Learning to generate physical ocean states: Towards hybrid climate modeling

Ocean General Circulation Models (OGCMs) require extensive computational resources to reach equilibrium states, while deep learning emulators, despite offering fast predictions, lack the physical interpretability and long-term stability necessary for climate scientists to understand climate sensitivity to greenhouse gas emissions and mechanisms of climate variability such as tipping points. We propose to take the best from both worlds by leveraging deep generative models to produce physically consistent oceanic states that can serve as initial conditions for climate projections.Training on ocean variables from idealized numerical simulations, we develop methods to physically constrain the generation of states and assess through both physical metrics and numerical experiments the viability of this hybrid approach, combining the computational efficiency of deep learning with the physical accuracy of numerical models.