Deep learning to represent sub-grid processes in climate models

2018/06/30 by Stephan Rasp, Michael S. Pritchard, Pierre Gentine · 76 citations
Computer Science · Earth and Planetary Sciences · Environmental Science · Mathematics · Physics and Astronomy · #Artificial intelligence #Artificial neural network #Atmospheric and Environmental Gas Dynamics #Climate change #Climate model #Climate variability and models #Climatology #Cloud computing #Computer science #Context (archaeology) #Deep learning #Forcing (mathematics) #Geography #Geology #Grid #Meteorological Phenomena and Simulations #Meteorology #cs.LG #physics.ao-ph #stat.ML

paper · pdf · doi:10.1073/pnas.1810286115

published in Proceedings of the National Academy of Sciences 115(39), 9684-9689 (National Academy of Sciences) · View official PNAS version at https://doi.org/10.1073/pnas.1810286115

openalex publication_date 2018/09/06 · arxiv created 2018/09/07 · arxiv updated 2022/06/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05

Abstract

The representation of nonlinear sub-grid processes, especially clouds, has been a major source of uncertainty in climate models for decades. Cloud-resolving models better represent many of these processes and can now be run globally but only for short-term simulations of at most a few years because of computational limitations. Here we demonstrate that deep learning can be used to capture many advantages of cloud-resolving modeling at a fraction of the computational cost. We train a deep neural network to represent all atmospheric sub-grid processes in a climate model by learning from a multi-scale model in which convection is treated explicitly. The trained neural network then replaces the traditional sub-grid parameterizations in a global general circulation model in which it freely interacts with the resolved dynamics and the surface-flux scheme. The prognostic multi-year simulations are stable and closely reproduce not only the mean climate of the cloud-resolving simulation but also key aspects of variability, including precipitation extremes and the equatorial wave spectrum. Furthermore, the neural network approximately conserves energy despite not being explicitly instructed to. Finally, we show that the neural network parameterization generalizes to new surface forcing patterns but struggles to cope with temperatures far outside its training manifold. Our results show the feasibility of using deep learning for climate model parameterization. In a broader context, we anticipate that data-driven Earth System Model development could play a key role in reducing climate prediction uncertainty in the coming decade.

