SEA-ViT: Sea Surface Currents Forecasting Using Vision Transformer and GRU-Based Spatio-Temporal Covariance Modeling

TL;DR

SEA-ViT is a novel deep learning model combining Vision Transformer and GRUs to accurately forecast sea surface currents using high-frequency radar data, accounting for complex spatio-temporal and climatic dependencies.

Contribution

The paper introduces SEA-ViT, integrating ViT and GRUs for improved spatio-temporal sea surface current forecasting with climate index incorporation.

Findings

Enhanced forecasting accuracy demonstrated on 30-year dataset.

Effective modeling of climate influences like ENSO phases.

Open-source code and pretrained models provided.

Abstract

Forecasting sea surface currents is essential for applications such as maritime navigation, environmental monitoring, and climate analysis, particularly in regions like the Gulf of Thailand and the Andaman Sea. This paper introduces SEA-ViT, an advanced deep learning model that integrates Vision Transformer (ViT) with bidirectional Gated Recurrent Units (GRUs) to capture spatio-temporal covariance for predicting sea surface currents (U, V) using high-frequency radar (HF) data. The name SEA-ViT is derived from ``Sea Surface Currents Forecasting using Vision Transformer,'' highlighting the model's emphasis on ocean dynamics and its use of the ViT architecture to enhance forecasting capabilities. SEA-ViT is designed to unravel complex dependencies by leveraging a rich dataset spanning over 30 years and incorporating ENSO indices (El Ni\~no, La Ni\~na, and neutral phases) to address the intricate relationship between geographic coordinates and climatic variations. This development enhances the predictive capabilities for sea surface currents, supporting the efforts of the Geo-Informatics and Space Technology Development Agency (GISTDA) in Thailand's maritime regions. The code and pretrained models are available at \url{https://github.com/kaopanboonyuen/gistda-ai-sea-surface-currents}.