simpler parametric models. Our study also indicated that the presence of the 

 wave may either enhance or reduce the bottom shear stress due to the mean 

 currents. Further study using this model is needed, as well as wel 1 -designed 

 field or laboratory measurements, to elucidate the complex dynamics within the 

 bottom boundary layer. 



Entrainment and deposition of the Mississippi Sound sediments was studied 

 by means of a rotating annular flume. It was found that bottom shear stress, 

 salinity, sediment type, and time history of bottom sediments have great 

 influence on the entrainment rate of sediments. The water content (or bulk 

 density) alone was found to be insufficient to characterize the erodability 

 (or stability) of the Mississippi Sound sediments. Bacteria, macrofauna, and 

 organic matter may be important in affecting the erodability of the sediments. 

 Any model of sediment dispersion should include the time-history of the bottom 

 sediment as a parameter. 



More flume studies should be performed to better understand the effect of 

 various parameters on the erodability of sediments. The result will lead to a 

 better understanding of the bed stability in the area, thus a better planning 

 of dredging and disposal activities. Dredging or disposal at an unstable site 

 should be avoided. On the other hand, depending on the parameters, disposal 

 of sediments at a new site may have a stabilizing or de-stabilizing effect. 



Dispersions of sediment due to tidal currents, wind-driven currents, and 

 waves have been studied. Waves are found to be generally more effective in 

 causing entrainment of sediments. Model simulation of two events in September 

 1980 showed reasonable agreement with data. 



Measurements of flow and concentration data usually consist of a finite 

 set of data in a random field. To achieve meaningful model comparison with 

 data, a long time series of data is required at any given point to obtain the 

 proper mean value and variance. In addition, models capable of resolving the 

 variances as well as the mean variables should be used. 



211 



