In the present work a weakly nonlinear wave model originally developed by Rebaudengo Lando` et al (1996) is applied to the transformation of wave spectra from offshore to nearshore, and subsequently, it has been systematically applied to the derivation of long-term time series of spectral wave parameters on decreasing depth from corresponding offshore wave data. The derived long-term series of nearshore parameters have been used as input to a new method, recently developed by Stefanakos & Athanassoulis (2006), for calculating return periods of various level values from nonstationary time series data. The latter method is based on a new definition of the return period, that uses the MEan Number of Upcrossings of the level x* (MENU method), and it has been shown to lead to predictions that are more realistic than traditional methods. To examine the effects of bottom topography on the nearshore extreme value predictions, Roseau (1976) bottom profiles have been used for which analytical expressions are available concerning the reflection and transmission coefficients. A parametric (JONSWAP) model is used to synthesize offshore spectra from integrated parameters, which are then linearly transformed based on the previous transmission coefficient to derive first-order nearshore wave spectra. Second-order random sea states have been simulated by following the approach of Hudspeth & Chen (1979) (see also Langley 1987, Lando et al 1996), exploiting the quadratic transfer functions on decreasing depth to calculate the second-order nearshore spectra. Finally, wave parameters are extracted from the nearshore spectra by calculating the first few moments.
FORMULATION AND APPLICATION OF FREQUENCY-DOMAIN NONLINEAR WAVE TRANSFORMATION MODEL INCORPORATING INCIDENT AND REFLECTED WAVE COUPLING
