Nonlinear Ocean Wave Forecast via Data Assimilation Based on Dynamic Averaging Scheme

Phase-resolved reconstruction and forecasting of the ocean wave field are essential for ship operations and route optimization. Currently, X-band marine radar is widely used to detect statistical characteristics of the wave field, such as significant wave height and peak wave period. However, due to radar shadow effects, reconstructing and predicting the surface elevation around the ship's position is challenging. To address this, a dynamic averaging and evolution algorithm based on real-time radar images has been developed for the deterministic prediction of waves near the ship's position, which is within the blind zone. In this dynamic averaging scheme, the shadowed radar images are used as assimilation data to enhance the numerical forecasting results and improve wave prediction accuracy. The high-order spectral (HOS) model is employed as an evolution model for nonlinear waves. The combined dynamic averaging and evolution algorithm, along with the HOS model, has been tested in various scenarios with different ship speeds and sea states. Results indicate that the proposed approach reliably forecasts waves under rough sea conditions.