Correction of Air-Sea Heat Fluxes in the NEMO Ocean General Circulation Model Using Neural Networks

Abstract. The atmospheric forcing and the heat exchanges between the ocean and the atmosphere represent one of the major sources of uncertainty for numerical ocean reconstructions and predictions. Air-sea heat fluxes may suffer from inaccuracies in meteorological fields, sea surface variables, and bulk formulations, which have a strongly non-linear dependence on the ocean state. Here, state-dependent errors of the heat fluxes are learned by artificial neural networks (ANN) from a dataset of heat flux correction terms, derived in turn from previous sea surface temperature nudging experiments. The pre-trained model predictors include stationary fields, atmospheric forcing data, ocean state, and stratification indices. Variable importance scores emphasize the dependence of the air-sea heat flux errors on the wind forcing. The pre-trained model of heat flux correction is then used to adaptively correct the fluxes online, in a series of global ocean experiments performed with the NEMO (Nucleus for European Modelling of the Ocean) ocean general circulation model, augmented with ANN inference capabilities. Results indicate the positive impact of the correction procedure, beyond the training period, e.g., in independent observation-poor and -rich periods, leading to the same dynamical and subsurface signature as in nudging experiments. Prediction experiments also indicate the method's potential for operational forecast applications. The method may also be adopted in coupled long-term reanalyses, long-range predictions, and projections.