth Minister and published under the EPBC Act, the Conservation Advice is especially relevant to any assessment of whether an action will have or is likely to have a significant impact on a listed threatened species for the purposes of 18(3) of the EPBC Act. In the absence of any directly conflicting evidence or specific disagreement, I will proceed on the basis that the Conservation Advice is a reliable and accurate source of general information about the Southern Greater Glider, including its taxonomy, biology and ecology, distribution, and threats to its survival.

Conservation status

26    The Greater Glider was previously listed in the vulnerable category of the threatened species list under the EPBC Act effective from 5 May 2016. With effect from 5 July 2022, the Southern Greater Glider was recognised as a separate species and was listed in the endangered category following an assessment by the Scientific Committee against the applicable listing criteria, based on an observed, estimated, inferred, projected or suspected population reduction, the causes of which may not have ceased, may not be understood, or may not be reversible. The Conservation Advice states that “[t]he main factors that make the species eligible for listing in the Endangered category are an overall rate of population decline exceeding 50 percent over a 21-year (three generation) period, including population reduction and habitat destruction following the 2019–20 bushfires”. The supporting evidence on which the Scientific Committee relied is set out in its listing assessment, which is included as an attachment to the Conservation Advice. It may be noted that the Scientific Committee assessed the species as ineligible for listing as vulnerable, endangered or critically endangered under other criteria dealing with geographic distribution, total population size and number of mature individuals. As population viability analysis had not been undertaken for the full species, there was insufficient data to determine the eligibility of the species for listing in any category under the criterion dealing with the probability of extinction in the wild.

Distribution

27    The Conservation Advice describes the Southern Greater Glider as the largest gliding possum in eastern Australia. It has a broad distribution “from around Proserpine in Qld, south through NSW and the ACT, to Wombat State Forest in central Vic”. Its area of occupancy has decreased substantially since European settlement, mostly due to land clearing, and is continuing to decline due to further clearing, fragmentation impacts, edge effects, bushfire, climate change and some forestry activities.

28    In relation to important populations and sub-populations, the Conservation Advice states:

Given its Endangered status, all populations of the greater glider (southern and central) are important for the conservation of the species across its range. Due to the species’ low fecundity and limited dispersal capabilities, areas where the species has become locally extinct are not readily recolonised. Coastal populations may be important for maintaining genetic diversity, as they are geographically distinct from inland populations (DoEE 2016b).

29    The Strathbogie State Forest supports a high density of the Southern Greater Glider, with the total population across the Strathbogie Ranges having been estimated to be approximately 69,000: see generally J L Nelson et al, “Estimating the density of the Greater Glider in the Strathbogie Ranges, North East Victoria” (2018), Arthur Rylah Institute for Technical Environmental Research: Technical Report Series No. 293 (Nelson 2018). Referring to a comparison of survey data collected between 1983 and 2017, the Conservation Advice notes that the population of Southern Greater Gliders in the Strathbogie State Forest appears not to have declined over a 34-year period to the same extent as has been observed elsewhere in Victoria, and that surveys “found relatively high densities of gliders”.

Habitat

30    The Southern Greater Glider is an arboreal nocturnal marsupial, predominantly solitary and largely restricted to eucalypt forests and woodlands of eastern Australia. It is typically found in highest abundance in taller, montane, moist eucalypt forests on fertile soils, with relatively old trees and abundant hollows. Gliders have an estimated longevity of 15 years, and females give birth to a single young from March to June.

31    During the day, the Southern Greater Glider shelters in tree hollows, with a particular preference for large hollows in large, old trees. While both live and standing dead trees are used for denning, the species prefers to use live hollow-bearing trees when adequate numbers are available. Most hollow-bearing trees used for denning by arboreal and scansorial mammals are at least 100 years of age. However, the size and age at which suitable hollows develop depends on tree species and climate. Multiple dens are used by each individual glider, typically within a relatively small home range (1 to 4 hectares). Densities of gliders vary significantly across the glider’s range, and average densities have been found to range from 0.6 to 2.8 individuals per hectare in Victoria.

32    The availability of tree hollows is identified in the Conservation Advice as a key limiting resource and the probability of the occurrence of the species is positively correlated with the availability of tree hollows. The Conservation Advice notes that “[l]arge hollow-bearing trees are in rapid decline in some landscapes primarily due to timber production practices and bushfires that prevent trees growing to an age when they might produce hollows”, and that this is a concern for recovery of the species. In addition, the abundance of hollow-bearing trees may be an overestimate of the actual number that are suitable for occupation by wildlife, as only one in every three to five hollow-bearing trees within montane ash forests is occupied by arboreal marsupials. The Conservation Advice states that “[a] decline or loss of hollow-bearing trees reduces the numbers of greater gliders in the landscape”.

33    The Conservation Advice states that the Southern Greater Glider is sensitive to bushfire and is slow to recover following major fires. Thus, “[o]ver the longer term, repeated disturbance such as intense or too-frequent fires degrades greater glider habitat by changing the composition, structure and nutrient profile of forests”, including the destruction of live and dead hollow-bearing trees, particularly in young forests. While unburnt areas provide critical refuges for gliders in regions heavily impacted by fires, Southern Greater Gliders have limited dispersal capabilities and are slow to recover and recolonise burnt sites following fire.

34    The Southern Greater Glider’s diet mostly comprises eucalypt leaves supplemented by buds and flowers. It feeds from a restricted range of eucalypt species, including Narrow-Leaved Peppermint (Eucalyptus radiata) in Victoria. The tree species favoured by Southern Greater Gliders varies regionally. It favours forests with a diversity of eucalypt species, due to seasonal variation in growth and nutrient content of its preferred tree species. Approximately 85 percent of the Southern Greater Glider’s water requirements are provided by consumed leaves.

35    The Conservation Advice describes habitat critical to the survival of the Southern Greater Glider as follows:

Habitat critical to survival for the greater glider (southern and central) may be broadly defined as (noting that geographic areas containing habitat critical to survival needs to be defined by forest type on a regional basis):

•    large contiguous areas of eucalypt forest, which contain mature hollow-bearing trees and a diverse range of the species’ preferred food species in a particular region; and

•    smaller or fragmented habitat patches connected to larger patches of habitat, that can facilitate dispersal of the species and/or that enable recolonization; and

•    cool microclimate forest/woodland areas (e.g. protected gullies, sheltered high elevation areas, coastal lowland areas, southern slopes); and

•    areas identified as refuges under future climate changes [sic] scenarios; and

•    short-term or long-term post-fire refuges (i.e. unburnt habitat within or adjacent to recently burnt landscapes) that allow the species to persist, recover and recolonise burnt areas.

Habitat meeting any one of the criteria above is considered habitat critical to the survival of greater glider (southern and central), irrespective of the current abundance or density of greater gliders or the perceived quality of the site. Forest areas currently unoccupied by the greater glider (southern and central) may still represent habitat critical to survival, if the recruitment of hollow-bearing trees as the forest ages could allow the species to colonise these areas and ensure persistence of a subpopulation.

Threats

36    The Conservation Advice identifies the key threats to the Southern Greater Glider as frequent and intense bushfires, inappropriate prescribed burning, climate change, land clearing and timber harvesting – noting that “[t]here are synergies between these threats, and their combined impact needs to be considered in the recovery of the species”.

37    The threat arising from “inappropriate fire regimes” is addressed as encompassing both extensive severe bushfires and high frequency fires. Dealing with the former, the Conservation Advice states: “Substantial population losses or declines have been documented in and after high severity bushfires … Losses can occur as a result of direct mortality due to lethal heating or suffocation from smoke, or indirect mortality due to the loss of key habitat features and resources”. As to “high frequency fires”, the Conservation Advice notes that “[f]requent fire can decrease the availability of hollow-bearing trees in the landscape, and change the floristic composition and nutritional profile of glider habitat”. In that context, the Conservation Advice specifically addresses planned burning:

Too intense or frequent planned burning may contribute to population losses or declines in the southern part of the greater glider’s range. Bluff (2016) reported that hollow-bearing trees (HBTs) affected by fire during planned burns were 28 times more likely to collapse than HBTs that were not burnt. Parnaby et al. (2010) found that following low intensity prescription burns in the Pilliga forests (NSW), mean collapse rates for burnt HBTs were 14-26%. This was consistent with the collapse rate of 25.6% found by Bluff (2016). A survey following a planned burn at Tallarook Range in the Central Highlands (Vic) in 2021 found that a large number of potential greater glider habitat trees were burnt, with “many destroyed” (N. Stimson 2021, pers. Comm. 26 June).

There is increased pressure from some parts of the community to undertake more hazard reduction burning, follow[ing] the severe bushfires of 2019-20.

38    It may be observed that the language used in the Conservation Advice here specifically refers to planned burning that is “[t]oo intense or frequent”, rather than necessarily treating all planned burning as a threat to the Southern Greater Glider.

39    The Conservation Advice also refers to physical disturbances associated with firefighting operations, including the construction of roads and control lines, earthworks, tree removal and backburning. In relation to planned burning, the Conservation Advice relevantly notes:

In Vic, loss of HBTs due to mechanical site preparation works associated with prescribed burning (which primarily occurs in foothill forests close to settled areas) may reduce suitable habitat for the greater glider (southern and central). Trees that are assessed as potentially hazardous (if they were to catch fire) are routinely removed from the perimeter of planned burns on public land in Vic. They are also removed from bushfire control lines during and after bushfire suppression activities (DELWP n.d). Although not all hazardous trees are hollow bearing, many are, or are likely to be trees that form hollows more quickly (J Nelson 2021. pers comm 16 April).

40    The conservation and recovery actions contemplated in the Conservation Advice relevantly include the following:

•    Re-assess and revise current prescriptions used for prescribed burning to ensure that the frequency and severity of fires in greater glider habitat are minimised, in order to mitigate the risk of further population declines and loss of hollow-bearing trees. Measures to reduce risk from future bushfires should be strategic, incorporate adaptive management, and include a risk assessment that considers trade-offs between fire control efficiency and environmental damage.

•    Implement and enforce measures to reduce direct mortality and loss of hollow-bearing trees during site preparation and execution of prescribed burns, including rake hoeing around the base of trees.

•    Ensure that eucalypt forests and the impacts of disturbance (including fire) are managed to prevent them transitioning to less nutritious, hotter, and/or more fire-prone plant communities, and to ensure that food tree species preferred by the greater glider (southern and central) continue to be the dominant canopy trees.

•    Protect and maintain sufficient areas of suitable habitat, including denning and foraging resources and habitat connectivity, to sustain viable subpopulations throughout the species’ range.

41    The Conservation Advice also addresses survey and monitoring priorities both in relation to Southern Greater Gliders and hollow-bearing trees, as well as information and research priorities including “the development of guidelines for fire management by assessing the impacts of fire management and different fire regimes (including frequency and intensity) on habitat, subpopulation size and hollow availability”.

Action Statement under the FFG Act

42    As mentioned above, the Southern Greater Glider is also classified as endangered under the FFG Act. Section 19 of the FFG Act relevantly requires the Secretary to prepare an action statement for any listed taxon of fauna as soon as possible after that taxon is listed, having regard to any management advice given by the Scientific Advisory Committee and any other relevant nature conservation, social and economic matters. The action statement must set out what has been done to conserve and manage that taxon and what is intended to be done, and may include information on what needs to be done: s 19(2).

43    An action statement for the Southern Greater Glider was published in 2019: Greater Glider (Petauroides volans subsp. volans): Action Statement No 267 – Flora and Fauna Guarantee Act 1988 (Action Statement). At the time the Action Statement was issued, the Southern Greater Glider was listed in the category of vulnerable (facing a high risk of extinction in the wild in the medium-term future) before its elevation in June 2023 to the category of endangered (facing a very high risk of extinction in the wild in the near future).

44    The contents of the Action Statement are consistent with the Conservation Advice. It states that Southern Greater Gliders are “forest dependent and prefer older tree age classes in moist forest types” and that they are “obligate users of hollow-bearing trees for shelter and nesting, with each family group using multiple den trees within its home range”. The Action Statement continues: “Greater Glider density varies proportionally to the availability of hollow-bearing trees and do not persist in areas of forest where such trees are absent”, and notes that “Lindenmayer et al. (1990) found the abundance of Greater Gliders in the Victorian Central Highlands was positively correlated with forest age and the density of hollow-bearing trees”.

45    The Action Statement contains the following map showing the distribution of the Southern Greater Glider in Victoria:

46    Under the heading “Threats”, the Action Statement states:

The key threats to the Greater Glider can be summarised in terms of elevated mortality, habitat degradation and the risks associated with small, fragmented populations, including genetic decline. Factors contributing to elevated mortality and the loss of hollow-bearing trees include bushfire, planned burning, drought, timber harvesting and hyper-predation (SAC 2017). There is some evidence to indicate that climate change in the form of more extreme droughts and higher temperatures might result in a reduction in quality or availability of food. Increased morbidity or mortality might also be associated with heat stress. As populations decline and become more isolated, they are more prone to the effects of small population size and potentially genetic decline. This may result from habitat fragmentation due to land management practices or contraction of suitable habitat due to climate change. Fragmentation and isolation impact on the ability of Greater Gliders to recolonise suitable habitat and reduce genetic exchange between sub-populations.

47    In relation to “social and economic issues”, under the sub-heading “Bushfire Risk”, the Action Statement provides:

Fuel management practices across Victoria’s public and private lands aim to reduce the impact of bushfires to communities at risk and provide for the regeneration or preservation of forest assets including ecological and cultural values. This includes undertaking fuel management by planned burning, slashing and mulching. Forest Fire Management Victoria (FFMVic) further reduces risk on public land through early detection and rapid suppression of bushfires. Other works such as upgrading fire towers, building new bridges and improving roads make suppression more achievable and safer, helping to reduce the impact on community and the environment. These activities can require the removal of hazardous trees to ensure that staff can undertake bushfire management work safely. While these activities may have an impact on Greater Glider populations and their habitat, the protection of life and property is an overriding priority of the Victorian Government. Recognising that hazardous trees may also be important habitat for arboreal mammals, wherever possible FFMVic undertakes values checking before undertaking on-ground works to mitigate impacts to habitat.

48    Under the sub-heading “Animal Welfare”, the Action Statement notes:

The management of public land, including timber harvesting and bushfire prevention and suppression, has the potential to result in injury and death to many animals each year. As a large and readily observable mammal, the impact of these activities on the Greater Glider arouses concern.

49    The Action Statement addresses existing conservation measures, including in relation to “fire management”. In that regard, the Action Statement provides:

The primary objectives of DELWP’s fuel management program is to minimise the impact of major bushfires on human life and other values – including the environment, as well as maintaining or improving the resilience of natural ecosystems. DELWP’s fuel management processes are designed to consider forest values and how they can be protected through strategic planning and the way we operationally deliver any fuel management activities. Planning and delivery of this work is guided by expert knowledge and advice to ensure forests are managed for a diverse range of ecological, cultural and built values.

The Greater Glider is recognised as a threatened species in DELWP’s processes of strategic and operational bushfire management planning. In strategic bushfire management planning, DELWP will take account of Greater Glider habitat and colonies when designating different fire management zones in the landscape.

Operational planning includes additional checks of values that may be impacted by fuel management activities. Nominated burns are tested against known data surveys which encompass a range of information, to determine if they overlap habitat areas and ranges of vulnerable species. If a potential impact is flagged, biodiversity experts within DELWP recommend options to minimise impact. Examples of mitigation measures include:

•    clearing fine fuels from around hollow-bearing habitat trees with rake hoes to ensure they do not burn;

•    burning at a lower intensity;

•    burning during particular seasons;

•    undertaking mechanical treatment rather than planned burning; and

•    avoiding using heavy machinery or chemicals in certain areas.

