Figure 5.10 shows that the rate of damage increased with wave height, and 

 damage became more scattered as the damage increased. Although the wave heights 

 were reduced when the depth was increased approximately midway through the run, the 

 damage continued to increase. This was due to the fact that, when the depth was 

 increased, several stones higher on the structure (at the new still water level) were 

 displaced down into the damage area, which was centered at the lower still water level. 

 This can be seen in Figure 5. 10 as small jumps in eroded area after the depth change. 

 These small increments of damage approximately correspond to one stone displaced. 



Although the structure visually appeared to stabilize at various times during 

 Series A', Figure 5.10 indicates that the structure never really stabilized or reached 

 equilibrium, where further waves produced no additional damage. Thompson and 

 Shuttler (1976) noted this for their riprap tests. It is also clear that an extraordinarily 

 large number of waves was required to induce failure. For Series A', approximately 

 60,000 waves at significant wave heights appreciably greater than the no-damage wave 

 height were required to fail the structure. Again, failure was defined as exposure of the 

 underlayer through a hole at least D„5o in diameter. A higher mean damage value at 

 failure was recorded for Series A' than was noted by Van der Meer (1988). Van der 

 Meer quoted a failure damage value of S=S, whereas i^= 13 at failure for Series A'. It 

 was noted during the series that the waves had a difficult time moving the second layer 

 of stones. This led to high damage values at failure, with most of the first layer of 

 stones moved before the second layer began moving. Most other recent damage testing 



97 



