Towards fully differentiable neural ocean model with Veros
By: Etienne Meunier , Said Ouala , Hugo Frezat and more
Potential Business Impact:
Helps predict ocean changes by learning from data.
We present a differentiable extension of the VEROS ocean model, enabling automatic differentiation through its dynamical core. We describe the key modifications required to make the model fully compatible with JAX 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. Our implementation is available online.
Similar Papers
Differentiable Physics-Neural Models enable Learning of Non-Markovian Closures for Accelerated Coarse-Grained Physics Simulations
Computational Physics
Speeds up complex science predictions by 1000x.
Case study of a differentiable heterogeneous multiphysics solver for a nuclear fusion application
Computational Physics
Makes fusion energy research faster and easier.
Ensembles of Neural Surrogates for Parametric Sensitivity in Ocean Modeling
Atmospheric and Oceanic Physics
Makes ocean computer models more trustworthy.