Fire management and greenhouse gas emissions

What greenhouse gases are produced by savanna fires?


It is thought that frequent and intense fires are making some parts of the savanna net emitters of CO2

When fire burns grass, leaves or wood, it emits smoke (tiny particles of ash etc.) and gases which are produced by combustion. The main greenhouse gases produced by such fires are carbon dioxide (CO2), methane and nitrous oxides.

In stable landscapes the carbon dioxide emitted by fire is re-absorbed to a greater or lesser extent by the new plant growth that follows fire, particularly in the next rainy or growing season. There is emerging evidence, however, that for large areas of far northern Australia such as Arnhem Land, the northern Kimberley and western Cape York Peninsula, these landscapes are not stable as more plant material is being burnt by wildfire than is growing back again afterwards, and consequently frequent late dry season fires are slowly reducing the plant biomass and creating a net release of CO2 into the atmosphere – however, the nature of this process is complex and the quantities of CO2 released and absorbed by savanna fires have yet to be precisely determined1.

Bushfires also emit other greenhouse gases, principally methane and nitrous oxide, which unlike CO2 are not re-absorbed by the landscape to any great extent. (Methane and nitrous oxide are mostly broken down in the atmosphere with only a small amount of methane being re-absorbed by the soil). These emissions are therefore easier to count as being net greenhouse gas emissions and are included in Australia’s National Greenhouse Gas Inventory as being produced by savanna burning.

How significant are Methane and Nitrous Oxide as greenhouse gases?
Greenhouse gases are those gases in the atmosphere that trap the heat coming from the earth and so keep that heat in the atmosphere. The key greenhouse gases make up a very small part of the atmosphere: CO2 comprises around 380 molecules in every million, Methane comprises only around 1.7 molecules in every million and Nitrous oxide make up only around 0.32 molecules in every million. These gases are stable in the atmosphere for many years and despite their low concentrations they can alter the balance of radiation that warms the planet.

A given amount of methane gas is around 25 times more effective at trapping heat over a 100 year period than the equivalent amount of CO2 but because there is much less methane in the atmosphere it has around 30% of the impact that CO2 does on the radiation balance. Similarly a  given quantity of nitrous oxide is around 300 times more effective at trapping heat over a hundred year period than CO2 but due to its lower concentrations in the atmosphere it has around 10% of the greenhouse impact2.


How significant are the greenhouse gas emissions produced by savanna fires?


Fires in Australia's tropical savannas make a significant contribution to greenhouse gas emissions.

The Australian Greenhouse Office’s National Greenhouse Gas Inventory estimates that accountable greenhouse gas emissions from savanna fires contribute less than 2% of Australia ’s total emissions. Following rules adopted by the Kyoto Protocol, the Inventory only accounts for methane and nitrous oxide as fire-related greenhouse gas emissions. 

Savanna fires are also a very significant source of the most common greenhouse gas — CO2. It has been estimated that the burning of savannas in northern Australia releases up to 218 million tonnes of CO2 3 which persist in the atmosphere for periods up to seven months of the year over the fire season.

This amount is equivalent to 38.5% of Australia ’s total greenhouse gas emissions in 2004. Although there are considerable timing uncertainties concerning the degree to which this CO2 contributes to climate forcing, it is clear that the gases produced by savanna fires are an important factor in Australia’s greenhouse budget.


How does strategic fire management reduce greenhouse gas emissions?

firebreak map Arnhem Land

Fire breaks recorded by satellites (MODIS imagery) during the 2005 fire season on the Arnhem Land Plateau. Early dry season fire breaks  shown in green (black arrow) were put in along rivers and tracks and succeeded in stopping the late dry season wildfires shown in pink (red arrow) that came in from the east before July 1. Source: WA Dept. Land Information.

