Savanna Explorer > Arnhem Land > Fire > Impact of Wildfire

Arnhem Plateau heathland after an intense fire


Arnhem Plateau woodland after 2004 wildfires showing damage to entire tree canopies

While the population of the western plateau was severely disrupted and in decline by the end of the nineteenth century, a number of families in the eastern plateau continued nomadic hunter gatherer life styles up into the 1960s.

The descendants of these remaining groups included people like Wamud Namok who retained considerable traditional knowledge and who continued to visit the plateau. What he and other elders saw in the late 20th century dismayed them: a plateau country that was plagued by frequent intense wildfires which were slowly degrading the cultural and natural values.

These relentless, intense fires have been damaging rock art sites, destroying valuable traditional food plants as well as damaging many distinctive plant communities along with the animals that depend on them.

Satellite evidence

As well as the testimony of senior elders, another strand of evidence for the impact being wrought by fire on the plateau comes from satellite images. Since the early 1990s images of country blackened by fire (fire scars) have been available for the whole of Australia from NOAA (US national Oceanic and Atmospheric Administration) weather satellites. The image for 1999 is shown in the upper part of the figure below, where the blue areas mark scars from the generally less intense, patchier fires that burnt from January – July and the red areas mark scars from the generally more intense, more widespread wildfires that burnt from August to December.

The bar graph shows that in 1999 more than half of Arnhem Land was burnt, mostly by late fires (orange bars). While 1999 was a bad year, on average for the 10 years 1995–2004 around 40% of the area was burnt mostly by late fires. These data show that there has been a new long-term pattern of fire established in the plateau dominated by frequent extensive fire mostly late, intense fires.

fire history graph_arnhem_land

Fire regimes on the Arnhem Land Plateau: pale orange bars show extent of early fires, orange bars, late fires, brown bars the total extent of fires as a percentage of the area of the Arnhem region.

This represents a new fire regime — and a more destructive regime than the former one with more patchy, diverse fires.1

This satellite data is complemented by evidence on the ground from fire researchers who were able to support the observations of the elders after a series of field trips to the plateau.2  

Impacts on selected species

Have these impacts on the landscape translated into measurable declines in any plant or animal populations on the plateau? This is not an easy question to answer as there has been little detailed research on the plants and animals of the plateau. Add to this the fact that changes in numbers of species may be due to a number of factors, not just changed fire regimes. Nevertheless there are a number of species where numbers have declined on the plateau and where the change in fire regimes is thought to be a major factor. Some of these are described below.3


Cypress Pine Callitris intratropica: Declining
  • woodland species;
  • obligate seeder – only sprouts from a seed;
  • high mortality in hotter fires (with seedlings susceptible to mild fires);
  • can survive mild fires every 2-8 years, but not more frequent or more intense;
  • “long-term outlook in this area .. is bleak”4


Brush-tailed rabbit-rat  Conilurus penicillatus  hanging on
  • woodland species;
  • shelters mostly in hollow logs;
  • diet mostly comprises seeds of perennial grasses;
  • broad-scale studies demonstrate significant correlation with low fire frequency;
  • preferred fire regime – infrequent fire

Arnhem Land rock-rat  Zyzomys maini: Declining
  • stone country species;
  • dependent on seeds & fruits of mostly woody species;
  • single hot fire caused major drop in population, reduced subsequent breeding, and response evident >1 yr post-fire;
  • preferred fire regime – no fire.


Northern Quoll  Dasyurus hallucatus  Declining
  • lowland and (mostly) stone country species;
  • shelters in rock crevices and hollow logs;
  • main cause of mortality is predation;
  • predation rates much higher in extensively and intensively burnt areas;
  • broad-scale studies show highly significant correlation with low fire frequency;
  • preferred fire regime – infrequent and small-scale fire;

What has caused the change in fire regime?

The underlying cause of these changed fire patterns is the recent change in people’s activities and settlement patterns across northern Australia.

The climate-driven pattern of rain and growth of vegetation (usually January–April) followed by the drying of vegetation and then fire (usually May–December) has likely been a feature of Australian tropical savannas for millions of years and the entire extent of northern Australia has been occupied for tens of thousands of years by indigenous people who used, and in various places still use fire as a tool for customary and contemporary purposes.

