TM No. 377 



and analysis. The type of velocity data obtained were unique, and there 

 was no clearcut approach to the data analysis procedures. This preliminary 

 statement of the problem must therefore be based on the experience gained 

 in actually pursuing the studies without apologies for deviations from the 

 original concept. 



When the wind blows over the ocean, it imparts momentum and kinetic 

 energy to the water surface by exerting a net stress in the direction of 

 the wind. This stress results from the frictional drag produced by the wind 

 on the water surface, and from wind pressure forces acting upon sloped regions 

 of the free surface (i.e., the windward and leeward sides of the waves). The 

 effects of this net stress are partitioned within the water column and are 

 assumed to be directly or indirectly responsible for various observable phe- 

 nomena; e.g., surface waves, wind currents, and turbulent mixing over a wide 

 spectrum of wave numbers. It is obvious that the wind generation of surface 

 waves and the turbulent or eddy mixing within the surface layers of the ocean 

 are intrinsically related; for both phenomena are manifestations of the verti- 

 cal transfer of wind-imparted momentum moving through the upper layers of the 

 water. Thus, one should examine the effects of the more regular quasi-oscil- 

 latory particle motions of the waves, since the waves themselves are a mani- 

 festation of wind stress. For this reason, it was decided to apply the Rey- 

 nolds stress concept to the analysis of the water particle motion within the 

 dynamic wind wave regime and, specifically, to probe the following related 

 questions: (l) is there a measurable Reynolds stress in ocean waves? (2) 

 If it exists, is this stress essentially controlled by detectable interactions 

 of the predominant motions of the time and length scales of the waves them- 

 selves? (3) How are the wave motions related to the more easily observable 

 parameters such as wave height and wind speed? 



Categories of Wave Motion -- In this formulation of the problem the 

 sea surface, which is under the influence of the usual forces of wind stress, 

 is portrayed as a strongly turbulent fluid regime in which the motions are, 

 for the most part, rotational (i.e., possessing eddy viscosity), so that the 

 totality of motions can be associated with a definite spectrum at a given 

 instant. This leads to a consideration of three approximate frequency bands 

 containing the characteristic motions of the sea surface. 



The first band or category includes the mean or quasi-steady motions. 

 It is usually assumed that the degree of isotropy of the broad distribution 

 of turbulent motions increases directly with frequency and inversely with 

 the eddy scale of motion. This certainly appears to be the case, since the 

 greatest deviation from isotropic flow (i.e., rectilinear flow) is repre- 

 sented by the long period and gently meandering motions of the semi-diurnal 

 tides and so-called drift currents. These flows are predominantly horizontal 

 and can be relatively steady and unidirectional over periods of from 1 to 10 

 minutes. It is porbable that these quasi -rectilinear flows are strongly af- 

 fected by both winds and bottom stress and by the gross topography. 



10 



