Detrended fluctuation analysis of significant wave height time series

Abstract

Mean daily significant wave height time series are analysed by means of the detrended fluctuation analysis (DFA) method to examine the existence of long-range correlations, or long memory. The results indicate that the scaling behavior of significant wave height fluctuations is not constant over the considered time scales but there is a decrease of the scaling exponent with increasing time scale. Two crossover times have been identified, indicating three different scaling behaviors. Fluctuations associated to time scales between 10 and 100 days, approximately, show long-range correlation while fluctuations above and below this range present white and Brownian noise-like behavior, respectively.