waters of the Gulf were used for the fine tuning of their model. Their 

 study confirmed that diurnal tide in the Gulf is primarily a 

 co-oscillating tide driven by adjoining Atlantic Ocean and Caribbean 

 Sea. 



The water level response for a given tidal constituent is usually 

 expressed in terms of the surface displacement c (Schureman, 1941): 



C = F(t) A(X,*) cos [ ojpt + X - G (X,*)] (21) 



where \ is the longitude, * is the latitude, A is the mean amplitude 

 over 18.6 years and G the Greenwich phase or epoch at given position 

 (X,<t), (Dq is tidal frequency, x is the astronomical argument, while F 

 is the nodal factor, a slowly varying function of time. Tides at 

 particular stations are characterized by A and 6 for individual 

 constituents. In our study, A's and G's for 5 constituents (01, Kl, 

 PI, S2 and M2) along the open boundaries of our grid are supplied from 

 Reid and Whitaker's model. Surface displacements at the open boundary 

 stations are determined from a linear combination of those due to the 

 five tidal constituents. 



Tidal Currents off the Mississippi Coast 



As a first example, tides during 20 Sept. to 25 Sept. 1980 are 

 computed with our three-dimensional model. The surface displacements 

 at four stations (see Figure 4 for locations) within the Mississippi 

 sound are compared with measured data in Figure 5. Notice that the 

 measured data have been filtered such that variations due to 

 short-period oscillations on the order of a few hours or less are not 

 included. Initially, the diurnal tides are predominant. Towards the 

 end of the five-day period, the diurnal tides become somewhat less 

 predominant while the semi-diurnal tides became gradually more 

 apparent. Good agreement is found at all stations. 



In this simulation, a relatively large time step of 12 minutes was 

 used for both the external and the internal modes. Seven grid points 

 are used in the vertical direction. A relatively smooth bottom with a 

 roughness length, z^, of 0.1 cm was assumed. A parabolic length scale, 

 A, was assumed in tne vertical direction. 



The tide-driven horizontal currents at mid-depth are shown in 

 Figure 6 for two stations in the Mississippi Sound. Currents on the 

 order of 30 cm/ sec exist at both stations. Again, reasonable agreement 

 is found between data and model results. 



The horizontal velocity field at 1 m depth, after 3 days of 

 simulation, is shown in Figure 7. Relatively large currents exist at 

 the various tidal inlets and in the area between the Ship Island and 

 the Chandelier Island. Except in these areas, at this instant of time, 

 bottom shear stress generated by the tidal currents are generally less 

 than 0.8 dyne/cm2. Hence little sediment resuspension is expected. 

 However, during strong spring tides, such as those during the period of 

 12 June to 16 June, 1980, relatively stronger currents and bottom shear 

 stresses in excess of 0.8 dyne/cm^ could prevail within the tidal 



Sheng 

 276 



