This work presents the use of artificial neural networks (ANNs) to predict sea surface temperatures (SSTs) in the Western Mediterranean Sea from 1960 to 2005. The ANNs are trained with meteorological variables as input and concurrent satellite-derived SSTs as target.
Our results with ANNs are able to predict well both the seasonal and the interannual variability of SST in the Western Mediterranean Sea, with values mostly within the accuracy of the in situ and satellite SSTs used for validation. The ANN results within a
selected area of the Western Alboran Sea allow to predict SST maps for time coordinates before SST satellite-availability. The presence or collapse of the western Alboran gyre is detected in that area with good agreement by both the ANN predictions
and concurrent results from a 3-D circulation model. The same methodology allows to reconstruct incomplete SST satellite images.
GARCIA GORRIZ Elisa;
GARCIA-SANCHEZ Joan;
2007-06-07
AMER GEOPHYSICAL UNION
JRC35383
https://publications.jrc.ec.europa.eu/repository/handle/JRC35383,
10.1029/2007GL029888,
| Name | Country | City | Type |
|---|
This document is only visible at the Commission level.
You are not authorized to publish or distribute it outside the European Commission.
This is a public document. You can share this publication.
Datasets
| ID | Title | Public URL |
|---|
Dataset collections
| ID | Acronym | Title | Public URL |
|---|
Scripts / source codes
| Description | Public URL |
|---|
Additional supporting files
| File name | Description | File type |
|---|