Machine Learning for Stochastic Parameterization: Generative Adversarial Networks in the Lorenz '96 Model

TL;DR

This paper introduces a novel stochastic parameterization method using GANs for the Lorenz '96 model, demonstrating improved performance over baseline methods in capturing system dynamics at weather and climate scales.

Contribution

It develops the first GAN-based stochastic parameterization for the Lorenz '96 model, advancing data-driven uncertainty modeling in sub-grid process simulations.

Findings

GAN models outperform baseline in reproducing Lorenz '96 dynamics

Models with better forecast skill also simulate climate regimes more accurately

GAN configurations effectively capture spatio-temporal correlations

Abstract

Stochastic parameterizations account for uncertainty in the representation of unresolved sub-grid processes by sampling from the distribution of possible sub-grid forcings. Some existing stochastic parameterizations utilize data-driven approaches to characterize uncertainty, but these approaches require significant structural assumptions that can limit their scalability. Machine learning models, including neural networks, are able to represent a wide range of distributions and build optimized mappings between a large number of inputs and sub-grid forcings. Recent research on machine learning parameterizations has focused only on deterministic parameterizations. In this study, we develop a stochastic parameterization using the generative adversarial network (GAN) machine learning framework. The GAN stochastic parameterization is trained and evaluated on output from the Lorenz '96 model, which is a common baseline model for evaluating both parameterization and data assimilation techniques. We evaluate different ways of characterizing the input noise for the model and perform model runs with the GAN parameterization at weather and climate timescales. Some of the GAN configurations perform better than a baseline bespoke parameterization at both timescales, and the networks closely reproduce the spatio-temporal correlations and regimes of the Lorenz '96 system. We also find that in general those models which produce skillful forecasts are also associated with the best climate simulations.