Abstract. The performance of two methods for quantifying whitecapping
dissipation incorporated in the Simulating Waves Nearshore (SWAN) wave model is evaluated for waves
generated along and off the US east coast under energetic winter storms
with a predominantly westerly wind. Parameterizing the whitecapping effect
can be done using the Komen-type schemes, which are based on mean spectral
parameters, or the saturation-based (SB) approach of van der Westhuysen (2007), which is based on local wave parameters and the saturation level
concept of the wave spectrum (we use “Komen” and “Westhuysen” to denote
these two approaches). Observations of wave parameters and frequency spectra
at four National Data Buoy Center (NDBC) buoys are used to evaluate simulation results. Model–data
comparisons show that when using the default parameters in SWAN, both Komen
and Westhuysen methods underestimate wave height. Simulations of mean wave
period using the Komen method agree with observations, but those using the
Westhuysen method are substantially lower. Examination of source terms shows
that the Westhuysen method underestimates the total energy transferred into
the wave action equations, especially in the lower frequency bands that
contain higher spectral energy. Several causes for this underestimation are
identified. The primary reason is the difference between the wave growth
conditions along the east coast during winter storms and the conditions used
for the original whitecapping formula calibration. In addition, some
deficiencies in simulation results are caused along the coast by the
“slanting fetch” effect that adds low-frequency components to the 2-D wave
spectra. These components cannot be simulated partly or entirely by
available source terms (wind input, whitecapping, and quadruplet) in models
and their interaction. Further, the effect of boundary layer instability
that is not considered in the Komen and Westhuysen whitecapping wind input
formulas may cause additional underestimation.