Predictions from Boussinesq-equation-based models for the evolution of
breaking surface gravity waves in shallow water are compared with fie
ld and laboratory observations. In the majority of the 10 cases invest
igated, the observed spectral evolution across the surf zone is modele
d more accurately by a dissipation that increases at high frequency th
an by a frequency-independent dissipation. However, in each case the p
redicted spectra are qualitatively accurate for a wide range of freque
ncy-dependent dissipations, apparently because preferential reduction
of high-frequency energy (by dissipation that increases with increasin
g frequency) is largely compensated by increased nonlinear energy tran
sfers to high frequencies. In contrast to the insensitivity of predict
ed spectral levels, model predictions of skewness and asymmetry (stati
stical measures of the wave shapes) are sensitive to the frequency dep
endence of the dissipation. The observed spatial evolution of skewness
and asymmetry is predicted qualitatively well by the model with frequ
ency-dependent dissipation, but ij predicted poorly with frequency-ind
ependent dissipation. Although the extension of the Boussinesq equatio
ns to breaking waves is ad hoc, a dissipation depending on the frequen
cy squared (as previously suggested) reproduces well the observed evol
ution of wave frequency spectra, skewness, and asymmetry.