Burning and Wildfire


Prescribed Burning and WildFire in South West Forests

F H McKinnell

Consultant, Forestry and Natural Resources Management



Although this paper deals with the fire management policies of CALM and the former Forests Department, there is a parallel story in the management of fire on private land, which is the responsibility of local government

Of course, there will always be wildfires in WA forests. We will never eliminate them, but their impact can be kept low, and fire suppression costs can be kept low, by proper preventative measures. The prime preventative measure we have is fuel reduction burning.

A fundamental premise of forest and rural fire management in Australia is that the reduction of fuels leads to lower fire intensity when fires do occur (as they always will, whether from lightning, arson or accidental escapes). Lower fire intensity means that the fires are easier and safer to control and the damage they cause is also reduced. A fact that is often overlooked is that fire suppression in light fuels is much cheaper than in heavy fuels, even taking into account the cost of fuel reduction burning.


The Historical Perspective

Following the formation of the Forests Department in 1919, a policy of general fire exclusion in State forest was adopted, although not without some misgivings at the time. At its core, this approach was a consequence of the European training of the early foresters, where forest fire was regarded as nothing but a disaster, and total fire exclusion was the only way to ensure the survival of the forest ecosystem. Not all staff of the Department held this view, some seeing that the forests had been subject to aboriginal burning and lightning for millennia, without apparent damage.

Actually, the fire management policy was not one of total fire exclusion. The forest was divided up into blocks of about 4000 ha for management purposes, and within the blocks, compartments of around 300 ha were laid out by a forest track system. Around the perimeter of each compartment was a strip about 100 m in width in which fuel reduction burning took place every 3-5 years.  The intention was that this fuel-reduced buffer would either prevent a fire entering the interior of the compartment (if the burn was recent) or allow a control action to be carried out in relatively light fuel. The inside of the compartment was kept unburnt. A huge effort was directed into establishing this compartment system, taking advantage of sustenance labour available during the Depression. However, the system became increasingly untenable as the difficulties of carrying out burns in light fuel against 15-20 year old fuels became apparent. The commencement of World War II and the consequent lack of manpower meant that any possible change was put on hold, although senior staff still subscribed to the fire exclusion approach.

This policy was maintained until 1953, when A C Harris was appointed Conservator of Forests. He was a professional forester, trained in Australia, not Europe, and had worked for many years in the northern jarrah forest. Harris was convinced that regular broad area prescribed burning was necessary to keep forest fuels at a level where fire would cause minimal damage. His cause was aided by an assessment of the (1950) Plavins wildfire, which, although fortunately not a large one, caused very severe damage to the forest when it burned through heavy fuels. It was clear to all concerned that fire exclusion was an unsustainable policy.

Harris changed the Departmental fire management policy to one of regular broadcast burning the whole forest, except for areas withheld to protect regeneration. This change could not be made overnight. The development of procedures for successful broad area burning was slow, and there was still a reluctance to cause any apparent damage at all from a burning operation. Any leaf scorch over about 2 metres above ground was viewed with disfavour. Much of the early burning was done in winter. This meant that those burns that were carried out were at the lowest end of the fire intensity scale, were very patchy and removed comparatively little of the accumulated litter. Their effectiveness in fuel reduction was therefore low. In addition, the large areas of heavy fuel accumulation then in existence meant that the overall approach was very cautious. Monitoring procedures were minimal or non-existent and many compartments were recorded as being burnt when, in fact, only the edges of the compartment had been affected. Overall, the implementation of the new burning policy proceeded very slowly. The forest estate still carried huge fuel loads, and the scene was set for large fires in years to come. The 1961 fire season was the Forests Department’s Annus Horribilis. George Peet and Jim Williamson have described in detail the Dwellingup fire, the main event of the year, in an earlier presentation at this seminar.

The response of the Department to the situation was dramatic: a specialised fire management unit was set up, more resources were directed to prescribed burning, planning and monitoring procedures were expanded and a fire behaviour research and development program was commenced. The effect of these changes was an immediate increase in the annual area prescribe burnt and a marked improvement in the quality of the field operations.

