Eif
xf er =
(1)
Eof
where Eif is the energy per frequency transferred inside and Eof is the incident
energy per frequency from outside the breakwater.
If the breakwater is working well, the overall transfer rate should be low. Figure 4,
bottom plot, shows that for the initial hours of the storm, the xfer is less than .15 for all
wave periods <10 seconds for IN002. This means less than 15% of the outside energy
is making it into the harbor at those frequencies. The long period xfer is <1.0 which
indicates that the outside long period energy (>20 seconds) is still more than that on
the inside.
After the first six hours of the storm the Hm0 stays around 2 meters for the next 8
hours and the Tp has shifted from 6 to 8 seconds at IN001 (Figures 2&3, top plots). Lit-
tle incident energy is evident for wave periods >20 seconds. On the inside at IN002,
figures 2&3,bottom plots, show that the bimodal spectrums have also grown. Since
long period harbor oscillations cause ship movement, the energy for periods >20 sec-
onds is of particular interest. There appears to be a relationship between the total
energy of the spectrum and energy >20 second.
Transfer factors for 1/6/03 1300 to 1/7/03 000 GMT are shown in figures 5&6,
bottom plots, for IN002. For some of the periods > 20 seconds, the transfer factor is
greater than 1.0. Another observation from these plots is that the transfer factors for
waves <10 seconds appear to be of the same magnitude throughout the storm.
Figure 7 shows the directional wave statistics for January 2003. This plot is an
example of the wave climate for IN001 and shows that there were numerous times
during the month that the Hm0 exceeded 0.5 meter. In almost all of these cases, the Tp
was greater than 5 seconds. Figure 8, bottom plot, shows the mean transfer rates for
each frequency at IN002 for outside wave records with Tp >5 seconds. The dashed
line represents a one standard deviation band. On average, the breakwater reduces the
energy transferred for all frequencies. As expected, the breakwater performs best for
short period waves. Figure 9 shows the short period portion of the Figure 8. Mean
transfer rates for waves between 5 and 10 seconds is less than 5%.
Portions of the breakwater has undergone repair. Figures 8&9,top plots, illustrate
mean transfer rates for a gage placed in front of a repaired section. Energy is, on aver-
age, reduced for all frequencies. In particular, energy below 10 seconds is reduced by
50% when compared to results from a gage placed in front of an non-repaired section
(Figures 8&9, bottom plots).
For more information, contact: James P. McKinney or Margaret A. Sabol, CEERD-
HC-SO
References
[1] P. D. Welch, "The Use of Fast Fourier Transform for the Estimation of Power
Spectrum: A Method Based on Time Averaging Over Short, Modified Peri-
odogams," IEEE Transactions on Audio and Electroacoustics, June 1967.
2