Spectral analysis of sea surface elevations produced by big storms: The case of hurricane Wilma

Abstract

Hurricane Wilma was the most extreme hydro-meteorological event ever to occur in the Caribbean Sea. During its occurrence, the sea surface elevation was recorded close to Puerto Morelos, Quintana Roo, Mexico. The equipment used which features both, acoustical and pressure sensors to measure the water depth, was deployed approximately at a depth of 20 m so it gathered invaluable information that helps understanding the nature of this cyclone and provided unique data related to the hurricane from which various different analyses can be performed. Wavelet and Hilbert Huang Transform (HHT) spectra are the frequency domain analysis techniques presented in this paper. Together with the well-known and elsewhere reported simultaneous appearance of Sea and Swell waves, several other sea surface oscillations were produced by the cyclone. These wave dynamics were found in the results of the wavelet and HHT analyses of the free sea surface elevations time series before, during and after the hurricane passed the observation point (a total of 13 days). The wavelet analysis showed the time evolution of the energy distribution over an important range of frequencies. Two main features can be highlighted: the large number of wave periods recorded when the eye of the cyclone was close to the measuring point and the occurrence of a very long wave, with a period of around 109 h containing great energy. In turn, the HHT analysis showed its robustness, being capable of detecting long oscillations within time series shorter than the oscillation itself.