50    The Action Statement notes that research was being carried out by the Department of Environment, Land, Water and Planning (DELWP) (a predecessor of DEECA) to investigate “the impact of fuel management on Greater Glider as part of its monitoring, evaluation and reporting program for bushfire risk management while loss of hollow-bearing trees resulting from planned burning has been assessed for sites in East Gippsland (Bluff 2016)”. The “intended management actions” listed in the Action Statement include:

Determine the most appropriate fire management practices to ensure viable populations of Greater Glider and their habitat (including key features such as hollows in large old trees). This includes developing feasible, cost-effective measures to mitigate any significant impacts of planned burning on Greater Glider populations and their habitat. Incorporate these measures into strategic, tactical and operational fire management plans.

Fire management

General comments

51    The Secretary led a great deal of evidence in relation to Victoria’s approach to fire management, including the framework and context in which fire management activities are conducted. While much of this evidence was uncontroversial and unchallenged, it has limited direct relevance to most of the critical factual issues for determination in the proceeding.

52    As mentioned above, the case does not raise questions about the content of the broader fire management regime in Victoria or the efficacy of planned burning within that regime, nor any challenge to the public policy in relation to the mitigation of bushfire risks (whether to lives, property or environmental values). Those matters do not have a direct bearing on the question whether or not the conduct of the planned burns as proposed in the Strathbogie State Forest will have or is likely to have a significant impact on the Southern Greater Glider. Apart from that question, the defence under s 43B of the EPBC Act requires some consideration of the evidence in relation to the history of planned burns in the Strathbogie State Forest, including the planned burn areas, and the Melbourne Corporation issue requires some consideration of the purposes or objects of fuel reduction burns as works for the “prevention or suppression of fire”.

53    The evidence about the framework of policies and procedures governing fire management in Victoria was given by Mr Chris Hardman, who is the Chief Fire Officer (CFO), Forest and Fire Operations Division, DEECA and Mr Shaun Lawlor, Regional Manager, Forest and Fire Operations for the Hume Region of Forest Fire Management Victoria (FFMVic). In the circumstances, it is necessary to refer to this evidence in order to provide background and context to the questions to be determined in the proceeding.

54    I will deal separately with the evidence of Mr Lawlor and Mr Hardman in relation to the actual conduct of planned burns, including the measures that are taken to keep fire under control and ensure that it does not behave in an unplanned manner, and their predictions or expectations based on their experience as to the likely fire behaviour in the proposed planned burns. In so far as the latter involved opinion evidence, its admissibility was not challenged, and it may be taken as having been accepted that their evidence in that regard was based on their specialised knowledge from training, study or experience: see Evidence Act 1995 (Cth), s 79.

55    As CFO, Mr Hardman leads DEECA’s fire management operations under the Forests Act and is responsible for fire prevention, preparedness and response operations on behalf of the Secretary. As set out in further detail below, the Secretary is responsible under the Forests Act (and its predecessor legislation) for managing the risk of bushfire in State forests, national parks and other protected land in Victoria: see, currently, Forests Act, s 62(2). The CFO also represents DEECA in Victorian and national emergency management arrangements.

56    As the Regional Manager, Forest and Fire Operations for the Hume Region, Mr Lawlor is responsible for the tactical and operational planning and delivery of regional forest fire programs, including bushfire response, planned burning, and other forest management in the Hume Region.

Fire management in Victoria

57    Victoria has a particularly high bushfire risk when compared to other parts of the world and other parts of Australia due to its landscape, vegetation types and weather patterns. This risk is increasing as a result of climate change. Victoria has hot dry summers with very little rain. In summer, Victoria is generally subject to hot, dry northerly wind conditions, followed by south-westerly wind changes. When a bushfire ignites, it is common for a northerly wind to push a narrow bushfire from north to south and, when the wind changes to a south-westerly direction, the bushfire will move from west to east and have a broad and dangerous front.

58    Victoria has experienced an increase in scale of bushfires in the last 20 years, as evidenced by the increased number of megafires in the State. A megafire is a fire which burns over one million hectares. There have been three megafires in the last 20 years, being the 2003 Eastern Victorian alpine bushfires, the 2006-2007 Great Divide Complex bushfires in eastern Victoria, and the 2019-2020 bushfires in East Gippsland and other locations (known as the Black Summer bushfires). By comparison, in the 152 years leading up to 2003, Victoria experienced two recorded megafires (Black Thursday in 1851 and Black Friday in 1938-1939).

59    Victoria has a risk-based approach to bushfire management, which seeks to reduce impacts from bushfire including on people, property, the economy and the environment. This risk-based approach ultimately recognises that the protection of human life is the highest priority.

60    DEECA delivers programs to manage bushfire risk through FFMVic, which was established in 2013 and is responsible for all forest fire management activities in Victoria. FFMVic is led by DEECA, but also draws on capability from Parks Victoria, VicForests and Melbourne Water. FFMVic works with the Country Fire Authority (CFA) and Fire Rescue Victoria to deliver fire management services in Victoria.

61    Fuel management activities in Victoria are undertaken in accordance with the Code, which is a Code of Practice made by the Victorian Minister for Environment under Pt 5 of the Conservation Forests and Lands Act 1987 (Vic). The Code sets the objectives for bushfire management on public land across Victoria, and provides a risk analysis framework that assists DEECA to achieve those objectives. The Code does not prescribe the operational detail for the achievement of bushfire management outcomes and objectives. The operational detail is specified in bushfire management manuals and guidelines, which must be publicly available and consistent with the Code.

62    For administrative purposes in relation to the delivery of fire management services, Victoria is divided into six regions: Barwon South West, Grampians, Loddon Mallee, Gippsland, Hume and Port Phillip. The Strathbogie State Forest is located within the Hume region. There are four districts within the Hume region: Upper Murray, Ovens, Goulburn and Murrindindi. The Strathbogie State Forest is located within the Goulburn district.

63    Each of the six regions in Victoria has its own Bushfire Management Strategy, which describes the bushfire risk for that region, taking account of its unique environment and landscape. The Bushfire Management Strategies identify actions that can be taken to protect communities, infrastructure and the environment from the threat of bushfire, and provide the direction and options for fuel management across a region and for each district.

64    Fuel management is a key strategy for reducing the risk of bushfire in Victoria. While there are several layers to how the Bushfire Management Strategies are implemented on the ground, the objective is to reduce bushfire fuels and fuel hazard, particularly the leaves, bark, twigs and shrubs that cause the greatest risk of a bushfire spreading. Fuel management includes planned burning, mechanical treatment activities (such as mowing, slashing and mulching), and non-burn treatment activities such as spraying.

65    Victoria’s fuel management program includes “residual risk targets” which are set by the CFO and implemented across the regions and districts. A residual risk target is the percentage of bushfire risk remaining after fuel in forests has been reduced, either through fuel management activities or as a result of bushfires, and before any fire suppression activities are undertaken. Bushfire risk is the risk of destruction of property from bushfire, which is also used in this context as a proxy for risk to human life. The State-wide residual risk target is 70%. The residual risk target for each region and district varies depending on factors including vegetation, topography, and the population and assets in the area. The target for the Hume region is 69%, and for the Goulburn district is 75%.

66    The specific activities that comprise a region’s fuel management program are listed in a “Joint Fuel Management Program” for that region, which is created and implemented jointly between FFMVic and the CFA. A Joint Fuel Management Program sets out a rolling three-year program of planned fuel management activities, including planned burns. The Joint Fuel Management Program for the Hume Region refers to the regional residual risk target of 69% and notes that, without any fuel management, the projected residual risk will increase to 73%. Through the conduct of the activities programmed in the Joint Fuel Management Program, the region works towards managing local bushfire risk by identifying planned burns and other works that will meet regional and State-wide targets.

67    Fire Management Zones are “areas of public land where fire is used for specific asset, fuel and overall forest and park management objectives”: see the Code at [122]. There are four different classes of Fire Management Zone: Asset Protection Zones, Bushfire Moderation Zones, Landscape Management Zones and Planned Burning Exclusion Zones.

68    As will be discussed further below, the four planned burns that are the subject of this proceeding are within an area that is zoned as a Bushfire Moderation Zone. According to the Code (at [128]), a Bushfire Moderation Zone “aims to reduce the speed and intensity of bushfires”, and “complements the [Asset Protection Zone] in that the use of planned burning … is designed to protect nearby assets, particularly from ember spotting during a bushfire”. The Code continues (at [129]):

Where practicable, the [Bushfire Moderation Zone] will aim to achieve ecological outcomes by seeking to manage for ecologically desirable fire regimes, provided bushfire protection objectives can still be met. This may include using other fuel management methods.

69    This can be compared to the Landscape Management Zone, within which planned burning is used for three broad aims, namely:

•    bushfire protection outcomes by reducing the overall fuel and bushfire hazard in the landscape

•    ecological resilience through appropriate fire regimes

•    management of the land for particular values including forest regeneration and protection of water catchments at a landscape level.

70    The key principles of the fuel management strategy for the Hume Region, as set out in its Bushfire Management Strategy, include to “focus fuel management activities within Asset Protection Zone and Bushfire Moderation Zone, where fuel hazards are reduced to an acceptable level whilst protecting ecological assets where possible, and to select fuel management regimes that meet the target reduction in risk to life and property, reduce risk of large fires to other important values (such as water catchments) and minimise the negative impact on ecosystem resilience”. Fuel management activities in the Landscape Management Zone are undertaken “where there is a clear bushfire risk reduction objective or ecological outcome”, otherwise such activities are minimised within the Landscape Management Zone to reduce negative impacts on ecosystem resilience. Another key principle is to “avoid fuel management within high value ecological areas (HVEAs), particularly within [the Landscape Management Zone]”.

71    The use of Fuel Management Zones is illustrated in the document published by DEECA titled “Hume Bushfire Management Strategy 2020: Bushfire Risk Engagement Areas” (2020) at p 43 using an example of zoning in areas around Jamieson. Among other things, this includes the placement of Landscape Management Zone “in areas further from communities, where fuel management can reduce overall fuel and bushfire hazard in the landscape”, and where “[m]aintaining ecosystem resilience is equally important, and we apply fire regimes that protect ecosystem values (for example, old-growth vegetation that provides critical habitat for Greater Glider and Powerful Owl)”.

Fuel management burning in Victoria

72    Section 4 of the Code addresses planned burning in the following terms:

111. Planned burning is the deliberate introduction of fire into the landscape to:

•    modify fuel hazard, bushfire hazard and damage potential

•    contribute to ecological objectives

•    contribute to regeneration following timber harvesting activities.

112. Planned burning is the most effective technique for managing fuel hazard over large areas. Other localised treatments include ploughing, mulching, herbicide application, chain rolling, grazing, mowing and slashing.

113. Planned burning assists bushfire suppression actions by reducing the intensity and severity of bushfires.

73    Consistently with the primary objectives for bushfire management on public land, the Code identifies the “outcomes” of fuel management including planned burning as “[r]educed impact of major bushfires on human life, communities, essential and community infrastructure, industries, the economy and the environment”, and states that “[h]uman life will be afforded priority over all other considerations”.

74    Prior to 2016, planned burning on public land in Victoria was administered to meet annual hectare-based bushfire fuel management targets, consistent with the recommendations of various Commissions of Inquiry and Royal Commissions. However, since 2016, Victoria has employed a risk reduction target (also referred to as a residual risk target), as noted above. The State-wide target of 70% “means that bushfire fuels have been reduced to the point where impacts to life and property are reduced by about a third of the maximum risk”: DEECA, “Safer together: A new approach to reducing the risk of bushfire in Victoria” (2015). The regional residual risk targets and the delivery of the regional Bushfire Management Strategies are aggregated to assess the State-wide target of 70%.

75    In this context, the Secretary measures bushfire risk by using a model called Phoenix RapidFire to simulate a set of bushfires across the landscape using real time information about weather, topography, vegetation, fire characteristics (such as flame height, ember density, spotting distance, convection column strength and intensity), in order to show how the fire will spread and behave under specified conditions, and thereby predict bushfire behaviour both with fuel reduction and without fuel reduction. In broad terms, the Secretary models how fires would behave on a “worst case” scenario (assuming weather conditions such as those prevailing on Black Saturday and fuel at its maximum risk level), and then runs modified risk scenarios seeking to reduce the number of houses or “address points” that are lost down to the applicable residual risk target. In this way, the Secretary can determine the risk reduction value of a particular planned burn or burns. The results of this modelling are used to predict how different fuel management activities will affect bushfire risk across landscapes, so as to inform fuel management priorities and determine strategies for reducing bushfire risk.

Federal and State legislative frameworks

EPBC Act

76    The EPBC Act commenced operation on 16 July 2000. The objects of the Act are set out in s 3(1) and relevantly include:

(a)    to provide for the protection of the environment, especially those aspects of the environment that are matters of national environmental significance; and

…

(c)     to promote the conservation of biodiversity; and

(ca)     to provide for the protection and conservation of heritage; …

...

77    Section 3(2) of the EPBC Act provides that, in order to achieve those objects, the EPBC Act, inter alia:

(a)     recognises an appropriate role for the Commonwealth in relation to the environment by focussing Commonwealth involvement on matters of national environmental significance and on Commonwealth actions and Commonwealth areas; and

...

(d)     adopts an efficient and timely Commonwealth environmental assessment and approval process that will ensure activities that are likely to have significant impacts on the environment are properly assessed; and

(e)     enhances Australia’s capacity to ensure the conservation of its biodiversity by including provisions to:

(i)     protect native species (and in particular prevent the extinction, and promote the recovery, of threatened species) and ensure the conservation of migratory species; …

...

78    The EPBC Act binds the Crown in each of its capacities: s 4.

79    Chapter 2 of the EPBC Act is headed: “Protecting the environment”. The simplified outline of the Chapter set out in s 11 states that it “provides a basis for the [Commonwealth] Minister to decide whether an action that has, will have or is likely to have a significant impact on certain aspects of the environment should proceed”, and that “[i]t does so by prohibiting a person from taking an action without the Minister having given approval or decided that approval is not needed”.

80    Division 1 of Pt 3 of Ch 2 of the EPBC Act contains requirements for environmental approvals by imposing conditional prohibitions on the taking of actions that have or will have, or are likely to have, a significant impact on various “matters of national environmental significance”: see the heading to Div 1; and the object set out in s 3(1)(a). While the term “matters of national environmental significance” is undefined, it is clear from the context and structure of the Division and from references elsewhere in the Act that it refers to the matters with which the Subdivisions contained in Div 1 are concerned, namely: World Heritage properties; National Heritage places; Ramsar wetlands; listed threatened species and ecological communities; listed migratory species; nuclear actions; actions in or affecting Commonwealth marine areas; actions in or affecting the Great Barrier Reef Marine Park; certain actions involving unconventional gas development or large coal mining development impacting on water resources; and actions prescribed by the regulations. Each Subdivision contains its own civil penalty provisions and offence provisions in respect of a contravention of the prohibitions, together with exceptions including where an approval has been given under Pt 9 of the EPBC Act in relation to the taking of the action.

81    Relevantly to the present case, Subdiv C of Div 1 deals with listed threatened species and ecological communities. Section 18 prohibits a person from taking an action that has or will have, or is likely to have, a significant impact on a listed threatened species that is included in the extinct in the wild, critically endangered, endangered or vulnerable categories, and imposes civil penalties for contravention by an individual or a body corporate. Section 18A creates corresponding criminal offences.

