the prediction of these currents has been a major difficulty. Since 

 there is abundant wave energy to suspend the sediment, even weak 

 currents may lead to significant sediment transport. 



In the field there may not be a constant mean current throughout the 



water column at each point. The surface layer of the sea has a mean 

 motion dominated by the wind and Stokes drift. In the case of breaking 

 or near breaking the latter can be a substantial fraction of the wave 

 speed, e.g., when breakers pass over a sandbar and the return flow is in 

 a channel. Nielsen's (1979) observations of the effect of gently 

 spilling breaking waves should be noted. Turbulence from wave breaking 

 may affect sediment transport once it has diffused to the lower boundary 

 of the fluid. 



The midlayer of water will largely move under its own inertia and 

 pressure gradients from the mean surface and mean wave stresses. The 

 bottom boundary layer not only responds to the stress from the mean flow 

 and waves but also to pressure gradients so that, in a general three- 

 dimensional situation, all three layers have different directions. Even 

 such a well-defined problem as the mean current profile for currents and 

 waves in the same direction has no satisfactory solution yet. Brevik 

 (1980) and Kemp and Simons (1982) both find that an increase in wave 

 height "steepens" the velocity profile outside the wave boundary layer. 

 However, this effect does not seem to be found by Bakker and Van Doom 



(1978) in their Figure 3b. 



Different modes of transport occur depending on the bed features, 

 or lack of them. These features are well documented for currents or 

 waves alone (Allen, 1968; Nielsen, 1979), but for the combination of 

 waves and currents there is little except for Tanaka, Ozasa, and 

 Ogasawara's (1973) measurements of ripples. Theoretical models (based on 

 extensive experiments) for sediment movement in oscillatory flow, with 

 or without a mean current, have been developed by Nielsen, Svendsen, and 

 Staub (1978) and Nielsen (1979). As is the case for structures, it is a 

 triple interaction (waves, current, and bed) which requires study. 



Three-dimensional laboratory experiments with a controlled flow 

 have been performed by Bijker (1967) and Rasmussen and Fredsjie (1981). 

 Bijker's measurements were for waves at right angles and at oblique 

 incidence to a current; the author notes that the measurements of 

 sediment transport were only partly successful in his current-dominated 

 regime. Rasmussen and Freds^e made more accurate measurements in a 

 wave -dominated regime with waves at right angles to the current. They 

 find reasonable agreement with a theory in which bedload is calculated 

 using their recent theory (Freds^e and Rasmussen, 1980), and suspended 

 load is found from concentration and velocity profiles based on Nielsen 



(1979) and Freds^e (1981). 



64 



