H D /L S , relative wave height at the toe H D /d s , relative berm length B/L x , 

 relative water depth at the toe d s /L s , relative water depth at top of berm 

 di/L^ , and relative berm depth d 1 /d s are presented in Plates 1-3. 



20. Stability number shows no significant trend with increasing values 

 of H D /L S or H D /d s (Plate 1) . For the range of conditions and berm armor 

 stone weights presented herein, wave conditions representative of nonbreaking 

 waves at the toe did not cause berm stone damage. Hence, the guidance 

 developed from this test series is strictly limited to breaking wave design 

 conditions. From Plate 1 it is seen that all but one data point represented 

 N s values for H D /d s greater than 0.7. A similar indication is given by the 

 narrow band of high H D /L S values (Plate 1) typical of breaking waves. The 

 lack of a strong trend in N s with increasing values of wave steepness and 

 relative wave height is possibly due to the fact that all reported test 

 conditions are breaking waves. It is very likely that an increase in stabil- 

 ity number would be realized for relative wave height and wave steepness 

 values associated with nonbreaking waves. 



21. The range of relative berm lengths tested was rather narrow. (For 

 consistency with Tanimoto, Yagyu, and Goda (1982), what is commonly referred 

 to by many individuals as berm width is being referred to herein as berm 

 length and is defined as a horizontal distance measured across the berm crest 

 and normal to the structure crest.) For all 2-D tests, the berm length B 

 was equal to 3t , which defines the length of three armor stones set side by 

 side (Figure 5). Tanimoto, Yagyu, and Goda (1982) showed that berm length 

 relative to incident wave length (relative berm length B/L x ) to be an 

 important parameter for stability of berm stone fronting impervious vertical 

 walls. A fixed crest length of approximately 0.4 ft was used in the 3-D 

 tests, while the 2-D tests used berm crests which were three stones long, 

 resulting in a rather narrow range of tested B/L 1 values. For this range, 

 no significant correlation between stability number and relative berm length 

 was noted (Plate 2) . This lack of trend, as compared to the one developed by 

 Tanimoto, Yagyu, and Goda (1982), is possibly due to the lower reflectivity of 

 rubble as compared to impermeable vertical structures and to the shortness of 

 the toe berm lengths tested relative to incident wave length. 



22. The stability number shows a general trend to increase with 

 increasing values of d s /L s and dj/I^ (Plates 2 and 3, respectively). This 

 phenomenon follows the logic that the longer the wave period the deeper the 



20 



