Simultaneous Nonlinear Wave and Ship Motion Forecast via Data Assimilation

A reliable near-future phase-resolved ocean wave forecast plays a crucial role in marine operations. With the development of the remote sensing and computational technologies, it is now possible to reconstruct the initial phase-resolved ocean surface from radar measurements and launch a nonlinear wave model such as the high-order spectral (HOS) method to predict the wave evolution in real time. However, due to the unavoidable errors in model configurations (e.g., initial conditions and physical parameters) and the chaotic nature of the nonlinear wave equations, the prediction by HOS can deviate quickly from the true dynamics. Recent studies, including those of the authors, have shown that this dilemma can be eased to some extent by incorporating the wave observational data into models via data assimilation methods such as ensemble Kalman filter (EnKF). In this work, we aim at the further improvement of wave prediction accuracy and simultaneous ship motion forecast. This is realized by coupling HOS, EnKF, and a Cummins-equation-based ship model, and including the observed ship motion as one additional data source. Through numerical testing, it is shown that the new integrated approach not only provides accurate ship motion forecast, but also uplifts the wave prediction accuracy compared to the state-of-the-art single-source (wave) data methods.