212 



wind direction, this shows that the longer waves have directional features which 

 the shorter waves have not. It also suggests that the distribution of slopes is 

 governed largely by very short waves. These examples are quoted to illustrate 

 how studies of the propagation of radar and radio waves can provide additional 

 information about the geometry of the sea surface. 



THE FLUX OF MOMENTUM ACROSS THE AIR-SEA BOUNDARY 



The flux of momentum from the atmosphere into the ocean maintains ocean 

 currents and waves. The physical mechanism of this process is not under- 

 stood. Direct measurements of this flux could be done in the following ways: 



1. Reynolds stresses just above the boundary 



2. Stresses at the boundary 



3. Reynolds stresses just beneath the boundary. 



Hot wire anemometry has now been perfected to the point where Reynolds 

 stresses have been successfully measured over land. The method has not yet 

 been used from a vessel. The instrument problem is to achieve a sufficiently 

 high frequency response so that all turbulent elements essential to the momentum 

 flux are included in the measurements. I am not aware of any attempts to meas- 

 ure Reynolds stresses beneath the water surface. This problem deserves con- 

 siderable attention. 



The "wind slope" of the boundary represents one of the oldest and most 

 powerful ways of measuring stresses at the boundary. This involves, however, 

 certain assumptions, and the method is not quite as straightforward as one would 

 wish. A suggestive set of measurements have recently been conducted by Van 

 Dorn. By recording the difference in up and downwind water level along an 800 

 foot basin to an accuracy of 0.1 mm, he achieved laboratory precision under 

 natural conditions. His results, in general, confirm Keulegan's laboratory 

 work. They indicate that at low wind speeds the transfer is due principally to 

 viscous stresses across a very thin laminar boundary layer; at high wind 

 speeds, the transfer results mainly from the difference in pressure on the wind- 

 ward and lee sides of little waves. In order for the latter process to be opera- 

 tive, the turbulent wind just outside the laminar boundary layer must exceed the 

 velocity of the waves. The effect of a detergent spread on the water is to great- 

 ly suppress the little waves and the normal stresses associated with them. The 

 stress increases somewhat during a hard rain. 



Van Dorn's results indicate therefore the existence of two processes of 

 about equal importance. They will have different effects on waves and currents, 

 and they should be studied separately. Would it not be possible to measure di- 

 rectly the shear in the laminar boundary layers, both above and beneath the sur- 

 face? Can methods be devised for measuring the pressure difference across the 

 little waves? 



DISCUSSION 



The principal need, I believe, is for measurements whose interpretation 

 does not depend on any preconceived theory. Two types of measurements sug- 

 gest themselves. One type can give information in great detail about the indi- 

 vidual roughness elements of the boundary, and of motion very close to the 

 boundary. The other can give statistical descriptions of the processes involved. 

 The study of the sun's glitter, and of reflection of other electro-magnetic waves 

 from the sea surface, falls into the second category. Further work along the 

 former line is contemplated by Van Dorn, and along the latter line by Duntley, 

 Burt and Sekera. 