Field studies and remote-sensing data have shown that early dry season fires emit less greenhouse gases (Carbon dioxide, nitrous oxides and methane) per area affected than the more intense, late dry season fires4. This is mainly because the earlier fires:

  • Are not as intense and burn less of the grassy fuel than a more intense fire would — so plants that are burned are often only partially consumed by the fire, and the fire often leaves parts of the plant unburnt
  • Do not burn the entire grass layer — often large patches of grass and litter fuels are unburnt by an early dry season fire.
  • Usually stay in the grass layer, whereas the intense fires typical of the late dry season can move into the upper canopy and can consequently consume the additional biomass of organic matter in tree trunks and branches. Smouldering stumps and wood are known to emit more methane and nitrous oxide gas than grass fires.5

Early dry season fires tend to be more easily stopped by roads, small creeks and rivers or dew cover and so tend to burn less country than a late dry season fire.

Furthermore, if early dry season fires are used to create fire breaks — or strips of already burnt country — in the landscape, this can limit the spread of late dry season wildfires. So the major way in which wildfires can be controlled and greenhouse emissions reduced is through this strategic early dry season burning.


Won’t all the emissions abated in this way just go straight back into the air if there is a big wildfire in the future?

It is possible that despite the best efforts of fire managers, a large wildfire may burn a very large percentage of the WALFA Project area at some point in the future. If, for example, a large area of the plateau had remained unburnt for five years due to careful fire management but was then burned by an intense fire, would five years’ worth of fuel go up in smoke, producing a huge pulse of greenhouse emissions equivalent to all emissions produced if the site had been burnt more frequently? The answer is that it would not and that such fires would only have minor impacts on the overall greenhouse emissions provide they were infrequent – and this is an important point underlying the WALFA strategy.


Decomposers like these termite colonies limit the grass and leaf litter available for burning

Unlike southern forests, the fire-prone northern savannas do not accumulate large amounts of fuel in the form of litter as in the humid tropics litter is rapidly decomposed by organisms including bacteria, fungi and termites, and the available fuel for burning tends to level out after 2-3 years. So if an area of humid tropical savanna is left unburnt for five years or even ten years, most of the grass and leaf litter produced in this period will have been decomposed, leaving only a small proportion available as fuel for fire (see graph below).

This decomposition releases less greenhouse gases over a longer period of time than burning does. For example, when termites digest plant material they can release significant quantities of methane, but much of this methane is then re-absorbed by bacteria in the soil.5


What might be the long-term impact on greenhouse emissions from the landscape of better fire management?

The humid tropical savannas of far northern Australia are dynamic landscapes in which vigorous growth in a lush wet season is balanced by high rates of decomposition and regular fires. Many plants and animals are adapted to regular fire and it has an important place in Indigenous culture, so the overall goal of the WALFA project is not to remove fire from the landscape, nor is it to guarantee there will never be very large wildfires, but rather the goal is to shift the overall long-term pattern of fire – the fire regime – away from one dominated by frequent intense fires to one with less frequent intense fires and a much more varied pattern of burning.

Under this new fire regime, while there will probably be a somewhat greater biomass of trees and grass, the more significant change is that relatively more plant material in the West Arnhem Land Plateau will be decomposed rather than burnt resulting in less greenhouse gas emissions each year than has been the case with more frequent fires.



1. Cook, G.D., Liedloff, A.C., Eager, R.W., Chen, X., Williams, R.J., O’Grady, A.P. & Hutley, L.B. 2005, 'The estimation of carbon budgets of frequently burnt tree stands in savannas of northern Australia, using allometric and isotopic discrimination', Australian Journal of Botany, 53, 621–630.

2. See table 2.14, Changes in Atmospheric Constituents and in Radiative Forcing, Chapter 2, Fourth Assessment Report, IPCC 2007

3. Australian Greenhouse Office, 2006, National Greenhouse Gas Inventory 2004. Australian Greenhouse Office, Canberra.

4. Russell-Smith, J., Edwards, A., Cook, G.D., Brocklehurst, P., Schatz, J. 2004, Improving greenhouse emission estimates associated with savanna burning in northern Australia: Phase 1. Final Report to the Australian Greenhosue Office. Tropical Savannas CRC, Darwin.

5. Cook GD (2008) Fuels, fires and greenhouse gases. In ‘Managing fire regimes in north Australian savannas – ecology, culture, economy’ (eds J Russell-Smith, PJ Whitehead, P Cooke).CSIRO Publications, Melbourne. (in preparation)