Generalizable neural-network parameterization of mesoscale eddies in idealized and global ocean models

TL;DR

This paper introduces a physics-informed neural network parameterization for mesoscale eddies that generalizes across different ocean model resolutions and configurations, improving energy representation in simulations.

Contribution

It proposes a dimensional scaling approach to enforce physics constraints, enhancing neural network generalization in ocean modeling.

Findings

Improved energy representation in ocean simulations.

Enhanced generalization across grid resolutions and depths.

Better performance compared to baseline parameterizations.

Abstract

Data-driven methods have become popular to parameterize the effects of mesoscale eddies in ocean models. However, they perform poorly in generalization tasks and may require retuning if the grid resolution or ocean configuration changes. We address the generalization problem by enforcing physics constraints on a neural network parameterization of mesoscale eddy fluxes. We found that the local scaling of input and output features helps to generalize to unseen grid resolutions and depths offline in the global ocean. The scaling is based on dimensional analysis and incorporates grid spacing as a length scale. We formulate our findings as a general algorithm that can be used to enforce data-driven parameterizations with dimensional scaling. The new parameterization improves the representation of kinetic and potential energy in online simulations with idealized and global ocean models. Comparison to baseline parameterizations and impact on global ocean biases are discussed.