Oceanic eddy detection and lifetime forecast using machine learning methods

• Published in: Geophysical research letters 2016-12, Vol.43 (23), p.12,234-12,241
• Main Authors: Ashkezari, Mohammad D., Hill, Christopher N., Follett, Christopher N., Forget, Gaël, Follows, Michael J.
• Format: Article
• Language: English
• Subjects: Altimeters; Altimetry; Anomalies; Artificial intelligence; Aversion learning; Components; Detection; Eddies; eddy; eddy lifetime; Geophysics; Identification; Identification methods; Learning algorithms; Learning behaviour; Machine learning; Marine; Mathematical models; ocean; Ocean circulation; Oceanography; remote sensing; Robustness; Serial learning; Velocity; Vortices
• ISSN: 0094-8276; 1944-8007
• DOI: 10.1002/2016GL071269

We report a novel altimetry‐based machine learning approach for eddy identification and characterization. The machine learning models use daily maps of geostrophic velocity anomalies and are trained according to the phase angle between the zonal and meridional components at each grid point. The trained models are then used to identify the corresponding eddy phase patterns and to predict the lifetime of a detected eddy structure. The performance of the proposed method is examined at two dynamically different regions to demonstrate its robust behavior and region independency. Key Points The phase angle between the components of the geostrophic velocity anomalies is proposed to identify eddy structures Machine learning models are trained to detect the phase patterns of eddies; we demonstrate that this method is largely region independent The proposed method eliminates the previously reported issue of eddy misclassification and is also capable of estimating the eddy lifetime