Acquisition of the Significant Wave Height From CFOSAT SWIM Spectra Through a Deep Neural Network and Its Impact on Wave Model Assimilation.

The wave numerical simulation accuracy can be improved by assimilating remotely sensed wave observations. In addition to the nadir, significant wave height (SWH), the Surface Waves Investigation and Monitoring (SWIM) onboard Chinese‐French Oceanic SATellite (CFOSAT) provides two additional columns of wave spectra observations within wavelengths from 70 to 500 m. A model based on a deep neural network (DNN) is developed to retrieve the total SWH from the partially wave spectra observed by SWIM. The DNN model uses the parameters from both the SWIM spectra and the nearest nadir as the inputs, and the DNN is trained on the SWH from cross‐matched altimeter observations. The DNN‐based acquisition of the SWH is verified to achieve a high accuracy. A set of assimilation experiments are performed based on MFWAM and show promising results. Compared to the assimilation of SWIM nadir SWHs only, the addition of the newly obtained SWIM SWH notably enhances the positive impacts of assimilation, not only proving the effectiveness and accuracy of the DNN model but also demonstrating the unique potential of SWIM in wave assimilation. Plain Language Summary: Assimilation is an effective method of improving the accuracy of wave models by using remotely sensed wave heights. The Surface Waves Investigation and Monitoring (SWIM) onboard Chinese‐French Oceanic SATellite (CFOSAT) is a new and unique instrument that can provide more wave observations than traditional altimeters. Therefore, SWIM has the potential to have positive impacts on wave forecasting. However, the additional observations from SWIM can cover only partial wave energy spectra within wavelengths of 70–500 m. Therefore, a model for retrieving the total wave height from partially observed wave spectra is critical before we can assimilate SWIM observations into wave models. A model for retrieving the total wave height is developed based on a deep neural network, a widely used technique in artificial intelligence. Assimilation experiments are carried out to test the effectiveness of the newly obtained total wave heights. Satisfactory results are found, indicating that the total wave heights from our model significantly improve the wave simulation accuracy. These findings also demonstrate the capability of SWIM as an effective wave monitoring instrument that can likely exert highly positive impacts on the accuracy of wave forecasts. Key Points: A deep neural network (DNN) model is developed for the acquisition of significant wave heights from Chinese‐French Oceanic SATellite (CFOSAT) Surface Waves Investigation and Monitoring (SWIM) spectraAdditional positive impacts are found in the assimilation of the Surface Waves Investigation and Monitoring (SWIM) box significant wave height (SWH) from the DNN compared to the assimilation of nadir SWH only [ABSTRACT FROM AUTHOR]