82    The central provision for the purposes of this case is s 18(3) of the EPBC Act, which provides:

Endangered species

(3)     A person must not take an action that:

(a)     has or will have a significant impact on a listed threatened species included in the endangered category; or

(b)     is likely to have a significant impact on a listed threatened species included in the endangered category.

Civil penalty:

(a)     for an individual—5,000 penalty units;

(b)     for a body corporate—50,000 penalty units

83    Section 18A(2) relevantly provides that a person commits an offence if the person takes an action that is “likely to have a significant impact” on a listed threatened species (other than one included in the extinct category or a conservation dependent species), which is punishable on conviction by imprisonment for a term not more than 7 years, a fine not more than 420 penalty units (or up to 5 times that amount for a body corporate), or both. For such purposes, the element whether the species is a listed threatened species is subject to strict liability: s 18A(2A).

84    Section 19 provides a range of exceptions or defences to the civil penalty and offence provisions, namely: if an approval is in operation under Pt 9; if Pt 4 lets the person take the action without such an approval; if there is in force a decision of the Minister under Div 2 of Pt 7 that the relevant subsection of ss 18 or 18A is not a “controlling provision” for the action; and if the action is one whose authorisation is subject to a special environmental assessment process as described in s 160(2).

85    The term “action” is relevantly defined to include “an activity or series of activities”: s 523(1)(d). It may be noted that a decision by a government body, including a State or an agency of a State, to grant a governmental authorisation for another person to take an action is not itself an action: s 524(2). While a decision to grant a governmental authorisation is removed from the operation of the EPBC Act, the subsequent physical acts or activities pursuant to the authorisation or by way of implementation of the decision can be actions within the meaning of s 523: see Secretary, Department of Primary Industries, Parks, Water and Environment v Tasmanian Aboriginal Centre Inc (2016) 244 FCR 21 at [77]-[79] (Allsop CJ, Griffiths and Moshinsky JJ). The present case is concerned with the actions proposed to be taken by the State of Victoria or its agencies in conducting the planned burns, rather than the grant of any governmental authorisation. Neither the applicant nor the Secretary or the Attorney-General suggested that s 524(2) had any relevance to the present case.

86    The term “impact” is relevantly defined under s 527E as follows:

527E Meaning of impact

(1)     For the purposes of this Act, an event or circumstance is an impact of an action taken by a person if:

(a)     the event or circumstance is a direct consequence of the action; or

(b)     for an event or circumstance that is an indirect consequence of the action—subject to subsection (2), the action is a substantial cause of that event or circumstance.

(2)    For the purposes of paragraph (1)(b), if:

(a)    a person (the primary person) takes an action (the primary action); and

(b)    as a consequence of the primary action, another person (the secondary person) takes another action (the secondary action); and

(c)    the secondary action is not taken at the direction or request of the primary person; and

(d)    an event or circumstance is a consequence of the secondary action;

then that event or circumstance is an impact of the primary action only if:

(e)    the primary action facilitates, to a major extent, the secondary action; and

(f)    the secondary action is:

(i)    within the contemplation of the primary person; or

(ii)    a reasonably foreseeable consequence of the primary action; and

(g)    the event or circumstance is:

(i)    within the contemplation of the primary person; or

(ii)    a reasonably foreseeable consequence of the secondary action.

(Emphasis in original)

87    The term “listed threatened species” is defined in s 528 to mean a native species included in the list referred to in s 178, which requires the Minister by legislative instrument to establish a list of threatened species divided into the following categories:

(a)     extinct;

(b)     extinct in the wild;

(c)     critically endangered;

(d)     endangered;

(e)     vulnerable;

(f)     conservation dependent.

88    Section 179 of the EPBC Act addresses when a native species is eligible to be included in each of the categories listed in s 178(1). As noted above, the Southern Greater Glider has been included on the list of threatened species in the endangered category from 5 July 2023, having previously been included in the vulnerable category. Those categories are dealt with by s 179(4) and (5), which provide:

(4)     A native species is eligible to be included in the endangered category at a particular time if, at that time:

(a)     it is not critically endangered; and

(b)     it is facing a very high risk of extinction in the wild in the near future, as determined in accordance with the prescribed criteria.

(5)     A native species is eligible to be included in the vulnerable category at a particular time if, at that time:

(a)     it is not critically endangered or endangered; and

(b)     it is facing a high risk of extinction in the wild in the medium-term future, as determined in accordance with the prescribed criteria.

(Emphasis in original)

89    Part 4 of the EPBC Act deals with cases in which environmental approvals are not needed, which generally provides an exception to the prohibitions contained in Div 1 of Pt 3. These include actions covered by Ministerial declarations in relation to the accreditation of management arrangements or authorisation processes either under State or Territory laws pursuant to bilateral agreements between the Commonwealth and the State or Territory, or under Commonwealth laws; Ministerial declarations in relation to bioregional plans; conservation agreements; forestry operations undertaken in accordance with a Regional Forest Agreement; actions in accordance with zoning plans in the Great Barrier Reef Marine Park; and actions with prior authorisation. The only provision contained in Pt 4 which has any potential relevance to the facts of the present case is s 43B (under Div 6 – Actions with prior authorisation), which deals with the lawful continuation of uses of land, sea or seabed that were occurring immediately before the commencement of the EPBC Act. The Secretary relies on s 43B as providing an exemption from the requirements of s 18(3) in relation to the conduct of planned burning in the Strathbogie State Forest.

90    Part 9 of the EPBC Act is contained in Ch 4, which sets out processes for environmental assessments and approvals by the Commonwealth Minister. These processes generally involve three stages: deciding whether approval is needed and identifying the applicable “controlling provisions” (Pt 7); assessing the relevant impacts of the action (Pt 8); and deciding whether or not to approve the action, and on what conditions (Pt 9).

91    The environmental assessment and approval process contained in Ch 4 of the EPBC Act is generally enlivened by a referral to the Commonwealth Minister by the proponent (generally the person proposing to take the action, or a party to a contract or agreement under which the action is proposed to be taken) or by a State or an agency of a State that is aware of the proposed action. The Commonwealth Minister also has a power to request a proponent to refer the proposal: s 70. A proposed action is a “controlled action” if the taking of the action by the person without approval under Pt 9 would be prohibited by a provision of Pt 3, which is referred to as a “controlling provision” for the action: see s 67. A proponent may refer the proposed action to the Commonwealth Minister for a decision whether or not the action is a controlled action even if the proponent does not think that the action is a controlled action: s 68(2). Thus, the referral mechanism can potentially be used as a way to obtain a decision by the Commonwealth Minister that the action is not a controlled action.

92    Once a proposed action is referred to the Commonwealth Minister, s 75(1) provides that the Minister must decide whether the action the subject of the referral is a controlled action and which provisions of Pt 3 (if any) are controlling provisions for the action. In so deciding, the Commonwealth Minister must consider all adverse impacts (if any) the action has or will have, or is likely to have, on the matter protected by each provision of Pt 3: s 75(2)(a). The Commonwealth Minister must not consider any beneficial impacts the action has or will have, or is likely to have, on the protected matters: s 75(2)(b). It is an offence for a proponent to take the action while a decision is pending on whether or not it is a controlled action: s 74AA.

93    The “relevant impacts” of a controlled action must be assessed in accordance with Pt 8 of Ch 4, which prescribes a range of different assessment approaches. The relevant impacts are the impacts that the action has or will have, or is likely to have, on the matter protected by each provision of Pt 3 that the Commonwealth Minister has decided is a controlling provision for the action.

94    After the controlled action has been assessed, the Commonwealth Minister must decide whether or not to approve the taking of the action for the purposes of each controlling provision: ss 130(1), 133. The Commonwealth Minister can attach conditions to the approval of the controlled action if he or she is satisfied that the conditions are necessary or convenient for protecting, or repairing or mitigating damage to, a matter protected by a relevant controlling provision: s 134. In deciding whether or not to approve the taking of an action, and what conditions to attach to an approval, the Commonwealth Minister must consider certain mandatory considerations set out in s 136(1) and (2), including matters relevant to the controlling provision or provisions, and “economic and social matters”. The Commonwealth Minister may have regard to whether the proponent is a suitable person to be granted an approach, having regard to the person’s environmental history: s 134(4). In relation to decisions about threatened species, the Commonwealth Minister must not act inconsistently with Australia’s obligations under specified international conventions, or with a recovery plan or a threat abatement plan: s 139(1). If the Commonwealth Minister is considering whether to approve a proposed action for the purposes of a subsection of s 18 that has or will have, or is likely to have, a significant impact on a particular listed threatened species, the Commonwealth Minister must have regard to any approved conservation advice for the species or community: s 139(2). Otherwise, the Commonwealth Minister must not consider any matters that he or she is not required or permitted by Div 1 of Pt 9 to consider: s 134(5).

95    It must be emphasised that the present case does not directly concern any aspect of the assessment and approval processes set out in Ch 4 of the EPBC Act. Rather, the applicant’s contention is that the Secretary is required to follow those processes by obtaining approval from the Commonwealth Minister under Pt 9 before carrying out the proposed planned burns, in order to avoid a contravention of s 18(3) of the EPBC Act. In the absence of any such approval, the applicant seeks injunctive relief under s 475 of the EPBC Act to restrain what it contends is a threatened contravention of the EPBC Act. Section 475 is in the following terms:

475 Injunctions for contravention of the Act

Applications for injunctions

(1)     If a person has engaged, engages or proposes to engage in conduct consisting of an act or omission that constitutes an offence or other contravention of this Act or the regulations:

(a)     the Minister; or

(b)     an interested person (other than an unincorporated organisation); or

(c)     a person acting on behalf of an unincorporated organisation that is an interested person;

may apply to the Federal Court for an injunction.

Prohibitory injunctions

(2)     If a person has engaged, is engaging or is proposing to engage in conduct constituting an offence or other contravention of this Act or the regulations, the Court may grant an injunction restraining the person from engaging in the conduct.

...

Meaning of interested person—organisations

(7)     For the purposes of an application for an injunction relating to conduct or proposed conduct, an organisation (whether incorporated or not) is an interested person if it is incorporated (or was otherwise established) in Australia or an external Territory and one or more of the following conditions are met:

(a)     the organisation’s interests have been, are or would be affected by the conduct or proposed conduct;

(b)     if the application relates to conduct—at any time during the 2 years immediately before the conduct:

(i)     the organisation’s objects or purposes included the protection or conservation of, or research into, the environment; and

(ii)     the organisation engaged in a series of activities related to the protection or conservation of, or research into, the environment;

(c)     if the application relates to proposed conduct—at any time during the 2 years immediately before the making of the application:

(i)     the organisation’s objects or purposes included the protection or conservation of, or research into, the environment; and

(ii)     the organisation engaged in a series of activities related to the protection or conservation of, or research into, the environment.

(Emphasis in original)

96    The jurisdiction of this Court to grant an injunction under s 475 of the EPBC Act is separate from the administrative processes established under Pts 7, 8 and 9 of the EPBC Act. As summarised above, the latter involves the referral of a proposed action to the Commonwealth Minister, who must then consider and determine whether the action is a controlled action and, if so, what the controlling provisions are. Such a determination will require the Commonwealth Minister to make factual findings as to whether the proposed action has or will have, or is likely to have, a significant impact on a protected matter, such as a listed threatened species within the meaning of s 18 of the EPBC Act. When determined administratively in that context, such findings will not involve matters of jurisdictional fact which constitute objective pre-conditions on the powers conferred on the Commonwealth Minister by the EPBC Act: see e.g. Anvil Hill Project Watch Association Inc v Minister for Environment and Water Resources (2008) 166 FCR 54 at 60 (Tamberlin, Finn and Mansfield JJ). Accordingly, while a decision made by the Commonwealth Minister that an action is (or is not) a controlled action is amenable to judicial review, the Court in any such judicial review proceedings would not be required itself to make findings of fact on the evidence as to whether or not the action is a controlled action under any of the provisions of Pt 3 of Ch 2 of the EPBC Act, that is, whether the action would be prohibited by the relevant provision in the absence of an approval under Pt 9.

97    In contrast, an application to this Court for an injunction under s 475 of the EPBC Act raises a justiciable issue whether a person “has engaged, is engaging or is proposing to engage in conduct constituting an offence or other contravention of this Act or the regulations”. While the grant of any injunctive relief remains subject to discretionary considerations, it is clear that the Court is required to determine on the evidence before it whether the past, current or future conduct of the person contravenes a provision of the EPBC Act. Relevantly to the present case, this includes the question whether the action that the Secretary proposes to take in conducting the planned burns in the Strathbogie State Forest has or will have, or is likely to have, a significant impact on the Southern Greater Glider as a listed threatened species in the endangered category within the meaning of ss 18(3) and 18A of the EPBC Act. On an application for a prohibitory injunction under s 475(2), this is essentially a question of fact: see e.g. Friends of Leadbeater’s Possum at [1292], [1342] ff; see also Booth v Bosworth (2001) 114 FCR 39 (Branson J); Minister for Environment and Heritage v Greentree (No 2) (2004) 138 FCR 198 at [192] (Sackville J); Australian Brumby Alliance Inc v Parks Victoria Inc (2020) 277 FCR 559.

98    For present purposes, it is unnecessary to explore fully the relationship between, on the one hand, the administrative processes of environmental assessment and approval by the Commonwealth Minister and, on the other hand, the enforcement provisions under the EPBC Act, including the jurisdiction to grant injunctions in respect of contraventions of the Act. At least in circumstances where a person threatens to engage in, or to continue to engage in, conduct that contravenes the EPBC Act, this Court may be required to determine the factual and legal questions as to whether that conduct constitutes an action that, in the absence of an approval under Pt 9 or some other statutory exception or defence, would contravene an applicable statutory prohibition under Pt 3 of the EPBC Act. The fact that the Commonwealth Minister may be able to determine similar questions in the context of a referral or an environmental assessment and approval under Ch 4 of the EPBC Act, including whether a proposed action is a controlled action in respect of one or more provisions in Pt 3 for which approval is required, does not detract from or qualify the Court’s task in hearing and determining an application for an injunction under s 475 of the EPBC Act. Section 67A contains a prohibition on taking a controlled action without an approval under Pt 9, with a note indicating that a person can be restrained from contravening that section by an injunction under s 475. A “controlled action” is defined for such purposes by s 67 in terms that are not expressly confined to an action that the Commonwealth Minister has decided is a controlled action under s 75. In any event, even if s 67A were to be read as referring to actions that the Commonwealth Minister has decided are controlled actions, the jurisdiction conferred by s 475 is capable of being enlivened in appropriate circumstances by a threatened contravention of a prohibition in Pt 3, such as ss 18 or 18A of the EPBC Act.

99    It may be accepted that any injunction that might be granted by the Court under s 475 in such circumstances would ordinarily be conditional in nature, so as to be expressed to restrain the action unless and until an approval were granted under Pt 9 or some other statutory exception or defence were established: compare Booth at [120] (Branson J). On any referral of a proposal, it would be matter for the Commonwealth Minister to determine on the available material whether the proposed action is a controlled action and which provisions are controlling provisions for the action, as well as whether or not the taking of the action should be approved or conditionally approved in the light of the assessment of its relevant impacts. However, at least in circumstances where there has been no referral of the proposed action to the Commonwealth Minister under Pt 7 of the EPBC Act, and a person is threatening to “break the law” by taking such action without an approval under Pt 9, an application may be brought to this Court to restrain the threatened contravention of the EPBC Act.

