verification, and probably narrow-banded spectra as well, the model is inca- 

 pable of predicting the full extent of the shoaling peak prior to breaking, 

 probably due to the use of linear wave theory. This limitation, coupled with 

 any errors in the incipient breaking criterion, may cause the location of the 

 break point to be erroneously specified (e.g., Izumiya Test 9). However, the 

 surf zone transformation part of the model seems to be very accurate for plane 

 beaches and quite reasonable for stepped and barred bathymetry. Consequently, 

 the model is usually capable of simulating the general form of the wave height 

 variation across the breaker zone even though there may be a slight horizontal 

 shift between observed and simulated distributions. This is substantiated by 

 the results presented above. Inclusion of this breaking scheme into the 

 model, despite its shortcomings, is an obvious improvement over methods which 

 specify the surf zone wave height to be proportional to the local water depth 

 everywhere, especially for complex bathymetry involving stepped or barred 

 beaches. The accuracy of this model for use in predicting surf zone charac- 

 teristics under wide-banded spectral wave conditions is questionable. The 

 model assumes wave breaking begins at a well defined point. This assumption 

 is most valid under extremely narrow-banded spectral conditions but is vio- 

 lated otherwise. 



35 



