| description abstract | Visible sea surface images are analyzed to determine the distribution of the average length of breaking crests per unit sea surface area per unit speed increment ?(c). The ?(c) distribution offers a scale-dependent description of wave breaking that is valuable for understanding wave energy dissipation, momentum flux from the wave field to the surface currents, and air?sea fluxes of gas and sea salt aerosols. Two independent processing techniques for determining ?(c) from video images are implemented. In particular, the importance of the definition of the velocity of a breaking event is considered, as a single value, as a function of time, or as a function of space and time. The velocity can furthermore be defined as the full translational velocity or as the velocity normal to the breaking front. The ?(c) distributions resulting from various definitions of velocity, sensitivity to thresholds, observational resolution, and the effect of surface currents and long wave orbital velocity are presented. The appropriateness and limitations of the comparison of the first moment of ?(c) with the breaking rate are discussed. Two previous field observations of ?(c) give qualitatively different results: Melville and Matusov found an exponential form for ?(c), whereas Gemmrich et al. obtained a function that peaks at intermediate speeds and is up to an order of magnitude higher than that of Melville and Matusov. Both results can qualitatively be reproduced using the current dataset by employing the definitions of breaking velocity used in the previous studies. The authors argue that the current optimal interpretation of breaking speed resolves the breaking velocity as a function of both space and time and considers the velocity orthogonal to the breaking crest. | |
