Spectral Analysis of Waves Produced by HSC for Coastal Management Benassai G. 1 , Piscopo V. 2* , Scamardella A. 2 1 The University of Naples “Parthenope”, Department of Engineering, Centro Direzionale Isola C4, 80143 Naples, Italy 2 The University of Naples “Parthenope”, Department of Science and Technology, Centro Direzionale Isola C4, 80143 Naples, Italy guido.benassai@uniparthenope.it, vincenzo.piscopo@uniparthenope.it, antonio.scamardella@uniparthenope.it *Corresponding author: Tel.: +39-081-5476590/5476684; fax: +39-081-5476414 Abstract The present study focuses on wake wash management in the Bay of Naples by time variant spectral analysis, establishing more reliable operational strategies and mitigative measures. A time frequency analysis of several wake wash signals due to high speed craft (HSC), both catamarans and monohulls, operating in the Gulf of Naples is carried out in order to derive the energy per unit area distribution versus time and better characterize the wake wash event and the relevant spectrum. Subsequently a new procedure, based on spectral analysis, is proposed to evaluate both wake wash height and energy density, that also allows to exclude the wind sea wave component, that cannot be deduced by classical procedures based on time history analysis. In this respect, in fact, the influence of wind sea waves may lead to more reliable coastal management guidelines and wake wash measures, even if the measurement campaign has been carried out under calm met-ocean conditions. A comparison with wake wash values determined by time history analysis is finally carried out, showing that the spectral analysis can be applied with good confidence for coastal management purposes, leading to more accurate results and more reliable operational strategies. Key-words: Spectral analysis; Wake wash; High Speed Craft (HSC); International wake wash Rules 1. Introduction Wake wash waves generated by high speed crafts may seriously damage the coastal environment and jeopardize people’s safety and property (Parnell and Kofoed-Hansen, 2001; Soomere, 2005, 2006). High speed vessels, in fact, generate wake wash waves with patterns different from those ones produced by conventional ships and described by the classical Kelvin wave model. Besides, the relevant energy characterizes high length waves, slightly visible in open sea, but rather evident when impacting the coast in shallow waters, where they become higher and higher up to the breaking down. In order to manage the coastal risk induced by wake wash, scientific community provided new measurement programmes (Croad and Parnell, 2002) and analysis techniques for wake wash prediction, based on fluid mechanics. Anyway it was found (Parnell et al., 2007) that wake wash analyses are difficult to apply, as wake waves vary in their basic parameters even during a single wake event, so that a significant variability in data records for the same vessel at different times but at the same locations is generally recognized. In other words, it is difficult to determine a general characterization of such wakes near the coast, due to the transient and nonlinear nature of this phenomenon, and the fact that wake impact is influenced by the local bathymetry and coastline configuration. Furthermore, differently from the classical (deep-water) Kelvin wave, the wakes excited at relatively high speeds may contain a substantial amount of non-stationary components. Owing to the finite spatial extension of such wakes (Torsvik et al., 2009 a,b , Didenkulova et al., 2013), the non-stationary components frequently dominate in the nearshore, in particular within the close proximity of the ship track. Under these transient conditions, the use of traditional stationary-process (ergodic) statistical methods fails to provide adequate information about the wake properties and impact. Sheremet et al. (2012) suggested the usage of a time-frequency representation (TFR) to analyse ship wake events, demonstrating that TFR methods can be applied to the analysis of ship wake signals, even in the presence of a broad-band wind-wave background. In the present study the signals of fast ferries wakes recorded in the Gulf of Naples during Spring and Summer 2007 are reanalysed, in order to provide essential new information about the structure of wakes from fast ferries that cannot be extracted by classical methods. The wake wash signal is analysed by means of a time-frequency technique (windowed Fourier Transform), which permits to identify the relevant properties of wakes generated by both catamarans and mono-hulls, the former presenting a first group of longer leading waves followed by a secondary, shorter and less energetic wave group, the latter exhibiting only a group of leading waves, which are generally shorter than those ones produced by catamarans (Begovic et al., 2007 a,b ). Subsequently the wave energy per unit area is plotted versus time, in order to determine the interval when maximum energy and so wake wash occurs. Finally, wake wash wave spectra are derived in order to extend frequency domain analysis to coastal management purposes. In this respect, in fact, as wake wash values required by International Rules are generally overestimated, in terms of both height and energy density, by conventional time history analysis of wave height and period, the more comprehensive frequency domain analysis permits to obtain more realistic values, as it allows to remove the wind sea wave disturbances from the wake wash signal. In this respect a new procedure, based on frequency domain analysis, is proposed to evaluate both wake wash height and energy density. Subsequently, the proposed procedure is tested by a comparison with relevant results obtained by time history analysis (Benassai et al., 2013) and finally the influence on wake wash management, in terms of minimum distance of navigation tracks from coastline is discussed.