Land Theme Report
Australia State of the Environment Report 2001 (Theme Report)
Prepared by: Ann Hamblin, Bureau of Rural Sciences, Authors
Published by CSIRO on behalf of the Department of the Environment and Heritage, 2001
ISBN 0 643 06748 5
Physical changes to natural habitats (continued)
Fire: a hot topic [L Indicator 2.6]
Fire has shaped much of the Australian vegetation (White 1994). Aboriginal peoples used fire widely to manage the vegetation (Langton 1998). It is probable that in Australia's wetter areas under Aboriginal management wet sclerophyll forest replaced closed forest, and open savanna replaced open forest across much of its pre-Aboriginal distribution. In arid regions controlled patch burning produced a characteristic mosaic of low groundcovers, shrubs and grasses.
Some of the changes that are well documented include:
- lower frequency of burning, associated with higher grazing intensity, in arid and tropical rangelands. In the northern tropics, this led to the build up of massive fuel reserves and huge wild-fires. In more arid areas this had led to less grassland, more bare soil and more shrubs. In both cases it is more difficult to use fire as a management tool.
- increase in weed species, such as 'woody weed' regrowth of native species that are inedible to most herbivores.
- introduction of weeds that require fire for their control, including rubber vine (Cryptostegia grandiflora) and prickly acacia (Acacia nilotica).
- very frequent burns in regions surrounding roads, and metropolitan and urban centres.
- build up of understorey in forests, resulting in massive wild-fires in the period 1900s to 1970s. Since then 'cool prescribed burns' have been introduced by some forestry agencies to reduce wild-fire risk.
- regular burning of crop stubbles and cane from the 19th century to 1970s. Now partially reduced by the adoption of stubble-mulching in some areas.
The Department of Land Administration in Western Australia (DOLA) has been conducting real-time fire monitoring of the Kimberley region in north-western Western Australia since 1993 using satellite thermal signals from night images, local detailed studies using Landsat image analysis, and ground-truthing data. The results show that the area burnt has increased every year since monitoring began (Table 18). The reason suggested for the greater extent of fire in the latter part of the decade has been the effect of wetter seasons, but increased fire-stick management by Aboriginal people now occupying larger areas of their traditional lands in the Kimberleys may also have been significant.
|Year||Area (million ha)|
A Part season or year only.
B IBRA province 1, covering all of north-western Australia.
|Year||Fire-affected area (km2)|
|1 April 1998 - 31 March 1999||312 000|
|1 April 1999 - 31 March 2000||711 000|
|Total||1 023 000|
http://www.rss.dola.wa.gov.au / for maps of hot spots and fire scars
Source: Craig et al (2000).
The monitoring system reported above was gradually extended to cover all of Western Australia and the Northern Territory by 1997 and it has been extended since 1997 to cover the whole of Australia at a resolution of 1 km2.
For the first time a moderately accurate continental picture of where fire is occurring, how extensive such fires are and how frequently fires recur exists. Over 1 million km2were burnt in the two years of monitoring, and in the northern tropical regions very extensive contiguous areas (>50 000 km2) were burnt each year. The total area burnt represents 13% of the continent. This area is known to be an underestimate as small and day only fires are not detected by the method used.
The total area burnt doubled in the second year, probably because of the increased biomass production associated with the La Nia (wetter) phase of the Southern Oscillation. Ninety per cent of the area burnt each year occurred in the north and north-west tropics and in the arid interior. Thus, although a very large number of fire hot spots were detected, elsewhere the areas burnt were generally 1-10 km2in size.
As biomass increases and then dries after the rainy season there is an increase in the number of fires in March to August in northern Australia. Lightning strikes are probably responsible for many of the very large burnt areas that occur in more remote and inaccessible regions, but in some areas of the Kimberleys burns occur every season, each year, and much of this burning comes from deliberate firing. The picture that emerges from seven years of monitoring in the Kimberleys is that large parts of this region are being burnt at least twice a year, including late burns at the end of the dry season.