100    I note for completeness that, in Tasmanian Aboriginal Centre at [93], the Full Court observed that the EPBC Act “reposes in the Minister the decision whether or not to approve, for the purposes of each controlling provision for a controlled action, the taking of the action, if there is a referral” and that “[t]he Act does not contemplate an ex ante decision in exercise of judicial power of what would be or would not be a contravention of the Act if action (as yet not fully identified) took place”. The Full Court therefore considered that “questions of assessment of impact, such as dealt with in Pt 8, may best be seen as the task of the Commonwealth executive not judiciary, at least in the context of an ex ante analysis involving a party who does not threaten to break the law”. The concluding qualification is important, in that it acknowledges the role of injunctive relief under s 475 of the EPBC Act to restrain an imminent contravention of the Act. The jurisdiction to grant such relief is borne out by past decisions of this Court. The Full Court in Tasmanian Aboriginal Centre acknowledged at [94] that the primary judge had granted an injunction in that case because the respondent “was threatening to go ahead with the project with significant dispatch”. As the Full Court’s decision illustrates, it might in some circumstances also be possible for the Court to grant declaratory relief in the light of which the respondent may consider whether a referral under s 68 of the EPBC Act is required. However, an “interested person” within the meaning of s 475 (such as the applicant in the present case) cannot unilaterally initiate a referral under Pt 7 of the EPBC Act.

101    By way of defence to the applicant’s claim that the planned burns constitute an action or actions the taking of which without approval under Pt 9 would be prohibited by s 18(3), the Secretary relies on s 43B of the EPBC Act, which provides as follows:

43B Actions which are lawful continuations of use of land etc.

(1)    A person may take an action described in a provision of Part 3 without an approval under Part 9 for the purposes of the provision if the action is a lawful continuation of a use of land, sea or seabed that was occurring immediately before the commencement of this Act.

(2)     However, subsection (1) does not apply to an action if:

(a)     before the commencement of this Act, the action was authorised by a specific environmental authorisation; and

(b)     at the time the action is taken, the specific environmental authorisation continues to be in force.

Note: In that case, section 43A applies instead.

(3)     For the purposes of this section, neither of the following is a continuation of a use of land, sea or seabed:

(a)     an enlargement, expansion or intensification of use;

(b)     either:

(i)     any change in the location of where the use of the land, sea or seabed is occurring; or

(ii)     any change in the nature of the activities comprising the use;

that results in a substantial increase in the impact of the use on the land, sea or seabed.

102    In broad terms, the Secretary contends that s 43B provides an exemption from the requirement to obtain an approval under Pt 9 on the basis that the planned burns in the Strathbogie State Forest involve a lawful continuation of a use of land that was occurring immediately before the commencement of the EPBC Act on 16 July 2000.

103    Finally, s 158 of the EPBC Act sets out a further exemption to the application of the prohibitions under Pt 3. Under s 158, a person proposing to take a controlled action may apply in writing to the Commonwealth Minister for an exemption from a specified provision of Pt 3. This power is only indirectly relevant to the issues in this case, in so far as there was some evidence that the State of Victoria and its agencies have previously been granted exemptions under s 158 from the application of Pt 3 of the EPBC Act in relation to certain activities for the prevention and suppression of bushfires in 2009 and 2020.

Forests Act

104    The Forests Act confers on the Secretary the exclusive control and management of “all matters of forest policy”: s 5(1)(a). The “land manager” (relevantly either Parks Victoria or the Secretary) shall protect State forests, which include reserved forests under s 42(1) and proclaimed protected forests, and shall have the control and management of State forests: s 18.

105    The Secretary has responsibility for all matters relating to fire in any State forest (s 18AA(a)). Section 61A provides that there must be a CFO of the Department of Environment and Primary Industries (now known as DEECA). The Secretary may authorise the CFO in writing to perform any of the Secretary’s functions in respect of fire related activities in every State forest, national park or on protected public land (ss 29, 61B(a)).

106    Section 62(2) of the Forests Act imposes a statutory obligation on the Secretary to carry out work in State forests for the prevention and control of fire:

(2)     Despite anything to the contrary in any other Act or law, the Secretary must carry out proper and sufficient work in State forests, national parks and on protected public land—

(a)     for the immediate prevention and suppression of fire; and

(b)     for the planned prevention of fire.

107    In performing his or her functions and exercising his or her powers under s 62(2)(a), the Secretary is required to collaborate and consult with Emergency Management Victoria: s 61D.

108    Section s 62A(1) provides that the Secretary may apply and use fire in a State forest or national park or on protected public land for land and resource management: that is, as part of silvicultural activities; to control pest animals and pest plants; or to maintain, manage, protect or enhance the ecology of, or land or vegetation. In applying and using fire in a State forest or national park, or on protected public land, the Secretary must have regard to any relevant Code of Practice: s 62A(2).

Code of Practice for Bushfire Management on Public Land

109    As mentioned above, planned burning on public land in Victoria is undertaken in accordance with the Code, the purposes of which include setting “the objectives for bushfire management on public land, and strategies and actions across the prevention, preparedness, fuel management (including planned burning), response and recovery spectrum to achieve those objectives”, and providing “a risk analysis framework that assists the Department to effectively and efficiently achieve bushfire management objectives”. The primary objectives for bushfire management on public land in Victoria are:

•    To minimise the impact of major bushfires on human life, communities, essential and community infrastructure, industries, the economy and the environment. Human life will be afforded priority over all other considerations.

•    To maintain or improve the resilience of natural ecosystems and their ability to deliver services such as biodiversity, water, carbon storage and forest products.

110    Among other things, the Code identifies a number of “[p]rinciples for bushfire management on public land”. The first two of those principles are as follows:

Protection of human life as the highest priority

The protection of human life (emergency services personnel and the community) will be given priority over all other obligations in bushfire management.

Ecosystem resilience

Fire will be used deliberately in the landscape to meet legislative obligations and land management objectives, which promote ecological resilience and facilitate ecosystem services at a landscape level.

111    The principles also acknowledge that “[b]ushfire risk cannot be eliminated” and that “[d]ecisions about bushfire management will be based on risk analysis and will be transparent”.

112    Section 4 of the Code deals with fuel management including planned burning. The overview of this section of the Code states:

Under the Forests Act 1958, the Secretary may use fire for land and resource management and is responsible for carrying out proper and sufficient work in State forests, national parks and on protected public land to prevent and suppress bushfires. Fuel management activities will contribute to meeting this objective.

113    The overview then addresses planned burning as set out in the extract set out earlier in these reasons at paragraph 72.

114    The strategies identified in s 4 of the Code relevantly include reducing the fuel hazard on public land, and using planned burning on public land for ecosystem resistance. The actions include undertaking strategic planning for fuel management including planned burning, and using Fire Management Zones to implement local bushfire management objectives.

The parties’ contentions

The applicant’s case

115    By its originating application dated 14 June 2023, the applicant relevantly sought the following substantive relief:

1.    An injunction under s 475 of the Environment Protection and Biodiversity Conservation Act 1999 (Cth) (the EPBC Act) to restrain the Respondent from conducting planned burns in the Strathbogie State Forest.

2.    …

3.    A declaration that planned burns in the Strathbogie State Forest are required to be referred under s 68 of the EPBC Act.

116    The applicant’s amended statement of claim dated 26 October 2023 did not seek any declaratory relief, but rather sought an injunction under s 475(2) “restraining the respondent from undertaking or authorising the planned burning in the Strathbogie State Forest”. The applicant did not press any claim for a declaration in either its opening or closing submissions.

117    In the amended statement of claim, the applicant relevantly alleged that the planned burns in the Strathbogie State Forest constitute an action that will have, or is likely to have, a significant impact on the Southern Greater Glider, in that fire leads to a long-term decrease in the size of an important population of Southern Greater Glider, reduces the area of occupancy of an important population of Southern Greater Glider, and adversely affects habitat critical to the survival of the Southern Greater Glider. The particulars of this allegation set out in the amended statement of claim were as follows:

The decline of the Southern Greater Glider is accelerated by fire including because:

(a)    Some Southern Greater Gliders are killed directly by fire, being between 1-9 per cent of animals in the path of fire.

(b)    The Southern Greater Glider is dependent on hollow-bearing trees and low-severity planned burns cause hollow-bearing trees to decline by up to 26 percent.

(c)    Fire causes loss of food resources.

118    On 13 November 2023, pursuant to orders made by consent on 19 October 2023, the applicant filed further and better particulars of the allegation set out above, in the following terms:

(d)     Using the FRaME model described in the report of Adj. A/Prof. Philip Zylstra dated 13 November 2023, the burns proposed at Barjarg, Lima East, Strathbogie South, and Tallangallook can be expected to have the following severity:

(e)    The burns proposed at Barjarg, Lima East, Strathbogie South, and Tallangallook can be expected to cause a reduction in the number of hollow-bearing trees by between 9% and 47%. The fire mitigation measures lack the prescription of raking around hollow-bearing trees except in the small area that is in proximity to the control lines and burn boundaries, indicating that the overall risk of decline of hollow-bearing trees is at the higher end of this range.

(f)     Applying the 90th percentile severity of the table set out in paragraph (d), the burns proposed at Barjarg, Lima East, Strathbogie South, and Tallangallook can be expected to cause the following effects on Southern Greater Gliders:

(g)    These effects will kill some Southern Greater Gliders through partial thickness burns via asphyxiation following burns to their respiratory tracts.

(h)     High severity fires have been reported to cause a decline in Southern Greater Glider numbers by half in affected areas. A significant proportion of each of the fires proposed at Barjarg, Lima East, Strathbogie South, and Tallangallook will likely burn with high severity. This consequence is independent of the “intensity” of the fire itself.

(i)     The effects set out above constitute a significant impact on the Southern Greater Glider because this is an important impact on the species in that:

a.    The Risk Matrix in Table 2 of the Conservation Advice for the Southern Greater Glider identifies “inappropriate fire regimes” as having a “catastrophic” impact on the species as a whole.

b.    The nature of inappropriate fire regimes is described in Table 1 of the Conservation Advice for the Southern Greater Glider, where the following is stated under the heading “high frequency fires”:

Too intense or frequent planned burning may contribute to population losses or declines in the southern part of the greater glider’s range. Bluff (2016) reported that hollow-bearing trees (HBTs) affected by fire during planned burns were 28 times more likely to collapse than HBTs that were not burnt. Parnaby et al. (2010) found that following low intensity prescription burns in the Pilliga forests (NSW), mean collapse rates for burnt HBTs were 14-26%. This was consistent with the collapse rate of 25.6% found by Bluff (2016). A survey following a planned burn at Tallarook Range in the Central Highlands (Vic) in 2021 found that a large number of potential greater glider habitat trees were burnt, with “many destroyed” (N. Stimson 2021, Pers. comm. 26 June). There is increased pressure from some parts of the community to undertake more hazard reduction burning, follow the severe bushfires of 2019-20.

c.     Whilst the four proposed burns might not reach the level of “catastrophic” impact themselves according to the Risk Matrix in the Conservation Advice, by reasonable inference from the evidence as to the likely severity of the fires, the impact on the Southern Greater Glider reaches the level of “significant impact” in the sense of an important impact for the listed matter.

119    As will become apparent, these further particulars drew heavily on the report of the applicant’s expert witness Dr Philip Zylstra.

120    In opening submissions, counsel for the applicant submitted that the four proposed burns amounted to one “action” that is likely to have a significant impact on the Southern Greater Glider. The applicant relied on the Conservation Advice, including in relation to the taxonomy and distribution of the Southern Greater Glider, the significance of hollow-bearing trees as habitat, and the identification of key threats to the Southern Greater Glider including in particular “frequent and intense bushfires, inappropriate prescribed burning, climate change, land clearing and timber harvesting”. The applicant emphasised that the Conservation Advice recognises that “too intense or frequent planned burning may contribute to population losses or declines in the southern part of the greater glider’s range”. The applicant also relied on the Action Statement in relation to the Greater Glider issued under the FFG Act, including its acknowledgment that planned burns are among the factors that contribute to elevated mortality and the loss of hollow-bearing trees.

121    The applicant contends that the Strathbogie State Forest contains a large population of approximately 69,000 Southern Greater Gliders which is important to the conservation of the species. In the applicant’s submission, the area of the planned burns would comprise approximately 8% of the Strathbogie State Forest, potentially affecting approximately 5,600 gliders. The conduct of the planned burns would be likely to kill individual gliders in the affected areas, although the precise number of likely deaths is unknown. Further, it would cause the collapse of a substantial number of hollow-bearing trees in the areas reached by fire, and potentially cause structural damage to other hollow-bearing trees rendering them more vulnerable to future collapse. The applicant submitted that this amounted to a significant impact on the species.

122    As mentioned above, the applicant stressed in its opening submissions that “[t]he efficacy of planned fuel reduction burns as a tool in managing bushfire risks is not an issue in this case”. Rather, the applicant’s case focused on the impact of the four planned burns on the Southern Greater Glider. Among other things, the applicant contended that fire severity is relevant to: (i) the extent of injury to and mortality of gliders caused by heat, and that this could be approximated by the extent of “canopy scorch” caused by the fire; and (ii) the extent of collapse of hollow-bearing trees during or following the fire.

123    In relation to canopy scorch, the applicant relied in opening submissions on the model developed by its expert Dr Zylstra to predict flame height and canopy scorching, and contended that there would be some degree of canopy scorching to more than 50% of the areas in which the planned burns occurred, in contrast to the assumption on which the Secretary’s experts had proceeded of between 1% and 2% of the planned burn areas that would experience (high) canopy scorch. In this regard, the applicant took issue with the notion that it was possible to distinguish between degrees of canopy scorch (i.e. in terms of low, medium or high canopy scorch). The applicant submitted that Dr Zylstra’s predictions were consistent with past fire severity mapping of planned burns undertaken previously by the State, and that “[c]anopy scorching of the extent predicted by [Dr] Zylstra will cause injury and death to many [Southern Greater Glider] individuals due to direct heat exposure”. This was because canopy scorch was the best available “proxy” for heat and for flame height and length sufficient to give rise to a risk to gliders.

124    It will be necessary to return to these submissions below in the light of the oral evidence at the hearing and the applicant’s closing submissions which placed far less weight on Dr Zylstra’s modelling and greater emphasis on concessions that were said to have arisen from the evidence given by the Secretary’s witnesses.

125    In relation to the loss of hollow-bearing trees, the applicant submitted that there would be a substantial collapse of such trees that are reached by fire in the course of the planned burns, leading to the loss of dens used by the Southern Greater Glider.

126    In addressing the risks of both canopy scorch and the loss of hollow-bearing trees, counsel for the applicant placed some reliance in opening submissions on the timing of the planned burns during a period in which female gliders may be pregnant or nursing their young, and therefore more vulnerable and at risk of “hyper-predation”. The Secretary’s counsel objected to any reliance on such a contention, on the basis that it went beyond the scope of the case that had been pleaded and particularised by the applicant. In any event, this matter received little if any attention in the evidence, and there are therefore limits to the extent to which any weight can be placed on it.

