Date of Award
Master of Science in Ocean Engineering
Lester R. LeBlanc
In this work, time histories of North Pacific wave spectra estimated from NOAA Data Buoy Office (NDBO) measurements and spectra hindcast by the Spectral Ocean Wave Model (SOWN) are analyzed by multivariate statistical methods. Eigenfunctions are used to provide a reduced description of the variance in the wave data sets. Previous work has shown that five functions often represent more than 90 percent of the variance.
Eigenfunctions (EOFs) representing 13 years of hindcast and 11 months of buoy data are compared to evaluate the use of eigenfunctions to describe the spectra in remote locations, where buoy data is not available. The hindcast functions are used to reconstruct the buoy observations, and the root mean square error between the actual and reconstructed spectra is discussed.
The use of eigenfunctions may reduce the amount of information required for transmission by a buoy to construct the spectra. The eigenfunctions could be stored onboard the buoy and onshore. If the observed spectra is calculated onboard the buoy, a set of coefficients could be evaluated and transmitted to shore for use with the eigenfunctions and a mean to reconstruct the spectra. The functions could make a statistical description of the wave environment available to users without access to a large computer.
This study also provides a comparison of wave parameters and energy as calculated by the SOWM and NDBO observations.
Hinchey, Laurence R., "ANALYSIS OF A REDUCED REPRESENTATION OF OCEAN WAVE SPECTRA" (1986). Open Access Master's Theses. Paper 2395.