Multiscale Spatio-Temporal Attention Network for Sea Surface Temperature Prediction

Accurate prediction of sea surface temperature (SST), a crucial indicator of global climate and ecosystem changes, holds significant economic and social benefits. Deep learning has shown preliminary success in modeling the dynamic spatial-temporal dependencies within SST signals, yet it remains chal...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhenxiang Bai, Zhengya Sun, Bojie Fan, An-An Liu, Zhiqiang Wei, Bo Yin
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10844304/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849724385624064000
author Zhenxiang Bai
Zhengya Sun
Bojie Fan
An-An Liu
Zhiqiang Wei
Bo Yin
author_facet Zhenxiang Bai
Zhengya Sun
Bojie Fan
An-An Liu
Zhiqiang Wei
Bo Yin
author_sort Zhenxiang Bai
collection DOAJ
description Accurate prediction of sea surface temperature (SST), a crucial indicator of global climate and ecosystem changes, holds significant economic and social benefits. Deep learning has shown preliminary success in modeling the dynamic spatial-temporal dependencies within SST signals, yet it remains challenging to obtain precise SSTs due to the inherent variabilities across multiple temporal and spatial scales, driven by distinct physical processes. In this paper, we propose a novel multi-scale spatio-temporal attention network, named MUSTAN, tailored for the SST prediction problem. MUSTAN achieves multi-scale fusion through a progressive scale expansion paradigm, where sub-scale representations are iteratively merged with its counterpart scale units, enabling the propagation of fine-scale SST changes across broader scales. For each scale, MUSTAN introduces temporal attention to characterize dynamic SST patterns in different ocean regions, and spatial attention to capture intricate SST evolution interplay among these regions. Extensive experiments conducted on datasets from the Bohai Sea, Yellow Sea, and South China Sea consistently validate the effectiveness and superiority of our design, outperforming the state-of-the-art methods on SST prediction tasks.
format Article
id doaj-art-dde619a7fbd24bffa45ef86d634ea301
institution DOAJ
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-dde619a7fbd24bffa45ef86d634ea3012025-08-20T03:10:46ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01185866587710.1109/JSTARS.2025.353112210844304Multiscale Spatio-Temporal Attention Network for Sea Surface Temperature PredictionZhenxiang Bai0https://orcid.org/0009-0002-6381-5015Zhengya Sun1https://orcid.org/0000-0003-2381-6363Bojie Fan2An-An Liu3https://orcid.org/0000-0001-5755-9145Zhiqiang Wei4https://orcid.org/0000-0002-2830-8301Bo Yin5https://orcid.org/0000-0001-6318-0174Department of Computer Science and Technology, Ocean University of China, Qingdao, ChinaUniversity of Chinese Academy of Sciences, Beijing, ChinaDepartment of Computer Science and Technology, Ocean University of China, Qingdao, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin, ChinaDepartment of Computer Science and Technology, Ocean University of China, Qingdao, ChinaDepartment of Computer Science and Technology, Ocean University of China, Qingdao, ChinaAccurate prediction of sea surface temperature (SST), a crucial indicator of global climate and ecosystem changes, holds significant economic and social benefits. Deep learning has shown preliminary success in modeling the dynamic spatial-temporal dependencies within SST signals, yet it remains challenging to obtain precise SSTs due to the inherent variabilities across multiple temporal and spatial scales, driven by distinct physical processes. In this paper, we propose a novel multi-scale spatio-temporal attention network, named MUSTAN, tailored for the SST prediction problem. MUSTAN achieves multi-scale fusion through a progressive scale expansion paradigm, where sub-scale representations are iteratively merged with its counterpart scale units, enabling the propagation of fine-scale SST changes across broader scales. For each scale, MUSTAN introduces temporal attention to characterize dynamic SST patterns in different ocean regions, and spatial attention to capture intricate SST evolution interplay among these regions. Extensive experiments conducted on datasets from the Bohai Sea, Yellow Sea, and South China Sea consistently validate the effectiveness and superiority of our design, outperforming the state-of-the-art methods on SST prediction tasks.https://ieeexplore.ieee.org/document/10844304/Data modelingdeep learningmultiscale spatio-temporal predictionsea surface temperature (SST)
spellingShingle Zhenxiang Bai
Zhengya Sun
Bojie Fan
An-An Liu
Zhiqiang Wei
Bo Yin
Multiscale Spatio-Temporal Attention Network for Sea Surface Temperature Prediction
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Data modeling
deep learning
multiscale spatio-temporal prediction
sea surface temperature (SST)
title Multiscale Spatio-Temporal Attention Network for Sea Surface Temperature Prediction
title_full Multiscale Spatio-Temporal Attention Network for Sea Surface Temperature Prediction
title_fullStr Multiscale Spatio-Temporal Attention Network for Sea Surface Temperature Prediction
title_full_unstemmed Multiscale Spatio-Temporal Attention Network for Sea Surface Temperature Prediction
title_short Multiscale Spatio-Temporal Attention Network for Sea Surface Temperature Prediction
title_sort multiscale spatio temporal attention network for sea surface temperature prediction
topic Data modeling
deep learning
multiscale spatio-temporal prediction
sea surface temperature (SST)
url https://ieeexplore.ieee.org/document/10844304/
work_keys_str_mv AT zhenxiangbai multiscalespatiotemporalattentionnetworkforseasurfacetemperatureprediction
AT zhengyasun multiscalespatiotemporalattentionnetworkforseasurfacetemperatureprediction
AT bojiefan multiscalespatiotemporalattentionnetworkforseasurfacetemperatureprediction
AT ananliu multiscalespatiotemporalattentionnetworkforseasurfacetemperatureprediction
AT zhiqiangwei multiscalespatiotemporalattentionnetworkforseasurfacetemperatureprediction
AT boyin multiscalespatiotemporalattentionnetworkforseasurfacetemperatureprediction