127    In relation to the defence raised by the Secretary under s 43B of the EPBC Act, the applicant submitted that the conduct of planned burning on the relevant land comprising the burn areas does not fall within the exemption under s 43B, because planned burning is intermittent, irregular and infrequent, rather than a “continuation” of a use of land that was occurring immediately before the commencement of the EPBC Act. Further, the applicant submitted that most of the planned burn areas had never been the subject of fuel reduction burns, and that the current planned burns would involve an enlargement or expansion of fuel reduction burning in the Strathbogie State Forest, or a change in the location of where the use of the land is occurring that results in a substantial increase in the impact of the use of the land within the meaning of s 43B(3).

128    In relation to the Secretary’s reliance on the Melbourne Corporation principle to argue that s 18(3) is invalid in its application to s 62(2) of the Forests Act, or alternatively to the four planned burns, the applicant submitted that s 18(3) of the EPBC Act imposes a minimal burden on the State, in so far as it involves a procedural requirement to obtain approval rather than an absolute prohibition on planned burning, and that such a requirement has no bearing on the continued functioning of the State as a polity. The applicant further contended that s 18(3) of the EPBC Act cannot realistically be triggered by actions undertaken for the immediate prevention and suppression of fire under s 62(2)(a), and that planned burns under s 62(2)(b) of the Forests Act are not “immediate” works and involve sufficient time during the development of plans and before the conduct of the burn for the action to be referred for approval under Pt 9 of the EPBC Act.

The Secretary’s case

129    The Secretary filed an amended defence dated 27 October 2023. The Secretary relevantly pleaded that:

(a)    it is the “land manager” of the Strathbogie State Forest as defined in s 3(1) of the Forests Act;

(b)    it shall protect and, subject to the Forests Act, has the control and management of the Strathbogie State Forest pursuant to s 18 of the Forests Act; and

(c)    it must under s 62(2) of the Forests Act carry out proper and sufficient work in State forests for the immediate prevention and suppression of fire and for the planned prevention of fire.

130    The Secretary pleaded that it presently plans to carry out the planned burns in the four areas of the Strathbogie State Forest on a day or days in the period between 1 March 2024 and 31 May 2024, with the particular details of the conduct of those planned burns likely to change as further information is obtained and the plans are refined (including to adjust for environmental conditions on the day of a burn). The total size of the nominal burn areas is approximately 2,000 hectares, within which a smaller area will in fact be subject to burning as determined by specific operational planning and the identification of treatable fuels within each area. The Secretary pleaded that the object of the current planned burns is the reduction of fuel (comprising leaf and bark litter, twigs and like matter) in order to reduce the speed and intensity of bushfires, as part of a State-wide program for the planned prevention of fire, and that whether and when the planned burns go ahead will depend on conditions in autumn 2024.

131    The Secretary’s opening submissions make it clear that it is not part of the respondent’s case in this proceeding that “any immediate impacts on [Southern Greater Gliders] are to be netted off against the positive effects of the planned burns for the [Southern Greater Gliders]”, although the Secretary maintains that the purpose of the planned burns is to mitigate the risk of future bushfires (including for the benefit of the species).

132    The Secretary denies that the planned burns constitute an action that will have, or is likely to have, a significant impact on the Southern Greater Glider within the meaning of s 18(3). On the contrary, the Secretary submits that the impact of the planned burns on Southern Greater Gliders within each burn area is likely to be minimal, and that the evidence does not establish that any impact on individual gliders or their habitat is likely to have an impact (let alone a significant impact) on the listed species.

133    In this regard, in its opening submissions the Secretary submitted that the planned burns are not likely to produce the effects for which the applicant contends, which are said to rely on predictions and characterisations of the fire likely to be caused by the planned burns as “high severity”, and on eliding the fundamental differences between planned burns and high severity bushfires. The Secretary disputed the prediction made by Dr Zylstra and his “FRaME” model (as defined in paragraph 148 below) that the planned burns will produce “high severity” fire in a significant proportion of each planned burn area, having regard to the limits on the use of computer models to predict fire behaviour during planned burns, the various factors that can be manipulated by a knowledgeable operator of a prescribed burn in order to reduce the risk of undesirable fire behaviour, and the experience of senior officers in relation to the conduct and delivery of planned burns on past occasions. The Secretary submits that the planned burns are sophisticated operations conducted by personnel experienced in the use of ignition methods and other measures to control the intensity and spread of the fires.

134    The Secretary referred to the specific mitigation measures directed to the protection of gliders in the Delivery Plans for each planned burn. The Secretary contends that the available evidence suggests that arboreal animals such as the Southern Greater Glider regularly survive fires that do not significantly affect the canopy, and that the number of individual gliders that might be killed or injured through heat exposure is likely to be minor given the small area of canopy scorch that is predicted. The Secretary contends that a short-term reduction in the abundance of hollow-bearing trees in the burn areas would be unlikely to threaten the persistence of the population of Southern Greater Gliders, and that the areas that remain unburnt (such as south facing slopes and moist areas associated with drainage lines or creeks) will include the highest-quality habitat for gliders.

135    In its opening submissions, the Secretary submitted that there was a large population of Southern Greater Gliders in the Strathbogie State Forest which had continued without significant decline notwithstanding the influence of planned burning in the past. The Secretary therefore submitted that the evidence does not support a finding that the planned burns, separately or cumulatively, are likely to have any impact, or any significant impact, on the Southern Greater Glider as a species.

136    Further or alternatively, the Secretary pleads the following defences.

(a)    First, the Secretary says that the planned burns are a lawful continuation of a use of land that was occurring immediately before the commencement of the EPBC Act, and pursuant to s 43B of the EPBC Act are permitted to be carried out without an approval under Pt 9 of that Act. In this regard, the Secretary relies on the history of planned burning by State agencies and DEECA and its predecessors in the protection against bushfire, including the history of planned burns in the Strathbogie State Forest.

(b)    Secondly, the Secretary says that, to the extent that s 18(3) of the EPBC Act purports to prohibit the planned burns in the Strathbogie State Forest without an approval under Pt 9, it imposes a special disability or burden on the exercise of powers and fulfilment of functions of the State of Victoria which curtails its capacity to function as a government, and accordingly is invalid either in its application to the work pursuant to s 62(2) of the Forests Act or in its application to the four planned burns in the Strathbogie State Forest. By way of particulars of this defence, the Secretary says that it is an essential constitutional function of the government of the State of Victoria to protect its people and land from bushfire, including by managing and controlling the immediate prevention and suppression of fire and the planned prevention of fire.

137    As to the first of these two defences, the Secretary submits that it (and its predecessors) have long had statutory authority to engage in planned burning, which may amount to a relevant use of land for the purposes of s 43B of the EPBC Act. The Secretary submits that an action or activity need not be “continuous” or constant in order to amount to a lawful continuation of an existing use of the land, and that s 43B of the EPBC Act is capable of covering actions that are only carried out from time to time or that are inherently intermittent. It is submitted that the “land” that is the subject of the continuing use is a broader concept than the specific location on which each instance of the particular use is carried out, and is not limited to the precise areas which are the subject of a planned burn. In the present context, the Secretary submits that the “land” that was being used for planned burns immediately before the commencement of the EPBC Act is properly identified as the land comprising the Strathbogie State Forest, and to that end relied on evidence as to the history of planned burns in the Strathbogie State Forest since the 1960s, including the specific history of the current planned burn areas.

138    As to the second of the two defences, which relies on the principle recognised in Melbourne Corporation, the Secretary relied on and adopted the submissions of the Attorney-General. It may be noted that, as ultimately developed in written and oral submissions, the Attorney-General did not rely on the imposition of a special disability or burden on the State, in the sense of any targeted or discriminatory operation of s 18(3) of the EPBC Act. Rather, the focus of the submissions was on the practical operation of s 18(3) in its application to the Secretary so as to curtail or interfere with an essential constitutional function of the State.

The Attorney-General for the State of Victoria

139    The Attorney-General intervened in the proceeding pursuant to s 78A of the Judiciary Act in relation to the issue raised by the Secretary’s defence referred to in paragraph 136, namely whether s 18(3) of the EPBC Act infringes the Melbourne Corporation principle in its application to fire prevention and suppression works conducted by the Secretary under s 62(2) of the Forests Act or, alternatively, in its application to the four planned burns that are the subject of this proceeding.

Standing

140    The Secretary does not dispute that the applicant satisfies the definition of an “interested person” under s 475 of the EPBC Act and has standing to seek an injunction in relation to the proposed conduct of the planned burns.

141    In that regard, the applicant relied on evidence relating to its incorporation on 15 November 2019 under the Associations Incorporation Reform Act 2012 (Vic), its purpose under its Rules of “seeking to protect the Strathbogie Forest from inappropriate management, including native forest logging”, and its activities in relation to the protection of the Southern Greater Glider population in the Strathbogie State Forest. In relation to the latter, the applicant’s recent activities from about January 2023 include corresponding and meeting with officers from DEECA and FFMVic in relation to its concerns about the planned burns and their impact on Southern Greater Gliders (particularly as a result of the collapse of hollow-bearing trees), corresponding with State Ministers and members of Parliament in relation to the planned burns, conducting citizen surveys and making reports on the presence of individual gliders in parts of the Strathbogie State Forest, making freedom of information requests in relation to fuel load assessments carried out by DEECA in the planned burn areas, issuing media releases in relation to the planned burns and their impact on Southern Greater Gliders, holding public meetings, and maintaining a website and Facebook page containing publicly accessible information about its activities.

142    Having regard to its purpose and the applicant’s activities undertaken in respect of the Strathbogie State Forest, I find that the applicant is an “interested person” within the meaning of s 475(7) of the EPBC Act, on the basis that its interests are or would be affected by the proposed conduct, its objects or purposes include the protection or conservation of the environment, and it has engaged in activities related to the protection or conservation of the environment.

Overview of the evidence

The applicant’s evidence

143    The applicant relied on the following evidence:

(a)    an affidavit affirmed 14 June 2023 by its Chairperson, Mr Lobert (Lobert affidavit);

(b)    an expert report by Adjunct Associate Professor Philip Zylstra dated 13 November 2023 (Zylstra report), together with an expert report in reply dated 8 January 2024 (Zylstra reply report);

(c)    an expert report of Associate Professor Grant Wardell-Johnson dated 20 November 2023 (Wardell-Johnson report), together with an export report in reply dated 8 January 2024 (Wardell-Johnson reply report); and

(d)    an expert report of Dr Charles Meredith dated 19 November 2023 (Meredith report).

144    Each of the applicant’s expert witnesses gave oral evidence and was cross-examined.

Bertram Otto Lobert

145    Paragraphs 3 to 8 of the Lobert affidavit (and associated exhibits to that affidavit marked BL -2 to BL-9) were directed to establishing the applicant’s standing to seek the injunctive and declaratory relief sought in the originating application, including as an “interested person” who may apply for an injunction under s 475 of the EPBC Act, and I made an order under s 136 of the Evidence Act limiting their use to that question.

146    The Lobert affidavit also addressed the listing of the Southern Greater Glider as an endangered species under the EPBC Act and as threatened under the FFG Act, exhibiting copies of the Conservation Advice and Action Statement, and information from documents produced by DEECA and FFMVic about the planned burns in the Strathbogie State Forest more generally.

Adjunct Associate Professor Philip Zylstra

147    Dr Zylstra is an Adjunct Associate Professor at Curtin University in Perth, Western Australia. In 2011, he developed a forest fire behaviour model for the purposes of his Ph.D. at the Australian Defence Force Academy and the University of New South Wales. In addition to his research into fire behaviour, he has performed work for various public bodies in New South Wales in relation to fire management, including the planning and conduct of prescribed burns.

148    After developing the initial model for his Ph.D. thesis, Dr Zylstra worked on a project to “validate” the model, leading to the publication of a peer-reviewed journal article that he co-authored in 2016: P Zylstra et al, “Biophysical mechanistic modelling quantifies the effects of plant traits on fire severity: species, not surface fuel loads determine flame dimensions in eucalypt forests” (2016) PLoS One 11 e0160715. The model was subsequently further developed into the “Fire Research and Modelling Environment”, or “FRaME”, with the goal of producing “the first ever mechanistic fire effects model, capable of demonstrating direct or first-order fire impacts on fauna” (with the term “first-order” impacts referring to “immediate injury or death of individual animals caused by flames and/or smoke”). The Zylstra report refers to several ensuing reports and articles published by Dr Zylstra in which the FRaME model was used to predict fire behaviour and to assess the impacts on particular classes of fauna.

149    Among other things, Dr Zylstra considers that the FRaME model is better at predicting the height of flames than other models which are based primarily on surface fuel load. Dr Zylstra claims that FRaME is “a scientific tool developed with transparent methodology and peer-reviewed”, in contrast to the “McArthur-based industry-standard modelling” (which is explained in further detail below). Dr Zylstra considers that there is no relationship between fuel loads and the rate of spread of a fire. Instead, FRaME “models fire behaviour by calculating the flammability of leaves, and scaling this up to plants, plant strata and entire plant communities using the physics of convective heat transfer”. As Dr Zylstra accepts, this “requires descriptors of forest structure and composition that are not included in the Fuel Hazard Scoring system, so precise predictions of fire behaviour for these sites are not possible without additional field surveys”.

150    One of Dr Zylstra’s recent publications (P Zylstra, “Quantifying the direct fire threat to a critically endangered arboreal marsupial using biophysical, mechanistic modelling” (2023) Austral Ecol 48, 266-288) is said to have examined a prescribed burn conducted in Western Australia in which 77% of a population of critically endangered Western Ringtail possum (Pseudocheirus occidentalis) had been killed, and used FraME to model the causes and mitigating factors of mortality, so as to provide validation for the model by “correctly” predicting heavy mortality from burns and asphyxiation. It may be noted in passing that the particular study population considered in this article comprised 22 individual possums in an urban reserve south of Perth.

151    Dr Zylstra was instructed to describe the nature and intensity of the fire that is likely to occur in the Strathbogie State Forest pursuant to the proposed planned burns, in the highest degree of detail and with the greatest degree of particularity that he was able. The conclusions reached in the Zylstra report were summarised as follows:

a.     A significant proportion of each block will burn with high severity.

b.    High severity fire will likely kill a small to medium proportion of gliders through partial-thickness burns and via asphyxiation following burns to their respiratory tracts.

c.     Prescribed burns cause the collapse of a proportion of hollow-bearing trees that have previously been scarred by fire.

d.     Measures to mitigate the collapse of hollow-bearing trees focus only on trees immediately adjacent to control lines such as roads, protecting only the very small part of the area that is visible.

e.     Prescribed burns will cause scarring to the base of many hollow-bearing trees, making them vulnerable to collapse from future planned or unplanned fires.

152    Dr Zylstra considered that the scale or exact proportions of these impacts in terms of direct mortality and damage to habitat are unknown, but that they “may be quantified with detailed vegetation survey and data on animal hollow sites and behaviours”.

Predicted fire behaviour

153    In his report, Dr Zylstra took into account the prescriptions contained in the Delivery Plans and Operations Maps for each of the four planned burns, and made predictions of the likely fire behaviour using both his FraME model and the “industry-standard approach”, the latter of which is based on a model developed by Arthur Grant McArthur (see A G McArthur, “Fire behaviour in Eucalypt forests” (1967) (McArthur 1967)) (which is used for what is known as the McArthur meter) and is contained in a fire modelling tool called “Amicus” that was developed by the CSIRO. Dr Zylstra notes that the McArthur meter’s equations for flame height and rate of spread are also used in Phoenix Rapidfire modelling; these models are further discussed later in these reasons. The results of Dr Zylstra’s fire modelling using the “industry-standard approach” was that “fire intensity was predicted to be low to moderate”, and the predicted scorch height “was not sufficient to scorch the tree canopies”. This was contrasted with the modelled fire behaviour using FRaME.

