but underpredicts damage up to A'.^-SOOO. For a greater number of waves, this equation 

 overpredicts damage. 



The empirical coefficients in Equations 6.13 and 6.14 are recalculated in the 

 following using the present experimental data, which were obtained from tests much 

 longer in duration than those listed in Table 6.1. The wave heights and periods for the 

 present experiment were listed in Tables 4.3 and 4.4 and the experimental data are 

 summarized in Appendix B. In order to get a more gradual increase in S(t) it was 

 necessary to raise a and lower b. The final coefficients were a^ = 0.025 and b = 0.25 for 

 Equation 6.13 and Op = 0.022 and b = 0.25 for Equation 6.14 yielding the general 

 predictive relations for mean damage progression due to breaking waves for time and 

 frequency wave statistics, respectively, as 



5(0 = 5(r„) + 0.025(iV/C'' C^"'' " C^') M t„<t< /„,, (6.15) 



S{t) = S{t,) + 0.022{NJ'T;''' it'-'' - ?;■") for t„<t< r„,, (6.16) 



Figure 6.7 shows Equation 6.15 (dashed line) and Equation 6.16 (solid line) plotted 

 against the damage data of Figure 5.10 for Series A'. The mean damage is well 

 predicted by Equations 6.15 and 6.16 with a correlation coefficient of r = 0.99 for both 

 equations for this series. Similarly, Figures 6.8 and 6.9 show that the equations predict 

 damage well for Series B' and C, with correlation coefficients of r = 0.99 for both. 

 Equations 6.15 and 6.16 are therefore generalized for Series A', B', and C. The 



124 



