DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Causally-Informed Deep Learning to Improve Climate Models and Projections

Iglesias-Suarez, Fernando and Gentine, Pierre and Solino-Fernandez, Breixo and Beucler, Tom and Pritchard, Michael and Runge, Jakob and Eyring, Veronika (2024) Causally-Informed Deep Learning to Improve Climate Models and Projections. Journal of Geophysical Research: Atmospheres, 129 (4), pp. 1-16. Wiley. doi: 10.1029/2023JD039202. ISSN 2169-897X.

[img] PDF - Published version
[img] PDF - Published version

Official URL: https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JD039202


Climate models are essential to understand and project climate change, yet long-standing biases and uncertainties in their projections remain. This is largely associated with the representation of subgrid-scale processes, particularly clouds and convection. Deep learning can learn these subgrid-scale processes from computationally expensive storm-resolving models while retaining many features at a fraction of computational cost. Yet, climate simulations with embedded neural network parameterizations are still challenging and highly depend on the deep learning solution. This is likely associated with spurious non-physical correlations learned by the neural networks due to the complexity of the physical dynamical system. Here, we show that the combination of causality with deep learning helps removing spurious correlations and optimizing the neural network algorithm. To resolve this, we apply a causal discovery method to unveil causal drivers in the set of input predictors of atmospheric subgrid-scale processes of a superparameterized climate model in which deep convection is explicitly resolved. The resulting causally-informed neural networks are coupled to the climate model, hence, replacing the superparameterization and radiation scheme. We show that the climate simulations with causally-informed neural network parameterizations retain many convection-related properties and accurately generate the climate of the original high-resolution climate model, while retaining similar generalization capabilities to unseen climates compared to the non-causal approach. The combination of causal discovery and deep learning is a new and promising approach that leads to stable and more trustworthy climate simulations and paves the way toward more physically-based causal deep learning approaches also in other scientific disciplines.

Item URL in elib:https://elib.dlr.de/202881/
Document Type:Article
Title:Causally-Informed Deep Learning to Improve Climate Models and Projections
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Iglesias-Suarez, FernandoDLR, IPAhttps://orcid.org/0000-0003-3403-8245UNSPECIFIED
Gentine, PierreColumbia University, New York, NY, USAhttps://orcid.org/0000-0002-0845-8345UNSPECIFIED
Beucler, TomUniversity of Lausanne, Lausanne, Switzerlandhttps://orcid.org/0000-0002-5731-1040UNSPECIFIED
Pritchard, MichaelUniversity of California, Irvine, CA, USAhttps://orcid.org/0000-0002-0340-6327UNSPECIFIED
Eyring, VeronikaDLR, IPAhttps://orcid.org/0000-0002-6887-4885UNSPECIFIED
Date:19 February 2024
Journal or Publication Title:Journal of Geophysical Research: Atmospheres
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:Yes
Page Range:pp. 1-16
Keywords:climate modeling; causality; deep learning
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Earth Observation
DLR - Research area:Raumfahrt
DLR - Program:R EO - Earth Observation
DLR - Research theme (Project):R - Atmospheric and climate research
Location: Oberpfaffenhofen
Institutes and Institutions:Institute of Atmospheric Physics > Earth System Model Evaluation and Analysis
Deposited By: Iglesias-Suarez, Dr. Fernando
Deposited On:20 Feb 2024 13:02
Last Modified:20 Feb 2024 13:02

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.