This report describes the environments of Australia, points out the variety that exists in fire regimes, and provides statistics on the occurrences of fires. It also notes some of the social impacts of wildfires in recent years, presents an outline of the agencies involved with fires and touches on the community policies and infrastructure that have developed in response to fires. To keep the report brief, comprehensive referencing has been foregone. General Australian references to fires and their effects are: ‘Bushfires in Australia’ (Luke and McArthur 1978); ‘Fire and the Australian Biota’ (Gill et al. 1981); ‘The Ecology of Fire’ (Whelan 1995); and, the forthcoming book ‘Flammable Australia: the Fire Regimes and Biodiversity of a Continent’ (Bradstock et al. in press).
Fire environment, fire regimes and ecological role of fire
Australia is a large, geographically diverse continent straddling the tropic of Capricorn. It is relatively flat, dry and warm although it also has mountains of moderate height (particularly along the eastern seaboard), rainforests and ski-fields – as well as a large arid and semi-arid zone. This report covers the Australian mainland and its largest offshore island, the State of Tasmania. Australia has an area of 7.9 million km2 (Luke and McArthur 1978).
Across the tropical north is a large savannah region with a monsoonal (wet-dry) climate. In the southwest is a region of eucalypt forests, woodlands and shrublands with a largely Mediterranean climate. In the southeast, including Tasmania, there are forests and woodlands, but the climate there is quasi-Mediterranean with the dry summer conditions due to high evaporation rather than low rainfall. A large area of the continent is arid and semi-arid, both tropical and temperate, and is dominated by hummock grasslands and Acacia shrublands. Prominent, species-rich woody plant genera in Australia are Acacia and Eucalyptus. Among the vertebrate animals, marsupials are particularly noteworthy.
Australia is a fire-prone continent. Fires occur in rainforests and in deserts. Its people have ignited and lived with landscape fires for perhaps 60 000 years, the current antiquity considered for Aboriginal people in Australia (see Miller et al. 1999). Consideration of fires without consideration of people in Australia is sometimes difficult, sometimes impossible, often unwise.
Fire regimes and their effects are wide ranging. Fires occur with a mean interval of 1 to 2 years in parts of the savannah in the tropical north and of the order of 300 or more years in the temperate rainforests of the southeast. Fires occur at all times of the year in some part of the continent. Intensities of surface fires probably reach a maximum of the order of 100 000 kWm-1 (Gill and Moore 1990). Peat fires, which occur especially in Tasmania, can have long-term ecological significance, but are not widespread on a continental scale.
Australia has a large and diverse flora and fauna, much of it unique to the continent. This diversity, combined with the fact that there is a wide range of fire regimes and physical environments, means that there is an enormous range of potential fire regime effects. These effects include local extinction of flora and fauna, changes in water yield and quality, changes in pasture palatability, and woody plant encroachment.
Major wildfire impacts on people and property during the 1980s and 1990s
Most socio-economic damage from wildfires in Australia occurs in the southeastern part of Australia (see Cheney 1979 for ‘fire-hazard areas’), but more generally where forests and woodlands abut cities and their suburbs. In the 1980s, there were extensive fires in South Australia and Victoria; 76 people died and there was massive property damage including the loss of 2 463 houses and 30 000 stock (see Ramsay et al. 1996). In the 1990s, there were fires in Sydney, New South Wales, where 206 houses were destroyed and four people died (see Ramsay et al. 1996). The impacts of such circumstances can last decades or more for the people involved.
Most human deaths in wildland fires in the last decade have been those of firefighters. At least 52 have died on active duty since 1980: in South Australia (10), Victoria (19), and in NSW (23) (Paix 1999). In early 2000, three more firefighters were killed in Sydney.
Fire database: fire numbers, areas burned, and fire causes
Various problems arise in the use of statistics for forest fires in Australia. The use of differing classifications of vegetation types creates difficulties. For example, ‘forest’ in the Food and Agriculture Organization (FAO) definition includes trees at least 5m tall having a cover of more than 10 percent, but the Australian National Forest Inventory, while using the same 5m height threshold, requires a “projected foliage cover of overstory strata about equal to or greater than 30 per cent” (Resource Assessment Commission [RAC] 1992). The definition of ‘forest’ used by forestry organisations, for fire-reporting purposes, has been the vegetation of the land under forestry jurisdiction.
Forest fire statistics usually apply to data collected by State Forest Authorities and may not include the forests in National Parks, Crown Lands, and private property. Some State forestry organisations have been amalgamated into agencies with larger jurisdictions and the way that statistics are published may reflect this.
The area base for the statistics is continually changing so care should be taken in the use of fire statistics. Changes occur due to tenure and land-use changes. The area of forest in Australia around 1790 has been estimated as 69 million ha while about 43 million ha survived to about 1990 (RAC 1992).
