A Dynamic Forest Fire Spread Model Based on Cellular Automata Principles

The existing forest fire spread models based on cellular automata (CA) use fixed time steps to simulate the process of forest fire spread, which limits the model’s ability to reflect the actual characteristics of fire development. Furthermore, ignoring the significant impact of the firebreak on fire leads to poor accuracy and applicability. To address these issues, this paper proposes an algorithm for dynamic time step adjustment and establishes a model of fire crossing the firebreak. Firstly, the algorithm calculates the speed of fire spread based on the environmental factors that affect forest fire spread. Then, it dynamically adjusts the time step according to the speed of fire spread, which better aligns with the actual development trend of fire. Secondly, the model of fire crossing the firebreak is established by first calculating the probability of the entire firebreak being crossed using an empirical model, followed by dividing the firebreak into a number of cells based on the principle of CA and then calculating the probability of each cell being crossed at the current time step based on the size of the cell and the dynamic time step. This paper conducted numerous experiments using the controlled variable method to discuss the effectiveness of the model and the impact of different factors on fire spread. The results show that the proposed algorithm based on dynamic time step adjustment is effective, and vegetation type, wind, and terrain are critical factors in determining the spread of forest fires. The experimental results also demonstrate that firebreaks can effectively slow the spread of fire, and the time and location of the firebreak being crossed significantly impact the spread of fire. The model proposed in this paper has significant practical significance for analyzing the scope of forest fires, designing firebreaks, formulating firefighting strategies, and making emergency decisions.