suspended; adjacent particles would become suspended later with further 

 increase in orbital velocity as the wave crest nears coincidence with the 

 sand ripple crest. At this point the particle clouds boil upward rapidly, 

 move slightly ahead of the sand ripple crest, then reverse direction as 

 particles settle in the slower velocities under the wave trough. Ripple 

 observations show that particle clouds occur for sustained times (greater 

 than 5 minutes) at the same source points along the ripple crests, which 

 lends evidence to the hypothesis since time is required for the ripple 

 system to change, and thence give rise to shifted streamlines with con- 

 sequent shifts in particle cloud locations. Suspended-particle clouds 

 have been observed not only in the laboratory, but in ocean waves in the 

 nearshore zone. The varying distance between the particle cloud sources 

 and the intake nozzle apparently causes randomness in the quantity of sedi- 

 ment pumped, especially when pumping is within 6 inches of the bottom. 



4 . Sample Collection and Processing . 



a. Sample Pumping . A CERC laboratory study (Watts, 1953) indicated 

 that an average representative sample of the wave-induced suspension could 

 be obtained by pumping if the ratio of intake velocity to maximum orbital 

 velocity is about 2. During this study, intake velocities varied from 18 

 to 25 feet per second (see App. A) and maximum orbital velocities were 

 generally below 5 feet per second, so the intake velocity-orbital velocity 

 ratio of 2 was equaled or exceeded for the bulk of the data. An average 

 of 40 gallons of sediment-laden seawater was pumped for each sample, which 

 required 2.5 to 3 minutes of pumping through the 0.5-inch nozzle. 



b. Decanting Water-Sediment Mixture . In the suspended-sediment col- 

 lections described in this study, the water- sediment mixture was pumped 

 directly into a collection-decanting tank calibrated for volume versus 

 tank water level. Water levels in the tank were taken on completion of 

 pumping, using a Lory point gage, and recorded on the sampling data sheet. 

 The total volume in each sampling was obtained from a calibration graph, 

 and the equivalent saltwater weight for this volume was based on a specific 

 gravity of 64 pounds per cubic foot. The ovendry weight of the sediment, 

 decanted and reduced from the water-sediment mixture, was then divided 

 into the total weight to obtain the concentration by weight for the sam- 

 pling. 



Partial separation of the sediment from the pumped water- sediment 

 mixture was accomplished in the field, using the sediment extraction mech- 

 anisms shown in Figure 7. For the Nags Head data collection, most of the 

 samples were decanted with tank 2; for the Ventnor data, most were done 

 with tank 3. In using any of the decanting mechanisms, 5 minutes was 

 allowed for sediment to settle after the sample pumping had ceased. The 

 methods used to decant the water from the sediment in tanks I, 2, and 3 

 were as follows: 



(a) In tank 1, the sand which had settled out at the bottom 

 of a transparent plastic hose loop (Fig. 7) was flushed out by 

 lowering the discharge end of the hose below the elevation of the 



