where C is an empirical coefficient and the minimum and maximum values of 

 Km^ are 0.0 and 1.0, respectively. The recommended value of C is given by 



C = 0.51 - ^'l^ ^ ; < I < 3 2 (15) 



h — h — 



for smooth impermeable structures tested over the range 1. B/h £ 0.86 and 

 rough impermeable breakwaters tested over the range 0.88 1. B/h i. 3.2 (Fig. 15). 

 However, for submerged breakwaters tested with 1 on 15 fronting slopes, equation 

 (14) underestimates the wave transmission coefficient. For example, equation 

 (14) underestimates the wave transmission coefficient for BW14 when submerged 

 and the error increases as the breakwater becomes relatively more submerged 

 (Fig. 16). The data from BW14 and from Saville (1963) show that for submerged 

 breakwaters with 0.88 <_ B/h f. 3.2 and with a 1 on 15 fronting slope equation 

 (14) should be adjusted to 



K^^ = C (l - ~j - (1 - 2C) ^ ; |- < and 1 on 15 fronting slope (16) 



Figures 17 and 18 illustrate the observed and predicted wave transmission 

 coefficients for two of the rough impermeable breakwaters tested by Saville 

 (1963) for two values of crest width. Figure 17 shows the case of a structure 

 with a crest width-to-structure height ratio of 0.88; Figure 18 shows the same 

 information for a much wider structure with a width-to-height ratio of 3.2. A 

 scatter plot of observed and predicted transmission coefficients using Saville 's 

 (1963) data indicates the level of ability to predict Ky^ (Fig. 19). 



The above discussion shows that the breakwater freeboard and wave runup 

 have a major influence on the magnitude of the wave transmission by overtopping 

 coefficient. Breakwater crest width has a much smaller effect and only large 

 changes in breakwater crest width could be used to reduce the size of the 

 transmission coefficient for a given design situation. 



Wave transmission by overtopping coefficients may be predicted for imperme- 

 able structures using the computer program OVER (App . F) which applies methods 

 described in this section. 



c. Influence of a Breakwater on Other Wave Characteristics . The magnitude 

 of the wave transmission by overtopping coefficient, Ky^^, is generally the 

 most important parameter to determine for the design of an impermeable break- 

 water used to reduce wave height. However, in addition to reducing the average 

 wave height, the breakwater may also alter other characteristics of the waves, 

 such as spectral shape or wave height distributions. Since these additional 

 wave characteristics may be considered in some design problems, they are briefly 

 discussed below. 



The case of monochromatic waves incident on the structure is the condition 

 most often used to test wave transmission of laboratory breakwaters in previous 

 studies. This type of wave is similar to swell wave conditions in the prototype 

 where the incident wave height and period are approximately constant. Spectral 

 analysis of water level records for gages landward of the breakwater indicates 

 that a significant part of the wave energy of transmitted waves may be at 

 harmonic frequencies of the forcing wave (Saville, 1963; Goda, 1969). The 

 fraction of wave energy at the forcing period (Fig. 20) shows the same trend 



31 



