Abstract
The characteristics of wave spectra in the swell-dominated seas in the Chancay Bay, Peru, and off the coast of Hambantota, Sri Lanka, were studied based on in situ measurements. According to the characteristics of the spectral shapes, the measured wave spectra were divided into single- and double-peaked spectra. For the single-peaked measured spectra, many widely used wave spectrum models were adopted to fit the measured wave spectra. By comparing the fitting results with the measured wave spectra, these wave spectra were insufficient for describing the peak enhancement or the high-frequency tail shape of the measured spectra. The spectral shape parameters were found to play an important role in the expression of the spectrum models. The peak enhancement factor γ in the Joint North Sea Wave Project (JONSWAP) spectra could adjust the peak height of the spectrum and the shape parameter m of the spectrum high-frequency tail in the Wallops spectra made the wave spectrum more flexible. Therefore, an improved single-peaked wave spectrum was proposed by combining the JONSWAP and Wallops spectra. It performed better than other spectrum models in describing the swell and has advantages in both the wave characteristic parameters and wave spectral shapes. The spectrum peak width parameter σ was relatively consistent in the two measurement stations and both around 0.24. Moreover, the proposed model overcame the limitations of the excessively high estimated slope in the high-frequency range of the Wallops spectrum; the shape parameter m of the spectrum high-frequency tail is primarily distributed in the range (1, 5). For the double-peaked measured spectra, the Ochi–Hubble spectrum was found to work better than the Torsethaugen spectrum.
© 2022 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).