Clearly the increased damping makes a significant difference in the 

 results . 



Some insight into the performance of this breakwater may be gained 

 by following the theoretical results as a function of B/L. At very low 

 B/L the fixed-body transmission dominates. The trough at a B/L of 0.4 

 comes mainly from the interaction of the roll-generated wave and the 

 fixed-body transmission. For the next peak at B/L of 0.5 the roll- 

 generated wave dominates as the roll resonant frequency is encountered. 

 The next trough at B/L of 0.65 is a result of all three motion-generated 

 waves interacting with the fixed-body transmission making only a rela- 

 tively small contribution to the transmitted wave. The following peak 

 at B/L of 0.7 results from interactions among the motion-generated waves 

 which are of about equal magnitude. At a B/L of 0.86 the heave- generated 

 wave dominates again, but as B/L increases beyond this the effect of 

 heave and roll are rapidly decreasing while sway motion is becoming the 

 dominant wave-generating mechanism. In the region of B/L between 0.4 

 and 1.0, changing the physical properties of the breakwater can have 

 a marked effect in shifting the peaks and troughs by altering the heave 

 and roll resonant frequencies. 



Experience with linear ship motion theory has shown that the worst 

 agreement between predicted and measured motions occurs when rolling mo- 

 tions are considered (Salvesen, 1970) . This discrepancy is often over- 

 come by arbitrarily increasing the computed roll damping to compensate 

 for the viscous damping which is neglected. As indicated in Figure 10, 

 when damping is added the theory gives a better prediction where roll 

 motion plays a significant role. This places a significant restriction 

 on the theory requiring careful monitoring of predicted roll motion. 

 Where the theory predicts large roll motion, additional damping will be 

 required to obtain results comparable to measurements. 



Figure 11 shows the predicted fixed-body transmission coefficient 

 and the results of model tests. Agreement is quite close except at B/L 

 of 0.78. The peak in predicted transmission may be due to a resonance 

 of the waves within the well of the catamaran breakwater. There is 

 another peak near B/L of 1.4 indicating the presence of higher harmonic 

 resonances as well. Model tests show at least a slight hump in this 

 region suggesting that the theoretical prediction clearly overestimates 

 the effect of this phenomena, but that this probably is occurring in 

 real life. 



For the data measured in the field, the transmission coefficient is 

 defined as the square root of the transmitted wave spectral density 

 divided by the incident wave spectral density. Figure 12 shows the 

 transmission coefficient derived from the data obtained at the Tenakee,- 

 Alaska breakwater. The theoretically predicted transmission coefficient 

 with the computed hydrodynamic damping doubled is also shown for com- 

 parison. Details of the technique used in the spectral analysis of the 

 field data may be found in Section III. 



32 



