On the other hand, the turbulent flow pattern in the prototype and in the 

 model will be similar if the Reynolds criterion is observed 



V 



r 



~ £ — = 1 (2) 



where v r is the ratio of kinematic viscosities of the prototype fluid 

 and the model fluid. Combining Equations (1) and (2), one has 



3/2 



y = I (3) 



r ^r v ' 



i.e. , the prototype and model will be dynamically similar if the model 

 fluid is so chosen that its viscosity observes Equation (3). Since 

 water is used in the present model study, the result of the experiment 

 therefore corresponds to the performance of an ocean of, say, heavy oil. 

 One immediately sees that an ocean of water, and an ocean of oil cannot 

 behave in a similar manner if the viscosity is an important factor in 

 defining the flow pattern. 



Fortunately, on a normal beach the breaking of the waves is largely 

 due to an increase of wave amplitude, such that the wave form becomes 

 unstable. The tumbling over of the wave front and its breaking into a 

 large number of eddies are affected by the viscosity of the fluid only to a 

 a minor extent. Since the motion of sediment takes place in the area 

 where the eddies are created, the viscosity of fluid is a factor of 

 secondary importance only. This is no longer true for the sediment motion 

 along a submerged beach at the base of a cliff. It is common knowledge 

 that the viscosity of fluid plays a significant role in dissipating 

 turbulent energy into heat. In an ocean of water, the turbulence 

 generated at the cliff may extend to a significant distance from the cliff 

 before being completely destroyed by viscosity, while in an ocean of oil 

 this distance would be very much limited. Naturally the sediment transport 

 in the former will be larger than in the latter, due to the difference 

 in the area of the submerged beach which is being exposed to the action 

 of turbulence. The effect of viscosity on the diffusion of turbulence 

 from a line source into an unlimited body of fluid is rather involved 

 from the analytical point of view, although it can easily be studied 

 experimentally in the laboratory. For instance, by shaking a plate under 

 a large body of fluid, turbulence of controlled intensity can be generated 

 over the entire area of the plate. The diffusion and decay of the 

 turbulence thus created can then be studied by observing the trace of dye 

 particles or concentration distribution of minute foreign particles. 

 By using fluids of different viscosity, one can easily find out, at least 

 qualitatively, what effect viscosity has on this type of turbulent diffusion. 



From the discussion above, one may conclude that although the result of 

 the model study indicates that the motion of sediment along a submerged 

 beach at the base of a cliff is only a small percentage of the normal 

 littoral transport on a sloping beach, it is conceivable that the two may 



29 



