Internal Waves 



547 



during 1930 in the Atlantic Ocean, quite a number of periods were found 

 other than the tidal periods (see Ekman, 1942). Their existence seems to be 

 established, but cannot be explained. But they show the difficulties in the 

 interpretation of the phenomenon. They prove, furthermore, that a break- 

 down of the variations into single component waves is only possible when 

 the observations are very complete. 



It is of particular importance that the normal tidal current be eliminated. 

 This current ought to have the same direction and the same velocity in all 

 depths, according to the theory. This elimination is important for the in- 

 vestigation of internal waves as well as for the determination of the tidal 

 currents which are correlated to the surface tides. The customary method is, 

 to compute a mean current for the entire layer from the surface to the bot- 

 tom (see Defant, 1932, p. 164). This, of course, is only possible if the current 

 distribution for the entire depth of the ocean is known. But in genera] this 

 applies only for the upper layers. However, the formation of mean values 

 for the upper layer of the ocean ought to yield somewhat correct values, 



-lOm 



350 



Fig. 227. Left: vertical distribution of the amplitudes of the vertical displacements r\ of 



the internal waves of the l-4th order. The corresponding phases have been entered on the 



curves. Right: combination of the four waves and comparison with the observations (• for 



amplitudes ??, + for phases) (Lek and Fjeldstad). 



because just here the internal waves are particularly well developed on ac- 

 count of the greater density gradient. Sverdrup (1942, p. 595) showed the 

 theoretical foundation for this method. According to (XVI. 19) 



Un 



c n 



8z ' 



where u n presents the horizontal current velocity, c n the velocity of progress and 

 r\ n the vertical displacement of the wave of /?th order. As the values of r\ of every 

 internal wave disappear at the surface and the bottom of the sea, it has to be 



h 



fu a dz = 0. (XVI. 38) 



For waves at the boundary surface between two media this relation will be 



35< 



