NeuralOGCM: Differentiable Ocean Modeling with Learnable Physics

TL;DR

NeuralOGCM introduces a differentiable ocean modeling framework combining physics-based solvers with deep learning to enhance simulation speed, stability, and physical accuracy through end-to-end training.

Contribution

It presents a novel differentiable ocean model that integrates learnable physical parameters and neural corrections, enabling autonomous optimization and improved simulation fidelity.

Findings

Outperforms traditional models in speed and accuracy.

Maintains long-term stability and physical consistency.

Effectively captures subgrid-scale processes.

Abstract

High-precision scientific simulation faces a long-standing trade-off between computational efficiency and physical fidelity. To address this challenge, we propose NeuralOGCM, an ocean modeling framework that fuses differentiable programming with deep learning. At the core of NeuralOGCM is a fully differentiable dynamical solver, which leverages physics knowledge as its core inductive bias. The learnable physics integration captures large-scale, deterministic physical evolution, and transforms key physical parameters (e.g., diffusion coefficients) into learnable parameters, enabling the model to autonomously optimize its physical core via end-to-end training. Concurrently, a deep neural network learns to correct for subgrid-scale processes and discretization errors not captured by the physics model. Both components work in synergy, with their outputs integrated by a unified ODE solver. Experiments demonstrate that NeuralOGCM maintains long-term stability and physical consistency, significantly outperforming traditional numerical models in speed and pure AI baselines in accuracy. Our work paves a new path for building fast, stable, and physically-plausible models for scientific computing.