Cited by

• Explaining Weather Bulletins via ILP
• AI-Integrated Scientific Inquiry: A Practice-Centered Vision for Science Education
• Subgrid-Scale Parameterization in Burgers' Equation Using Structure-Preserving Neural Networks and Entropy Variables
• Numerical models outperform AI weather forecasts of record-breaking extremes
• A Diffusion-Based Framework for High-Resolution Precipitation Forecasting over CONUS
• Retrofitting Earth System Models with Cadence-Limited Neural Operator Updates
• Epistemic and Aleatoric Uncertainty Quantification in Weather and Climate Models
• Crowdsourcing the Frontier: Advancing Hybrid Physics-ML Climate Simulation via a 50,000 Kaggle Competition
• Improvement of a neural network convection scheme by including triggering and evaluation in present and future climates
• Applications of Machine Learning and Artificial Intelligence in Tropospheric Ozone Research
• A Framework for Hybrid Physics-AI Coupled Ocean Models
• Bridging Idealized and Operational Models: An Explainable AI Framework for Earth System Emulators
• Geo-Aware Models for Stream Temperature Prediction across Different Spatial Regions and Scales
• Beyond the Training Data: Confidence-Guided Mixing of Parameterizations in a Hybrid AI-Climate Model
• Machine Learning Workflows in Climate Modeling: Design Patterns and Insights from Case Studies
• Climate-Adaptive and Cascade-Constrained Machine Learning Prediction for Sea Surface Height under Greenhouse Warming
• Breaking the Statistical Similarity Trap in Extreme Convection Detection
• GEN2: A Generative Prediction-Correction Framework for Long-time Emulations of Spatially-Resolved Climate Extremes
• FedRAIN-Lite: Federated Reinforcement Algorithms for Improving Idealised Numerical Weather and Climate Models
• CERA: A Framework for Improved Generalization of Machine Learning Models to Changed Climates
• On Learning Closed-Loop Probabilistic Multi-Agent Simulator
• NeurIPS 2024 ML4CFD Competition: Results and Retrospective Analysis
• Resolution Revolution: A Physics-Guided Deep Learning Framework for Spatiotemporal Temperature Reconstruction
• Modeling Partially Observed Nonlinear Dynamical Systems and Efficient Data Assimilation via Discrete-Time Conditional Gaussian Koopman Network
• Stochastic and Non-local Closure Modeling for Nonlinear Dynamical Systems via Latent Score-based Generative Models
• Stabilizing PDE--ML coupled systems
• Artificial Intelligence for Atmospheric Sciences: A Research Roadmap
• Towards NoahMP-AI: Enhancing Land Surface Model Prediction with Deep Learning
• Reduced Cloud Cover Errors in a Hybrid AI-Climate Model Through Equation Discovery And Automatic Tuning
• Supporting renewable energy planning and operation with data-driven high-resolution ensemble weather forecast
• From Proxies to Fields: Spatiotemporal Reconstruction of Global Radiation from Sparse Sensor Sequences
• FESOM2-JAX v1.0: a differentiable shadow of the ocean-sea-ice model FESOM2, cast onto GPUs
• Quantum mechanical closure of partial differential equations with symmetries
• Climate Science and Control Engineering: Insights, Parallels, and Connections
• Mjölnir: A Deep Learning Parametrization Framework for Global Lightning Flash Density
• Fourier analysis of the physics of transfer learning for data-driven subgrid-scale models of ocean turbulence
• A Lightweight Neural Network for Accelerating Radiative Transfer Modeling in WRF
• Generative emulation of chaotic dynamics with coherent prior
• Learning data-driven discretizations for partial differential equations. [europepmc]
• Applying Machine Learning to Improve Simulations of a Chaotic Dynamical System Using Empirical Error Correction. [europepmc]
• A machine learning based prediction system for the Indian Ocean Dipole. [europepmc]
• Predicting clustered weather patterns: A test case for applications of convolutional neural networks to spatio-temporal climate data. [europepmc]
• Stable machine-learning parameterization of subgrid processes for climate modeling at a range of resolutions. [europepmc]
• Adversarial super-resolution of climatological wind and solar data. [europepmc]
• Analog Forecasting of Extreme-Causing Weather Patterns Using Deep Learning. [europepmc]
• Confronting the Challenge of Modeling Cloud and Precipitation Microphysics. [europepmc]
• Combining data assimilation and machine learning to infer unresolved scale parametrization. [europepmc]
• Climbing down Charney's ladder: machine learning and the post-Dennard era of computational climate science. [europepmc]
• Deep learning for bias correction of MJO prediction. [europepmc]
• Machine Learning Emulation of Gravity Wave Drag in Numerical Weather Forecasting. [europepmc]
• The AI gambit: leveraging artificial intelligence to combat climate change-opportunities, challenges, and recommendations. [europepmc]
• Data-driven prediction in dynamical systems: recent developments. [europepmc]
• Sensitivity of Air Pollution Exposure and Disease Burden to Emission Changes in China Using Machine Learning Emulation. [europepmc]
• Variability of ENSO Forecast Skill in 2-Year Global Reforecasts Over the 20th Century. [europepmc]
• Physics-informed deep-learning parameterization of ocean vertical mixing improves climate simulations. [europepmc]
• Stability analysis of chaotic systems from data. [europepmc]
• Large anomalies in future extreme precipitation sensitivity driven by atmospheric dynamics. [europepmc]
• Machine learning and the quest for objectivity in climate model parameterization. [europepmc]
• Knowledge-guided machine learning can improve carbon cycle quantification in agroecosystems. [europepmc]
• A numerical compass for experiment design in chemical kinetics and molecular property estimation. [europepmc]
• Deep learning for water quality. [europepmc]
• Neural general circulation models for weather and climate. [europepmc]
• Foundation models and intelligent decision-making: Progress, challenges, and perspectives. [europepmc]

Related

• Causally‐Informed Deep Learning to Improve Climate Models and Projections
• Finetuning AI Foundation Models to Develop Subgrid-Scale Parameterizations: A Case Study on Atmospheric Gravity Waves
• Machine-learned climate model corrections from a global storm-resolving model
• CondensNet: Enabling stable long-term climate simulations via hybrid deep learning models with adaptive physical constraints
• Neural General Circulation Models for Weather and Climate