Statistical Analysis of Non-Stationary Waves off the Savannah Coast, Georgia, USA

Abstract

The statistical characteristics of short-term wave records collected from a wave buoy deployed off the coast of Savannah, Georgia, USA are analyzed. The statistical distribution of observed surface elevation is derived and compared with theoretical estimations. The minimum length of time series for stationary conditions is specified using a synthetic approach. A new technique that considers the variability of different dimensionless parameters has been introduced to improve the theoretical distributions of wave crests, troughs and heights. The observed distributions of wave crests and troughs are computed and compared with the Rayleigh distribution as well as the Tayfun distribution. The characteristic wave heights of measured data are compared with those estimated from the Boccotti distribution over different time intervals. The empirical estimation of expected maximum wave heights as a function of the number of waves is examined using a statistical approach, and compared with results from both Boccotti and Tayfun models using the newly introduced technique. Conclusions concerning the efficacy and applicability of theoretical distributions to real ocean wave conditions are summarized.