565 



boundary conditions, the coast of Sylt, rather than 

 the main-land coast was deemed to be relevant as up- 

 wind boundary. This was based on the expectation 

 that the wave energy is propagating in a narrow 

 angular sector around the wind direction [e.g. , 

 Hasselman et al, (1973)] and since the coast of 

 Sylt is rather symmetric it should not cause asym- 

 metry in the wave field. But the coast to the North 

 and South of Sylt is strongly asymmetric. In fact, 

 the distance to shore in the direction of 155° (the 

 direction of the highest peak) is almost 2.5 times 

 the distance to shore in the direction of 65° (the 

 "symmetrical" direction, see Figure 4) . If this 

 asymmetry in the windward boundary is indeed the 

 cause, then it seems that the "ideal" generation 

 cases of JONSWAP may be contaminated to some degree 

 by asymmetric boundary conditions. Still, relating 

 this conclusion to the observed k-spectrum is largely 

 speculative as long as it is not substantiated with 

 more data. In particular the shapes of the k- 



spectra at locations closer to shore may give some 

 clues. 



The expectations regarding the directional dis- 

 tributions for the locally generated wind sea off 

 Noordwijk in the second sortie seem to be more 

 realistic, at least in an overall sense (Figure 14, 

 for k > kjj,) . Any skewness is hard to identify 

 through visual inspection of the plots due to the 

 small scale variations in the functions. These 

 probably stem from the statistical variability of 

 the estimates. The swell peak (k = 0.3 k^ to k = 

 0.6 kjjj) is unimodal and covers a narrow angular 

 sector with a half power width of about 35°. 



The directional distribution functions of the 

 spectrum in the third sortie seem to be strongly 

 skewed for the lower wave numbers (Figure 15, 

 K - 2 kjjj, say) but for higher wave numbers skewness 

 is hard to identify visually. As for the main di- 

 rection of the energy distribution, it varies almost 

 monotonously from approximately 80° at higher wave 



Noordwijk 760323 



k=311 k^ 



l<=S.32krr 



^v^vA/v^ 



k„= 5.88 « 10 -3 m" 



FIGORE 15. Normalized directional distribution functions of the k-spectrum off Noordwijk, 

 March 23rd, 1976. Directions are relative to true North. 



