Role of wave breaking in the air–sea interaction - Breaking and Dissipation of Ocean Surface Waves - page 398

whitecaps, their thickness, as well as bubble size and their concentration, rate of produc-
tion, and depends both on the breaking frequency of occurrence, i.e. how often the waves
break (see e.g. Sharkov , 2007 ) and the surface distribution of the bubbling patches due
to breakings of different scales (e.g. Zaslavskii & Sharkov , 1987 ), and on breaking sever-
ity, whose absolute strength defines the size of the bubbles produced ( Garrett et al. , 2000 ;
Manasseh et al. , 2006 ) and perhaps the thickness of the foam (e.g. Reul & Chapron , 2003 ).
The external structure of the whitecapping was broadly discussed in Chapters 3 and 5 .
The internal structure, before the onset of the remote-sensing (i.e. acoustic) methods of
bubble identification, was mainly studied either in the laboratory (e.g. Longuet-Higgins &
Turner , 1974 ) or in the surf zone (e.g. Blanchard & Woodcock , 1957 ; Raizer & Sharkov ,
1980 ). Field measurements, however, even though extremely difficult since in those days
they had to rely on high-resolution photography of the surface and underwater distribu-
tions of the bubbles, were also available and produced reliable quantitative estimates (e.g.
Kolobaev , 1975 ; Johnson & Cooke , 1979 ; Bortkovskii & Timanovskii , 1982 ; Bezzabotnov
et al. , 1986 ).
Bezzabotnov et al. ( 1986 ) observed the bubble structure in large enclosed seas, i.e. the
Baltic Sea and the Caspian Sea, and in the Pacific Ocean, at wind speeds from 9 to 20m s.

The main observed difference between the seas and the ocean, where the salinity is some
three times greater, is the volume concentration N V of air in the bubbles which was much
lower in the Baltic and the Caspian.
In Bezzabotnov et al. ( 1986 ), bubble structure was statistically and dynamically analysed
separately for active and passive whitecaps, as well as for surface and underwater (10 cm
below the surface) bubbles. For active breaking fronts, histograms of the bubbles sizes
are quite broad and asymmetric, whereas they are very narrow for the bubbles remaining
behind the breaking wave. The mean values of these histograms are also very different for
the active whitecaps this was about 1mm, and for the passive foam some 5

/

.

5mm. The

/

speed of horizontal bubble propagation reached up to 1m s in the breaking waves, which

apparently corresponds to the orbital velocity of the water particles at the propagating wave
crest, and in the residual foam the bubbles remained more or less stationary.
Bezzabotnov et al. ( 1986 ) explain the observations through the difference in lifetime
of small and larger bubbles. For the small surface bubbles it is of the order of 0

1s
( McIntyre , 1972 ; Bortkovskii , 1987a ), and by the time the larger bubbles surface those
smaller ones disappear. The authors also noticed a negative correlation, which reaches values
of

.

8, between bubble size and their horizontal propagation speed. In a way, this is
counter-intuitive since this means that slower-propagating breakers produce large bubbles,
at least among those which stay on the surface without being injected under the water first.
Below the surface, the structure of the two-phase medium, both the bubble-size dis-
tribution and void fraction, depends on the breaking severity. The latter, according to
Bezzabotnov et al. ( 1986 ), can be characterised by the area of the whitecap and orbital
velocities in the active breaker. Statistically, in their study the whitecap area depends on
the wind (in this regard, see review of the dependences of the whitecap coverage W on the
wind speed in Section 3.1 ).

− 0 . 6to − 0 .