Abstract's details

A wave investigation in the tropical Atlantic Ocean using satellite altimetry data and Hilbert transform based methods

Jean Luc Mélice (LOCEAN UMR CNRS/IRD/UPMC/MNHN, France)

CoAuthors

Sabine Arnault (LOCEAN UMR CNRS/IRD/UPMC/MNHN, France)

Event: 2014 Ocean Surface Topography Science Team Meeting

Session: Science Results from Satellite Altimetry: Regional and basin-scale processes and sea level rise

Presentation type: Type Poster

Contribution: not provided

Abstract:

The variability of the tropical Atlantic Ocean is investigated with weekly altimetric Absolute Dynamic Topography data from 1992 to January 2011 at 1/3° grid resolution. An Empirical Orthogonal Function analysis reveals three regions of high variability. The first region, between 3°N and 11°N, is characterised by the presence of strong eddies linked to the North Equatorial CounterCurrent retroflection in the vicinity of the brasilian coast. In the second region, we observed the presence of instability waves centered at 4°N between 40°W and 10°W. The third region, between 3°S and 3°N is characterised by the presence of equatorial Kelvin and Rossby waves.
We explore the characteristics of these waves with Hilbert transform based methods. These techniques allow us to estimate the instantaneous amplitude, period, and speed of the propagations. We show that these quantities are correctly estimated, and are physically valid if the signals are asymptotic, and are decomposed in their high (from 2 to ~8 weeks), and low (from ~8 to 52 weeks) frequency asymptotic components. For this purpose we develop a filtering technique based on the Empirical Mode Decomposition method. The estimated mean speed of the propagating waves are ~19 cm/s, for the eddies close to the brasilian coast, ~50 cm/s, for the instability waves at 4°N, ~181 cm/s, for the Kelvin waves at the Equator, ~61 cm/s for the Rossby waves at 3°N, and ~48 cm/s at 3°S. All of these waves are characterized by the presence of a strong annual cycle in their instantaneous amplitude, and speed.

 
Jean Luc Mélice
LOCEAN UMR CNRS/IRD/UPMC/MNHN
France
jlmelice@locean-ipsl.upmc.fr