Abstract
This study examines the performance of a state-of-the-art spectral wind wave model that uses a full solution to the nonlinear interaction source term. The situation investigated here is fetch-limited wind wave evolution, for which a significant observational database exists. The authors consider both the evolutionary characteristics such as the predicted development of wave energy and peak wave frequency with fetch, as well as the predicted local features of the directional wavenumber spectrum: the spectral shape of the dominant wave direction slice, together with the directional spreading function. In view of the customary practice of constraining the shape of the spectral tail region, this investigation required relaxing the constrained tail assumption. This has led to new insight into the dynamic role of the spectral tail region.
The calculations have focused on the influence of two of the source terms in the spectral evolution (radiative transfer) equation for the energy density spectrum—those due to wind input and to dissipation predominantly through wave breaking. While the form of the wind input source term exerts some influence, the major impact arises from the dissipation source term, for which the authors explore a range of variants of the quasi-linear form proposed by Hasselmann. Due to the nonlinear coupling of spectral components through the wave–wave interaction term, it is only possible to obtain a detailed physical understanding of spectral evolution through such numerical experiments.
The results point to basic shortcomings in the present source terms. These lead to predicted local spectral properties and fetch evolution characteristics that differ significantly from the available observations. It is concluded that further refinement of the dissipation source term is required to improve modeling capabilities for wind sea evolution.
