Deep Learning Prediction of 30-Day Sea Surface Height Anomalies in the Kuroshio–Extension Region Using a Conv3D Encoder–Decoder Framework

Kalpesh Ravindra Patil

Presentation information

[E] Oral

A (Atmospheric and Hydrospheric Sciences ) » A-TT Technology &Techniques

[A-TT49] Machine Learning Techniques in Weather, Climate, Ocean, Hydrology and Disease Predictions

Wed. May 27, 2026 3:30 PM - 5:00 PM 101 (International Conference Hall, Makuhari Messe)

Chairperson:Martineau Patrick(Japan Agency for Marine-Earth Science and Technology), Jayanthi Venkata Ratnam(Application Laboratory, JAMSTEC)

4:45 PM - 5:00 PM

[ATT49-12] Deep Learning Prediction of 30-Day Sea Surface Height Anomalies in the Kuroshio–Extension Region Using a Conv3D Encoder–Decoder Framework

*Kalpesh Ravindra Patil¹, Toru Miyama², Yasumasa Miyazawa³, Ratnam J. V.², Alessandro Damiani² (1. Researcher, APL, VAiG, JAMSTEC, 2. Senior Researcher, APL, VAiG, JAMSTEC, 3. Deputy Director, APL, VAiG, JAMSTEC)

Keywords:Sea Surface Height Anomaly, Kuroshio, Deep Learning, Northwest Pacific, Mesoscale Eddies, Persistence Model