Oceanographyproject

Today satellites provide a surface signature of the temperature with a high spatial frequency: ie a good horizontal resolution but a low vertical resolution. Thanks to the ARGO database collected by buoys making vertical surveys, one has precise but sparse knowledge of the vertical thermal structure of the ocean. The objective of this project is to develop a methodology to statistically combine additional information to obtain a 3D time series sufficiently resolved horizontally and vertically to follow the eddies. It is based on a regression classification method that allows both to classify the temperature profiles and to propose a linear model between the satellite observations and the information of the buoys according to its label.