Deep learning surrogate models of JULES-INFERNO for wildfire prediction on a global scale
This work addresses the bottleneck of slow wildfire risk forecasting for climate scientists and policymakers by providing a computationally efficient alternative, though it is incremental as it applies existing deep learning techniques to a specific model.
The researchers tackled the high computational cost of the JULES-INFERNO global wildfire model by developing deep learning surrogate models that reduce prediction time from several hours to under 20 seconds for 30-year forecasts on a laptop CPU, while maintaining high accuracy with an average error per pixel under 0.3% and SSIM over 98%.
Global wildfire models play a crucial role in anticipating and responding to changing wildfire regimes. JULES-INFERNO is a global vegetation and fire model simulating wildfire emissions and area burnt on a global scale. However, because of the high data dimensionality and system complexity, JULES-INFERNO's computational costs make it challenging to apply to fire risk forecasting with unseen initial conditions. Typically, running JULES-INFERNO for 30 years of prediction will take several hours on High Performance Computing (HPC) clusters. To tackle this bottleneck, two data-driven models are built in this work based on Deep Learning techniques to surrogate the JULES-INFERNO model and speed up global wildfire forecasting. More precisely, these machine learning models take global temperature, vegetation density, soil moisture and previous forecasts as inputs to predict the subsequent global area burnt on an iterative basis. Average Error per Pixel (AEP) and Structural Similarity Index Measure (SSIM) are used as metrics to evaluate the performance of the proposed surrogate models. A fine tuning strategy is also proposed in this work to improve the algorithm performance for unseen scenarios. Numerical results show a strong performance of the proposed models, in terms of both computational efficiency (less than 20 seconds for 30 years of prediction on a laptop CPU) and prediction accuracy (with AEP under 0.3\% and SSIM over 98\% compared to the outputs of JULES-INFERNO).