Another possible explanation for the present observations is nonlinear 

 interactions among wave trains during their generation and propagation to the 

 FRF directional gauge. Various interaction mechanisms have been proposed 

 by the WAMDI Group (1988), Freilich, Guza, and Elgar (1990), Resio and 

 Perrie (1991), and others. Testing these hypotheses requires more detailed 

 spatial measurements of S(f,6) than are currently available. However, an 

 experiment can and should be designed for this purpose if the current results 

 are to be fully explained. 



By way of guidance for model testing and engineering design, a representa- 

 tive example of S(f,6) for late stage growth and peak storm conditions is 

 shown in Figure 4. It is simply an average of the 19 observations from the 

 relatively stationary peak storm conditions of Event N. It contains the prima- 

 ry features noted above and represents a nearly stationary dynamic process 

 that continued for 56 hr. It should be noted that although the high-frequency 

 tail has a spectral density less than the spectral peak, the total energy in the 

 part of the tail that is bimodal (found from the area under the frequency spec- 

 trum over the high-frequency range of frequencies) is roughly 25 percent to 

 50 percent of the total energy. Because the bimodal tail occurs at peak energy 

 conditions, there is considerable energy in the tail region. 



The key point is that a wave generation and transformation model at conti- 

 nental shelf and larger scales clearly must have a result that looks qualitatively 

 like Figure 4 to represent conditions at the FRF with reasonable verity. Addi- 

 tionally, nearshore process models of the FRF will be improved if a spectrum 

 like that in Figure 4 is used to describe the wave boundary condition outside 

 the surf zone in high-energy cases. 



20 



Chapter 4 Discussion 



