Ocean Wave Modeling from Satellite Images Using Data Assimilation

Abstract

Due to wave-induced problems such as surface currents and buoyancy, internal ocean waves, especially large nonlinear ones, may have a major effect on ship and submarine operations. Thus, a predictive system is needed by the navy which will determine the potential effects of internal waves in their areas. Data assimilation is a process in which information from multiple sources is typically or entirely interpreted and adapted. Numerical prediction models are frequently used in weather forecasts for estimating future conditions of the atmosphere, which depend on exact initial state. The consideration is that because of the methodology or other factors, the actual conditions of model mostly differ from the observation. Therefore, data assimilation is considered initially to be a process in which the data observed are interpreted and processed and correspond to certain space and time distributions, so that the initial fields for numerical predictions are as accurate as possible. The data assimilation methods include function fitting, the stepwise correction, the optimal interpolation, the variability method, and the ensemble Kalman filtering. Although various methods of data assimilation are listed, statistical and mathematical analyses are ultimately used to find the final solution for such methods. This paper proposes a system to model ocean waves using ensemble Kalman filtering and image processing techniques.