154    To predict fire behaviour using FRaME, in the absence of actual data concerning the forest structure and composition in the planned burn areas, Dr Zylstra was required to construct “a series of model forests based on the dominant Ecological Vegetation Class” in each of the planned burn areas, namely herb-rich foothill forest, which he supplemented with site-specific details of fuel hazard values drawn from the burn Delivery Plans. This involved postulating a notional or hypothetical “stand” of vegetation that was populated with representative species for each of the applicable strata (canopy, midstorey, understorey and near-surface), in respect of which values were attributed for “cover” and height. The various “leaf traits and crown shape parameters” required for modelling in FRaME were taken from ratios for the relevant species in previous publications by Dr Zylstra. The manner in which this notional “stand” of vegetation was constructed, including in particular the selection by Dr Zylstra of representative species of canopy and midstorey trees, was the subject of some attention in his cross-examination.

155    Dr Zylstra then applied a “script” in FRaME in order to model fire behaviour by calculating 1,000 “predictions” for each burn block, with randomly varied values within defined ranges for plant species composition, canopy dimensions, slope and weather conditions. The code comprising this “script”, which was set out in an annexure to the Zylstra report, included parameters for slope, temperature, fuel moisture content and wind speed. Without reproducing here the detailed tables setting out the mean and maximum results or the figures showing the modelled flame heights and rates of spread against wind speed, the FRaME modelling predicted higher flames with a greater rate of spread in each of the planned burn areas than the values that were predicted by Dr Zylstra using the industry-standard tools.

156    Further, as depicted in Table 6 of the Zylstra report (set out immediately below), FRaME predicted higher fire severity, with crown scorch predicted for between 32% and 77% of each planned burn area (32%, 68%, 73% and 77% respectively) and crown fire of 1% in three of the four planned burn areas.

157    Addressing the interpretation of these differing fire behaviour predictions as between the industry-standard approach and the FRaME model, Dr Zylstra states in his report that “it is not possible to say with certainty which predictions are the most reliable, but there are several reasons to have greater confidence in those derived using FRaME”. These include the fact that FRaME has been peer-reviewed, whereas the claims made by McArthur (1967) regarding the influence of fuel load on fire behaviour and flame height are regarded as contrary to a range of peer-reviewed evidence. Dr Zylstra states that, in order to develop FRaME, he “developed models for leaf flammability and up-scaled this to plants and entire forests using simplified physics of heat transfer”, and “then tested this against a database of wildfire flame heights in diverse forests, finding that vegetation structure and composition accounted for most of the variability in flame heights”. The predictions differed “because the industry-standard expectations are derived from a measure that is unrelated to fire severity” (which, according to Dr Zylstra, refers to the effect of fire on vegetation).

158    Dr Zylstra accepted in his report that there was some uncertainty in the predictions of fire behaviour using the FRaME model. For example, he stated:

As already detailed, predictions using FRaME required estimates of structural and species traits, and these introduce uncertainty. For example, I estimated the heights of midstorey plants from a single photograph. If those plants were shorter in a physical burn block, then they would more often ignite, and if they were taller then they would ignite less often. There is reason, however, to expect that large flames causing crown scorch are possible. Past severity mapping shows some degree of crown scorch across much of the area as per Table 6. I also note that the elevated fuel hazard scores are very high (Table 2) for a forest described as having a typically open understorey. (Citations omitted.)

159    Accordingly, Dr Zylstra expressed the view that burning under low or moderate fire danger conditions will not guarantee low severity fire and that, “[a]lthough industry standard modelling predicts that crown scorch will not occur, this is dependent upon an already disproved relationship to fuel load, which has no bearing on fire severity”.

First-order impacts – external and respiratory burns

160    In the light of Dr Zylstra’s predictions regarding fire behaviour in the four planned burn areas, Dr Zylstra then addressed the impact of the planned burns on the Southern Greater Glider. As set out above, the ultimate conclusion expressed by Dr Zylstra in his report was that “[h]igh severity fire will likely kill a small to medium proportion of gliders through partial-thickness burns and via asphyxiation following burns to their respiratory tracts”. However, Dr Zylstra’s opinions on questions concerning the likely impact on individual gliders, the population of gliders in the Strathbogie State Forest and the Southern Greater Glider as a species were subject to an initial caveat:

To address these three questions adequately, I would require quantitative descriptions of the distribution of gliders in the proposed burn blocks, with details on the likelihood of their location and sheltering behaviour. Given this and the lack of precise detail on forest structure to enable better fire behaviour predictions, I am limited to more general answers.

161    In relation to the “risk environment” for the Southern Greater Glider, Dr Zylstra noted that there was a high risk posed by inappropriate fire regimes and habitat clearing. A risk assessment for the Southern Greater Glider carried out for DELWP (the predecessor to DEECA) in 2022 had assessed the risk posed by “increased fire frequency/intensity” as significant to high. Prescribed burning was listed as one of the mitigation controls, but its efficacy “was ranked as poor due to the ineffectiveness of prescribed burns under catastrophic conditions, and their impacts on hollow-bearing trees” (emphasis in original). Dealing with “first-order impacts”, the Zylstra report stated:

A NSW study found that first-order fire effects from even a single fire in ten years more than halved Southern Greater Glider numbers in their study area, arguing that such impacts only arose if high severity fire (crown scorch) occurred. The authors concluded that prescribed burns in glider habitat should only be conducted under “low/moderate fire weather conditions where canopy scorch does not occur.” In contrast, surveys of glider populations before and after burns in the Strathbogie Ranges have reported either minimal losses or slight increases, and the population appears to have remained stable. (Citations omitted).

162    Before coming to his predictions concerning the first-order impacts on gliders, Dr Zylstra acknowledged a number of “knowledge gaps”. These include that “[i]t has not been shown that crown scorch is synonymous with first-order impacts on Greater Gliders”, and that the recommendation to that effect “was taken from a personal comment that characterised fire behaviour across a large area with a single metric, and the assumption that this metric is adequate to explain the direct impact of fire on gliders”, referring to a 2018 paper by C M McLean et al, “The threatened status of the hollow dependent arboreal marsupial, the Greater Glider (Petauroides volans), can be explained by impacts from wildfire and selective logging” (2018) 415-416 Forest Ecology and Management 19-25 (McLean (2018)). Further, while crown scorch “may approximate the heat flux that will cause fatal injury to gliders”, Dr Zylstra noted that “gliders sheltering in hollows, coated in insulating fur, or climbing to tree tops may be at lower risk” and that “the threshold for injury to respiratory tracts may be lower than it is for foliage”. Conversely, however, Dr Zylstra considered that the data was “not yet sufficient to show that first-order effects of burning can be dismissed”, observing that “[a]lthough existing studies showing continued population growth of gliders at burn sites are promising, most of the survey transects were outside of the area mapped as burnt”.

163    In order to address these knowledge gaps, Dr Zylstra relied on FRaME to calculate first-order effects on gliders “accounting for animal morphology and behaviour”. Dr Zylstra used a script to model the likelihood of mortality due to burns of the body and respiratory tracts, based on assumptions that gliders would climb to a height of 23 meters (being 2 meters from the top of the mean tree height) and would shelter in hollows at mid-canopy height of 18.7 meters. The effects on gliders were modelled for the 90th percentile of modelled flame heights – that is, as I understand it from Dr Zylstra’s evidence in his cross-examination, taking the 900th highest flame from the results modelled for each burn area in the earlier section of the Zylstra report. The modelled first-order effects on gliders were set out in a series of graphs showing predictions generated by FRaME of the temperature over time in tree tops and hollows respectively in respect of air temperature and the temperature of the epidermis, papillary dermis and reticular dermis of an animal in the tree tops or sheltering in a tree hollow. The conclusions drawn from this analysis were set out as follows:

At the 90th percentile, exposed gliders in tree tops were expected to suffer partial-thickness burns and burns to the respiratory tract in all burn blocks, and full-thickness burns in the Lima East – Mt Albert block (Table 7, Fig. 6). Gliders sheltering in hollows were modelled to receive burns to respiratory tracts in all blocks. These results suggest that all gliders would be killed at the 90th percentile of modelled behaviour.

Second-order impacts – hollow-bearing trees

164    Dr Zylstra addressed the “risk environment” for second-order impacts by referring to previous estimates of collapse rates for hollow-bearing trees in Victorian prescribed burns. While the proposed mitigation for this impact was to rake fuels away from the base of trees, Dr Zylstra noted that such measures are “specifically focused close to control lines” and offered no protection to the main area of each burn.

165    Dr Zylstra opined that FRaME can be used “to calculate the likelihood of stem ‘girdling’ by fire, where heat penetrates to kill the cambium around the full stem”, and that he had added a function to FRaME to model fire scarring from flame contact with the tree bole after the fire front has passed. Dr Zylstra used published data in relation to wood density for the species Mountain Gum (Eucalyptus dalrympleana), together with information about bark thickness and assumptions about stem diameter. He then used FRaME to model the capacity for fire scar formation based on the 50th percentile of modelled flame heights (that is, taking the median or 500th highest flame from the results modelled for each burn area). The results were set out in a series of graphs showing the predicted temperature over time of the bark and at different increments to the cambium of the tree, with fire scarring occurring when the cambium is heated above 60 degrees Celsius. The Zylstra report concluded that “FRaME modelling predicted the capacity for fire scar formation in each block at the 50th percentile of modelled flame heights”.

166    Based on earlier research that found that hollow-bearing trees were more vulnerable to collapse if they had existing fire scars, Dr Zylstra expressed the conclusions set out above in paragraph 151 in relation to the risk of collapse of hollow-bearing trees in future planned or unplanned fires.

Zylstra reply report

167    In his reply report, Dr Zylstra addressed and responded to criticisms of his fire behaviour modelling, particularly by one of the respondent’s experts, Dr Trent Penman. In summary, Dr Zylstra stated that objections to his modelling “were predominantly grounded in hypothetical conjecture and unsupported claims of authority for older models”, and maintained that the FRaME predictions of canopy scorch and crown fire were consistent with mapping for previous prescribed burns.

168    Dr Zylstra considered that fire behaviour that is likely to impact Southern Greater Gliders mainly occurs away from control lines, where habitat trees do not receive protection, and that ground crews and photographers predominantly see only “the better-controlled fire close to control lines”. He stated that favourable post-burn monitoring of Southern Greater Gliders had almost entirely monitored areas that were not subject to crown scorch in which gliders had sought refuge. He therefore expressed his expectation that “impacts of prescribed burns on southern greater gliders are significantly higher than they are currently perceived to be”.

169    In defending the FRaME predictions as realistic, Dr Zylstra stressed that he had aggregated differing scorch categories (low, moderate and high) into the single category of “crown scorch”, based on the guidance contained in McLean (2018) that the use of prescribed fire as a tool to limit the spread and intensity of unplanned fires in habitats known to contain high densities of greater gliders “should only occur under low/moderate fire weather conditions where canopy scorch does not occur”. Dr Zylstra regarded this recommendation as referring only to a single aggregated category of “canopy scorch”, and he therefore adopted that approach (while keeping crown fire or “canopy burnt” separate). Accordingly, the FRaME predictions of crown scorch in 32% to 77% of the planned burn areas included “low canopy scorch”. Dr Zylstra noted that the category of low canopy scorch was not the lowest category of fire severity in burnt areas, as it was possible for fire to burn only leaf litter and low plants without scorching the tree canopy at all (which would be mapped from satellite photography as an unburnt area). I note here that the respondent’s witness Mr Lawlor maintained in his evidence that such areas would usually be classified as low canopy scorch, which included burnt areas in which there was 0% scorch, rather than as “unburnt”.

170    Dr Zylstra concurred with an observation made by Mr Lawlor that the maximum predicted flame heights using FRaME were extreme, but queried whether such possibilities were unlikely, referring to the Annual Review of the Hume Region Fuel Management Program 2022-2023 (2023 Hume Region Annual Review) in which 0.9% of planned burns areas in the Hume region were categorised as crown fire, aligning with the FRaME predictions. Dr Zylstra explained that his predictions of crown fires were “when a sufficient ‘ladder’ of understorey and midstorey plants was present to convey fire into the tree canopy”, and that this “occurred at low rates of spread in what is frequently referred to as ‘torching’, or ‘passive crown fire’ events” which “do not precipitate major runs of fire, because they involve the ignition of a single tree that then extinguishes” (in contrast to major fire runs resulting from active crown fires which “typically involve strong winds or steep terrain”).

171    The Zylstra reply report dealt with the range of options that are available to fire crews to control fire behaviour, and responded to Dr Penman’s claim that such techniques are sufficient to mitigate fire severity across each of the planned burn areas. Dr Zylstra queried the practicalities involved in many of those techniques, such as the manipulation of weather conditions, lighting patterns and topography. He provided a detailed discussion of such techniques in order to illustrate his general claim that the theoretical control of a prescribed burn diminishes significantly as fire spreads beyond ignition lines. He stated that “[t]he notion that light-up patterns and other skills of burn practitioners can be used to ensure low severity across burn blocks is not borne out by mapping”.

172    Dr Zylstra accepted in his reply report that it was “difficult to predict the fire behaviour of the proposed burns with accuracy”. However, he reiterated that his “purpose in providing predictions was not to state what the fire behaviour would definitely be, but to demonstrate the range of possibilities that could occur within the prescriptions for each burn”, in a context where “predictions made using the industry standard approach predict a very limited range of results that do not reflect the occurrence of canopy scorch and crown burn mapped for prescribed burns in the area”. Accordingly, he stressed that the purpose of his predictions “was therefore to determine whether first-order impacts on gliders are possible within the prescription envelope” (emphasis in original).

173    In relation to the inputs used for the FRaME modelling, Dr Zylstra notes that his approach provides a “significantly improved representation of each site” than the single value for fuel load or Forest Fire Danger Index (FFDI) on which the prescriptions for the planned burns are based. It is useful to interpose here a brief explanation of FFDI and Fire Danger Indexes more generally. The FFDI is a type of Fire Danger Index (FDI). The Australian Bushfire Glossary defines the term “Fire Danger Index” as “a relative number denoting the potential rates of spread, or suppression difficulty for specific combinations of temperature, relative humidity, drought effects and wind speed”. The FFDI is a summary statistic adopted by McArthur (1967) which forms part of the McArthur meter. In his oral evidence, Dr Zylstra described the FFDI as “a way of combining all the parameters together” (e.g. air temperature, wind speed, humidity and a drought factor) so as “to give some measure of how much fire risk there is in a certain day so that the hotter, drier, windier days come up as having a higher risk”.

174    In response to the criticism that FRaME is highly sensitive to small changes in the input variables, Dr Zylstra accepted that input errors could be made, the significance of which was not always apparent to users, but claimed that “these errors become less likely as the experience of the user increases”. In particular, Dr Zylstra claimed that his experience in developing and using such inputs over the last 20 years meant that he had “become adept at finding inputs that may be highly sensitive under some circumstances (e.g., preventing or facilitating crown fire)”, whereas “until FRaME is developed as a more user-friendly product, others do not have this option”. As will become apparent below, this self-confidence was perhaps belied by the discovery in the course of Dr Zylstra’s oral evidence of a significant error in the input of one of the variables (in relation to the range limits for dead fuel moisture content) that had been used to model the fire behaviour in the planned burn areas for the purposes of his expert report.

