Are we asking the right questions?
By Karl Just
Planned burning is widely used to reduce bushfire risk, but its effects on fuel can vary markedly between ecosystems and over time. Ecologist Karl Just argues for closer field assessment, stronger monitoring and greater protection of long-unburnt refuges.
South-eastern Australia is one of the most fire-prone regions in the world. Bushfire is an unavoidable part of living in this landscape, and reducing the risk to life and property is an urgent and legitimate public concern. Each year, Forest Fire Management Victoria burns thousands of hectares of public land as part of its fuel-management program. For many people, the logic seems straightforward: less vegetation means less fuel, and therefore less intense fire.
After 20 years working as an ecologist, including extensive monitoring of vegetation after fire, I have come to believe that the reality is much more complicated. I first became concerned about planned burning because of the damage I observed in native ecosystems. I also began noticing something that seemed counterintuitive: some planned-burn sites developed heavier, more continuous fine fuels within several years of burning, while some long-unburnt forest supported relatively low fuel loads.
This does not mean that all planned burning is ineffective, or that fire has no place in land management. It means that burning should not be treated as a universally beneficial action. Its effectiveness depends on the vegetation type, existing fuel structure, fire intensity, season, weather, time since the last fire and the location of the burn in relation to the assets it is intended to protect. We need to ask a basic question more often: at this particular site, will burning actually reduce risk over a meaningful period?
What makes up fuel hazard?
Fire risk cannot be understood simply by looking at whether an area has recently burned. Forest Fire Management Victoria uses the Overall Fuel Hazard Assessment Guide, which considers several components of fuel. These include bark fuel, elevated fine fuel and fuels closer to the ground, including the surface litter layer. The components interact to influence flame height, spotting, rate of spread and the ability of fire to move vertically into shrubs and tree canopies.
Bark is particularly important because loose or fibrous bark can generate embers and carry fire ahead of the main front. Stringybarks are an obvious example. Elevated fine fuels include shrubs, grasses, bracken, suspended leaves and fine twigs that can readily ignite and carry flame above the surface. Near-surface fuel consists of low vegetation and partly suspended material close to the ground, while surface fuel includes leaf litter, twigs and other material lying on the forest floor.
A burn may consume much of the surface litter and temporarily lower the overall hazard. However, what happens next is crucial. Fire can stimulate dense germination or resprouting of shrubs and other plants. It can open the canopy and understorey, allowing more sunlight to reach the ground. It can also disrupt the complex litter, fungi and soil matrix that helps break down organic material and retain moisture. Within a few years, a site may develop a dense layer of elevated or near-surface fine fuel that is more flammable than the vegetation present before the burn.
This is why the number of hectares treated tells us little about the duration or quality of the risk reduction. The relevant measure is not simply whether an area was burned, but how its different fuel components changed before the burn, immediately afterwards and over subsequent years.
What field observations suggest
My work has concentrated on the drier forests of central Victoria, including box-ironbark and related forest types. I am not applying these observations to grasslands or grassy woodlands, which can behave very differently and require their own evidence and management approaches.
In a number of dry forests proposed for planned burning, the existing fuel hazard appeared low. In others, where fuels were initially higher, burning reduced them for a short period but was followed by vigorous understorey growth. At several sites around Castlemaine, the contrast between recently burned and longunburnt patches was striking. Areas burned between 2015 and 2020 commonly developed dense fine fuels, whereas nearby longunburnt forest could have a relatively open understorey.
Kalimna Park in Castlemaine provided a clear visual example. Four years after planned burning, parts of the site carried abundant elevated vegetation. Long-unburnt areas nearby were often more open. Similar questions arose from observations at Graytown. These are field observations rather than the final word on the subject, but they are sufficient to show why broad assumptions are unsafe. The response of vegetation needs to be measured at the scale at which management decisions are made.
Long-unburnt vegetation may have several characteristics that keep elevated fuels lower. Fire-stimulated shrubs may be less abundant. A more intact canopy can reduce solar radiation and wind near the ground. Established litter, fungi and soil organisms contribute to decomposition and moisture retention. Mature vegetation also tends to have a more complex structure than the even-aged regrowth that can follow fire.
Importantly, the apparent fuel benefit of a burn may be brief. At some sites, fuel is reduced for only two or three years before rising to levels equal to or greater than those recorded before burning. If a site then needs to be burned every few years to retain the benefit, the practical and ecological costs become enormous. Frequent burning across large areas would be expensive, expose communities to more smoke and profoundly alter habitat for native plants, animals and fungi.
The ecological costs matter to beekeepers
For beekeepers, the consequences of poorly targeted burning are not abstract. Forests and woodlands are living systems that produce nectar and pollen, shelter wild pollinators and provide the seasonal sequence of flowering on which managed bees also depend.
The preparation for a planned burn can cause significant damage before the fire is lit. Control lines, machinery access and the removal or modification of trees may disturb soil and vegetation. Where detailed pre-fire biodiversity surveys are absent, important plants, habitat trees or sensitive areas may not be identified and protected.
