Multiscale Measurements of Ocean Wave Breaking Probability

Abstract

Recent numerical model studies of nonlinear deep water wave group evolution suggest that wave breaking onset is associated primarily with a threshold behavior linked to the nonlinear wave group hydrodynamics. Motivated by these findings, a recently published probability analysis of observed dominant ocean wind wave breaking events reported a threshold behavior using the significant wave steepness as a measure of the mean nonlinearity of these waves. The present study investigates whether a similar threshold dependence in terms of an appropriate spectral measure of wave steepness, the spectral saturation, may be found for the breaking probability of shorter wind waves above the spectral peak. Extensive data records of open ocean whitecap breaking wave occurrences for wind speeds up to 18 m s?1 were analyzed for breaking probability dependence on spectral saturation in spectral bands with center frequencies ranging from 1 to 2.48 times the spectral peak frequency. Results are based on the measured ratio of passage rates past a fixed point of breaking crests to total crests for different wave scales. An extension of the zero-crossing method for counting wave crests was developed. Using this method the authors found that in any spectral subrange within the observed range of frequencies, a strong correlation exists between breaking probability and an appropriate mean spectral steepness parameter and that this correlation is characterized by a robust threshold behavior, just as was reported previously for the spectral peak waves. Further, to offset the influence of increasing directional spreading of the waves above the spectral peak frequency, an empirical directional spreading function was used to normalize the azimuth-integrated spectral saturation. Under this normalization, the spectral saturation threshold for breaking onset appears to have a common level over the frequency range investigated. This study also examined the correlation of breaking probability with spectral peak wave age. The low correlation found for all spectral ranges investigated suggests that nonlinear wave hydrodynamics are more important than wind forcing for the breaking of these wind waves.