TY - JOUR
T1 - Integration of a Deep-Learning-Based Fire Model Into a Global Land Surface Model
AU - Son, Rackhun
AU - Stacke, Tobias
AU - Gayler, Veronika
AU - Nabel, Julia E. M. S.
AU - Schnur, Reiner
AU - Alonso, Lazaro
AU - Requena-Mesa, Christian
AU - Winkler, Alexander J.
AU - Hantson, Stijn
AU - Zaehle, Sönke
AU - Weber, Ulrich
AU - Carvalhais, Nuno
N1 - Funding Information:
This project has received funding from the European Union's H2020 research and innovation programme under grant agreement N.101003536 (ESM2025 ‐ Earth System Models for the Future). We should secondly acknowledge SeasFire (Lazaro), DeepCube (Christian) and USMILE (Alex). Stijn acknowledges support from the Max Planck Tandem group program. Also, we appreciate valuable comments for elevating the quality of the manuscript from two anonymous reviewers. This work used resources of the Deutsches Klimarechenzentrum (DKRZ) and the Max Planck Gesellschaft (MPG) under project ID mj0143. Further, data sets provided by the Max Planck Institute for Meteorology (MPI‐M) via the DKRZ data pool were used. Open Access funding enabled and organized by Projekt DEAL.
Funding Information:
This project has received funding from the European Union's H2020 research and innovation programme under grant agreement N.101003536 (ESM2025 - Earth System Models for the Future). We should secondly acknowledge SeasFire (Lazaro), DeepCube (Christian) and USMILE (Alex). Stijn acknowledges support from the Max Planck Tandem group program. Also, we appreciate valuable comments for elevating the quality of the manuscript from two anonymous reviewers. This work used resources of the Deutsches Klimarechenzentrum (DKRZ) and the Max Planck Gesellschaft (MPG) under project ID mj0143. Further, data sets provided by the Max Planck Institute for Meteorology (MPI-M) via the DKRZ data pool were used. Open Access funding enabled and organized by Projekt DEAL.
Publisher Copyright:
© 2024 The Authors. Journal of Advances in Modeling Earth Systems published by Wiley Periodicals LLC on behalf of American Geophysical Union.
PY - 2024/1
Y1 - 2024/1
N2 - 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.
AB - 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.
KW - deep learning
KW - DGVM
KW - fire
KW - hybrid modeling
UR - http://www.scopus.com/inward/record.url?scp=85170372294&partnerID=8YFLogxK
U2 - 10.1029/2023MS003710
DO - 10.1029/2023MS003710
M3 - Article
AN - SCOPUS:85170372294
SN - 1942-2466
VL - 16
JO - Journal of Advances in Modeling Earth Systems
JF - Journal of Advances in Modeling Earth Systems
IS - 1
M1 - e2023MS003710
ER -