Fourteenth International Conference on Climate Change: Impacts & Responses

Abstract

The knowledge of a free-burning fire’s potential rate of spread is critical for safe and effective wildfire control, in this work We are carrying out a data driven cellular automata simulation of wildfire propagation. Our model performs not only for the approximation of fire propagation, but also helps in finding critical zones in which fire would be most devastating and would spread to a large scale. We also demonstrate how we extract useful data from satellite imagery and use machine learning techniques in order to prepare inputs used later for simulation. We aim to locate critical zones and to estimate fire spread so that needed precautions can be taken to limit and control any eventual risks in real life, this can also help managers to rapidly predict the spread of fires, providing a decision basis for formulating effective fire extinguishing plans.