The following caveats apply generally but unevenly across the vegetation types and tenures of land in Australia:
The methods used for the establishment of a database of areas burned varies (e.g. ground observation, aerial photography or satellite imagery) and thus the accuracy of the data varies.
The diligence of reporting fires and the areas they burn may vary.
Data may, or may not, include areas burned by prescription. For example, satellite sensing of areas burned will not discriminate between causes of fires. The major cause of burning in some major forest areas is due to prescription (see Gill and Moore 1997 for south-western Australian eucalypt forests).
Data are often presented for short periods only.
Estimating the numbers of fires in Australia by any means is difficult even with a constant methodology. Data on High-Temperature Events (HTE) which represent vegetation fires depicted by satellite remote sensing gives a national coverage. Numbers of HTE’s for the past two years (1998-1999 and 1999-2000) are 115 000 and 230 000 per year for Australia as a whole (Western Australian Department of Land Administration [DOLA] – data courtesy of Mr R. Craig). However, the number of HTE is inflated, on the one hand, by multiple counts of single fires and deflated, on the other, by the many fires that are missed by the evening pass of the satellite used for the daily detection of HTE.
The areas of burned surface estimated from satellite imagery by DOLA (viz. ‘fire-affected areas’ or FAA), for the two years were 312 000 and 712 000 km2. This wide difference was apparently caused by large differences in the area of FAA in arid Western Australia (WA). These figures indicate an average proportion of the continent burned per year of 6.5 percent. This average contrasts with the extreme year of 1974-1975 when 15.2 percent of the continent was estimated to have been burned (Luke and McArthur 1978). In that same extreme year, a massive 33.5 percent of the entire Northern Territory was burned (Luke and McArthur 1978).
For ‘forests’ in the period 1956-1971 there was an average of 1945 fires per year with an average area of forest burned of 362 000 ha, or 1.8 percent of the total area (Luke and McArthur 1978). Forests in this case probably represent the vegetation managed by forestry agencies (see above). Taking the average area burned is a bit misleading, however, because the range in the area burned was from 53 000 to 1 061 000 ha (Luke and McArthur 1978). Estimates of areas of forest burned in more recent times are shown in Table 1. The Table is a summary of those produced by the National Greenhouse Gas Inventory for the period 1991-1996 (see Table 1 for the URL). The tables showed nominal, constant, figures for some years; where these extended for more than two years they were not included in Table 1.
It is suspected that the numbers in Table 1 were for ‘forests’ in the broad sense, whereas the data from the Forest Services quoted from Luke and McArthur (1978) were from forestry jurisdictions. Drawing comparisons between the two sets of figures would be unwise. For a detailed study of historical trends in areas burned and number of fires in southwestern Australian forests, see Gill and Moore (1997).
The Victorian Department of Natural Resources and Environment has posted the average figures for fire causes on public land over a 20-year period (http://www.nre.vic.gov.au/ ). Of the average 584 fires per year, 26 percent were caused by lightning, 25 percent were deliberately lit, and 26 percent were caused by agricultural sources and campfires combined. These fires did not burn areas proportionate to their numbers, however. Nearly half the area burned, on average, was due to lightning-caused fires; deliberate fires burned 14 percent, while agricultural sources and campfires, collectively, accounted for only 8 percent of the burned area.
Estimates of the areas burned by prescription are shown in Table 2. ‘Areas burned’ are usually the sum of the areas of burning blocks, not necessarily the actual area burned, which is less than that of the burning block (see Gill and Moore 1997).
Table 1. Estimated areas of forest land (ha) burned by wildfires in Australia. Source: National Greenhouse Gas Inventory 1991-1996 (<http://www.greenhouse.gov.au/index.html>). Abbreviations: NSW – New South Wales; WA – Western Australia; SA – South Australia; ACT – Australian Capital Territory.
Table 2.Estimated areas of forest land (ha) burned by prescribed fires in Australia. Source: National Greenhouse Gas Inventory 1991-1996; http://www.greenhouse.gov.au/index.html). Abbreviations: see Table 1.
Fire-suppression and fire-management organizations
There are many agencies in Australia concerned with suppression of wildland fires. In Table 3, agencies listed as “Fire Suppression Agencies” are those State government agencies that have as their major role the immediate suppression of any wildfire. Assisting them in suppression are, often, the State fire-management organisations and, less often, the metropolitan fire brigades. “Fire-management organisations” are government land management agencies which, if large enough, have a dedicated branch concerned solely with fire management. No private suppression organization is known to us, but many farmers, graziers and forest-plantation owners have their own fire suppression equipment.
