A researcher checking fire intensity measurements after a tropical savanna experimental fire

Dr Dick Williams uses buried electronic timers to measure the rate of spread of fires in northern Australia.

Kapalga fact sheet series: fire behaviour

CSIRO conducted one of the world’s largest fire experiments - the Kapalga Fire Experiment – over ten years at Kapalga Research Station in Kakadu National Park in the Northern Territory, Australia and this fact sheet is one in a series highlighting the significant findings of this fire research.

  • 5 September 2007 | Updated 14 October 2011

Fire in the Top End

Frequent fire is a natural and important part of the tropical savanna environment in northern Australia.

Every Dry Season, from May to October, up to 50 per cent of the landscape is burnt, and most fires are deliberately lit.

Fire is the most important land management tool for biodiversity conservation however; there is widespread confusion and concern over why so much burning takes place.

There is a pressing need to improve our knowledge and public understanding of the role of fire in the northern Australian environment. 

As part of the Kapalga Fire Experiment, CSIRO studied fire behaviour and its implications for conservation management.

What is fire behaviour?

Fire behaviour is a term used to describe the physical characteristics of fire, for example:

  • how fast they travel
  • how much energy is released
  • how high the flames are, etc.

One important and widely-used fire behaviour measure is fire intensity - the rate of energy release along the fire front.

It is defined as the product of three basic physical features:

The relatively high intensity of fires late in the dry season is now known to be due to a combination of more extreme fire weather, and drier and heavier fuel loads, compared with the early dry season.
  1. rate of spread
  2. fuel load
  3. heat released from fuel during combustion.

The relatively high intensity of fires late in the Dry Season (September – October) is now known to be due to a combination of more extreme fire weather, and drier and heavier fuel loads compared with the early Dry Season (May – June).

Knowledge of the seasonal changes in the determinants of fire behaviour in the tropical savannas of northern Australia is necessary to better understand the impact, or 'severity', of fires and fire regimes on the environment.

The ability to quantify and predict fire behaviour also allows people preparing to use prescribed fire to estimate the likelihood that they will be able to manage fire for their purpose.

For example, land managers and graziers may need to burn to manage grasses and biodiversity, and to avoid larger fires occurring later in the dry season.

They will want to know what kind of fire may occur, given the weather conditions and the type and moisture content of fuels at the time of ignition, and what impacts it will have.

In addition, fire management authorities require knowledge about fire weather and fire behaviour to determine fire dangers for particular weather conditions each day.

What determines fire behaviour?

Fire behaviour is determined by the quantity and quality of fuel, and prevailing fire weather conditions.

The major determinants of fire weather are:

  1. air temperature
  2. atmospheric humidity
  3. wind speed.

Measures of temperature, humidity and wind speed may be used in conjunction with soil and fuel moisture features to derive Fire Danger Indices, such as the commonly used Forest Fire Danger Index (FFDI).

Then, knowing where you are in the dry season (early, mid or late), soil moisture levels, FFDI and fuel loads you can determine the potential rate at which a fire might spread, and hence fire intensity.

Fire weather

Seasonal variation in fire weather for the Jabiru region in eastern Kakadu was explored using 11 years of meteorological data.

The average FFDI for Jabiru, was seen to increase steadily through the dry season.

Extreme FFDI values (>50) occur occasionally, but even so, the peak FFDI values for Jabiru are far less than the peak values of about 100 that have been recorded for eucalypt forests in southern Australia.

Fuel loads

Fuel load measurements at Kapalga were consistent with those of other savanna areas in Australia, both moderately moist (mesic) and semi-arid.

They varied depending on the time since fire and the time of year.

Key factors in seasonal differences were:

  • temperature and humidity
  • leaf litter build up
  • higher and drier fuel loads.

Pre-fire fuel loads ranged from two to five tonnes per hectare in annually burnt experimental sites.

Fuel loads were 20 - 50 percent higher in the late dry season (as a consequence of leaf build up) than in the early dry season, and were also drier.

Fuel loads in unburnt savannas were five to ten tonnes per hectare.

Fire intensity

A fire’s intensity is definied as the amount of energy released per unit length of fire front, in units of kilowatts per metre.

The measurement is sometimes called Byram fire-line intensity.

The mean fire intensity of early dry season fires was found to be one-third that of late fires.

The relatively high intensity of late dry season fires is due to a combination of more extreme fire weather and drier and heavier fuel loads in the later part of the dry season, compared with the early dry season.

Average fire intensity for early fires was 2 200 kW/m, compared with 7 700 kW/m for fires late in the dry season; maximum measured intensity was about 20 000 kW/m.

This compares with potential maximum intensities of 100 000 kW/m for crown fires burning under extreme conditions in south-eastern Australia.

'As part of the Kapalga Fire Experiment,' explains Dr Dick Williams, 'we measured the intensity of early and late dry season fires by measuring rate of spread and fuel loads.'

'We then related intensity to measures of fire severity after fires such as char and scorch height in trees, and the amount of ground burnt.'

Scorch height and char height are useful post-fire indicators of fire intensity.

Char height increases by about 0.5 metre for each 1 000 kW/m.

Fires above 2 000 kW/m consume the entire grass layer; above 4 000 kW/m fires scorch the tree canopies to a height of about 20 metres.

It is important to note that although late dry season fires are usually of higher intensity than early dry season fires, fires of equally high intensity can occur early in the dry season if the time since fire is two years or more and the weather is, as it can occasionally be, hot, dry and windy.

This is an important aspect that is being further investigated as part of the Burning for Biodiversity program.

Current research – burning for biodiversity

Results from the Kapalga Fire Experiment highlighted the need for further research focusing on the effects of fire frequency and time-since-fire on fire behaviour, biodiversity and ecosystem processes. 

To meet these needs a new fire research project, Burning for Biodiversity, has been developed at the Territory Wildlife Park near Darwin in the Northern Territory, as part of the Bushfire Cooperative Research Centre.

The project is a collaboration between:

  • CSIRO
  • the Northern Territory Government
  • Charles Darwin University, Darwin.

Read more about the Kapalga fire experiment.

  • Andersen A, Cook G, Williams R. 2003. Fire in Tropical Savannas, The Kapalga Experiment. New York: Springer-Verlag.