Another American scientist, George Byram, came up with the idea of measuring fires by their rate of heat release, or "intensity". [Incidentally, this is the same Byram of the Keetch-Byram Drought Index.] Just as apples, celery and steak have calories or, in more modern terms, joules, so do fuels. In fact, for our purposes, we can say that the number of joules per gram, or kilojoules per kilogram (kj/kg), are the same in all fuels whether the fuel be dried apples or grass. While there is argument about the value of this "constant" number (because of various technical reasons about the effectiveness of heat released in fire development and variation between fuels), let's say it is 20,000 kj/kg so we can do some sums in our heads or on the backs of envelopes. Those who prefer other numbers, usually lower ones, can reduce the answer by the appropriate percentage. Now Byram recognized that a fast-travelling fire would release heat faster than one travelling slowly and one with more fuel would release more heat. This led to the formula for fire intensity (Byram 1959):

fire intensity = kj/kg multiplied by rate of spread in metres per second (m/sec) multiplied by dry fuel load per square metre (kg/m2).

For convenience lets say a fire was moving at 1 m/sec (3.6 km/hr), about half the speed of a rapid walk but quite fast for a fire. Now let us assume that the fuel loading for this fire is 1 kg/m2 (10 t/ha), a moderate load for a forest but a relatively high one for a grassland. Then, the intensity would be 20,000 by 1 by 1 which gives 20,000 kW/m. The units are kilowatts per metre. This is interpreted as the number of kilowatts being released from the fuel per metre of fire edge. Notice that this is not kW per square metre of ground nor kW per square metre of flame surface. A single bar radiator has a rating of 1 kW.

Most people no nothing of kW/m. How can we communicate the nature of fire variation? "Fires ain't fires." Can we invent our own Richter-type scale to assist communication:

- Let's start with a range of intensities having a maximum value of 100 kW/m, call it the lowest step on our scale of fire intensity, and give it a value of 1.
- For the next level, Richter would have multiplied the values by 10 but this is too sharp an increase for us; an intermediate step is to multiply by the square root of 10 and round off to give us 350 kW/m. This is a convenient number as it is near the upper limit of the prescribed burning range for forests. Cheney (1978) suggested that the "fire intensities recommended for optimum prescribed burning [in forests] range from 60 to 250 kW/m while the maximum intensity recommended ... is 500 kW/m". From 100 to 350 is Level 2 on our scale.
- Level 3 would be 350-1,000 kW/m.
- Level 4 would be 1,000 - 3,500 kW/m, the upper value being about the limit of fire control in eucalypt forests (Luke and McArthur suggest 4,000).
- Level 5 would be in the range 3,500 - 10,000 kW/m.
- Level 6 would cover 10,000 - 35,000 kW/m.
- Level 7 would be 35,000 - 100,000 kW/m.

A scale of some sort would be useful in talking with the public or the media. Maybe a Richter-type scale is the way to go in describing the maximum intensity of a fire event (given that there is a wide variety of fire intensity in any one fire). Or, maybe a scientific "Red Baron" will shoot the idea down in flames.

- Byram, G.M. (1959). Combustion of forest fuels. In: K.P.Davis (ed) Forest Fire: Control and Use. McGraw-Hill, New York.
- Cheney, N.P. (1978). Guidelines for fire management on forested watersheds, based on Australian experience. FAO Conservation Guide 4, FAO, Rome.
- Gill, A.M. and Moore, P.H.R. (1990). Fire intensities in Eucalyptus forests of southeastern Australia. Proceedings of the International Conference on Forest Fire Research, Coimbra, Portugal. Paper B24.
- L
- uke, R.H. and McArthur, A.G. (1978). Bushfires in Australia. AGPS, Canberra.

*Malcolm Gill
Division of Plant Industry, CSIRO
12 March 1998*