Wave parameters and functions in wavelet analysis with filtering

This is a sequel to Huang’s paper entitled ‘Wave parameters and functions in wavelet analysis’ (Ocean Engineering, 2004, 31(1), 111–125) preliminary study of wave parameters and functions in wavelet analysis. Both discrete wavelet transform and median filtering techniques are applied to remove spiky noise from wind wave data obtained from a coastal observation tower. Fourier band-pass filters are also applied to remove low frequency drift and high frequency fluctuations. The original data and processed data are then analyzed by the time-domain, frequency-domain, and Morlet wavelet transform procedures. Two non-dimensional wave parameters, wave generation index NIT and non-stationary index NIF, are calculated from the Morlet wavelet transform and normalized with respect to the number of waves in each record, and the functional dependency of these two parameter on various wave statistics is reexamined in this study. Regression results confirm the effectiveness of the normalized wave generation index in describing the wave growth and decaying phenomena. It is also found that the normalized non-stationary index is linearly correlated with the time-domain groupiness factor per wave group and therefore can be used as another basic descriptor of wind waves in wavelet analysis.