Spatially Extended Tests of a Neural Network Parametrization Trained by Coarse-graining

Abstract

General circulation models (GCMs) typically have a grid size of 25--200 km. Parametrizations are used to represent diabatic processes such as radiative transfer and cloud microphysics and account for sub-grid-scale motions and variability. Unlike traditional approaches, neural networks (NNs) can readily exploit recent observational datasets and global cloud-system resolving model (CRM) simulations to learn subgrid variability. This article describes an NN parametrization trained by coarse-graining a near-global CRM simulation with a 4~km horizontal grid spacing. The NN predicts the residual heating and moistening averaged over (160 km)^2 grid boxes as a function of the coarse-resolution fields within the same atmospheric column. This NN is coupled to the dynamical core of a GCM with the same 160 km resolution. A recent study described how to train such an NN to be numerically stable when coupled to specified time-evolving advective forcings in a single column model, but feedbacks between NN and GCM components cause spatially-extended simulations to crash within a few days. Analyzing the linearized response of such an NN reveals that it learns to exploit a strong synchrony between precipitation and the atmospheric state above 10 km. Removing these variables from the NN's inputs stabilizes the coupled simulations, which predict the future state more accurately than a coarse-resolution simulation without any parametrizations of sub-grid-scale variability, although the mean state slowly drifts.

References in corpus (4)

• Adam: A Method for Stochastic Optimization
• Deep learning to represent sub-grid processes in climate models
• Earth System Modeling 2.0: A Blueprint for Models That Learn From Observations and Targeted High-Resolution Simulations
• Using machine learning to parameterize moist convection: potential for modeling of climate, climate change and extreme events

Cited by in corpus (15)

• WeatherBench: A benchmark dataset for data-driven weather forecasting
• Neural General Circulation Models for Weather and Climate
• Physically Interpretable Neural Networks for the Geosciences: Applications to Earth System Variability
• Use of neural networks for stable, accurate and physically consistent parameterization of subgrid atmospheric processes with good performance at reduced precision
• Deep Learning Based Cloud Cover Parameterization for ICON
• Machine learning emulation of gravity wave drag in numerical weather forecasting
• Combining distribution-based neural networks to predict weather forecast probabilities
• Data-driven super-parameterization using deep learning: Experimentation with multi-scale Lorenz 96 systems and transfer-learning
• Machine Learning Emulation of 3D Cloud Radiative Effects
• Non-local parameterization of atmospheric subgrid processes with neural networks
• Causally-informed deep learning to improve climate models and projections
• Discovery of interpretable structural model errors by combining Bayesian sparse regression and data assimilation: A chaotic Kuramoto-Sivashinsky test case
• Interpretable multiscale Machine Learning-Based Parameterizations of Convection for ICON
• Subgrid-scale parametrization of unresolved scales in forced Burgers equation using Generative Adversarial Networks (GAN)
• Reduced Data-Driven Turbulence Closure for Capturing Long-Term Statistics