was varied alternately with the distance YG1 to obtain the approximate 

 magnitude of net inflow of sand from the west. Third, the transmission coef- 

 ficients of the breakwaters were adjusted to obtain the correct size of the 

 salients behind the detached breakwaters. Fourth, the longshore location of 

 the eastern detached breakwater was translated two grid cells to the east to 

 obtain better agreement between calculated and measured positions of the 

 easternmost salient. This small adjustment can be thought of as compensating 

 for the finite grid size and oversimplification of the detached breakwaters as 

 thin. Finally, the modelers "stepped back" from the calibration procedure and 

 examined the results to see if there was a reasonable balance among the 

 parameters and overall replication of the shoreline change and historic 

 transport rates. The calibration result is shown in Figure 42, and the 

 corresponding START and OUTPT files are given in Appendix D. 



393. Figure 42 shows good agreement between the measured and calculated 

 shoreline positions. The calculated CVE indicated that the mean absolute 

 difference between the two shoreline positions was 4 ft. The calculated 

 volumetric change was 4,400 cu yd compared with the measured 4,300 cu yd, 

 again, a very good result. 



394. If data are available, model predictions should be verified by 

 reproducing measured shoreline change over a time interval independent of the 

 calibration interval. Sensitivity testing should also be done with the 

 calibrated model, with emphasis placed on sensitivity testing if verification 

 data are not available. In the present case, shoreline position data were 

 available for verification, but wave data over the interval between shoreline 

 surveys were not. (Additional gage data are available for Lakeview Park and 

 Cleveland which could be used to develop a more extensive wave data base, 

 including examination of variability. Development of an expanded wave data 

 set was beyond the scope of this illustrative case study, however.) 



395. Verification was made for the 13-month interval between 9 October 

 1978 and 17 November 1979. As stated, only 1 year of wave data was available. 

 It is doubtful that the same wave conditions that resulted in a net gain of 

 4,300 cu yd during the calibration period would likely produce a net loss of 

 300 cu yd for the verification period if all other conditions were left 

 unchanged (although the shoreline shape and position did change). 



167 



