Conclusions. What else do we need to know about wave breaking? - Breaking and Dissipation of Ocean Surface Waves - page 413

these days and some of them have attempted to investigate different aspects of the wave-
energy dissipation directly during the breaking in progress. This topic was discussed in
Section 7.2 .
The main difficulty with respect to the use of the phase-resolvent models in system-
atic investigation and parameterising the dissipation due to wave breaking is their heavy
demand in terms of computational resources. It is possible to simulate a few periods of
the breaking in progress, but it is not feasible to simulate the evolution of nonlinear wave
trains leading to the breaking onset over tens of wave periods. Here, an obvious com-
promise can be employed. Such evolution is described very well by the fully nonlinear
potential models, which are less computationally expensive, as described in Chapter 4 .
As discussed in this chapter, such potential models, however, cannot go past the breaking
onset, mostly because their physics becomes inadequate: that is, they are not designed to
describe turbulence, bubbles and spray. This is what the two-phase models are designed
to do, and therefore the apparent solution would be to couple the output of the fully
nonlinear potential models with the input of the two-phase models - at the point of the
breaking onset.
In Section 7.3 , direct and indirect measurements of the dissipation were discussed.
These start with a description of dedicated laboratory experiments followed by a sub-
section which concentrates on and summarises the difference between signatures of the
modulational-instability and linear-focusing breaking. Measurement of the dissipation in
the field waves with a continuous spectrum are described, again followed by a subsec-
tion with the most important summary, which is mainly dedicated to the cumulative effect
imposed by the large waves on the small-scale breakers.
A subsection on the dissipation at extreme wind forcing followed. This demonstrates
that at very strong winds the dissipation apparently balances the excessive wind input, as
the spectrum level does not change significantly in response to such extreme forcing. Data
on the forcing of pre-existing waves by a hovering helicopter are presented to support the
argument. Such dissipation behaviour is quite different to the benign conditions, where it
is a hydrodynamic phenomenon that is determined by the wave properties (and the wave
spectrum) rather than by the wind.
The directional distribution of the dissipation (like the directional distribution of the
wind input in this regard) is usually subject to a speculative or residual-approach argument
and has hardly been measured. In the last subsection of Section 7.3 , experimental evidence
of the bimodal distribution of the dissipation function was presented.
The bimodal distribution means that the breaking dissipation is weakest in the main
wave-propagation direction and therefore the breaking tends to make the directional spread-
ing narrower. In Section 6.2 , it was noted that the frequency distribution of the breaking
severity is also smallest at the spectral peak. That is, the breaking appears to narrow down
the spectral peak both in the frequency and in the directional domains.
A large Section 7.4 was dedicated to actual implementation of the dissipation terms
in spectral wave models. For many years, these terms were treated as tuning knobs to
close the wave-growth balances. Following the recent advances, this section reviewed the