Wildland fires in Australia are fought largely by volunteers belonging to a local brigade. Luke and McArthur (1978) estimated that there were 300 000 volunteer fire-brigade members in Australia. Volunteers are coordinated, equipped and trained by paid employees of the suppression agencies. State governments are the primary source of funding for the suppression agencies, but local government and communities are often heavily involved as well.
Among the State Government agencies which have significant capability for firefighting, but are primarily land-management agencies, are those in Western Australia (Department of Conservation and Land Management), Victoria (Department of Environment and Natural Resources) and New South Wales (State Forests of New South Wales). For further information, see the web page of the Australasian Fire Authorities Council (http://www.ausfire.com)
Both fire-suppression agencies and land-management agencies may be involved in the production of educational and training literature for firefighters and the public. They advise householders and rural people about preparedness for fire.
Table 3. Australian agencies that have as their primary role the suppression of landscape fires. Abbreviations: see Table 1.
Use of prescribed fire to achieve management objectives
Prescribed burning is used widely in Australia. In forestry it is used primarily for crop protection, disposal of debris after silvicultural operations, and for the protection of human lives and property (e.g. see Williams and Gill 1995 for New South Wales, and Gill and Moore 1997 for southwestern Western Australia). In Western Australia, the extent of prescribed burning has been in gradual decline for many years (Gill and Moore 1997), possibly because of reactions by the public to smoke.
For land uses outside of forestry there are many reasons for prescribed burning. In the pastoral areas Leigh and Noble (1981) list, among other reasons, removal of top hamper (dead grass), extension of the growing season, control of woody weeds, assistance with the establishment of improved pasture species, fuel reduction, and nutrient release. In conservation areas, fires may be prescribed for the maintenance of natural values (Good 1981). In agricultural areas, fires may be used just before the harvest of sugar cane and just after the harvest of cereal crops (Johnson and Purdie 1981), although such practices have often been criticized in recent decades.
In most land uses where native vegetation is predominant there will often be some burning designated for the protection of human life and property.
Systems employed to reduce wildfire hazards and wildfire risks
Australian societies have many levels of interacting mechanisms pertaining to wildland fires. These differ widely from place to place but here we try to indicate a range of activities that, in various combinations, affect the responses of communities and governments to wildfires in Australia.
The infrastructure surrounding responses to fires in Australia includes:
Fire-detection systems (which may involve dedicated observers using fire towers and aircraft, citizens using mobile phones, and analysts using satellite systems)
Networks of safe outdoor cooking facilities for campers (e.g. gas barbecues)
Firefighting equipment (4-wheel drive vehicles including tankers; graders; bulldozers; air tractors; helicopters)
Networks of tracks, roads, fuel breaks, and buffer strips
Supply systems for the use of water-enhancing agents such as foams
Systems of regular fuel-condition assessment (models, ground observations, satellite data)
Systems for the assessment of fire behavior (models, ground observations, infra-red airborne systems)
Systems of assessment of weather conditions (models, ground measurements, radar, satellite imagery; )
Systems for data display and manipulation such as Geographic Information Systems
Communication systems (control rooms, radio networks, computer networks, phones, vehicle-tracking devices)
Training systems and ranking structures for firefighters (e.g. the Incident Command System [ICS])
A system for the integration of emergency services in disaster situations (e.g. through police, paramedics, caterers, structural and rural firefighters)
A system of recruitment, training, and record keeping for firefighters
A system for the recording of mapped and other data for each fire to enable analysis, retrieval, and review
A research program.
A suite of mechanisms exists in relation to the setting of policy and its implementation. Tools that assist in the education, communication, and establishment of responsibility for fire matters include Codes of Practice (see http://www.nre.vic.gov.au/), Fire-threat Analyses (e.g. Mueller 1993), Plans of Management, interagency fuel management plans, and published Annual Reports. Programs for the creation of public awareness of wildfires include Community Fireguard (see http://www.cfa.vic.gov.au ). There is a vast quantity of printed material available from firefighting and land-management agencies suited to community education (e.g. accessed through the home page of the Australasian Fire Authorities Council <http://www.ausfire.com/>).
There is a large and complicated set of legal documents affecting fire suppression and management in Australia. There are:
International treaties and agreements
Within Australia inter-government agreements
State laws and regulations
Local government regulations
There are Acts and Regulations pertaining to:
Set-up and responsibilities of government agencies
Protection of human life and property
Greenhouse gas emissions
Total fire bans; etc.
An ‘Environmental Impact Statement’ may be required for some activities. A source of detailed information on legislation and regulation may be found at <http://www.austlii.edu.au>.
