Detection and measurement of wave breaking - Breaking and Dissipation of Ocean Surface Waves - page 95

Smith et al. ( 1996 ) used differences in the backscattered signal of vertical and hori-
zontal polarisations at low grazing angles to distinguish spilling breaking events. They
demonstrated a clear association of wave-breaking occurrence with the group structure of
wave fields. A very interesting outcome was identification of the dispersion relationship
of the breaking waves. Smith et al. ( 1996 ) observed developing seas, and for the young
seas the dominant waves are expected to actively break ( Banner et al. , 2000 ). It was found,
however, that the dominant breaking frequency is about 25% above the spectrum peak
frequency. Even higher levels of upshifting were reported as a result of dedicated wave-
breaking microwave radar observations by Stevens et al. ( 1999 ). This is consistent with
the observed shrinkage of individual waves prior to breaking brought about by the modu-
lational instability, and the corresponding upshift of the wave frequency ( Liu & Babanin ,
2004 ; Babanin et al. , 2007b , 2009a , 2010a , see also Section 5.2 ).
Hwang ( 2007 ) investigated the correlation between dominant breaking and short-scale
breaking based on both radar and acoustic observations. He argued that the radar-sensing
techniques often process a very large number of breaking events and from these statistics an
average length scale of the breakers in wave systems emerges. Based on sea-spike observa-
tions, it is about one order of magnitude shorter than the dominant wavelength of the wave
field. Such evidence is very important, particularly because, as discussed previously, the
microwave backscatter can account for wave breaking not producing whitecapping, which
is a challenge, if not impossible for other wave-breaking detection techniques (with the
exception of the infrared sensing method as discussed later in this section). This obser-
vation is consistent with some derivations for the dissipation function ( Hwang & Wang ,
2004 ).
Another potential remote-sensing technique for estimating breaking occurrences was
suggested by Challenor & Srokosz ( 1984 ) and Srokosz ( 1986 ) for satellite altimeters.
Indeed, Srokosz ( 1986 ) proposed a statistical model to connect the breaking probability
with the fourth moment of the spectrum m 4 (see (3.47) and Section 3.8 ), and Challenor &
Srokosz ( 1984 )showedhow m 4 may be estimated from the altimeter radar return. Satellite
altimeter missions have now been flown, almost continuously, for some 25 years (see e.g.
Zieger et al. , 2009 ), and thus such a technique would allow us to obtain the global distri-
bution of the breaking probability and even its long-term trends, both global and regional,
if they exist. Like many other remote methodologies outlined in this section, however, this
interesting idea is still awaiting its implementation in practical studies of wave breaking.
A paper by Phillips et al. ( 2001 ) will be the last one mentioned in this radar-probing
sub-section because it effectively links two major remote-sensing techniques, the radar
backscatter methods which have been discussed above and optical observations of breaking
which will be discussed next. It undertook measurements of

, the average length of the
breaking front per unit area per unit speed interval introduced by Phillips ( 1985 ). Combined
with the Duncan ( 1981 ) hypothesis (3.27) , measurements of

( c )

( c ) can provide the total

energy dissipation:

b br ρ w
g

c 5

S ds = ( c ) dc . (3.30)