Fire-spread was measured on 121 grass fires in a 2500 ha experimental site in the Northern Territory, Australia. Selected plots were harvested to alter the height, load and bulk density of the fuel-bed. Fires were lit from a line and allowed to travel up to 400 m down-wind. Fire-spread was correlated with fuel, weather and fireshape variables using multiple regression techniques. Wind speed had most effect on fire-spread. The influence of the other variables was examined after a model for wind speed and moisture content had been fitted. Fuel load did not influence fire-spread. Fires in natural swards burnt 18% faster than fires in cut grass, but this increase could not be fully explained by changes in the height or bulk density of the fuel bed. Grass type characterised either by species group or by surface-area-to-volume ratio of the fuel particle, did not appear to significantly influence fire-spread. Differences in spread rates between the two grasses were attributed to differences in grass curing. The influence of grass curing appeared to be less than indicated by published models. Models of fire-spread in grasslands currently in use need to be revised. Ignition line length was a significant variable influencing fire-spread and this must be taken into consideration when using experimental fires to validate theoretical models or develop empirical models from field observations.

This paper describes a model to predict fire spread in grasslands from wind speed at 10 m, dead fuel moisture, and degree o f grass curing in three defined pasture types.The model was developed from spread measurements of experimental fires that were adjusted to their potential rate of spread at wide fronts. Extrapolations of the model were compared with spread data from 20 major wildfires in Australia.This model uses different functions to describe the relationship between rate of spread and wind speed above and below a critical wind speed of 5 km h-l. A linear relationship is used below 5 km h-' ; above 5 km h-' rate of spread is described by a power function of wind speed with an exponent of less than 1.

The development of grass fires originating from both point and line ignitions and burning in both open grasslands and woodlands with a grassy understorey was studied using 487 periods of fire spread and associated fuel, weather and fire-shape observations. The largest fires travelled more than 1000 m from the origin and the fastest 2-minute spread rate was over 2 m s-1. Given continuous fuel of uniform moisture content, the rate of forward spread was related to both the wind speed and the width of the head fire measured normal to the direction of fire travel. The head fire width required to achieve the potential quasi-steady rate of forward spread for the prevailing conditions increased with increasing wind speeds. These findings have important implications for relating small-scale field or laboratory measurements of fire spread to predictions of wildfire spread. The time taken to reach the potential quasi-steady rate of spread at any wind speed was highly variable. This time was strongly influenced by the frequency of changes in wind direction and the rate of development of a wide head fire.

Project Vesta was a comprehensive research project to investigate the behaviour and spread of high-intensity bushfires in dry eucalypt forests with different fuel ages and understorey vegetation structures. The project was designed to quantify age-related changes in fuel attributes and fire behaviour in dry eucalypt forests typical of southern Australia. The four main scientific aims of Project Vesta were: To quantify the changes in the behaviour of fire in dry eucalypt forest as fuel develops with age (i.e. time since fire); To characterise wind speed profiles in forest with different overstorey and understorey vegetation structure in relation to fire behaviour; To develop new algorithms describing the relationship between fire spread and wind speed, and fire spread and fuel characteristics including load, structure and height; and to develop a National Fire Behaviour Prediction System for dry eucalypt forests. These aims have been addressed through a program of experimental burning and associated studies at two sites in the south-west of Western Australia.

The relationship between woody fuel consumption and fireline intensity was assessed using data collected at controlled fires and wildfires in south-western Western Australia, central Victoria and south-eastern New South Wales. The combined dataset consisted of fires in a range of dry eucalypt forests. Fire behaviour varied from slow, self-extinguishing prescribed bums to intense, fastmoving fires burning under conditions of extreme fire danger. Fire line intensity ranged from 50 kW m-1 to > 31 000 kW m-1 • Woody fuel consumption ranged from 31% to I 00%, and generally increased with fire intensity. Percentage consumption was highest for small woody fuels where the diameter was between 0.6 em and 2.5 em. Fireline intensity had a statistically significant, positive relationship with the proportion of woody fuel consumed by both controlled fires and wildfires. Two generalised linear models (GLM) describing woody fuel consumption as a function of fire line intensity were developed, one applicable to the prescribed fire environment (with fire line intensities typically< 750 kW m-1 ) and the other to the full range of fire line intensities. The prescribed burning model produced the best fit and lowest error statistics.