175    Dr Zylstra stated that “the validation of FRaME demonstrated its accuracy in predicting broad variation in fire behaviour from a low severity surface fire on one extreme, through to an extreme severity crown fire on the other extreme”, which had value for the purpose for which FRaME was used in his expert report, namely “to detect whether large flames that were damaging to the greater gliders were possible”. While he accepted that FRaME did not always provide “precise predictions for minor changes in [fire] behaviour” and that “important questions remain around the correct prediction of small flames”, Dr Zylstra considered that this was a distraction from the purpose for which FraME had been used in his expert report.

176    More generally, the Zylstra reply report addressed other issues that had been raised in relation to the FRaME model, including the status of critical comments made by other scientists following Dr Zylstra’s initial journal publication in relation to FRaME and his response to that critique, and the low rate of adoption of the FRaME model by others.

Oral evidence

177    Dr Zylstra was cross-examined on the development and design of the FRaME model in relation to the parameters used for different “stands” of vegetation. He indicated that a number of input values in relation to different classes of vegetation were based on estimates or approximations, which introduced some “noise” in the data (that is, “error in either direction”, whether exaggeration or underestimate) that could render the modelling “slightly less precise”. The purpose of the model was to test the relevance of different leaf traits and other vegetation characteristics on flammability and fire severity, using the physics of convective heat transfer.

178    Dr Zylstra said that he had proceeded on the basis that the predominant class of vegetation in the planned burn areas was herb-rich foothill forest. In constructing a forest stand for the purposes of the FRaME modelling, Dr Zystra chose Mountain Gum (Eucalyptus dalrympleana) as the representative species for canopy trees, notwithstanding that it was not on the list of species contained in the Ecological Vegetation Class benchmark issued by the Department of Sustainability and Environment (a predecessor of DEECA) in 2009 for the Highlands-Northern Falls Bioregion in respect of Herb-rich Foothill Forest (EVC 23) (EVC Benchmark). Dr Zylstra said that the EVC Benchmark does not contain an exhaustive list of representative species, but rather only gives “benchmark” species. Dr Zylstra had instead relied on a different source (Nelson (2018)) in making his selection of Mountain Gum, on the basis that it is a species favoured by the Southern Greater Glider and he was “particularly interested in looking at those parts of the forest where the gliders are most likely to be”. In responding to a suggestion that Nelson (2018) had identified Narrow-leaf Peppermint (Eucalyptus radiata) as a dominant species which was considered to be preferred by the Southern Greater Glider, Dr Zylstra said that Nelson (2018) had not found evidence to confirm that belief and had noted a strong positive relationship between the abundance of Southern Greater Gliders and the basal area of both Mountain Gum (Eucalyptus dalrympleana) and Blue Gum (Eucalyptus globulus). Ultimately, his choice was influenced by the fact that he already had data on the Mountain Gum: thus, he said that “it’s a matter of which species do I have the best information on hand for myself to put that in”, and that he “chose a species that the paper said has a strong relationship with gliders and for which I have good inputs for without driving down to Strathbogie and measuring them myself”.

179    In this context, Dr Zylstra explained:

I’m not trying to create an authoritative model on that forest because I don’t have that data and I’ve made that quite clear in the report. All I’m trying to do is seeing if we can improve on the modelling that guides the prescription and that modelling is based on the FFDI and the fuel load.

…

So I suppose that’s the premise we’re starting from that the reason these burns have been designed in this way is because there is already a simplistic form of modelling that has said, “These conditions are safe. We know that it’s safe to burn under this wind, this temperature range, etcetera, because we’ve measured this fuel load, we’ve put it into this model and we’ve now got an FFDI that tells us it’s safe within that range.” And my question was: can I improve on that?

180    Dr Zylstra was further interrogated in some detail on the composition of the forest “stand” that he used in his modelling, and the data that was used for each species that was selected. The tenor of Dr Zylstra’s evidence was that, in populating his “stand”, he was “trying to approximate something that I can’t see … and trying to improve on a single number of fuel load”, and that he had generally selected common species on which he already had some data from his previous work. Dr Zylstra acknowledged that he had not visited the Strathbogie State Forest. He did not accept that the forest stand that he had constructed for modelling purposes was “hypothetical”, because he regarded it as being underpinned by evidence rather than pure hypotheses. Nevertheless, he described his forest stand as “theoretical”, in that it was “an approximation ... based on raw data”.

181    Dr Zylstra explained that he had used the FRaME model to test what would occur if his theoretical forest stand was ignited, by running random variations within defined parameters for different vegetation structures and plant traits under varying weather conditions. The code for each of those variables generally contained a specified value together with a given range and standard deviation. The datasets in respect of vegetation in the planned burn areas were chosen by Dr Zylstra based on available data and his own experience, and involved some degree of subjectivity. He derived the variables for conditions such as temperature, wind and fuel moisture content from the prescribed ranges contained in the Delivery Plans for the planned burns. Dr Zylstra did not use actual data in relation to the topography of the planned burn areas, because he considered that the prescriptions for the planned burns were designed to limit the influence of slopes – for example, “your intention is … to try to not have fire governed by the terrain of the site but that you take some degree of control of that with your lighting pattern”. Accordingly, his model excluded steep slopes and instead used a narrow range of slope in order to “keep things close to level ground”.

182    Dr Zylstra confirmed that, when running the random variations to simulate fire behaviour, the FRaME model did not correlate or link any of the different variables, such as temperature and fuel moisture content. Dr Zylstra considered that there was insufficient information to support such an approach – “for example, you can have a hot day with still a higher dead fuel moisture content because it rained the day before” – and that he was “trying to use the raw numbers that are given in the prescription” without introducing too many additional assumptions. He agreed that the prescription set the range or outer limits, rather than stating what the conditions were likely to be on the day that the burns were conducted.

183    It is convenient here to set out an extract from the code that was used by Dr Zylstra in FRaME to model fire behaviour in each of the planned burn areas:

#BUILD PARAMETER FILES

if (a == 1) {

temp <- 21.5

tempRange <- 7

} else {

temp <- 22.5

tempRange <- 9

}

base.params <- buildParams(Structure = ST, Flora = FL, default.species.params = TR, a = a)

#MODEL PROBABILISTIC FIRE BEHAVIOUR

probFire_Frame(base.params, Structure = ST, Flora = FL, a = a, db.p“th = "out”mc.db",

slope = 0, slopeSD = 5, slopeRange = 10,

temp = temp, tempSD = 5, tempRange = tempRange,

DFMC = 0.125, DFMCSD = 0.05, DFMCRange = 7,

wind = 10, windSD = 10, windRange = 20,

jitters = 1000, l = 0.1, Ms = 0.01, Pm = 1, Mr = 1.001,

updateProgress = NULL, testN = 10, threshold = 1)

184    While this is only a small extract from a lengthy and complicated computer script, it was addressed in oral evidence as the part of the code that primarily deals with the variables that were derived from the Delivery Plans for the planned burns. Thus, if the Delivery Plan prescribes a temperature range between 18 to 27 degrees Celsius, this indicates a range of 9 degrees Celsius around a mid-point of 22.5 degrees Celsius. The standard deviation was chosen by Dr Zylstra, and could potentially be greater than the range parameters (which would have the effect of “flattening” the bell curve of the results).

185    The number of “jitters” indicates the number of times the simulation was run, namely 1,000. Although it was not entirely clear from Dr Zylstra’s oral evidence, this seems to have been treated as representative of the possible fire behaviour at 1,000 possible points or locations within each of the planned burn areas, each with different variations of weather conditions and plant conditions, rather than the possible fire behaviour throughout the relevant burn area under different conditions on 1,000 possible occasions. Thus, the percentages for the area in which there was canopy scorch set out in Table 6 represented the percentage of randomly generated runs of the model that resulted in canopy scorch in the theoretical constructed stand of forest vegetation. On the other hand, Dr Zylstra accepted that it was difficult to express the scenarios in terms of a percentage of the area, as opposed to a percentage of modelled cases.

186    There might be no material difference between these two approaches for the purposes of the conclusions sought to be drawn by Dr Zylstra. However, it seems to indicate some potential limitations on the practical use of the results of such modelling. For example, while there might be some variation in the conditions present in different parts of each burn area on any particular day, those conditions are unlikely to be entirely unconnected or random. Similarly, the ignition of a single theoretical forest stand on 1,000 different occasions is not necessarily indicative of the way in which fire is likely to behave across each entire planned burn area on any given occasion.

187    Another issue affecting Dr Zylstra’s modelling arose in the course of his cross-examination, when it became apparent that he had made an input error in the line of code relating to the parameters for dead fuel moisture content or “DFMC”. The relevant Delivery Plans prescribed a range for DFMC between 9% and 16%. As set out above, this was entered by Dr Zylstra into the code as “DFMC = 0.125, DFMCSD = 0.05, DFMCRange = 7”. Converting decimals into percentages, this represented a value for DFMC of 12.5% with a standard deviation of 5%. However, instead of entering a value of 0.07 for the DFMCRange (which would translate to a range of 7%, that is, between 9% and 16% in accordance with the applicable prescription), Dr Zylstra inadvertently entered a value of 7 (which would translate to a percentage of 700%). The effect of this error was to remove any range limits in the variations for DFMC. As a result, while there was still a standard deviation of 5%, approximately one-third of the 1,000 variations run by Dr Zylstra using FRaME would have been outside that standard deviation and would have used a value for DFMC that was outside the prescription in the Delivery Plans. Further, because the standard deviation for DFMC was greater than the prescribed range, some of the DFMC values within the standard deviation would still have been outside the range for the planned burns.

188    Dr Zylstra clarified that, on an examination of his data, he estimated that approximately 26% of the scenarios modelled by FRaME for the purposes of his report had used a DFMC below the prescribed minimum of 9%. While there were also some modelled scenarios that used a DFMC above 16%, Dr Zylstra assumed that this would have resulted in lower flammability (and therefore less severe fire). Dr Zylstra maintained that an underestimate in the DFMC would not have significantly affected the results or his conclusions and “was inconsequential”, on the basis that he considered that the overall results demonstrated that DFMC was not a significant driver of flame heights.

189    The error in the DFMC range came to light in relation to a “case study” that was included in the Zylstra report to illustrate a scenario of possible crown fire in the Lima East planned burn area. This case study was identified by Dr Zylstra from the 1,000 variations that were modelled by FRaME for that area. Dr Zylstra acknowledged that this particular scenario was at the more extreme end, representing the 1% of modelled variations that involved crown fire (that is, where flames ignite the canopy). The conditions applicable to the selected case study included a DFMC of 8% (that is, below the prescribed range of 9-16%), and a temperature of 26 degrees Celsius (that is, towards the upper end of the prescribed range of 18-27 degrees Celsius). Dr Zylstra said that, while an unrealistically low DFMC could produce a result with “too big a surface flame”, the FRaME model included “hard” limits on the length of surface flames based on what experiments had shown was possible. The predicted surface flame length in the case study was 0.7 meters, which he regarded as “quite a modest surface flame”, indicating that the low value for DFMC did not have a significant impact on the fire behaviour prediction. The total height that was reached by flames in this case study was 30.1 meters. Dr Zylstra explained that this was not referring to the length of a continuous flame from the ground, but rather that the flame climbed up the vegetation to that height so as to produce fire burning in the top of the tree. When asked why he had included this particular case study in his report, Dr Zylstra said that it was to show that it was possible that a large flame could occur in 1% of cases, albeit that this was “not typical of the fire behaviour through the block”.

190    Dr Zylstra was taken in cross-examination to some of his correspondence with the applicant’s solicitors in connection with his instructions to prepare an expert report. In an email dated 25 October 2023, Dr Zylstra referred to the “limitations” that were “going to be an issue”, in that “[t]he letter of instruction is asking for specifics of fire behaviour and its effects on gliders … and these are the things I said that I wouldn’t be able to provide without a detailed survey”. He advised that in order to provide answers “to a level of accuracy that is better than [FFMVic] already have”, he would “need to conduct surveys across diverse terrain and forests, then process and model them”. Dr Zylstra said that, if he had information on the forest structure, he could “use FRaME to model hypothetical scenarios that show fire behaviour and damage that their approach will miss” (emphasis in original), which “won’t give probabilities that they will kill gliders, but it will show that they can’t claim they won’t [kill gliders]”. In cross-examination, Dr Zylstra said that he had subsequently formed the view that he was “copping out a bit too easily” in the position taken in his email. While he had originally been “hesitant to make predictions”, he ultimately decided that a detailed survey was not necessary in order for him to improve on FFMVic’s predictions which were based entirely on fuel load. This was primarily because he became aware that he could use information on the fuel hazard scores in the planned burn areas, together with the general descriptions of the forest structure taken from available publications such as the EVC Benchmark.

191    In relation to his assessment of the impacts of the predicted fire on the Southern Greater Glider, Dr Zylstra accepted that his training and expertise was not “glider-focused” and that he had sourced relevant facts about Southern Greater Gliders from elsewhere. He had modelled the likelihood of mortality by reference to the 90th percentile of flame height in order to represent a scenario at the “worse end” in which there was a “genuine possibility of negative outcomes for the gliders”. Dr Zylstra said:

I could have taken the – the median possibility, but the question was not, are gliders likely to suffer damage, you know, over half the block. The question was, is there a real genuine risk to the gliders? And the genuine risk … is likely to happen in those worse ends of fire behaviour. So I didn’t want to take the most extreme example, but I went … back to the 90th percentile. So you’ve got about – roughly, you can estimate a one in 10 chance of … this occurring. So it’s not an insignificant chance, but it’s a real chance that gliders in these situations may suffer these outcomes.

192    The flame heights at the 90th percentile in each planned burn area would be below crown height, because there would not be fire in the canopy. However, the Zylstra report did not set out the data in relation to the precise flame height at the 90th percentile, nor the input variables or detailed predictions produced by FRaME for that particular scenario. Dr Zylstra considered that the most important “drivers” of flame height were wind speed and vegetation structure, and that temperature and DFMC had much smaller effects. In modelling effects on Southern Greater Gliders from the effects of heat, Dr Zylstra used the single scenario at the 90th percentile of flame height as “a point that is towards the bad end but not completely unlikely”. Dr Zylstra accepted that it might be better to have been able to model the effect on Southern Greater Gliders for a wide range of fire scenarios and then take a percentile of effects on the Southern Greater Gliders, but that this would not be logistically feasible because it took a long time to run each of the models.

193    Dr Zylstra’s conclusions in relation to the effects of fire on Southern Greater Gliders assumed that a glider was in a tree or a hollow that was reached by fire. His opinion that a “small to medium proportion” of gliders would be killed by fire appears to have been based on his selection of the 90th percentile (at which Dr Zylstra’s modelling indicated that gliders would be killed), in that “10 per cent is a small proportion” and he couldn’t rule out that Southern Greater Gliders would also be killed at slightly lower percentiles with less severe fire behaviour. In re-examination, Dr Zylstra said that the level of heat at the 90th percentile was well above the threshold that would cause mortality to gliders, and that it would need a significant reduction in flame height not to reach that threshold. Accordingly, although he did not quantify the number of gliders that would be killed by fire, he gave an estimate of a “small to medium” proportion. This seems to proceed on an implicit assumption that the various scenarios modelled by FRaME can be spread across the burn area so that, for example, 10% of the area would have flame heights at or above the 90th percentile with corresponding effects on the gliders that happen to be in those locations. It also implicitly assumes an even distribution of gliders across the planned burn areas. There did not appear to be any direct evidence to support either of those assumptions.

