Since April 2023, CIMA Research Foundation has been collaborating with the Basque Centre for Applied Mathematics (BCAM) in the study and development of technologies for analyzing and forecasting wildfires.
The collaboration with BCAM, lasting a year and a half, focuses on three main objectives. The first objective is the development of an Ensemble Forecasting procedure in fire simulation tools, with a particular emphasis on those based on specific mathematical algorithms known as Cellular Automata (CA)1.
The second objective involves studying the different parameterizations in the models of the wildfire spotting phenomenon, which occurs when secondary fires ignite in surrounding areas due to embers carried by wind and hot air columns from the main fire.
Finally, the third objective aims to integrate some of the investigated procedures and parameterizations into PROPAGATOR, CIMA Research Foundation’s simulation tool for wildfire dynamics. This will enable testing its applicability in operational scenarios.
Goals and expected results
- Feasibility analysis of Ensemble Forecasting for wildfire propagation simulators based on Cellular Automata.
- Development and analysis of physics-based parameterizations, to be eventually integrated into the PROPAGATOR code.
- Simulation of relevant case studies (wildfires where “Fire Spotting” played a key role in the burned area development) to verify that the introduced modules align with the real wildfire dynamics.
- Comparative study of three spotting parameters integrated into PROPAGATOR, to build a reference framework for comparing spotting models used in operational scenarios.
CIMA Research Foundation’s contribution
Within the collaboration, CIMA Research Foundation’s role is to facilitate the use of PROPAGATOR, adapting it for Ensemble Forecasting techniques and compiling the results in a technical report. Additionally, we collaborate and provide technical support for integrating spotting study systems into PROPAGATOR. Lastly, we contribute by conducting case study simulations and comparing model performance with actual observed data to verify reliability.
The joint research with BCAM has led to scientific publications such as
Fire-spotting modelling in operational wildfire simulators based on Cellular Automata: A comparison study, published in the prestigious journal “Agricultural and Forest Meteorology.”
- A Cellular Automaton (CA) for wildfire propagation is a mathematical model that simulates how fire spreads across a landscape. It works like a grid of cells, each representing a part of the terrain (such as a patch of forest, a meadow, or an already burned area).
The key idea is that each cell follows simple rules to determine whether it will catch fire, remain intact, or extinguish, based on the state of neighboring cells and other factors, such as wind and humidity. These rules are applied repeatedly, updating the grid step by step, and over time, complex patterns emerge that mimic the real behavior of fire.
CA models are used because they allow researchers and risk management professionals to quickly and effectively simulate phenomena like wildfire spread, helping them predict the evolution of a fire and make informed decisions. ↩︎
