Laboratory measurements of deep-water breaking waves

The results of laboratory experiments on unsteady deep-water breaking waves are reported. The experiments exploit the dispersion of deep-water waves to generate a single breaking wave group. The direct effects of breaking are then confined to a finite region in the wave channel and the influence of breaking on the evolution of the wave field can be examined by measuring fluxes into and out of the breaking region. This technique was used by us in a preliminary series of measurements. The loss of excess momentum flux and energy flux from the wave group was measured and found to range from 10% for single spilling events to as much as 25% for plunging breakers. Mixing due to breaking was studied by photographing the evolution of a dye patch as it was mixed into the water column. It was found that the maximum depth of the dye cloud grew linearly in time for one to two wave periods, and then followed a t 1/4 power law (t is the time from breaking) over a range of breaking intensities and scales. The dyed region reached depths of two to three wave heights and horizontal lengths of approximately one wavelength within five wave periods of breaking. A detailed velocity survey of the breaking region was made and ensemble averages taken of the non-stationary flow. Mean surface currents in the range 0.02-0.03 C (C is the characteristic phase speed) were generated and took as many as 60 wave periods to decay to 0.005 C. A deeper return flow due to momentum lost from the forced long wave was measured. Together these flows gave a rotational region of approximately one wavelength. Turbulent root mean square velocities of approximately 0.02 C were measured near the surface and were still significant at depths of three to four wave heights. More than 90 % of the energy lost from the waves was dissipated within four wave periods. Subsequently measured kinetic energy in the residual flow was found to have a t-1 dependence. Correlation of all the above measurements with the amplitude, bandwidth and phase of the wave group was found to be good, as was scaling of the results with the centre frequency of the group,. Local measures of the breaking wave were not found to correlate well with the dynamical measurements.