This project was the first, comprehensive investigation into the life cycle of fire behaviour, from initiation, through development, to steady-state behaviour. Knowledge gained will better enable fire authorities to understand the expected behaviour of fire, and the window of time they have to successfully attack new fires before they become too big for direct attack, and require extended or indirect strategies.
This project investigated factors that are important in influencing the occurrence of bushfires, the successful initiation of spot fires from firebrands, the rate of growth and development of new fires that successfully ignite to steady-state rate of spread, and the transitions of such fires through vertical fuel strata as they increase in intensity and size.
The project's research team was based at CSIRO Ecosystem Sciences. Its leader was Dr Andrew Sullivan, working with researchers Dr Peter Ellis, Jim Gould and Dr Matt Plucinski. Five major areas were investigated.
Fire occurrence modelling
This research developed models that predict the number and probability of human-caused bushfires per day in south west Western Australia from bushfire incident records and weather data. Significant predictor factors used in these models included fuel moisture, the number of recent bushfires, the day of the week (or type of day, such as public holidays) and rainfall.
Analysis shows that days with human-caused fires are more likely to occur on weekends and public holidays that coincide with days of low fuel moisture content, as well as days that follow periods of high fire activity.
Findings show that the models have reasonable accuracy (between 81 and 99%). Fire management agencies could use these models to inform their daily operational resource planning.
The study relied on fire incident records, and demonstrated the importance of high quality data for analyses and modelling.
Fire initiation from firebrands
This study calculated the probability of a flaming ignition in a dry sclerophyll forest fuel bed by standard flaming and glowing firebrands. The firebrand samples were selected to have the combustion and aerodynamic characteristics of small firebrands from the dry forests of southern Australia. Ignition was studied in a small purpose-built wind tunnel.
For flaming firebrands in the absence of wind, a fuel moisture content of 14% or less is necessary for ignition to commence, with a 50% probability of successful ignition once fuel moisture content is below 9%. In the presence of wind, the fuel moisture content can be as high as 21% before ignition commences, with a 50% probability of successful ignition once fuel moisture content is below 13%. At moisture contents less than about 7%, ignition is essentially guaranteed from any flaming firebrand equivalent to the standard sample.
For glowing firebrands, fuel-bed moisture content and wind speed were the key explanatory variables. The importance of wind in ignition from glowing samples suggested that air flow turbulence may play a large role in determining successful ignition from glowing firebrands.
Fire growth and development
The CSIRO Pyrotron (a bushfire wind tunnel) was used to experimentally study fires ignited from a range of sizes in a continuous bed of fine, dry eucalypt litter fuel, with rate of spread was measured precisely.
Analysis of the data resulted in a model that suggests that, under extreme fire weather conditions, a fire starting from a point in uniform, dry sclerophyll litter will take about 20 minutes to reach steady-state rate of spread. Under milder conditions, a fire may take an hour to reach steady state.
The results are most applicable to prescribed fire, but also provide important insights into fire development in eucalypt litter and indicate the importance of time in fire control, which increases with fire danger.
Vertical fire transition and propagation
This study investigated the growth of fire through the vertically separated, horizontally stratified structure of dry sclerophyll forests.
A model based upon the fundamental physics of conservation of energy, heat transfer and combustion was developed that described the escalating behaviour of a fire burning in the understorey of eucalypt forests as it ignites and spreads through higher strata of fuel.
Fire-suppression, resource-allocation framework
This framework provides a roadmap for linking the models developed for fire occurrence prediction, fire initiation probability, fire growth and fire transitions into a national bushfire suppression resource allocation tool for Australia. In particular, this tool could answer the critical questions asked by fire authorities, such as: How many resources are required for a given set of conditions? Where should such resources be placed? And how should the suppression actions be prioritised?
The increasing need for fire agencies to share suppression resources during peak demand periods makes it even more important to be able to anticipate short-term fire danger, fire load, and suppression-resource demands so that resources can be mobilised most effectively. Fire propagation and development information can be used with models of fire occurrence to better understand fire load. This information can assist decision making about resource requirements, allow fire authorities to better determine the transition between initial and extended attack, prioritise between fires, and generally improve planning.