Table 6 



Summary of Regression Results for Longshore Sand 



Transport Rate Density Equation 



Expression 



H^Vd + adH^/dx) 



H^Vd + adH^/dx + Ra v /V) 









Const. 





k ao A ) 



a 



fi 



N/(m - sec) 



r 2 



1.8 











-1.2(10 3 ) 



0.45 



2.5 











-9.9(10 2 ) 



0.51 



2.0 



20 







-7.7(10 2 ) 



0.66 



1.5 



20 



1.8 



-2.4(10 3 ) 



0.77 



height or energy dissipation as the waves moved toward shore. However, in 

 some cases the gradient was negative, indicating that broken waves were 

 reforming. 



46. Kraus , Ginger ich, and Rosati (1988) introduced the gradient of wave 

 height as a correction to the quantity H ims V in the form of 



^rms^(l + a d^rms/^ x ) in which the value of the empirical coefficient a was 

 determined by iteration to provide the best linear least squares fit. The 

 resultant plot and regression line are given in Figure 19. Visual agreement 

 and the correlation coefficient are considerably improved over Figure 18, 

 which involved only the product H^sV . 



47. The longshore current speed used in the analysis is the average of 

 a time -varying flow. The sand transport rate should depend on the range of 

 current speed as well as the average. As a measure of the range, Kraus, 

 Gingerich, and Rosati (1988) chose the coefficient of variation of the current 

 speed cf v /V , in which a v is the standard deviation of the speed during the 

 averaging interval. The coefficient of variation was conceptualized as 

 providing a correction to the leading term H ims V , and the quantity 



H ims V(l + a dH ims /dx + R o w /V) was used for regression. The result is shown 

 in Figure 20, and associated values of determined coefficients are given in 

 Table 6. Grouping of the data points about the regression line is improved 

 over previous plots, and the apparent necessity of using a nonlinear or power 

 law function of H ims V , as was suggested by Figure 18, is eliminated. 



46 



