CoAuthors

François Kaly (Laboratoire de Traitement de l'Information (LTI), Senegal); Sylvie Thiria (LOCEAN UMR CNRS/IRD/UPMC/MNHN, France)

The western tropical Atlantic ocean is a very energetic and highly variable region. It is one of the main contributors to the inter-hemispheric mass and heat transports. The aim of this study is to give a new picture of the space and time variability of this region using statistical tools applied to satellite measurements such as radar altimeters (TOPEX/Poséïdon/Jason series...), Soil Moisture and Ocean Salinity (SMOS) radiometer, the Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA) products and the European Center for Medium Range Weather Forecasts (ECMWF) wind stresses. The investigated variables are thus the Sea Surface Temperature (SST), the Sea Level Anomalies (SLA), the Sea Surface Salinity (SSS) and the wind stresses between 70°W-30°W, 0°N-15°N, from 2010 to 2016. The Self-Organizing Maps (SOM) approach, a clustering methodology based on non-linear artificial neural network, is combined with a Hierarchical Ascendant Classification to classify the different phenomena located in that area and to identify their characteristics. Three classes are identified and allowed us to focus on the dynamics of the North Brazil Current, and the North Equatorial CounterCurrent, respectively, and their links with the InterTropical Convergence Zone and the Amazon and Orinocco river runs off.