scatter widely. They concluded that the uncertainty was due to differing construction 

 techniques from test to test and different random number seeds being used to generate 

 the spectra. 



Font (1968) noted that the armor placement technique is important for 

 determining when initiation of damage occurs, but is less relevant for advanced damage, 

 which is similar to conclusions stated above by Thompson and Shuttier. It is anticipated 

 that the scatter in damage will be large for advanced damage primarily because of the 

 random nature of the interaction between irregularly shaped stones. Melby and Mlakar 

 (1997) showed how Carver and Wright's data could be used to compute the reliability 

 of a breakwater with respect to the zero-damage condition of armor stability and how 

 Van der Meer's relation could be used to compute reliability of a damaged mound. But 

 little work has been done on computing the reliability of a damaged rubble mound 

 through simulated damage using historical storm data. 



51 



