<o' 152 cm 



H, :118cm d :4.5m Hj/d =0.26 



Tp :4.8 s q3:0.92 d/gTp^ =0.020 



0p=1.52 q^^S.S d/Lp:0.163 



Hj^Ulcm 

 Tp :6.4 s 

 0n=1.37 



(i--52m Hs/d:0.33 



HyO.n d/gTp2:0.013 



q4:3.9 d/Lp:0.125 



(c) 



152 cm 



Hj :105 cm 

 T„ :8.8 s 



d :8.9 m 



q^ 3-2 



ItOsh- 



Hs/d:0.12 



d/gTp2: 0.011 



m^ 



^ e 



/JV-j^, 



Figure 18. Pen-and-ink strip-chart records and corresponding spectra 

 for some cases with five major spectral peaks: (a) Nags 

 Head record at 0400 e.s.t., 25 December 1968; analysis 

 at 0643 e.s.t., 25 December 1968; (b) Nags Head record 

 at 0800 e.s.t., 24 February 1969; analysis at 0642 e.s.t., 

 24 February 1969; (c) Huntington Beach record at 1900 

 e.s.t., 9 May 1972; analysis at 1840 e.s.t., 9 May 1972. 



small peak at high frequency. Small peaks at frequencies less than 0.05 

 hertz appeared in some of the Great Lakes spectra, especially spectra 

 from the Presque Isle pressure gages (^p. B) . These peaks were consid- 

 ered spurious and are not counted as major spectral peaks. Double-peaked 

 spectra are shown to be common at both the gulf and Great Lakes locations. 



It was noted that secondary spectral peaks do not always represent 

 independent wave trains. When waves are nonlinear, i.e., when Hg/d or 

 Hg/Lp are large, the wave profile will be nonsinusoidal. For such cases, 

 a secondary spectral peak at twice the dominant frequency is expected 

 even though a photo would show a single train of waves. The effect is 

 most evident for high, long waves at low tide; i.e., cases with high 

 values of Hg/d and low values of d/Lp. Appendixes A and B clearly 

 show that secondary peaks tend to occur systematically for high, long 

 shallow-water waves. However, the joint distribution table of number 

 of major spectral peaks versus significant height at each location shows 



41 



