1900 



1930 



2000 



Figure h Temperature recorded by thermistors 3 to 21 at the NEL Tower (4 Octo- 

 ber 7966, 1900-2000). (Each curve represents the variation of the temperature 

 from the mean for that numbered thermistor.) 



K. E. Kenyonll extended the theory to a description of the entire internal 

 wave spectrum which may result from the interaction of surface waves in the 

 ocean. He gave a computation of scattering of swell energy into internal wave 

 energy, using observations from the NEL Tower, from which he concluded that 

 this process may be of minor importance for the creation of internal waves. 



The following computations are similar to his, but we do not compute the 

 energy spectra, and derive only the equations for the amplitudes of the waves. 

 We come to different results, the reason for this being that Kenyon considered 

 surface wave spectra only for the case that the directionality of the spectrum is 

 given by cos'^a. This means that all surface waves are contained within an 

 angle of about ±30 degrees. From this narrow directional swell spectrum, 

 energy is fed to internal waves traveling mainly perpendicular to the swell. 

 This, obviously, does not agree with the observations mentioned above. Both 

 swell and internal waves travel mainly toward shore in the sea off Southern 

 California. 



Kenyon' s conclusion that surface waves with a narrow directional spectrum 

 do not feed considerable energy into the internal wave spectrum is supported by 

 K. Hasselmann. 12 This agrees with our results, but it will be shown here that 

 the process is about 100 times more effective for amplitude-modulated swell, 

 which may be described by two surface waves traveling in directions which in- 

 clude an angle between 90 and 180 degrees. This still leads to a progressive 

 wave, if the two amplitudes differ considerably in magnitude. If these two waves 

 are given by 



