This model describes the evolution of the wave energy spectrum in the North Sea and the Norwegian Sea. Waves are generated by wind, are propagating and are finally dissipated by internal friction and wave breaking, by interactions among the waves, and by friction at the sea bottom. In the model, these sources and sinks of wave energy are represented numerically by so called source terms. Advanced theories on the nature of the air-sea interaction and the exchange of momentum, heat and moisture at the sea surface have been developed by Vladimir Makin and co-workers, see list of publications. From this theoretical framework, a new parametrization of the wind input source term and the sea drag has been derived (Makin and Stam, 2003). This new parametrization has been inserted in NEDWAM and shown to yield good results, after retuning of the balance between the source terms. It will be implemented operationally in the near future. The new sea drag formulation will also be tested in the framework of the WAQUA/DCSM98 storm surge model and related high-resolution water level and current models.

The quality of wave forecasts is depends critically on the input from the atmosphere. At this moment, mesoscale versions of the atmosphere model HIRLAM are available operationally, employing horizontal grid spacings of 11-22 km. Additionally, downscaling methods have been developed, by which mesoscale surface wind fields from HIRLAM can be transformed to grids with a resolution of the order of 1-km. The introduction of the 11- and 22-km resolution HIRLAM models, replacing a version with 55 km resolution, on the whole appears to result in better wave forecasts. In situations with small-scale storms, the intensity and timing generally is much better in comparison with models with coarser grids.

The propagation scheme in NEDWAM is a first order upwinding scheme (WAMDI group, 1988; see also G.J.Komen et al., 1994, p 235-237). This scheme has a poor reputation for its dissipative properties (e.g. Rogers et al., 1999). Particularly low-frequency swell, which normally can propagate over long distances, is reduced significantly by this phenomenon, resulting in a systematic underestimation of swell. An extended study has been carried out by Rogers et al., in which a dozen propagation schemes have been tested and compared. A number of schemes appeared to be stable, computationally efficient and less dissipative than the upwinding scheme.

Measured wave directional spectra are used for determining an accurate initial condition. To this end a numerical technique called data assimilation is applied: available observations are used to adjust the initial condition of the model to the spectrum in the real world. Assimilation of wave observations in the NEDWAM model is particularly essential for making accurate forecasts of low-frequency swell (wave periods > 10 seconds). A data assimilation technique has been developed for NEDWAM (Voorrips et al., 1997). Wave spectra from buoys of the Dutch measuring network are now assimilated in the model, using this technique. The impact of this has been evaluated, comparing results before and after implementation (June 2000).

• Makin, V.K., and Stam, M., 2003. A new drag formulation in NEDWAM. KNMI Technical Report TR-250.
• Makin, V.K., 2003. A note on the parametrization of the sea drag. Boundary-Layer Meteorology 106, 593-600
• Kudryavtsev, V.N., and Makin, V.K., 2002. Coupled dynamics of short wind waves and the air flow over long surface waves, J. Geophys.Res.
• Makin, V.K., and Kudryavtsev, V.N., 2002. Impact of dominant waves on the sea drag. Boundary-Layer Meteorology, 103, 83-99
• Rogers, W.E., J.M. Kaihatu, N. Booij, L.H. Holthuijsen, 1999. Improving the numerics of a third generation wave action model. Naval Res. Lab. Rep. NRL/FR/7320-99-9695
• Voorrips, A.C., 1997. Optimal interpolation of partitions: a data assimilation scheme for NEDWAM-4. KNMI Scientific Report WR97-02.
• Voorrips, A.C., V.K. Makin and S. Hasselmann, 1997. Assimilation of wave spectra from pitch-and-roll buoys in a North Sea wave model. J.Geophys.Res., 102 (C3), 5829-5849.
• Komen, G.J., et al., 1994. Dynamics and modelling of ocean waves. Cambridge University Press
• WAMDI group, 1988. The WAM model - a third generation ocean wave prediction model. J. Phys. Oceanogr. 18, pp. 1775-1810

Koninklijk Nederlands Meteorologisch Instituut, Postbus 201, 3730 AE De Bilt, The Netherlands