height predictions. It is not a spectrum method. Rather, this model assumes 

 a single frequency sinusoidal wave where the usual procedure, when measured 

 deepwater wave spectra are available, is to set the amplitude equal to H Q /2 

 and the period equal to the significant wave period. In this study, instead of 

 applying the SPM equations to predict the wave height, the easier to use tech- 

 nique presented by McClenan (1975) was employed. The McClenan technique uti- 

 lizes a monogram which was constructed from the SPM equations and gives the 

 same results. The inputs to the monogram technique are the period, the deep- 

 water wave height, the deepwater wave angle, and the depths of interest. Figure 

 6 shows the comparison of the wave heights predicted by this technique with the 

 measured H s . There are no comparisons in the figure for d/H s _ ^ s < 2 because 

 the SPM method is applicable only for water depths greater than the breaker limit 

 (d b /H b -1.3). 



An examination of Figure 6 shows that the same segregation of the data into 

 swell, sea, and H1/3 > 1.8 meters, as in the previous comparison for the irregu- 

 lar wave technique, is appropriate for linear wave technique predictions. In 

 general, the following similar trends in the predictions are seen: 



(a) For swell waves there is significant overprediction. The 

 mean value of Hsp M /H s _ obs is 1.59 with a standard deviation of 

 0.35. 



(b) There is good agreement for the Baylor gage located in the 

 7.6-meter water depth. The mean is 1.04 with a standard deviation 

 of 0.08. 



(c) For sea with H s < 1.8 meters, the agreement between the 

 model and measurements is better than for swell but still overpre- 

 dicting. The mean is 1.25 with a standard deviation of 0.17. 



(d) For H Q > 1.8 meters, again the results are mixed. The 

 prediction agrees with the measurement for the Baylor gage in the 

 7.6-meter depth. The model overpredicts for the other gages. The 

 mean of Hgpjv(/H s _ obs is 1.51 with a standard deviation of 0.34. 



In general, in evaluating each of the two models from the comparison with 

 the gage measurements, it can be seen that the predictions of the irregular 

 wave-based model were about 10 percent better than those for the McClenan-SPM 

 model based on monochromatic waves. In addition the irregular wave model can 

 be used in much shallower water, while the monochromatic wave model is not 

 applicable in water shallower than the breaker depth. 



V . EVALUATION 



As indicated in the previous section, there are wide variations between the 

 model estimates and the field measurements of wave height. Two factors that 

 contribute to these variations are the bathymetry and the way the wave heights 

 are calculated. 



Bathymetric surveys at the FRF show that there is a depression near the 

 end of the pier. This causes some wave energy to be refracted away from the 

 gages, resulting in lower wave height measurements. The prediction techniques 



15 



