for 10 separate optimizations. Most of the cases, however, had a value of K 

 in the range of 1.1 - 1.9 10"^ m^'/N. The sum of squares was minimized with 

 respect to all profiles measured during the particular case, typically 

 encompassing 5-10 profile surveys per case. Figure 57 shows a representative 

 calibration run with the numerical model and a comparison with the measured 

 beach profile from the last profile survey of the simulated case (Case 6-1). 

 Beach profiles at selected time -steps from the model calculations are shown 

 together with the wave height distribution calculated at the last time step. 

 The optimal K- value for this case was 1.9 10"^ m^/N. As seen in Figure 57, 

 bar formation (size and location) and the amount of erosion on the foreshore 

 were well described by the numerical model. The small inshore bar was 

 purposefully neglected in the calibration simulation. This feature appeared 

 in the LWT experiment after 40 hr of run time, just prior to the last profile 

 survey. Measured wave heights are shown across the profile, indicating that 

 the wave height distribution was satisfactorily reproduced by the breaker 

 decay model . 



436. Transport rate distributions calculated at selected times are 

 shown in Figure 58. The magnitude of the transport rate decreased with time 

 as the profile approached an equilibrium shape in accordance with the behavior 

 of transport rate distributions directly inferred from the profile survey data 

 in Part V. Occasionally, the transport rate increased in the vicinity of the 

 break point compared with previous distributions, caused by movement of the 

 break point. As the break point moved offshore, energy dissipation increased 

 because of the decrease in depth occurring at the plunge point, and the 

 transport rate increased accordingly. 



437. It was not possible to relate K obtained from individual 

 calibrations to wave or sand characteristics with any significance for the 

 number of cases available for study. Qualitatively, the transport rate 

 coefficient seemed to decrease with increasing grain size and increase with 

 decreasing wave period. A wave period dependence of the profile time response 

 was also shown in the analysis in Part V of peak net cross -shore transport 

 rates calculated from the LWT data. 



438. Since it was not possible to relate the transport rate coefficient 

 to any physical property, it was desirable to achieve an optimal estimate of 



178 



