flume between tests, more tests could be accomplished and with more frequent 

 profiling. 



In the experiment, damage development was measured on two identical 

 breakwaters with seaward slopes of 1:2 exposed to depth-limited breaking waves on a 

 1:20 beach slope. The armor was limited to a traditional two-layer thickness of 

 uniformly sized stone and a widely graded riprap. The waves were run as sequences of 

 storms with varying irregular wave conditions and water levels. Seven unique long- 

 term damage test series were conducted. The first. Series A', was a long duration test 

 with varying water level and wave height. In this first series, the water level was 

 increased midway through the test. The wave heights were increased in five increments. 

 Damage was allowed to reach an observed apparent equilibrium level before the wave 

 height or water level was changed. The second and third series (B' and C) were 

 designed to yield insight into damage progression for variations in storm ordering. 

 During Series B', the water level was increased midway through the test, while for 

 Series C the water level was decreased midway through. For Series B' and C, wave 

 heights were increased in four increments, each after about 4,200 waves or 

 approximately 8 to 10 hr of prototype storm. Series D' and E' were similar to Series B' 

 except with different peak wave periods. Series F' and G' were similar to Series B' and 

 D' except that riprap was used as armoring. 



It was observed during the series that the damage never really stabilized or 

 reached an equilibrium. Also, Series B' and C indicated that the final damage was 



147 



