Towards fully differentiable neural ocean model with Veros

Etienne Meunier, Said Ouala, Hugo Frezat, Julien Le Sommer, Ronan Fablet

Keywords: Differentiable, Ocean, Simulation

TL;DR: Differentiable version of Veros model + applications to parameter calibration

Abstract: We present a differentiable extension of the VEROS ocean model \cite{veros}, enabling automatic differentiation through its dynamical core. We describe the key modifications required to make the model fully compatible with JAX’s autodifferentiation framework and evaluate the numerical consistency of the resulting implementation. Two illustrative applications are then demonstrated: (i) the correction of an initial ocean state through gradient-based optimization, and (ii) the calibration of unknown physical parameters directly from model observations. These examples highlight how differentiable programming can facilitate end-to-end learning and parameter tuning in ocean modeling. The source code for our implementation is available online.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 37