were chosen because they bracketed the 1 on 3.5 slope used in the small-scale 

 tests of this study. Only the small-scale tests witti a relative depth of 

 0.0144 and 0.0264 were used. The relative depth of 0.0144 gave the lowest 

 stability numbers and 0.0264 was in the range of relative depths tested in the 

 HRS study. Figure 12 shows the stability numbers, Ng, versus the Reynolds 

 number, Rg, which is described by equation (5). From Figure 12 it can be 

 hypothesized that riprap stability under irregular wave attack may be equiva- 

 lent to the monochromatic tests with the relative depths that yield the lowest 

 stability numbers. 



VI. RESULTS AND CONCLUSIONS 



The results of this study showed a reduction of about 20 percent in the 

 zero-damage stability numbers for a 1:10 (model: prototype) Froude scale model 

 from the expected prototype values. The reduction of stability in the model 

 appears to be related to the lack of penetration of the wave runup into the 

 filter layer and the improper modeling of the flow regime within the filter 

 layer. Figures 5 and 7 show that the difference between the small-scale tests 

 and the prototype tests decreased as the damage level increased indicating 

 that the scale effects decrease. The data points in Figure 7 indicate a 

 convergence of the damage trends, whereas the equation D' = aNg, as shown in 

 Figure 5, shows a crossing of the damage trends. The convergence of damage 

 trends seems reasonable because higher Reynolds numbers developed at higher 

 damage level and the viscous forces became less significant. The crossing of 

 the damage trends seems unreasonable and could be caused by the method used in 

 deriving the equation D' = aN^. The data points at the lower damage levels 

 could be exerting more influence on the equation than is justified. 



The following conclusions were reached: 



1. The tests at a 1:10 (model: prototype) Froude scale yield zero-damage 

 stability numbers about 20 percent lower than the prototype tests. This 

 indicates that scale effects in this study were less severe than those found by 

 Thomsen, Wohlt, and Harrison (1972). 



2. Scale effects were less severe at high levels of damage than at the 

 zero-damage level. 



3. The runup at the zero level was about 20 percent higher in the small- 

 scale tests than predicted by the prototype test. 



4. The shapes of the damage profile for the small-scale and the prototype 

 tests having the same relative depth were very similar. 



5. At the zero-damage level, wave period had less influence in the small- 

 scale tests than in the prototype tests. 



27 