This long period of occupation by Indigenous people created a landscape filled with patches of country that had been burnt in different ways – some patches with very little grassy fuel that were recently burnt, some with thicker grass burnt less frequently, some with fire sensitive plants that had never been burnt. These patches prevented wildfires becoming very large very often. The result has been plant and animal populations that are adapted to a variety of different fire types at various times of year and with varying frequency.5

In the last century, however, this situation has changed dramatically. The West Arnhem Plateau is now sparsely populated with Indigenous people living mostly in a few large regional settlements in the surrounding lowlands. There is limited fire management occurring on the lowlands to the east of the plateau but there are still plenty of places from which a fire can start such as from settlements and roads. Such fires often start through accidental ignition or from fires lit for other purposes that escape.  If these fires start in the late dry season in hot, windy weather they can become more intense, spreading into the tree canopy layer. Such wildfires will often be carried by the easterly trade winds of the late dry season into the Arnhem Land Plateau. Fires can also be carried into the plateau from the west. 6

2004 fire scars small

Satellite mapped fire scars in 2004. Green areas denote early fires; yellow, orange, pink and purple areas denote late fires. (MODIS imagery)

The image at right shows the wildfires of 2004: the coloured patches show country that satellites have registered as being burnt, the grey, green and yellow patches indicate fires in the early-mid dry season (often prescribed fuel-reduction burns) with the larger orange, pink and purple patches being country burnt by late dry season fires (often wildfires). As shown in 2004, much of the West Arnhem Plateau was burnt by wildfires.

While they wreak havoc on the ground these northern are emitters of Greenhouse Gases on a massive scale, often comprising over 40% of Northern Territory’s total annual emissions: more than the emissions of any other sector.7 If this sounds hard to believe, consider the scale of these fires – in 2004 an area greater than the entire state of Victoria was burned in the Northern Territory alone.8

Given that these greenhouse emissions come from fire patterns created by people, they can be considered anthropogenic – meaning that fire managers who can reduce the incidence of these wildfires could also reduce – or “abate” – the human-caused greenhouse emissions they create, and may thereby be eligible for payment for this environmental service.


1. Russell-Smith Jeremy, Yates Cameron, Edwards Andrew, Allan Grant E., Cook Garry D., Cooke Peter, Craig Ron, Heath Belinda, Smith Richard (2003) Contemporary fire regimes of northern Australia, 1997–2001: change since Aboriginal occupancy, challenges for sustainable management. International Journal of Wildland Fire 12, 283–297.

2. Jeremy Russell-Smith, Paul G. Ryan, David Klessa, Gordon Waight and Robert Harwood (1998) Fire Regimes, Fire-Sensitive Vegetation and Fire Management of the Sandstone Arnhem Plateau, Monsoonal Northern Australia The Journal of Applied Ecology, Vol. 35, No. 6, pp. 829-846

3. Woinarski, J. (2008). Biodiversity and Fire in Western Arnhem Land. In Managing fire regimes in north Australian savannas – ecology, culture, economy (eds J Russell-Smith, PJ Whitehead, P Cooke). CSIRO Publications, Melbourne. (in preparation)

4. Prior, L.D., Bowman, D.M.J.S., Brook, B.W. (2007) Growth and survival of two north Australian relictual tree species, Allosyncarpia ternata (Myrtaceae) and Callitris intratropica (Cupressaceae). Ecological Research, Vol 22, Issue 2, pp228-236.

5. Russell-Smith, J., Lucas, D., Gapindi, M., Gunbunuka, B., Kapirigi, N., Namingam,G.,  Lucas, K., Giuliani, P., Chaloupka, G. (1997) Aboriginal resource utilization and fire management practice in western Arnhem Land, monsoonal northern Australia: notes for prehistory, lessons for the future. Human Ecology, 25, 159–195.

6. Cooke, P.(2008). Social history of the Top End and western Arnhem Land - its influence on fire management. In Managing fire regimes in north Australian savannas – ecology, culture, economy (eds J Russell-Smith, PJ Whitehead, P Cooke). CSIRO Publications, Melbourne. (in preparation)

7. Australian Greenhouse Office (2007) Australia’s national greenhouse accounts: State and Territory Greenhouse Gas Inventories 2005. Department of Environment and Water Resources: Canberra.

8. As measured by the extent of fire scars mapped from MODIS (Moderate Resolution Imaging Spectroradiometer) Satellite images in the Victoria River District in late 2004.