Learning to generate physical ocean states: Towards hybrid climate modeling
This addresses the computational and interpretability challenges in climate modeling for climate scientists, though it appears incremental as it builds on existing hybrid approaches.
The paper tackles the problem of ocean general circulation models requiring extensive computational resources and deep learning emulators lacking physical interpretability by proposing a hybrid approach using deep generative models to produce physically consistent oceanic states as initial conditions for climate projections. The approach combines computational efficiency with physical accuracy to reduce computational burden and minimize drifts in baseline simulations.
Ocean General Circulation Models 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 abrupt % 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. We assess the viability of this hybrid approach through both physical metrics and numerical experiments, and highlight the benefits of enforcing physical constraints during generation. Although we train here on ocean variables from idealized numerical simulations, we claim that this hybrid approach, combining the computational efficiency of deep learning with the physical accuracy of numerical models, can effectively reduce the computational burden of running climate models to equilibrium, and reduce uncertainties in climate projections by minimizing drifts in baseline simulations.