The result was implementation of burning rotations based on fuel accumulation rates. In the western jarrah forest this meant a fuel reduction burn every 5 years or so, to maintain fuel loads at acceptable levels where direct attack was possible under moderate weather conditions. In the eastern jarrah forest, lower fuel accumulation rates meant that the rotation there was 7/8 years. In the karri forest minimum burn rotations were longer, due to the nature of the understorey. These regimes were varied according to local circumstances, such as areas of fire-sensitive regeneration or where fauna management plans called for variation.

Of course, the issue of resources for burning operations was a major factor, but the “can do” approach of the time addressed this problem and the outcome was the development of aerial ignition. This became a major tool for field staff, enabling larger areas to be burnt when fuel and weather conditions were just right. The use of multiple ignitions was able to address the problem in southern forests of different vegetation types drying out at different rates. This enabled the Department to achieve a dramatic reduction in fire hazards in the southern forests.

The test of all these changes was the multiple fire situation that arose with the passage of Cyclone Alby in April 1978. At one stage there were 60 wildfires running in State forest, but many could be left to run as they were located in areas of low fuel where they could do little damage. The efficacy of fuel reduction burning in enabling the successful control of these fires has been documented by Underwood, Sneeuwjagt and Styles in 1985.

There is no doubt whatever, that maintaining the forest estate in a low fuel condition does provide a very effective fire management regime that results in few large forest fires. Up to 1985, we had a very effective and stable organization responsible for forest management in Western Australia, one that rarely had to call on outside assistance to manage fires on State forest. The situation has changed dramatically since then.


What the Data Tell Us

The figure below shows the area of forest managed by the Forests Department and later by CALM, that was burnt each fire season by prescribed burns and by wildfire. The data cover the period from the 1960/61 fire season through to 2004/2005. Prescribed burn data from before 1960 are too unreliable to contribute to the story.

The catastrophic 1960/61 fire season is clearly evident, as is the very low level of forest area affected by wildfire over the next 25 years. For the next 10 years after 1961 the average area burnt was about 350,000 ha a year, dropping back to about 300,000 ha a year after 1971, reflecting the overcoming of the backlog of high fuel areas that existed. It is therefore reasonable to say that an average of 300,000 ha per year of fuel reduction burning is necessary to provide good protection of SW forests and their associated ecosystems against wildfires.

The graph also shows that after the formation of CALM in 1985, the level of prescribed burning progressively declined to less than two-thirds of the average under the Forests Department, with a particularly rapid drop to half that level, one-third of the level required for adequate fire protection of the forest estate, after the separation of CALM and the FPC in 2000. From the perspective of forest fire management, this was a destructive organisational change, as it removed a whole cadre of experienced fire managers from CALM. Although some arrangements were made to retain these resources “on hire”, getting the two organisations to work together has not been a satisfactory exercise.

At the same time, after 1985, the areas of forest covered by wildfires increased rapidly, to the extent that in 2003, the area burnt in wildfires was about the same as that covered by the fuel reduction burning program! The last time this happened was 1961! This is an incredible lapse in performance by the State agency charged with protection of the forest estate. Of course what these data cannot show is the fact, well known to the bush fire brigade personnel present here, that for much of the fire suppression operations in recent years CALM has been compelled to call on the assistance because it had insufficient resources itself.

There are several reasons for the progressive decline in prescribed burning after 1985:

* diversion of physical and financial resources away from the SW forest areas into other (long-neglected) areas for which CALM was responsible,

* conversion of significant areas of State forest to national park or conservation reserve and consequent development of areas management plans that reduced the extent, frequency and size of burns, for highly debatable “nature conservation” reasons,

* loss of focus on fire management due to the overwhelming pressure of the multitude of issues that the agency had to confront,

* progressively more onerous externally imposed controls, eg, over smoke generation,

* greater influence of ill-informed minority groups on Government policies, combined with greater direct political influence on agency activities.


