A Stochastic Approach to Short-Term Ocean Wave Forecasting: Preliminary Results Using Data From a Remote Sensing Imaging System

Abstract

There is a growing interest in the applications of real-time wave forecasting (RTWF), which consists in predicting physical quantities directly related to waves, such as the free-surface elevation, wave loads, or the motion of a ship, from a few seconds to several minutes in advance, and using measurements updated in real time. Unlike comparable RTWF methods found in the literature, which are based on the solution of the physical wave propagation equations, the present approach, known as SBP (Spectrum-Based Predictor), adopts a rigorous probabilistic view on the wave prediction problem, based on well-established, standard oceanographic assumptions. This paper presents an application of the SBP method to real wave field data coming from a stereoscopic camera system. To the best of the authors’ knowledge, this is the first time stereo wave data are employed to test RTWF algorithms. The data, recorded at a location in Korea, in the Yellow Sea, present some considerable challenges, such as strong current in excess of 1 m/s, steep waves with substantial non-linear components, and large directional spread in the high-frequency range. With some adjustments to the original SBP approach to account for current, several prediction configurations are tested, showing excellent agreement between the experimental prediction performance curves, and those expected from the SBP theory. With an observation range in the order of 100m, and in the wave conditions studied, reasonably accurate predictions can be achieved up to 20s ahead (approximately 3.5 peak wave periods).