A high-fidelity ensemble simulation framework for interrogating wildland-fire behaviour and benchmarking machine learning models

Background Wildfire research uses ensemble methods to analyse fire behaviours and assess uncertainties. Nonetheless, current research methods are either confined to simple models or complex simulations with limitations. Modern computing tools could allow for efficient, high-fidelity ensemble simulations. Aims This study proposes a high-fidelity ensemble wildfire simulation framework for studying wildfire behaviour, assessing fire risks, analysing uncertainties, and training machine learning (ML) models. Methods We present a simulation framework that integrates the Swirl-Fire large-eddy simulation tool for wildfire predictions with the Vizier optimisation platform for automated run-time management of ensemble simulations and large-scale batch processing. All simulations are executed on tensor-processing units to enhance computational efficiency. Key results A dataset of 117 simulations is created, each with 1.35 billion mesh points. The simulations are compared to existing experimental data and show good agreement in terms of fire rate of spread. Analysis is performed for fire acceleration, mean rate of spread, and fireline intensity. Conclusions Strong coupling between wind speed and slope is observed for fire-spread rate and intermittency. A critical Froude number that delineates fires from plume-dominated to wind-dominated is identified and confirmed with literature observations. Implications The ensemble simulation framework is efficient in facilitating large-scale parametric wildfire studies.