How can we predict fire spread?

Fire spread is a complex phenomena. Accurate data describing many variables are required in order to make an accurate prediction. Two strategies exist for predicting fire spread. The empirical approach attempts to isolate and measure the effects of each variable using experimentation. Predictive equations are subsequently developed from these experiments. A physical approach describes fire spread as heat transfer between burning and unburned fuel using series of coupled differential equations to solve for predicted fire spread. More details about each approach are mentioned below.

Empirical Models of Fire Spread

Fully empirical models rely on statistical correlation between variables known to influence fire spread, such as windspeed, slope, and fuel moisture content, with field observations of rates of spread. Empirical methods are incorporated into the national operational models of fire spread used in Canada, the Canadian Fire Behavior Prediction Model (Forestry Canada Fire Danger Group, 1992), and in Australia, the McArthur grassland and forest fire meters (Noble, et al, 1980). Unfortunately, the primary weakness of any empirical model, including both of these systems, is that the predictions made are not easily extrapolated beyond the conditions they were correlated to.

Physical Models of Fire Spread

Physical models of fire spread estimate the flux between burning and unburned fuel in order to determine the rate of fire spread. The prevailing assumption of this approach is that all heat transfer involved in the combustion reaction satisfies the conservation of energy. The conservation of energy is expressed as an equation in the figure to the right. This equation states that, under steady-state conditions, the rate of fire spread, R, in m/s, is equal to the ratio of the heat received by unignited fuel ahead of the fire, q, in J/s-m2, over the heat required to ignite the fuel at the leading edge of the fire, Q, in J/m3. The total energy flux received by the unignited fuel, q, is equal to the sum of the individual u energy fluxes received due to heat transfer via radiation, convection and conduction. The total energy required to ignite a unit volume of fuel, Q, is equal to the sum of the heat required to bring the individual v components of the fuel bed from ambient temperature to ignition temperature. A good source of information about the conservation of energy is Williams, 1982. See Weber, 1991 for more information on physical models of fire spread.