PCA-Enhanced Adaptive NVAR Framework for High-Resolution Sea Surface Temperature Forecasting in the East Sea

2026-06-10

AI Summary

This study addresses the high computational cost of traditional numerical ocean models and the error accumulation plaguing existing deep learning approaches in long-term forecasting of high-dimensional spatiotemporal ocean data. To overcome these limitations, the authors propose a reduced-order forecasting framework that integrates singular value decomposition (SVD) with adaptive next-generation reservoir computing (Adaptive NVAR). The method first compresses sea surface temperature fields into low-dimensional dominant modes via SVD and then models their temporal evolution using Adaptive NVAR to reconstruct forecast fields. This work presents the first successful application of Adaptive NVAR to a real-world ocean system, demonstrating consistently superior performance over standard NG-RC/NVAR across multiple forecast horizons. The proposed approach significantly reduces prediction errors while enabling efficient, stable, and scalable real-time sea surface temperature forecasting.

Abstract

Accurate forecasting of sea surface temperature (SST) in regional seas such as the East Sea is crucial for monitoring marine ecosystems, assessing climate risks, managing fisheries, and conducting naval operations. Traditional numerical ocean models provide reliable predictions but are computationally expensive and often unsuitable for real-time forecasting. Many deep learning methods also struggle with high-dimensional spatiotemporal ocean data and experience error accumulation over longer forecasting periods. This study builds on our previously proposed Adaptive Next-Generation Reservoir Computing (Adaptive NVAR) framework, initially introduced and tested on synthetic dynamical systems, and extends it to ocean forecasting. We present a reduced-order forecasting framework that combines Singular Value Decomposition (SVD) with Adaptive NVAR to predict SST dynamics in the East Sea. SST fields are compressed into a low-dimensional representation using SVD, which extracts dominant modes of ocean variability. Adaptive NVAR models the temporal evolution of these latent states, and the predicted states are reconstructed into SST forecasts. We evaluate the framework using regional ocean datasets and compare it with the standard NG-RC/NVAR. Results show that Adaptive NVAR consistently achieves lower forecasting errors across multiple prediction horizons. In addition, SVD reduces computational complexity, resulting in a fast and scalable framework suitable for real-time ocean forecasting.

Problem

• sea surface temperature forecasting
• high-dimensional spatiotemporal data
• real-time ocean forecasting
• error accumulation
• computational complexity

Innovation

• Adaptive NVAR
• Singular Value Decomposition
• Sea Surface Temperature forecasting
• Reduced-order modeling
• Reservoir computing