A posteriori learning of quasi-geostrophic turbulence parametrization: an experiment on integration steps

TL;DR

This paper demonstrates that jointly learning a subgrid-scale turbulence model with the dynamical solver using an  posterioriased loss function results in stable, realistic simulations of quasi-geostrophic turbulence, addressing challenges in 2D flow modeling.

Contribution

It introduces a novel approach of a posteriori learning for turbulence parametrization that enhances stability and realism in 2D flow simulations.

Findings

Stable and realistic quasi-geostrophic turbulence simulations achieved.

Joint learning with the dynamical solver improves model performance.

Addresses challenges of backscatter in 2D flow modeling.

Abstract

Modeling the subgrid-scale dynamics of reduced models is a long standing open problem that finds application in ocean, atmosphere and climate predictions where direct numerical simulation (DNS) is impossible. While neural networks (NNs) have already been applied to a range of three-dimensional flows with success, two dimensional flows are more challenging because of the backscatter of energy from small to large scales. We show that learning a model jointly with the dynamical solver and a meaningful \textit{a posteriori}-based loss function lead to stable and realistic simulations when applied to quasi-geostrophic turbulence.