Role of wave breaking in the air–sea interaction - Breaking and Dissipation of Ocean Surface Waves - page 364

Direct measurements in this field in general, however, are rare, particularly in the ocean,
and particularly so are the measurements of the spray-production function and spray's
effects on the stress. Therefore, the experimental knowledge is not always convincing and
conclusive (see also Section 9.1.3 on further discussions of boundary-layer research in
extreme conditions). Models, on the other hand, do help to reveal and analyse the physics,
but without firm experimental guidance they have to rely on assumptions and indirect
evidence and sometimes lead to conflicting conclusions or even contradictions.
Indeed, the measurements available were either conducted in the laboratory or, if in the
field, primarily in surf zones (see O'Dowd & de Leeuw , 2007 , for a recent review). Here,
we would like to point out measurements by Bortkovskii ( 1977 , 1980 ) which escaped this
review and are particularly valuable as they were conducted in the open sea. Bortkovskii
( 1980 ) concluded that the structure of the two-phase (air-water) whitecapping medium
defines both the spectrum of the bubbles in the water and of the droplets in the air which
result from collapse of the bubbles. In any case, the typical range of the wind speeds
involved in similar measurements is less than 10m s

(e.g. Geever et al. , 2005 ). These, therefore, mostly deal with film droplets as the spume
production at such wind speeds is marginal if any.
Measurements at much stronger wind speeds are hampered by apparent practical diffi-
culties. These are due to both the need for specialised and sensitive equipment and for that
equipment to be enduring enough to withstand the harsh conditions, and due to the actual
ability to conduct experiments in such conditions.
The equipment and measurement techniques for field observations of spray production
are being developed and becoming more robust. These include both indirect methods, such
as the whitecap method, refractory measurements and surf-zone profiles (see O'Dowd & de
Leeuw , 2007 ), and direct measurement of the spray (e.g. Koga , 1981 ; Koga & Toba , 1981 ;
Nilsson et al. , 2001 ; Geever et al. , 2005 ). The latter estimates the fluxes through the eddy
covariance by using the sonic anemometer and a condensation particle counter combined.
Consistency of the measurements and their reliability, however, is still an issue. A num-
ber of experimental data on spume generation are available (e.g. Lai & Shemdin , 1974 ;
Monahan & MacNiocaill , 1986 ; De Leeuw , 1993 ; Smith et al. , 1993 ; Wu , 1993 ; Andreas ,
1998 ), but review of the spray-production function by Andreas ( 2002 ) (see also Schultz
et al. , 2004 ) exhibited a variation by approximately six orders of magnitude. Using such
estimates for testing and calibration of the analytical models is clearly a problem.
The discrepancies are not necessarily due to inaccurate measurements as such. As in
the case of the sea drag parameterisations discussed in Section 9.1.1 above, the fact that
spray production can be affected by many processes other than just the wind speed, would
certainly contribute to the scatter, and the scatter will not be possible to reconcile unless
these additional factors and features are taken into account. O'Dowd & de Leeuw ( 2007 )
argued that, at least for the film spray, sea-surface temperature is important too (see also
Bortkovskii , 1997 ). Apart from the physical sources of spray, chemical source functions
should be taken into consideration as well. Correspondingly, O'Dowd & de Leeuw ( 2007 )
report reduction of the discrepancy down to a factor of two.

/ s, with few exemptions up to 20m /