The size of burn issue is a very important one. The smaller the area to be burned the higher the costs and the physical resources required, and, more importantly the less time available for other burns when weather and fuel conditions are suitable. The main drive for smaller burn areas has been the supposed better biodiversity conservation value of “mosaic” of areas of different fuel age

After the split of CALM and the FPC, the ethos of CALM was changed significantly towards a priority on biodiversity conservation. It now has a strong environmentalist outlook. While the fuel reduction burning policy was retained, its implementation was even further restricted by management policies calling for the establishment of a fine-grained mosaic of fuel ages and greater retention of older fuel ages. All this has been done without any strong scientific foundation at all, and with a definite grudging recognition of the need to restrict the incidence and extent of wildfire. The fact that the previous prescribed burning regime already had created a micro-mosaic of fuel ages was ignored and the history of Aboriginal burning was denied.

While some may dispute my contention of a lack of scientific foundation for the current policy, no one has ever been able to demonstrate that the previous 1961-1985 fire regime has caused any adverse effects on biodiversity. It has just been assumed that it was bad because it was done by foresters. In addition, there is a lingering addiction in environmentalist circles to the Eurocentric notion that any fire in the forest is an evil thing.

In a related matter, many area management plans now specify the designation of significant areas of forest as No Plan Burn Areas. Why we need these areas is a good question, because absence of fire is most certainly not any sort of natural condition for forest in this part of the world. All we can be certain of is that some day such areas will be cooked by a high intensity wildfire and any supposed special biodiversity values will be lost for some time, if not forever.

In recent weeks CALM staff have argued that they have made a significant impact on fuel, and that 33 percent of the forest area carries fuel aged 5 years or less. This has, however, been achieved only with the large contribution from wildfires. It still leaves 67 percent of the forest area in a high fuel condition- that is, nearly 70 percent will permit crown fires to take place. This is not a happy prospect for those who have to fight the fires.

I hope the foregoing does not convey the impression that I am just being critical of CALM – the agency does have a number of real obstacles to improved performance in this area, including:

* large areas of regeneration in jarrah forest that are being held for 20 years before the first fuel reduction burn,

* large areas of karri regrowth that are very difficult to burn safely before that are 25+ years old,

* large areas of forest rehabilitated after bauxite mining which carry very heavy fuels,

* the smoke issue,

* lack of on-ground resources,

* difficult seasonal conditions.


To some extent these problems are being used to justify lack of action. They are real problems, but there seems little inclination to attack them with a can-do outlook. For example, the jarrah regeneration can be safely burned at age 10, or even less, and this has been demonstrated many years ago. It may mean using techniques such as night burning, but it can be done. Furthermore, it is only a proportion of cutover coupes where the jarrah regeneration is deficient anyway.

The first post-regeneration burn in karri poses more difficult problems, but at least the pattern of occurrence of karri in a matrix of jarrah-marri forest enables the maintenance of broad areas of low fuel between patches of karri regrowth. We have shown, in Big Brook, that at about 30 years of age, the first burn can be done relatively easily. Large-scale clearfelling started again in karri in 1967, so CALM should now be burning 1500-2500 ha a year of regrowth of that age.

The rehabilitated bauxite mine areas pose complex problems. Much of this area carries very heavy scrub fuels, some exotic species used there are difficult to handle, being either fire sensitive or producing massive regeneration after a burn. Nevertheless, ways have to be found to blend in the fire management of the mined forests with the surrounding matrix of jarrah. This issue does not appear to be among CALM’s research priorities, but surely has to be something that must be solved before there can be any consideration of Alcoa handing back the forest to the State. BFF sees no concrete action being taken on this important issue.

