Plate 
If the spectrum can be described through the dominant wave, 
then the breaking of the dominant wave affects the spectrum every- 
where and sets the scale of the maximum level that the energy of the 
water surface can attain, Breaking of the wave occurs when the maxi- 
mum accelerationay, of the wave, thatis, the acceleration of the 
crest of the wave, reaches some limiting value determined by the 
acceleration of gravity g, such that 
Bs amit ak (5) 
where & in the case of a breaking Stokes wave is equal to 0.5 (Longuet 
-Higgins (1969)). Plate et. al. (1969) (see also Plate and Nath (1969)) 
have shown that the assumption equation 5 leads toa relation for the 
spectral peak given to: 
Gl (ava Saints HF? (6) 
Sai(hy sae 
with Ay = ae cae . We note that there exists a constant value 
of £4 if a is a constant. The assumption#~@then leads to a value 
for a . By putting Sg (1) = 2.74 (using Mitsuyasu's value as an 
average) and £ = 1.5. 107-2 (as given by Mitsuyasu for the field) one 
obtainsa ~ 0.15, whichis of the same order, but considerably smal- 
ler than the upper limit 0.5. 
A constant a@ or A , anda similarity spectrum whose shape 
does not depend on the size of the dominant wave, has a very impor- 
tant consequence for modeling of the ocean surface in a laboratory 
tank, As Plate and Nath (1969) have shown, it implies that wind-gene- 
rated waves can be used in the laboratory to exactly duplicate in sha- 
pe and reduced magnitude the wave spectra of the ocean surface, pro- 
vided that the Froude Fr number 
ese ae (7) 
is the same in model and prototype. The condition equation 7 can be 
satisfied in the laboratory, and thus it is possible to perform model 
studies of vibrating non-linear structures subjected to wind waves. 
Unfortunately, the exact correspondence of the field-and laboratory 
is not quite assured, as will be discussed in the remainder of the 
paper. 
[376 