Having an impact on land-use practices has been the widespread membership of the ‘National Landcare Program’, a government-based program through which funding is provided to the community (<http://www.landcare.gov.au>). Though not specifically addressing wildfires, the projects undertaken by community groups can involve fires. The farming community is heavily involved in this movement. Community-based ‘Park Care’ groups exist in the Australian Capital Territory; their involvement in the management of conservation areas includes fuel manipulation. There is a range of Community organizations that may affect policy such as various National Park Associations (State-based) and Conservation Councils.
Responding to wildfires in various ways are, among others, pastoralists in Northern Territory savannahs, traditional (Aboriginal) owners in desert grasslands of northern South Australia, managers of eucalypt forests of southwestern Australia, National Park managers in the semi-arid mallee shrublands of NSW, pine-plantation foresters in southeastern Queensland, and urban-interface residents in Tasmania. In this variety of circumstance, there is no single, planned, integrated system. Rather, there has developed a set of responses varying from the very simple to multi-faceted, multi-level, multi-agency mechanisms that vary regionally in accord with the diverse environments, tenures, population densities, and resource supplies present in different parts of the nation.
We would like to thank Mr R.H.D. McRae of the Australian Capital Territory Rural Fire Service for his comments on the draft manuscript.
Malcolm Gill and Peter H.R. Moore
Centre for Plant Biodiversity Research
CSIRO Plant Industry
GPO Box 1600, Canberra, ACT
Fax: ++61-2-6246 5249
Tel: ++61-2-6246 5116
Bradstock, R. A., J. E. Williams, and A. M. Gill. (eds.). Flammable Australia: the fire regimes and biodiversity of a continent. Cambridge University Press, Cambridge. In press.
Cheney, N. P. 1979. Bushfire disasters in Australia 1945-1975. In: Natural hazards in Australia (R. L. Heathcote, and B. G. Thom, eds.), 72-93. Australian Academy of Science, Canberra.
Gill, A. M., and P. H. R. Moore. 1990. Fire intensities in eucalypt forests of south-eastern Australia. International Conference on Forest Fire Research, Coimbra, Portugal, Proceedings. B.24, 1-12.
Gill, A. M., and P. H. R. Moore.1997. Contemporary fire regimes in the forests of south-western Australia. Report to Environment Australia. 115pp.
Gill, A. M., R. H. Groves, and I. R. Noble. 1981. (eds.). Fire and the Australian biota. Australian Academy of Science, Canberra.
Good, R. B. 1981. The role of fires in conservation reserves. In: Fire and the Australian biota (A. M. Gill, R. H. Groves, and I. R. Noble, eds.), 529-549. Australian Academy of Science, Canberra.
Johnson, R. W., and R. W. Purdie. 1981. The role of fire in the establishment and management of agricultural systems. In: Fire and the Australian Biota (A. M. Gill, R. H. Groves, and I. R. Noble, eds.), 497-528. Australian Academy of Science, Canberra.
Leigh, J. H., and J. C. Noble. 1981. The role of fire in the management of rangelands in Australia. In: Fire and the Australian biota (A. M. Gill, R. H. Groves, and I. R. Noble, eds), 471-495. Australian Academy of Science, Canberra.
Luke, R. H., and A. G. McArthur. 1978. Bushfires in Australia. Australian Government Publishing Service, Canberra.
Miller, G. H., J. W. Magee, B. J. Johnson, M. L. Fogel, N. A. Spooner, M. T. McCulloch, and L. K. Aycliffe. 1999. Pleistocene extinction of Genyornisnewtoni: human impact on Australian megafauna. Science 283, 205-208.
Mueller, C. 1993. Wildfire threat analysis. In: The burning question: fire management in NSW (J. Ross, ed.), 103-110. Centre for Continuing Education, University of New England, Armidale, Australia.
Paix, B. 1999. Improving burnover protection for Australian bushfire appliances. In: Bushfire’99 Proceedings, 307-319. Charles Sturt University, Albury, NSW.
Ramsay, G. C., N. A. McArthur, and V. P. Dowling. 1996. Building in a fire-prone environment: research on building survival in two major bushfires. Proc. Linn. Soc. NSW. 116, 133-140.
Resource Assessment Commission (RAC). 1992. Forest and Timber Enquiry Final Report, Volume 1. Australian Government Publishing Service, Canberra.
Whelan, R. J. 1995. The Ecology of Fire. Cambridge University Press, Cambridge, UK.
Williams, J. E., and A. M. Gill. 1995. Impact of fire regimes on the forests of eastern New South Wales. NSW National Parks and Wildlife Service Forest Issues 1, 68p.
Figure 1. High-intensity fire in a eucalypt stands under investigation by the C.S.I.R.O. Bushfire Research Unit in the 1980s. The New South Wales wildfires of 2001-2002 (Christmas Fires) have called for strengthening national wildland fire research capabilities. Photo: Courtesy C.S.I.R.O.