Meeting smoke emission standards has been a severe constraint on burning in the northern jarrah forest for some time, and threatens to become an issue in southern forest areas as well. CALM has done a good job of confining the smoke haze over Perth to within prescribed limits, but the cost has been the accumulation of high fuel loads very close to the Metropolitan area. Not many people realise how close we came to a Canberra-style disaster last summer. Urban residents must accept the fact that some short-term instances of thin smoke from a few burns are preferable to long-term dense smoke events from bushfires. But we have seen no public education campaign to achieve this outcome. The public has to realise that some smoke is the natural condition of our environment in Australia. Clear, smoke-free skies are what is unnatural.

BFF has urged the Government to relax those smoke emission standards to permit more fuel reduction burning to take place, and has been advised by the Minister for Environment that she supports this, but we do not see any apparent directive to CALM and the EPA to act on this advice. As fire smoke is a natural part of our environment, it should be removed from designation as air pollution.

Lack of on-ground resources has also been a problem for CALM. It is pleasing to see that the Government has provided more funds for additional manpower in the field, and this should enable a greater burn program to be achieved in coming years. However, the resources are now concentrated in a very few centres across the forest zone, and despite improved roads and improved equipment, access times to heavy fuel areas are too slow, as shown in the 2001 Muller report. As we all know, fast attack is the key to successful suppression.

In a number of areas, there are very high fuel loads on private property as well. This has the effect of obliging CALM to devote some of its forces to fires that threaten those areas, rather than fires in the forest. This actually happened in the 2005 Perth Hills fire. BFF has seen no sign of a concerted effort by FESA to compel offending Shire Councils to fulfil their obligations with respect to fire management in the areas for which they are responsible. We note that the old Bush Fires Board did just that after the 1961 fire season, with very effective results.

BFF believes there are also internal problems in CALM that reduce its ability to achieve adequate levels of prescribed burning. The internal structure is not conducive to rapid diversion of local manpower resources to burning at the limited times when suitable conditions exist. Other work programs at times have priority so that reduced resources are available for burning. The current purchaser-provider financial management system is also a clumsy and inappropriate way of running an agency with such diverse functions as CALM.

One reason advanced for CALM’s poor burning performance in recent years is that we have had a succession of dry years, which has restricted field operations. We certainly have had several years of below average rainfall, but it is hard to see how this resulted in poorer burning performance. Figure 2 below shows the rainfall for the previous year for Perth, which obviously has a strong influence on the area that can be burnt, plotted against total area covered by prescribed burning. Of course, the rainfall picture is much more complex than this, but the Perth data can give a crude measure of the suitability of the season for burning. It is interesting to see that there was a period of very similar rainfall between years 21 and 29 in the graph (1977-1988), and the last 10 years. In the first period the Forests Department burned an average of about 280,000 ha a year, but in the second period CALM managed an average of only 136,000 a year. It’s a fair indication that it was not the season that was stopping them, but other factors such as those described in the previous paragraphs.



Despite a fillip in the areas prescribed burnt over the last 3 years, the Southwest forest estate remains in a dangerously high-fuel condition. Nearly 70 percent of the forest is more than 5 years old since the last burn, and in jarrah, we know from hard experience that fuel older than this will sustain a crown fire in jarrah forest. We also know that we can do nothing at all to stop a crown fire in heavy forest fuel.

To provide an adequate level of fire protection – as good as that we had 20 years ago- CALM needs to overcome the backlog caused by its poor performance over the last 15 years by achieving 300,000 ha a year for several years and thereafter maintain an average of 250,000 ha a year. However, the level of prescribed burning is only part of the problem. The burning program needs to be arranged in large strips across the forest, designed to arrest a major wildfire, and the strips must be at least 3 km deep. The fine scale mosaic approach cannot provide this function.

The CALM fire management policy needs to focus more on fire prevention through fuel reduction burning, rather than relying on very expensive and relatively ineffective suppression operations. Forest fire management rates too low in CALM’s priorities. Those who argue that it is now a biodiversity/recreation management agency have to realise that all their biodiversity and recreation objectives are lost if they don’t win the fire management battle first.