194    Dr Zylstra did not specifically perform any modelling in relation to the collapse rates of hollow-bearing trees, but relied on estimates derived from other sources. In modelling the risk of scarring to Mountain Gums, he had selected the 50th percentile or median flame height on the basis that the height of flames was less important when looking at scarring to the base of hollow-bearing trees.

195    Dr Zylstra was also cross-examined at some length on his experience and qualifications, and some challenges were made to his independence from the applicant. Those matters were not ultimately relied upon as going to the admissibility of Dr Zylstra’s evidence. In so far as they have a potential bearing on the weight to be given to Dr Zylstra’s opinion, I will return to these matters below.

Adjunct Associate Professor Grant Wardell-Johnson

196    Dr Wardell-Johnson is an Adjunct Associate Professor in the School of Molecular and Life Sciences in the Faculty of Science and Engineering at Curtin University. He has qualifications in the fields of forest ecology and botany, including specialised knowledge in relation to fire in native forests. His previous experience includes providing expert opinion evidence in relation to the impact of timber harvesting operations in East Gippsland on the Southern Greater Glider and the Yellow-bellied Glider.

197    Dr Wardell-Johnson was asked to provide an opinion on the characteristics and distribution of the Southern Greater Glider, and the likely impact of planned burning on individual gliders, the population of Southern Greater Glider in the Strathbogie State Forest and the Southern Greater Glider as a species.

198    Dr Wardell-Johnson had not previously prepared any reports on the Southern Greater Glider in the Strathbogie State Forest, but for the purposes of his report he visited the Forest including the proposed burn areas in order to conduct “field reconnaissance”. Among the points which are said by Dr Wardell-Johnson to have emerged from this field visit are:

(a)    “the vegetation is (in general) in very good condition, with many old-growth characteristics”, and “[i]t is therefore unsurprising that there is a healthy population of [Southern Greater Gliders] (as found by Nelson et al 2018) in Strathbogie Forest”;

(b)    “Strathbogie Forest includes an array of vegetation types (including Damp Forest and Herb-rich Foothill Forest) with abundant mature trees and old-growth characteristics that provide excellent habitat for [Southern Greater Gliders]”; and

(c)    “[t]he initial prescribed burns would certainly kill some [Southern Greater Gliders] (though the numbers would be difficult to predict without detailed knowledge of their whereabouts or prior knowledge of the outcomes of the fires)”.

199    After describing the species of Southern Greater Glider and its distribution, Dr Wardell-Johnson notes in his report that “[c]learing, fire, logging and fragmentation have long been recognised as the major threats to the [Southern Greater Glider]”. Dr Wardell-Johnson describes the Southern Greater Glider’s habitat and reproductive cycles, as well as where in Victoria the Southern Greater Glider is found. Dr Wardell-Johnson states:

Within Victoria, the [Southern Greater Glider] is distributed throughout the forested parts of the east, including inland and southern falls of the Great Dividing Range, as well as the Strzelecki and Strathbogie Ranges. In eastern Victoria, the [Southern Greater Glider] is absent from high altitude alpine and sub-alpine habitats, Wilson’s Promontory and cleared areas. In western Victoria, [Southern Greater Gliders] are found in forested areas, west to Daylesford, but are absent elsewhere in western Victoria, including large areas of potentially suitable habitat such as the Otway Ranges and Grampians.

200    Dr Wardell-Johnson states in his report that “[t]he forests and woodlands of the Strathbogie Ranges have been subject to a long history of disturbance and modification”, and that “[f]ire has also influenced the Strathbogie landscape, predominantly through fuel reduction burning but also with some smaller scale bushfire events”. However, a large proportion of the Strathbogie State Forest has not been impacted by a large-scale bushfire for a significant time period, so that “it is likely that there are large proportions of older growth state vegetation which have not been fire affected since the 19th century, with some areas exhibiting centuries-old trees”.

201    Dr Wardell-Johnson states that the Southern Greater Glider “is found on a variety of forested areas of the Strathbogie Ranges on highly fertile soils such as areas of Damp Forest (usually south-facing slopes at higher elevations) and Herb-rich Foothill Forest (approximately 60% of the public land estate of Strathbogie forest – easterly and southerly aspects of moderate to high rainfall and well-drained soils – often on the lower slopes or gullies)”. He refers to Nelson (2018), which estimated the population of gliders in the Strathbogie Ranges to be about 69,000, with mean densities of 2 to 4 individuals per hectare across the study area. Dr Wardell-Johnson considers that the Nelson (2018) study “supported the contention that a large and regionally important population of the Greater Glider occurs in the Strathbogie Ranges”, and “suggested that the Strathbogie Ranges population has not declined to the extent that has been observed elsewhere in Victoria”. Dr Wardell-Johnson described the Strathbogie State Forest as “an extremely important conservation area for [Southern Greater Gliders]”, and “effectively an island of high-quality habitat (for Southern Greater Gliders) on the western margins of the distribution of this species”.

202    While Dr Wardell-Johnson was not able to provide a map showing the locations of individual Southern Greater Gliders within the Strathbogie State Forest, which would require a targeted survey, he stated that it is known that there are Southern Greater Gliders within each of the planned burn areas. According to Dr Wardell-Johnson, each of the planned burn areas have substantial proportions of habitat suitable for Southern Greater Gliders. Dr Wardell-Johnson considered that the mitigation measures contained in the Delivery Plans to reduce the impact on gliders, at least in their application to protected areas along boundary tracks, “would be inconsequential in terms of safeguarding habitat trees or of individuals, populations, or the species from the planned prescribed burn operations”. Rather, he described these as “basic operational fire control measures that are usually implemented to prevent escape of prescribed burning operations and to ensure they are implemented as scheduled”, which “would have negligible bearing on the safety of [Southern Greater Gliders] in relation to these proposed operations”.

203    In response to questions about the likely impact of the planned burns on the Southern Greater Glider, Dr Wardell-Johnson referred in his report to a risk assessment that had been carried out by Dr Jenny Nelson in 2022 for Victoria’s Regional Forest Agreement areas, which had nominated three risks to the Southern Greater Glider associated with fire, namely: bushfire, increased fire frequency/intensity; and inappropriate fire regimes – prescribed burning (J Nelson, Victoria’s Regional Forest Agreements: Risk Assessment for the Southern Greater Glider (2022). Dr Wardell-Johnson largely summarised aspects of Dr Nelson’s assessment, before concluding:

Some [Southern Greater Gliders] will die as a direct result of the proposed prescribed burns (i.e., burnt, injured, or suffocated) though it remains uncertain how many. This will only be known by having an accurate census prior to the fires and at various times after the fires. Further, some [Southern Greater Gliders] will die soon after the fire from predation, starvation, or exposure (see Wardell-Johnson and Robinson 2022, Tyndale-Briscoe and Smith 1969). Again, it would be uncertain as to how many, and actual number would again rely on an accurate census having been taken prior to and at various times after the fire.

…

The processes associated with the deaths of [Southern Greater Gliders] following the proposed burns would only be known by conducting detailed habitat measures prior to and following the fires (with appropriate controls and replication to ensure validity and reliability of the findings – see earlier background information, Green 1979). Considerable focus is directed by land management agencies towards examining the impacts of prescribed burning on [Southern Greater Gliders] and their habitat (see Macak et al. 2021). However, the reliability and validity of research to inform understanding of the effects of fuel management on threatened species can be fraught (see Wardell-Johnson et al. 2004). Research design requires considerable thought if the results of research endeavour are to be both valid and reliable.

204    In his reply report, Dr Wardell-Johnson stated that, although many precautions are taken to implement planned burns according to their prescriptions and safeguards, the natural world is changeable. Thus, weather and other conditions can change during the implementation of the fire, and “despite the best intentions, much can go wrong after the fire (e.g., if not fully blacked out along edges, or if very strong winds carry sparks beyond boundaries following implementation of the burn etc)”, adding to uncertainty and the risk of impact on Southern Greater Gliders.

Oral evidence

205    Dr Wardell-Johnson confirmed that his expertise in forestry management and disturbance ecology included publications on fire as a disturbing agent. He has been involved in providing expert evidence in previous court proceedings in relation to the impacts of timber harvesting on the Southern Greater Gilder. He did not accept that his previous work and experience in relation to the impact of fires was confined to wildfires or intense fire, as opposed to low intensity planned burns.

206    Dr Wardell-Johnson described that the purpose of the field visit he undertook in the Strathbogie State Forest was to familiarize himself with the ecology and vegetation of the area, as he had not previously visited the Forest.

207    Dr Wardell-Johnson accepted that the primary causes of the decline in Southern Greater Gliders in New South Wales and Victoria had been significant logging of areas of habitat containing hollow-bearing trees and the effects of the “Black Summer” bushfires in 2019-2020. He did not suggest that this decline was caused by planned burning. Nevertheless, he considered that the planned burns would be a form of disturbance that will be detrimental to Southern Greater Gliders. He considered that it was important to ensure that areas of suitable habitat were protected from “any management activity … that might tip the balance or not make the environment as suitable for the species”.

208    In relation to the risk or threat to the Southern Greater Glider from inappropriate fire regimes, Dr Wardell-Johnson accepted that there were a range of fire intensities or severities. His opinion that some gliders would die as a result of the planned burns was based on an assumption that the fire is “on the higher level of severity” or “on the higher end of the spectrum”, which “would probably be because of flames going up the trunks of trees or … being carried through the vegetation”. Nevertheless, he considered that this would “undoubtedly” happen because planned burns vary in their intensity and severity, so that there would be some patches or “intense pockets” that burn with higher levels of severity. While Dr Wardell-Johnson had some experience with prescribed burns in Western Australia, he had no operational experience of conducting planned burns in Victoria in autumn.

209    Dr Wardell-Johnson considered that most of the gliders that he predicts might die “won’t be from the immediate effects of the flames”, but rather as a result of “microclimate changes” and an increased risk of predation in the months following the fire which might put the gliders “on a downward spiral”. Dr Wardell-Johnson accepted that there was considerable uncertainty as to the number of Southern Greater Gliders likely to be affected by the planned burns, and he considered that surveys should be conducted involving the tagging of individual gliders to determine the extent and causes of mortality. He maintained that there would be an impact on Southern Greater Gliders, but it was difficult to quantify that impact without further surveys.

Dr Charles Meredith

210    Dr Meredith is a scientist who specialises in forest and fire ecology, with particular experience in the management of endangered species. He has been involved as a consultant in multiple ecological and impact assessments of forest sites in Victoria, New South Wales and Tasmania, including many within the range of the Southern Greater Glider. Dr Meredith was asked to provide his expert opinion on questions relating to the characteristics and distribution of the Southern Greater Glider, including its presence in the Strathbogie State Forest, and the likely impact of planned burning on the Southern Greater Glider.

211    Dr Meredith referred to the Action Statement for the Greater Glider as providing “an excellent summary of the species’ ecology”. Among other things, he stated that the Southern Greater Glider “shelters and breeds in tree hollows, generally in tall older or dead eucalypt trees”, showing “a preference for larger tree hollows and for living trees with large hollows”, and that they “will have more than one den tree within their home range, typically five to six”, which provide both shelter from predators and a thermally stable environment on hot summer days. Dr Meredith noted that the Southern Greater Glider is “very sensitive to overheating”, and that it has a low reproductive rate.

212    Dr Meredith stated in his report that the Southern Greater Glider occurs in suitable habitat throughout the Strathbogie State Forest, which supports some of the highest recorded densities of the Southern Greater Glider in Victoria, and where the population appears not to have declined significantly in the last 20 years (in contrast to other parts of Victoria). Again, this appears to be consistent with, if not drawn from, sources such as the Conservation Advice and the Action Statement.

213    Dr Meredith observed that the Southern Greater Glider is highly sensitive to high intensity fire and that, while the planned burns are not intended to be of high or even moderate intensity, “the reality on the ground is that fire intensity within a planned burn can be highly variable and will often include localised (and sometimes quite large) areas that burn at higher intensities”. In support of that statement, Dr Meredith cited the Zylstra report, as well as his own personal observations. Although Dr Meredith accepts that he is not an expert in modelling fire behaviour, he relied on his own observations of the outcomes of planned burns “over nearly 50 years” to state that “[w]hile some planned burns do go largely to plan, many don’t”. In particular, a burn may be more intense than planned – “[m]ost frequently, these higher intensities are localised at sites where the vegetation is more flammable than average, conditions were not as predicted or where dead timber or tree bases have ignited and burnt at high temperatures for a considerable period”.

214    In his report, Dr Meredith opined that “[f]ire intensities that are too high can directly kill [Southern Greater Gliders] through burning them, fatally damaging their respiratory tracts through heat or smoke inhalation, or through physical injury caused when they are in a falling tree that has collapsed after being weakened by the burning through of its trunk”. In his oral evidence, however, Dr Meredith accepted that the number of gliders that would be directly killed as a result of a planned burn is likely to be low.

215    Dr Meredith continued:

Fire can kill [Southern Greater Gliders] indirectly through the enhanced rates of loss of habitat trees caused by trees collapsing after being damaged by fire. Even quite mild fires can lead to tree loss if they ignite the base of a tree which then slowly burns until the tree is significantly weakened. Denning trees for the [Southern Greater Glider] are critical components of their habitat. Den trees must have suitable hollows and, while some may be dead stag trees, the best den trees are thought to be tall, old trees in good condition that would normally be expected to stand for many decades. Such trees are a scarce resource (much depleted in Victoria through timber harvesting, land clearance and extreme fires) and cannot be replaced in the short to medium term as young regrowth trees do not have large hollows.

216    Accordingly, Dr Meredith considered that, if the planned burns proceed, “there is an appreciable risk of a highly significant impact on the [Southern Greater Glider] population within the Strathbogie State [F]orest”. In his oral evidence, Dr Meredith qualified this by removing the word “highly”. Dr Meredith concluded that “[g]iven the apparent importance of this population (high density, stable), this would constitute a significant risk of impact at the State level and potentially at the national level”.

Oral evidence

217    During his cross-examination, Dr Meredith addressed the significance of the population of Southern Greater Gliders in the Strathbogie State Forest and its importance to the species:

What was in my mind is very much that this [the Strathbogie population] is a particularly important population. We know that it has been at a higher density than other areas for decades, and we know that it still is a higher density than other areas when other areas are largely declining. So that makes this a particularly important sub-population. We don’t know why any of those factors are occurring or what factors lie behind it, which increases the risk, if you like, of doing the wrong things. We don’t know what’s going on. But that’s what was in my mind about it being a significant – potentially significant impact at the state and possibly national level. So it’s the basis that the Strathbogie population seems to have some special characteristics that will make it long-term very important for the conservation of the species in Victoria.

218    Dr Meredith accepted that there were significant differences between planned fuel reduction burns and large, severe wildfires, particularly in relation to intensity. He agreed with the view expressed by the respondent’s expert Dr Swan that a low-intensity fire that mostly affects understorey vegetation will have little direct impact on Southern Greater Gliders through exposure to heat. This was consistent with Dr Meredith’s opinion in an earlier discussion paper that, while there was limited available data, arboreal mammals “are probably little affected by low-intensity fires if the trees are not burnt”. Dr Meredith also agreed with Dr Swan’s opinion that individual gliders in areas affected by low severity fire generally survive the initial fire event, although he considered that there may be subsequent impacts from changes to vegetation. He agreed that the food resources of the Southern Greater Glider were unlikely to be affected by planned burns that do not significantly affect the tree canopy, and that any reduction in foraging resources in areas of localised canopy scorch is likely to be transient, in terms of weeks or months.