4.7 Article

Assessment of SWAN and WAVEWATCH-III models regarding the directional wave spectra estimates based on Eastern Black Sea measurements

Journal

OCEAN ENGINEERING
Volume 272, Issue -, Pages -

Publisher

PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.oceaneng.2023.113944

Keywords

Directional wave spectra; Spectral wave estimates; SWAN; WaveWatch III; Black Sea; Wave modelling

Ask authors/readers for more resources

The study evaluates the accuracy of SWAN and WAVEWATCH-III (WWIII) models for directional wave spectra estimates based on measurements conducted in the eastern Black Sea. It is found that the SWAN model fits better the average variance density shape at all frequency ranges, but slightly overestimates in the 0.13-0.17 Hz range. On the other hand, the WWIII model shows better correspondence in the 0.13-0.17 Hz frequency range, but underestimates the lower frequency energies and overestimates higher frequency ranges. Therefore, the SWAN model is recommended for spectral density estimation at seasonal and annual scales.
The spectral wave models are often calibrated or validated based on bulk wave data derived from spectral in-formation such as the significant wave height (Hs) and mean period (Tm). However, the precision in the hind-casted spectral data was not evaluated in many seas, such as the Black Sea, due to the lack of spectral wave measurements. As part of this study, a spectral wave measuring buoy was installed near the cape Small Utrish to collect the spectral measurement data. The study objective is to assess the SWAN and WAVEWATCH-III (WWIII) models regarding the directional wave spectra estimates based on measurements conducted in the eastern Black Sea. For this purpose, the estimated spectral density distribution and spectral mean direction in terms of spectral frequency bins are assessed for both SWAN and WWIII outputs. The biases in variance densities are firstly calculated for both models at each frequency based on 14 710 spectral measurements, as well as the biases in the annual and seasonal averages. In addition, the correlation coefficients and root mean squared errors at each frequency are calculated and discussed. The accuracies in the mean direction at each spectral shape over the whole measurement period are secondly evaluated. Thirdly, the annual and seasonal averaged two-dimensional wave spectrums are discussed compared to measurements. The performances of both models are lastly verified based on bulk data measured at 8 locations on different coasts of the Black Sea. The results show that SWAN output fits better the average variance density shape at all frequency ranges f > 0.08Hz. However, it slightly overestimates the variance density in the 0.13-0.17 Hz range. In contrast to the SWAN model, WWIII data show only better correspondence in the 0.13-0.17 Hz frequency range but underestimate the lower frequency energies and overestimate higher frequency ranges. The SWAN model is therefore recommended for the spectral density estimate at seasonal and annual scales.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available