A CELLULAR-AUTOMATON MODEL OF WILDFIRE PROPAGATION AND EXTINCTION

We propose a new model to predict the spatial and temporal behavior of wildfires. Fire spread and intensity were simulated using a cellular automaton model. Monte Carlo techniques were used to provide fire risk probabilities for areas where fuel loadings and topography are known. The model assumes predetermined or measurable environmental variables such as wind direction and magnitude, relative humidity, fuel moisture content, and air temperature. Implementation of the model allows the linking of fire monitoring using remotely sensed data, potentially in real time, to rapid simulations of predicted fire behavior. Calibration of the model is based on thermal infrared remotely sensed imagery of a test burn during 1986 in the San Dimas experimental forest. The model and its various implementations show distinct promise for real-time fire management and fire risk planning.