Fire can scorch canopies, damage tree health and affect flowering and nectar production. The response varies among eucalypt species, fire intensities and seasons, but a burn that damages mature flowering trees can reduce a food resource extending well beyond the burn perimeter. Loss of hollow bearing trees also removes habitat for birds, bats, mammals and invertebrates. Hot burns may kill animals unable to escape, while repeated fire can simplify understorey vegetation and remove fire sensitive plant species.
The impacts below ground are easily overlooked. Fungi and soil organisms are fundamental to decomposition, nutrient cycling, plant health and the rebuilding of forest structure. When fire consumes the litter layer and exposes soil, erosion can increase and the microclimate at ground level can become hotter and drier. These changes influence the vegetation that returns, and therefore the future fuel profile as well as the future availability of forage.
None of this means that ecological values should be weighed against community safety as though only one can be chosen. Ecological condition can itself influence fire behaviour. Retaining mature, relatively open and moist long-unburnt patches may sometimes contribute both to biodiversity conservation and to a more varied, less continuous fuel landscape.
A major monitoring gap
If planned burning is being undertaken to reduce fuel hazard, its effectiveness should be demonstrated through robust measurement. Yet information obtained through freedom-of-information requests has sometimes shown fewer than two formal fuel-hazard assessments for burns covering more than 100 hectares. Such limited sampling may miss substantial variation in vegetation, topography, burn intensity and post-fire response.
To examine large sites more systematically, I developed a method using a one-hectare grid. Sampling across the grid makes it easier to see patterns rather than relying on one or two locations that may not represent the burn area. It also allows burned and unburned patches to be compared and changes to be followed over time.
Victoria has a Bushfire Monitoring Program, but program-level reporting and computer modelling cannot replace adequate field measurements at individual sites. Models are only as reliable as their assumptions and input data. A tally of hectares burned is a measure of activity, not necessarily a measure of reduced risk.
Every significant burn should have enough pre-fire assessment to describe the existing hazard and enough post-fire monitoring to show what changed. Monitoring should continue beyond the first year because the initial reduction is not the whole story. Measurements at two, three, five and, where possible, ten years after fire would provide a much clearer picture of whether risk reduction persists or whether dense regrowth creates a new hazard.
Strategies that make a difference
The location of planned burning is as important as its ecological effect. Burning remote bushland may contribute little to the immediate protection of houses and communities, particularly if the reduction lasts only a short time. Risk-reduction work should be strategically connected to the places where it can alter fire behaviour, improve suppression opportunities or reduce exposure around assets.
Alternatives and complementary measures deserve greater investment. These include careful preparation around properties; appropriate management of vegetation close to houses; welldesigned access and defendable spaces; rapid detection and suppression of ignitions; and targeted, manual reduction of particular fuels where this can be done without triggering broad regrowth.
Long-unburnt patches should be identified, monitored and maintained as part of this strategy. They are important refuges for wildlife and fire-sensitive plants, and they provide reference areas against which the effects of burning can be assessed. In some forest types, expanding and connecting these mature patches may help retain a mosaic of vegetation structures rather than repeatedly resetting extensive areas to young, dense regrowth.
This is not an argument for doing nothing. It is an argument for spending resources where they produce a demonstrable benefit. Proper assessment may show that burning is justified in one location, unnecessary in another and likely to be counterproductive in a third.
What communities can ask for
Members of the public, Landcare groups, naturalists and beekeepers can all help improve the quality of decision-making. When a planned burn is proposed, ask for the pre-burn fuel-hazard data and the assessment of ecological values. Ask how many plots were measured, where they were located and how well they represent the burn area. Ask what outcome is expected, how long the reduction is predicted to last and what postburn monitoring will test that prediction.
Communities can also learn the principles of fuel-hazard assessment and make their own observations. Consistent photographs, mapped plots and repeat measurements can provide valuable local evidence. Emerging tools such as multispectral imagery from drones may help reveal variation in vegetation condition, burn coverage and regrowth, although these tools should complement, not replace, careful work on the ground.
Concerns and evidence can be raised with local fire-management staff, elected representatives and the media. The most constructive message is not that all burning is wrong, but that public safety and environmental stewardship both require decisions to be based on site-specific evidence and transparent monitoring.
A more honest measure of success
Planned burning has become an established feature of bushfire policy, but familiarity should not shield it from scrutiny. The success of a program cannot be judged by the area burned. It should be judged by whether risk was reduced, where that reduction occurred, how long it lasted and what ecological costs accompanied it.
In many dry forests, a burn may provide a short window of reduced fuel followed by a longer period of increased fine fuel. In some long-unburnt areas, fuel structure may already be relatively favourable. Each site therefore needs careful assessment before fire is prescribed, followed by meaningful monitoring afterwards.
The central question is simple: did this burn make the community safer? Answering it requires evidence from the forest, not merely confidence in a model or a target expressed in hectares. By protecting long-unburnt refuges, improving property-level preparedness, strengthening rapid response and applying fire only where its benefits are likely to persist, we can pursue bushfire safety without needlessly degrading the ecosystems on which bees, other wildlife and people depend.