Modelling wave growth in narrow fetch geometries: The white-capping and wind input formulations

This paper investigates the performance of three different wave model source term packages in narrow fetch geometries. The packages are used to model the sea state in a complex coastal system with narrow fjords on the west coast of Norway. The modelling system is based on the Simulating WAves Nearshore (SWAN) wave model that is forced with winds from a nested atmospheric model and wave spectra from a regional wave model at the boundaries. The performances of the recent ST6, and two older SWAN white-capping and wind input packages, are evaluated by comparing modelled spectra and integrated wave parameters against five wave buoys. The comparison covers long-term statistics and two case studies of narrow fetch geometries (i) without swell and (ii) with swell-wind sea conditions. SWAN’s original saturation-based approach performs best in the fjord system. In narrow fetch geometry without swell, all packages overestimate the wave energy. ST6 shows the highest sensitivity to fetch geometry and local wind changes. The results indicate that the ST6 white-capping is too weak to balance its strong wind input.