Journal article
A comparison of methods for estimating directional spectra of surface waves
M Donelan, A Babanin, E Sanina, D Chalikov
Journal of Geophysical Research Oceans | AMER GEOPHYSICAL UNION | Published : 2015
DOI: 10.1002/2015JC010808
Abstract
Three methods of estimating the directional spectra of water waves are intercompared. The Maximum Likelihood Method (MLM) and the Maximum Entropy Method (MEM) require stationarity of the time series and yield only the frequency-direction spectra. The Wavelet Directional Method (WDM) does not require stationarity and also yields the wave number-direction spectra and is suitable for event analysis. The comparison includes three cases of wind-generated waves on a large lake and two cases of model-generated waves with different directional spreading. The comparisons of the frequency-direction spectra show that the Wavelet Directional Method yields the best estimates of the directional spectra.
Grants
Awarded by ONR
Funding Acknowledgements
This work was begun while M.D. was a Visiting Professor in the Centre for Ocean Engineering, Science and Technology. He acknowledges a Visiting Professor grant of Swinburne University, and A.B. acknowledges ONR grant N00014-13-1-0278. The numerical simulations were performed at Swinburne University of Technology's supercomputer facilities "Green machine" and "G2." The wave data used may be obtained by writing to the first author (mdonelan@rsmas.miami.edu).