Large numbers of hot spots occur throughout the length of the eastern interior plains, although the fire-scars are small compared with those in the north. There is also a wide arc of small hot spots through the grain producing areas of the continent during the summer and autumn. Thus, despite surveyed statistics that indicate over 75% of farmers practise stubble retention, it is clear that significant amounts of agricultural lands are burnt each year.
Fire is a confused and emotive issue, with insufficient scientific information and rational discussion within different parts of the community to provide an effective base for long-term sustainable management. Natural fires (called 'wild-fires' by some agencies) are as much part of the Australian environment as are drought and flood. Settled rural and urban communities must and do fear large-scale fires because of their destructive impact. Emergency and bushfire control services exist to suppress and contain fires that may threaten lives and property, whereas environmental scientists and some Aboriginal communities expect to use fire as a regular management tool.
Many landholders are ambivalent in their attitude to fire. Cropping farmers know the benefits of stubble and trash retention, but have problems of diseases, pests (such as mice plagues) and seeding through thick stubble which are easier to manage after burning. Forestry managers fear the devastation of wild-fires, but prefer to clear undergrowth with 'cool' burns in autumn or winter, to lessen the fuel risk in summer. The benefits and adverse impacts from fires are therefore considered in relation to specific ecosystem processes in the following section. Legislation on the control of fires currently reflects the old view of southern Australian settlers and urban dwellers, who saw fire only as a threat to life and property. There is a deficiency in the legislation for the use of fire as a management tool, particularly in those parts of Australia where suppression of fire has led to significant weed and bio-fuel accumulation and ecological change.
Figure 39: Seasonal occurrence of fires (hot spots) across Australia for 1998 to 2000.
Source: DOLA (1998-2000)
The Tropical Savannas Cooperative Research Centre has drawn together our knowledge on the effects of frequency and timing of burning, which clearly shows the catastrophic effect of frequent wild-fires, particularly late burns (Jacklyn and Russell-Smith 1998). The influence of the burning on different biogeographic provinces and biodiversity is covered separately in the Biodiversity Theme Report. Here we are concerned with the implications for erosion, carbon cycling and weed suppression.
Experiments conducted between 1990 and 1994 at Kapalga Research Station, Kakadu National Park compared early dry-season burn, three burns through the dry season, a late dry-season burn, and no burning. The late dry-season fires (September to November) were found to be the most damaging to ecosystem integrity (Williams et al. 1999, 1996).
Runoff in the early part of the wet season occurred only from the late-burnt area, and as the wet season progressed sediment concentrations from this area were two to three times that of other treatments, Sediments contained high concentrations of nitrogen, phosphorus, iron and manganese. Within a season or two of such practice, the nutrient-bearing surface soil is lost, with an absolute loss of soil fertility on sloping or rocky terrain. Water quality is also reduced, predisposing billabongs and creeks to algal blooms in the dry season (Saint and Russell-Smith 1997).
CSIRO's Tropical Ecosystems Research Centre has estimated that savanna fires in northern Australia release an estimated 80 million tonnes of carbon dioxide (CO2) per year, more than that emitted by all transport (70 million tonnes).
However, the regeneration of vegetation in the following wet season is considered to take up an equivalent amount (Cook et al. 1995, Hurst et al. 1996). This is an area that warrants further research.
Several weeds of national significance - rubber vine (Cryptostegia grandiflora ), mesquite bush (Propsopis chilensis) and prickly acacia (Acacia nilotica) - can be effectively controlled by intense fires, and the reintroduction of planned intense fires on all pastoral properties in north-eastern Queensland has been advocated as an essential element in weed control. The current situation is that there are more late dry-season fires than early ones and some areas are not being burnt at all in the southern Kimberley; Victoria River District; and the Barkly Tableland, north-western Queensland. Parts of eastern Cape York have not had any burning for over a decade. In these areas woody weeds have become too thick to allow effective grazing. Effective use of fire thus remains a significant challenge to landholders.