Work

Modelling wildfire propagation using the stochastic level-set method

Year: 2025

Type: article

Abstract: The modelling of wildfire spread involves managing uncertainties from various sources, typically addressed through probabilistic rather than deterministic approaches. Natural random variability in the dynamics of the fire front can be directly captured by integrating stochastic noise into the fire front tracking model. To achieve this, a Gaussian noise is introduced into the level-set model using a stochastic level-set approach in the wildfire spread simulator “ spark ”. The incorporation of stochasticity into the fire spread model enables the simulation to capture variability in fire front growth. The model is calibrated and validated by comparing the stochastic fire spread simulations to observed fire data. The developed model is also compared to a specific case of fire spread simulation created using the Wildland-Urban Fire Dynamic Simulator ( wfds ). This comparative analysis provides a conclusive evaluation, highlighting the performance and capabilities of the stochastic approach in capturing the uncertainties and complexities of fire behavior.

Source: Communications in Nonlinear Science and Numerical Simulation

DOI: 10.1016/j.cnsns.2025.109305

Authors: Sahar Masoudian, Jason Sharples, Duncan Sutherland, Zlatko Jovanoski, James Hilton

Institution: UNSW Canberra

Language: English

Topic: Fire effects on ecosystems

Subfield: Global and Planetary Change

Field: Environmental Science

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $3,710

Funder: University of New South Wales Canberra at Australian Defence Force Academy