Integration of a Deep-Learning-Based Fire Model Into a Global Land Surface Model

Fire is a crucial factor in terrestrial ecosystems playing a role in disturbance for vegetation dynamics. Process-based fire models quantify fire disturbance effects in stand-alone dynamic global vegetation models (DGVMs) and their advances have incorporated both descriptions of natural processes and anthropogenic drivers. Nevertheless, these models show limited skill in modeling fire events at the global scale, due to stochastic characteristics of fire occurrence and behavior as well as the limits in empirical parameterizations in process-based models. As an alternative, machine learning has shown the capability of providing robust diagnostics of fire regimes. Here, we develop a deep-learning-based fire model (DL-fire) to estimate daily burnt area fraction at the global scale and couple it within JSBACH4, the land surface model used in the ICON-ESM. The stand-alone DL-fire model forced with meteorological, terrestrial and socio-economic variables is able to simulate global total burnt area, showing 0.8 of monthly correlation (rm) with GFED4 during the evaluation period (2011-2015). The performance remains similar with the hybrid modeling approach JSB4-DL-fire (rm = 0.79) outperforming the currently used uncalibrated standard fire model in JSBACH4 (rm = -0.07). We further quantify the importance of each predictor by applying layer-wise relevance propagation (LRP). Overall, land properties, such as fuel amount and water content in soil layers, stand out as the major factors determining burnt fraction in DL-fire, paralleled by meteorological conditions over tropical and high latitude regions. Our study demonstrates the potential of hybrid modeling in advancing fire prediction in ESMs by integrating deep learning approaches in physics-based dynamical models. We develop a fire-vegetation model based on a hybrid approach integrating artificial intelligence (AI) techniques into physics-based models. Given the weather conditions, vegetation states, and human factors, our model estimates daily burned area fraction. The spatiotemporal variations in burned area are closely reproduced, especially over fire-prone regions, such as Africa, South America, and Australia. Our model is able to represent regional variations in the drivers of fire occurrence, showing different importance of input predictors for different regions. This approach shows the possibilities of using deep learning (DL) models to provide in-depth fire predictions in Earth system models. Deep neural networks (DNN) can accurately predict global burnt area fraction on a daily scaleIntegration of the DNN into a physics-based land model significantly improves the estimation of biomass burnt damage in vegetation dynamicsThe DNN accounts for regional fire variations by assigning varying degrees of importance to each predictor