3. The Duration of Irregular Wave Crests . 



The larger waves in an irregular wave train have the greatest nonlinearity, 

 so the duration of the crests of these waves tends to be small relative to wave 

 period. Figure 27 illustrates this trend by presenting the mean relative crest 

 duration for various probability levels of exceedance for waves traveling over 

 a flat bottom. For exampl e , 5 percent of the largest crests have a mean dimension- 

 less crest duration T(,/T = 0.406, whil e the average value for all of the wave 

 crests (at the 100-percent level) have l^/T = 0.473. 



VI. COMPARISON OF RESULTS FOR PROTOTYPE DATA 



The techniques presented in the previous section are to be used for the 

 future prediction of wave characteristics in field situation. In this section, 

 field data collected at the FRF in Duck, North Carolina, are analyzed in the 

 same manner as the laboratory data previously described and comparisons are 

 presented to demonstrate the applicability of the predictive techniques. 



1. Field Conditions . 



Four Waverider buoys were used to obtain the prototype data in an area 

 where the bottom contours are essentially straight and parallel. Table 10 gives 

 the locations of the gages and the mean water depths. The data were collected 

 in the third week of October 1980 during a severe storm that produced wave 

 heights exceeding 4 meters. The data set has widely varying significant wave 

 height and period of peak energy density combinations resulting as the storm 

 moved into and through the study area. Table 10 summarizes the range of wave 

 conditions used in this report. 



2. Prototype Verification . 



The field data were analyzed in the same manner as the laboratory data used 

 in formulating the predictive parameters. The observed crest elevations were 

 compared with those predicted by equation (8), using values of A* for m = 0.0 

 (Table 8), since the bottom slope was very gentle. Table 11 shows the results 

 of this comparison for decreasing exceedance probabilities. It should be noted 

 that the predicted crest elevations are slightly conservative, with the coef- 

 ficient of variation approximately 8.5 percent. Figures 28 and 29 illustrate 

 the predictive ability of equation (8); the diagonal line indicates perfect 

 agreement. The scatter decreases for higher probability of exceedance levels, 

 in part, because the data for higher probability levels are more statistically 

 reliable. These figures also show that the scatter from the prediction line 

 increases slightly as Hg/d increases. This increase in conservatism with 

 increasing wave height-to-depth ratio may be due to the fact that the waves 

 are interacting with a bottom that is slightly sloping and comprised of a 

 porous sandy material (rather than a rigid floor used in the laboratory 

 experiments). The directional characteristics of the prototype waves could 

 also be influencing these results because a spectrum of wave directions was 

 present, even though the primary wave direction was along the line of the 

 Waveriders. 



44 



