the Internal tide beam propagate across an ocean basin. It was on the 

 basis o£ this result that Rattray et dl. assumed the internal wave 

 generated at the continental shelf to be a progressive wave. 



In October of 1968 we attempted to measure the internal tide system 

 off the coast of Vancouver Island. Two vessels were used, each having 

 a string of temperature sensors spanning the beam of wave motion. 

 At this time the analysis of the resulting data has not been completed. 



DIFFRACTION OF INTERNAL WAVES BY A KNIFE-EDGE BARRIER 



Before looking at the Rattray model it is instructive to consider 

 a simplified model. Consider an internal wave confined to a channel of 

 a given depth. Then assume that a knife-edge barrier is inserted to 

 extend from the bottom, partially blocking the channel. The region 

 of fluid above the barrier will appear as a source region for the 

 internal-wave motion existing in the lee of the barrier. If we wait long 

 enough the transients should disappear and remaining will be signals of 

 the frequency of the incident wave. The details are given by Larsen 

 (1969) and we need not go into them here. 



Now the energy flux of the wave system is given by the correlation 

 of the pressure and velocity field. In Fig. 2 the magnitude and 

 direction of energy flux are plotted at selected points. The most 

 illuminating feature of this energy flux is its small magnitude in the 

 central portion of the wave field. There exists an interaction of the 

 pressure fields associated with each of the normal modes such that a 

 cancellation of the energy flux occurs in the central portion of the 

 beam. This is not a resonant interaction between modes, but a modi- 

 fication of the energy flux brought about by the presence of multiple 

 modes . 



The modification of the energy flux due to multiple modes may be 

 illustrated by the following simple model. Consider a channel containing 

 waves of the first two modes. In Fig. 3a we show the energy flux associated 

 with a single mode. The energy flux for the first mode has a zero in 

 the center of the channel and maximum at the channel boundaries. The 

 energy flux associated with the second mode has a peak at the center as 

 well as at the boundaries. 



In Fig. 3b we show the energy flux for a combined system of the first 

 two modes. The modes have Identical amplitudes and phases. Because of 

 the pressure field interaction the energy flux is no longer a simple sum 

 of the energy fluxes of the individual modes. There does exist a vertical 

 energy flux at some places. The energy flows along a sinusoidal curve 

 with no net vertical flux of energy. 



392 



