Hoffman (1978) reports on the occurrence of multipeaked spectra from 

 hindcasts generated for a grid point in the North Atlantic Ocean by the 

 Spectral Ocean Wave Model of the U.S. iNavy Fleet Numerical Weather Cen- 

 tral, Monterey, California. The criterion for accepting a secondary 

 peak was that its peak energy be at least 30 percent of the energy in 

 the highest peak and that it be reasonably separate in frequency from 

 the main peak. About 25 percent of 572 cases considered were multi- 

 peaked. There was a tendency for multipeaked spectra to be more common 

 during low wave conditions than during high wave conditions. 



It is of some engineering interest to know whether a spectrum with 

 a certain Tp value is more likely to contain major secondary peaks 

 than a spectrum with a different Tp value. Evidence at many locations 

 shows that spectra associated with certain ranges of Tp do indeed ex- 

 hibit a preference for multiple peaks. At Atlantic City, Virginia Beach, 

 and Nags Head on the Atlantic coast and at Huntington Beach on the Pacific 

 coast, multipeaked spectra are more common for long and short Tp cases 

 than for intermediate Tp cases (shown in Fig. 19), Multipeaked spectra 

 are least common for Tp values between about 8 and 10 seconds at the 

 three Atlantic coast locations. At Huntington Beach multipeaked spectra 

 are least common for Tp values of 12 to 13 seconds. At Lake Worth and 

 Naples, there is no clear evidence of a relationship between Tp and 

 the occurrence of secondary spectral peaks. Data on secondary peaks from 

 Pt. Mugu and the Great Lakes locations were not considered because of 

 difficulties with the compensated pressure spectrum and because of the 

 small number of Great Lakes cases. 



Additional insight on the occurrence of secondary spectral peaks can 

 be obtained from a detailed look at their characteristics for a repre- 

 sentative Atlantic coast location and the single Pacific coast location. 

 For example. Figure 19 shows that 70 percent of the Nags Head spectra 

 with Tp between 6 and 7 seconds have major secondary peaks; Figure 20(a) 

 shows in a histogram that the periods of those secondary peaks range from 

 2 to 17 seconds. The figure also shows that secondary peak periods are 

 fairly evenly distributed over the range of 3 to 17 seconds. The rela- 

 tively low percentage of secondary peaks at 7 to 8 seconds shown in Fig- 

 ure 20(a) is artificially low because the CERC analysis procedure gives a 

 nonuniform resolution in wave period which, tends to decrease the number 

 of cases in the 7- to 8-second intervals and increase the number of cases 

 in the 6- to 7- and 8- to 9-second intervals. The procedure also precludes 

 any cases with intervals of 11 to 12, 13 to 14, 15 to 16, and 17 to 20 

 seconds. 



Although the histograms in Figures 20 and 21 show how often secondary 

 peaks occur, no indication is given of how much energy is contained in 

 the secondary peaks. Therefore, another curve was added to the figures 

 to show the average energy for secondary spectral peaks in each period 

 interval. The energy associated with each peak was estimated by the pro- 

 cedures discussed in Section III and shown in Figure 7. Relatively high- 

 energy secondary peaks tend to occur in Figure 20(a) at periods slightly 

 less than Tp and at periods of 12 to 14 seconds. The long-period sec- 

 ondary peaks generally represent swell waves. 



44 



