Predicting Sudden Stratospheric Warmings Using Video Prediction Methods

Abstract Sudden Stratospheric Warmings (SSWs) are weather phenomena occurring in polar regions, and have a profound impact on mid‐latitude cold waves. In this paper, within a deep learning framework, we introduce video prediction techniques into SSW events forecasting for the first time. We develop...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuhao Du, Jiankai Zhang, Xinyuan Cheng, Yixiong Lu, Douwang Li, Wenshou Tian
Format: Article
Language:English
Published: Wiley 2025-04-01
Series:Geophysical Research Letters
Subjects:
Online Access:https://doi.org/10.1029/2024GL113993
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849397299201966080
author Yuhao Du
Jiankai Zhang
Xinyuan Cheng
Yixiong Lu
Douwang Li
Wenshou Tian
author_facet Yuhao Du
Jiankai Zhang
Xinyuan Cheng
Yixiong Lu
Douwang Li
Wenshou Tian
author_sort Yuhao Du
collection DOAJ
description Abstract Sudden Stratospheric Warmings (SSWs) are weather phenomena occurring in polar regions, and have a profound impact on mid‐latitude cold waves. In this paper, within a deep learning framework, we introduce video prediction techniques into SSW events forecasting for the first time. We develop a Global Attention Motion Decoupled Recurrent Neural Network (GMRNN) to better capture the detailed changes of the polar vortex. Through experiments on representative SSW events in 2018, 2019, and 2021, our model can stably predict SSW events 20 days in advance and accurately capture the morphological changes of the stratospheric polar vortex. Furthermore, we compared our model with baseline models, including PredRNN, MotionRNN, and the sub‐seasonal to seasonal (S2S) integrated forecast models from ECMWF, CMA, and ECCC. The results indicate that our model outperforms these models across various evaluation metrics, compare with ensemble prediction results GMRNN's Structural Similarity increased by approximately 11.2%, and the Anomaly Correlation Coefficient increased by approximately 9.5%. The GMRNN model exhibits superior stability and possesses prediction potential over a longer period.
format Article
id doaj-art-b84ca5da134746c49ef5d1202cbb4499
institution Kabale University
issn 0094-8276
1944-8007
language English
publishDate 2025-04-01
publisher Wiley
record_format Article
series Geophysical Research Letters
spelling doaj-art-b84ca5da134746c49ef5d1202cbb44992025-08-20T03:39:04ZengWileyGeophysical Research Letters0094-82761944-80072025-04-01528n/an/a10.1029/2024GL113993Predicting Sudden Stratospheric Warmings Using Video Prediction MethodsYuhao Du0Jiankai Zhang1Xinyuan Cheng2Yixiong Lu3Douwang Li4Wenshou Tian5College of Atmospheric Sciences Lanzhou University Lanzhou ChinaCollege of Atmospheric Sciences Lanzhou University Lanzhou ChinaCenter for Information and Language Processing University of Munich (LMU) Munich GermanyEarth System Modeling and Prediction Centre China Meteorological Administration Beijing ChinaCollege of Atmospheric Sciences Lanzhou University Lanzhou ChinaCollege of Atmospheric Sciences Lanzhou University Lanzhou ChinaAbstract Sudden Stratospheric Warmings (SSWs) are weather phenomena occurring in polar regions, and have a profound impact on mid‐latitude cold waves. In this paper, within a deep learning framework, we introduce video prediction techniques into SSW events forecasting for the first time. We develop a Global Attention Motion Decoupled Recurrent Neural Network (GMRNN) to better capture the detailed changes of the polar vortex. Through experiments on representative SSW events in 2018, 2019, and 2021, our model can stably predict SSW events 20 days in advance and accurately capture the morphological changes of the stratospheric polar vortex. Furthermore, we compared our model with baseline models, including PredRNN, MotionRNN, and the sub‐seasonal to seasonal (S2S) integrated forecast models from ECMWF, CMA, and ECCC. The results indicate that our model outperforms these models across various evaluation metrics, compare with ensemble prediction results GMRNN's Structural Similarity increased by approximately 11.2%, and the Anomaly Correlation Coefficient increased by approximately 9.5%. The GMRNN model exhibits superior stability and possesses prediction potential over a longer period.https://doi.org/10.1029/2024GL113993sudden stratospheric warmingsdeep learningvideo prediction
spellingShingle Yuhao Du
Jiankai Zhang
Xinyuan Cheng
Yixiong Lu
Douwang Li
Wenshou Tian
Predicting Sudden Stratospheric Warmings Using Video Prediction Methods
Geophysical Research Letters
sudden stratospheric warmings
deep learning
video prediction
title Predicting Sudden Stratospheric Warmings Using Video Prediction Methods
title_full Predicting Sudden Stratospheric Warmings Using Video Prediction Methods
title_fullStr Predicting Sudden Stratospheric Warmings Using Video Prediction Methods
title_full_unstemmed Predicting Sudden Stratospheric Warmings Using Video Prediction Methods
title_short Predicting Sudden Stratospheric Warmings Using Video Prediction Methods
title_sort predicting sudden stratospheric warmings using video prediction methods
topic sudden stratospheric warmings
deep learning
video prediction
url https://doi.org/10.1029/2024GL113993
work_keys_str_mv AT yuhaodu predictingsuddenstratosphericwarmingsusingvideopredictionmethods
AT jiankaizhang predictingsuddenstratosphericwarmingsusingvideopredictionmethods
AT xinyuancheng predictingsuddenstratosphericwarmingsusingvideopredictionmethods
AT yixionglu predictingsuddenstratosphericwarmingsusingvideopredictionmethods
AT douwangli predictingsuddenstratosphericwarmingsusingvideopredictionmethods
AT wenshoutian predictingsuddenstratosphericwarmingsusingvideopredictionmethods