Machine Learning Global Simulation of Nonlocal Gravity Wave Propagation

TL;DR

This paper introduces a novel machine learning approach using Attention U-Net to simulate atmospheric gravity wave fluxes globally, addressing limitations of traditional local parameterizations in climate models.

Contribution

It presents the first global simulation of gravity wave fluxes with ML trained on the WINDSET dataset, emphasizing the importance of nonlocal, data-driven schemes.

Findings

ML model captures global gravity wave fluxes effectively

Global nonlocality improves simulation accuracy

Demonstrates potential to replace traditional parameterizations

Abstract

Global climate models typically operate at a grid resolution of hundreds of kilometers and fail to resolve atmospheric mesoscale processes, e.g., clouds, precipitation, and gravity waves (GWs). Model representation of these processes and their sources is essential to the global circulation and planetary energy budget, but subgrid scale contributions from these processes are often only approximately represented in models using parameterizations. These parameterizations are subject to approximations and idealizations, which limit their capability and accuracy. The most drastic of these approximations is the "single-column approximation" which completely neglects the horizontal evolution of these processes, resulting in key biases in current climate models. With a focus on atmospheric GWs, we present the first-ever global simulation of atmospheric GW fluxes using machine learning (ML) models trained on the WINDSET dataset to emulate global GW emulation in the atmosphere, as an alternative to traditional single-column parameterizations. Using an Attention U-Net-based architecture trained on globally resolved GW momentum fluxes, we illustrate the importance and effectiveness of global nonlocality, when simulating GWs using data-driven schemes.