This is a terrific paper highlighting the perils faced by Australia’s many threatened species and why we need to be very careful about the amount of fire that is impacting the continent’s landscapes.

Human-induced climate change is driving rapid increases in the severity of fire weather conditions across much of southern Australia, such that catastrophic fire seasons are becoming increasingly common, with enormous negative impacts for our native plant and animal species, and ecological communities (such as fire-sensitive rainforests).

For example, following the catastrophic ‘Black Summer’ fire season of 2019/2020 – unprecedented in terms of the area burnt in a single fire season – a large number of plant and animal species needed to be rapidly added to the Australian Government’s threatened species lists.

It has long been suspected the fires are eroding the area of long-unburnt habitat, critical for the survival of very many native species, across southern Australia. The new paper in PNAS provides alarming evidence to support this suspicion. The authors provide very robust evidence that the area of long-unburnt (>30 years) habitat in southern Australian reserves has steadily declined over the last 40 years (from about 61% in 1980 to 36% in 2021). They also found that the areas affected align strongly with the distribution of a large number of threatened species known to need long-unburnt habitat for their survival.

These alarming findings present a management conundrum, because one of the foremost tools used by fire managers (including in reserves) is deliberate burning under mild fire weather conditions (known as prescribed burning or hazard reduction burning). Broad-scale prescribed burning has been shown to be useful for reducing the intensity of bushfires. However, it tends to increase the area that is burnt each year – which is likely, in turn, to decrease the amount of long-unburnt habitat. Under a changing climate, we will need new fire management approaches (moving beyond broad-scale prescribed burning) that specifically target the conservation of long-unburnt habitat.

This is a well-executed and comprehensive study that points to a concerning increase in fire impacts on Australia’s conservation reserves over the last 40 years that are negatively affecting conservation values. Fire impacts were higher in larger and more remote reserves and increased more strongly in conservation relative to state forests. Results differed in Western Australia, with fire frequency and proportion burnt declining since 1980 due to the effectiveness of prescribed burning, and a decline in the area of prescribed burns over time.

These findings point to the effects of removing Australia’s Indigenous people and their skilled use of fire from southern Australian landscapes and the currently low level of active management in conservation reserves, including regular low intensity fire, particularly in remote areas. Governance changes are needed to empower and support Indigenous people to manage reserves, with more investment in active management to ensure our reserves are more resilient to future climate change. Investment in active management of reserves will be imperative if they are to continue to provide conservation values in future.   

This research echoes results reported by other authors, which point to significant shifts in fire regimes across southern Australia. These results are consistent with those of similar studies in other fire-prone parts of the world. Anthropogenic climate change has long been expected to shift wildfire regimes towards more frequent and extreme fire events across many parts of the globe, including southern Australia and this study provides further confirmation that these shifts are already underway.
 
Bushfires in Australia are burning more frequently over larger areas and with higher severity. Indeed, extreme bushfires including fire thunderstorms have exhibited a marked increase over the last two decades. The intensity and rapid advance of these bushfires have dire consequences for threatened wildlife, and broader global impacts on Earth’s systems.
 
The research highlights the impacts on ecosystems in regions that are not accustomed to high frequency and/or high severity fire, such as high-country areas.

While the study does not specifically address whether fires are becoming less fragmented, my hunch is that there is a trend toward less fire fragmentation, which means that animals are less likely to find refuge during fire events.

South-eastern Australia has been impacted by significant bushfires over the last 20 years. The nature of wildfires in the area is unprecedented in post-colonial Australia. There has been considerable work examining the drivers of these fires, with climate being the most significant driver. It is likely that these patterns will continue in the future with the forecast changes in fire weather.
 
Fire managers are faced with the challenges of working in uncertain conditions and an ever-changing landscape. Fire management can not be held responsible for saving the last population of any species, as this is the work of conservation agencies. Furthermore, long unburnt habitat is only required for a subset of species and we require a mix of fire ages in the landscape to ensure the survival of diverse environments across the landscape. The challenge for the future is to ensure the right balance of fire ages are available to create resilient landscapes and systems.

There are many different ways of understanding fire, including the traditional knowledge of Indigenous Australians, scientific observations and models, and the lived experience of fire managers and communities. This impressive data-driven paper uses agency fire records to describe changes in fire frequency over the last 40 years across the vast and highly diverse woody landscapes of southern Australia.
 
They find that there has been a steady decline in the extent of long unburnt vegetation, which has important consequences for biodiversity and threatened species conservation. It is important to note that the picture emerging from this paper is complex, with different regions seeing different patterns in fire frequency, and the extent of long unburnt and recently burnt vegetation. As the authors also note, there is more to fire than just frequency – we care about the severity and the seasonality of fire too, among other things.
 
Nevertheless, this is an important study that helps build our understanding of recent fire patterns, which will support efforts to mitigate bushfire risk and conserve biodiversity.

Doherty and co-authors find a rapid and large decline in old growth, unburnt vegetation in most conservation areas of southern Australia, mountains in particular. These findings build on a large body of evidence now showing that increasing human-caused fire is a critical conservation issue. Australia has one of the worst records of animal extinction globally, especially mammals. Conservation scientists agree that a major reason has been changes in fire management in the last 200 years, giving us more severe fires- that burn more vegetation and leave few critical pockets of unburnt shelter- in drier inland habitats with sparse trees, so that mammals are exposed to foxes and cats after fires.

