operation. Such analysis should include such considerations as type of prime 

 power source (turbine, reciprocating, etc.); its inherent balancing and 

 vibration; method of mounting (isolation mounts or coupled solidly to 

 structure); exhaust method (muffler vs. direct exhaust into air); and location 

 and direction of exhaust outlet. The analysis should also consider platform 

 structural features, such as dimensions of legs and other elements \vhich might 

 serve as underwater radiators. Other features for study are the mass, 

 thickness, and damping of structural members. For example, one quiet platform 

 had legs filled with concrete, which might be a factor. This platform also 

 had excellent muffling of its engine exhaust, while in contrast the relatively 

 noisy semi-submersible platform had unmufflod exhaust stacks directed down 

 toward the ocean surface. Of course, it may be misleading to make simple 

 comparisons between these platforms, since they differ in many ways. The 

 semi-submersible, for example, possesses two lar-ge submerged hulls, which 

 could serve as excellent underwater sound radiators. 



Understanding the differences in noise radi.=!tion by the various 

 platforms will require critical analyses of sound and vibration data from a 

 large number of platforms, sampling a wide variety of types, construction 

 details, and operating and environmental conditions. 



VI. CONC LUSIONS 



The following must be considered tent.it ivp, as they are derived fi^om a 

 very elementary, initial analysis of data from a \.'cry small number of 

 platforms, and from calculations using many assumptions, some of which are not 

 yet validated. 



G-34 



