Forest (or grassland) fires are naturally ignited by lightning, or by sparks from another fire in the vicinity. The likelihood that a cloud-to-ground lightning strike actually sets off a fire depends on the amount of forest fuel available, and the dryness of this fuel (i.e. the rain history and current soil moisture). This likelihood, or lightning ignition efficiency, varies slowly from day to day. An example (for the western USA on 24 June 1999) is shown in Fig 1.
Once a fire has been ignited, the Haines Index can be used as a measure of the potential of fire growth. This index is based on low-level stability and is calculated by adding a temperature term and a moisture term (1). Values from 1 to 3 are assigned the temperature term depending on the magnitude of the low-level static stability. The moisture term also receives values from 1 to 3, depending on the dryness of the boundary layer. The resultant Haines Index varies from 2 to 6. A value of 2 indicates moist, stable air while a 6 indicates dry, unstable air. The potential for large fire growth or extreme fire behavior is very low when the index is 2, but high when the index is 6. The Haines index has proven to be a useful predictor for fire occurrence and severity in Idaho (2) and Arizona (1). The Haines index for 24 June 1999 is shown in Fig 2.
The Haines index needs to be combined with forecast surface wind speeds, humidity, rainfall and temperature to assess the forest fire danger. These three atmospheric indices, plus a drought factor (which depends on the rain history), have been combined into the Forest Fire Danger Index (FFDI). This index was empirically derived in Australia.
The outcome is a fire danger classification. This variable, ranging from low to extreme (Fig 3) is widely disseminated, in national forests for instance.