This paper provides convincing evidence that now climate change is driving severe fires in our protected areas of dense forest and alpine vegetation. Many animals in Australia and around the world need old growth vegetation or mountain habitat that is sensitive to fire. They will decline faster because of increasing droughts and subsequent fires under climate change. Mountain species are already under threat by other mechanisms of climate change- many cannot physically tolerate heat, and their envelope of suitable cool, wet habitat is rising so their distribution becomes rapidly smaller until it disappears.

I strongly agree with the authors that agencies in Australia and around the world need to plan to protect critical unburned wildlife habitat when we are faced with megafires, not just human infrastructure.

Fire is increasing in its frequency and severity in many regions globally. This is due to climate change and associated changes in weather patterns, in some cases, invasive plants such as gamba and buffel grass, and other changes made to landscapes by people. To survive during and after fire, many species rely upon unburnt areas of habitat. It’s long been thought that such areas are becoming increasingly rare, and the work of Tim Doherty, William Geary, and colleagues, which examined fire histories in over 400 locations in southern Australia, show this to be the case. Areas that had not burnt within the previous 30 years decreased by 25% from 1980 to 2021.

Their findings have significant ramifications for the protection of biodiversity under a rapid warming and more fire-prone world. Just as importantly, these results highlight the need to consider changes to prescribed burning and fire management policy and practices. Fire policy and management should seek to maintain, and in some regions, ideally increase the area of unburnt habitat, in order to support wildlife populations and aid their recovery and expansion following fire. The common practice of ‘clean-up’ burns—where managers will often target and burn any unburnt patches following wildfires or prescribed burns, as a means of attempting to reduce the risk of future fires—must be increasingly scrutinised and reduced.

This study provides an important and rare research contribution spanning both SW and SE Australia. While studies of fire focused on vegetation and climate are relatively more common in southeastern Australia, fewer studies have addressed the southwest and very few have done both.

Spanning all of southern Australia and many ecosystem types, the research paints a worrying picture of fire becoming more common in places it was not historically, with widespread consequences for a range of threatened animals across many taxa.

The effects of the 2019-2020 Black Summer bushfires dominate their results but they also lay down important findings and foundation for future research. Key nuances around ecosystem types (heath, woodland, dry forest, wet forest) are obvious given very different patterns of fire (see Fig 2). With many species in serious decline, the clear climate, fire regime, and fire management across the regions offer powerful opportunities for learning and providing the urgent knowledge necessary for conservation efforts. Their research draws a clear line underneath the importance of publicly available, nationally coherent datasets to drive research and knowledge.

This study is very important because we need to have more scientific information on fire regimes and their relationship with native flora and fauna.

However, the following are my key concerns:

1. The study emphasises maintaining larger unburned regions and focuses on animals requiring habitats in unburned regions. Maintaining unburned forests could be desirable for many ecosystems across the world but not for Australia where plant and animal species rely upon wildfires to complete their life cycle. The study recognises fire regimes for study regions but without descriptions of desirable fire regimes and how they are altered. A decline in unburned areas and increased fire frequencies on a decadal scale are not indicative of an altered fire regime.
    
2. Vegetation types and their burn requirements are not described. Australian vegetation types have different fuel accumulation rates and fire frequency requirements, e.g. wildfires in rainforests are less frequent and undesirable. NDVI cannot differentiate such vegetation types. Comparison with the western USA is inappropriate because most Australian vegetation requires shorter fire intervals.
    
3. The study uses the southern part of Australia and excludes other parts because they lack data. This is not true, apart from the availability of other data sets states such as Queensland have better vegetation maps e.g. the ‘regional ecosystem’ data has a species-level description and their relationship with burn.
    
4. The study mentions maintaining an unburned state for ‘old growth vegetation’, but such vegetation survives low-intensity burns e.g. rangers often burn grasses and other understorey vegetation in eucalyptus forests to reduce fuel load and to improve eucalyptus forests' health. Queensland’s state department has maintained experimental plots since the 1960s with different burn treatments demonstrating adverse effects of no burning and too frequent burning. Unburned plots show poor health in the top parts of the vegetation with crowded understorey vegetation.

This study should have emphasised the adverse effects of high-intensity burns covering larger extent and top canopies, and the favourable effects of patch mosaic burning (maintaining patches of burned and unburned regions) to maintain desirable fire regimes. Achieving longer unburned (>30 years) areas over larger regions will potentially lead to undesirable high-intensity and large-extent wildfires.

Threatened fauna survive against the odds in southern Australia’s increasingly vulnerable and compromised landscapes. Decreases in long-unburnt habitat and increases in recently burnt vegetation further compromises the quality of habitats, and further increases vulnerability.

The authors renew calls for an emphasis on fire management that protects vulnerable species but provide few practical suggestions.

Current land management depends on outdated broad-scale prescribed burning policies. This approach completely fails threatened species and compromises the safety of rural communities.

There is now overwhelming evidence in eucalypt forests that long unburnt (and unlogged) vegetation provides the greatest protection from fire. The development of overstorey shelter through the natural processes of self-thinning, self-pruning, and succession reduces the severity and potential for forests to burn. This understanding is now well-developed across a wide range of vegetation communities. In these areas, vegetation burnt as recently as three years previously, burns with more severity than long-unburnt vegetation.

The scientifically outdated ‘protection burning’ philosophy that drives Australian land management agencies must be redressed as a matter of urgency.

Increasingly, cases are likely to be brought before the courts to challenge such damaging fire management policies.

The risk to threatened species and rural communities is increased by prescribed burning. It is imperative that long unburnt vegetation is maintained. Early detection and rapid response facilitated by updated technology must drive the new paradigm in forest policy and management.