Energy dissipation across the wave spectrum - Breaking and Dissipation of Ocean Surface Waves - page 279

in the course of the Lake George experiment (see Section 3.5 ) which was specifically
designated to field investigations of the source functions under moderate-to-strong wind
forcing.
The new wind-input parameterisation of Donelan et al. ( 2006 ), for example, follows the
theoretical form proposed by Jeffreys ( 1924 , 1925 ), based on the wind-sheltering argument,
and incorporates two newly observed features of wind-wave coupling, i.e. dependence of
the growth increment on wave steepness (which makes the input function nonlinear in
terms of the wave spectrum) and full air-flow separation in strong wind-forcing situa-
tions (which leads to a relative reduction of the wind input). Additionally, it accommo-
dates enhancement of the wind input over breaking waves ( Babanin et al. , 2007b ,see
also Section 8.3 ), which becomes essential when the breaking rates are high. The param-
eterisation was able to reconcile experimental outcomes for fractional wind-wave growth
rates , which previously appeared incompatible, that is low values of the sheltering

coefficient in well-developed oceanic conditions ( Hsiao & Shemdin , 1983 ) and two-and-a-
half-times as high magnitudes of the sheltering for strongly forced and steep young waves
in the laboratory ( Donelan , 1999 ).
As was mentioned above, Tsagareli et al. ( 2010 )for S in and Babanin et al. ( 2010c )for
S ds employed additional constraints when testing the source functions, which constraints
allowed calibration of these functions separately, before they are put together in a wave
model to reproduce overall wave growth and overall wave-spectrum evolution. For S in ,
this is the total integrated wind input which must agree with independently observed or
known magnitudes of the wind stress. Within this approach, a new technique - a dynamic
self-adjusting routine - was developed for correction of the wind-input source function
S in at each step if the constraint was not observed. This correction involves a frequency-
dependent adjustment to the growth rate

γ( f )

, based on extrapolations from field data.
The model results also showed that light winds require higher-rate adjustments of the wind
input compared to strong winds.
Even though this topic is dedicated to the dissipation, we shall briefly review the wind-
input constraint and its implementation in Tsagareli et al. ( 2010 ). It is important, first
of all, as the first step to the dissipation-calibration constraint. Secondly, it is a key item
of the new methodology of separate source-function calibrations, intended to replace the
bundled-calibration method according to Komen et al. ( 1984 ).
So, the momentum flux

γ( f )

across the water surface (3.7) was considered the key boundary
parameter for calibrating the wind-input source function. The generation of surface waves
by the action of wind is due to work done by the wind stress exerted on this surface. Wind
stress is a result of the air-sea interaction, i.e. 'friction' of the air flow against the water
surface, and reflects the strength of this interaction. Physically, it is the drag force per unit
area exerted on the interface by the adjacent layer of the airflow. Therefore, wind stress
determines the exchange of momentum between the atmosphere and the water.
Significant stresses arise within the near-surface atmospheric boundary layer because
of the strong shear of the wind between the slowly moving air near the water surface and
the more rapidly moving air in the layer above (see e.g. Komen et al. , 1994 ).Closetothe

τ