SwinNowcast: A Swin Transformer-Based Model for Radar-Based Precipitation Nowcasting

Precipitation nowcasting is pivotal in monitoring extreme weather events and issuing early warnings for meteorological disasters. However, the inherent complexity of precipitation systems, coupled with their nonlinear spatiotemporal evolution, poses significant challenges for traditional numerical w...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhuang Li, Zhenyu Lu, Yizhe Li, Xuan Liu
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/9/1550
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850135976381251584
author Zhuang Li
Zhenyu Lu
Yizhe Li
Xuan Liu
author_facet Zhuang Li
Zhenyu Lu
Yizhe Li
Xuan Liu
author_sort Zhuang Li
collection DOAJ
description Precipitation nowcasting is pivotal in monitoring extreme weather events and issuing early warnings for meteorological disasters. However, the inherent complexity of precipitation systems, coupled with their nonlinear spatiotemporal evolution, poses significant challenges for traditional numerical weather prediction methods in capturing multi-scale details effectively. Existing deep learning models similarly struggle to simultaneously capture local multi-scale features and global long-term spatiotemporal dependencies. To tackle this challenge, we propose SwinNowcast, a deep learning model based on the Swin Transformer architecture. Through the novel design of a multi-scale feature balancing module (M-FBM), the model dynamically integrates local-scale features with global spatiotemporal dependencies. Specifically, the multi-scale convolutional block attention module (MSCBAM) captures local multi-scale features, while the gated attention feature fusion unit (GAFFU) adaptively regulates the fusion intensity, thereby enhancing spatial structure and temporal continuity in a synergistic manner. Experiments were performed on the precipitation dataset from the Royal Netherlands Meteorological Institute (KNMI) under thresholds of 0.5 mm, 5 mm, and 10 mm. The results indicate that SwinNowcast surpasses six state-of-the-art approaches regarding the critical success index (CSI) and the Heidke skill score (HSS), while markedly reducing the false alarm rate (FAR). The proposed model holds substantial practical value in applications such as short-term heavy rainfall monitoring and urban flood early warning, offering effective technological support for meteorological disaster mitigation.
format Article
id doaj-art-1bc29346758b41a4a5dc327d10d2a9b8
institution OA Journals
issn 2072-4292
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-1bc29346758b41a4a5dc327d10d2a9b82025-08-20T02:31:16ZengMDPI AGRemote Sensing2072-42922025-04-01179155010.3390/rs17091550SwinNowcast: A Swin Transformer-Based Model for Radar-Based Precipitation NowcastingZhuang Li0Zhenyu Lu1Yizhe Li2Xuan Liu3School of Electronics and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaSchool of Electronics and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaSchool of Artificial Intelligence, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaSchool of Artificial Intelligence, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaPrecipitation nowcasting is pivotal in monitoring extreme weather events and issuing early warnings for meteorological disasters. However, the inherent complexity of precipitation systems, coupled with their nonlinear spatiotemporal evolution, poses significant challenges for traditional numerical weather prediction methods in capturing multi-scale details effectively. Existing deep learning models similarly struggle to simultaneously capture local multi-scale features and global long-term spatiotemporal dependencies. To tackle this challenge, we propose SwinNowcast, a deep learning model based on the Swin Transformer architecture. Through the novel design of a multi-scale feature balancing module (M-FBM), the model dynamically integrates local-scale features with global spatiotemporal dependencies. Specifically, the multi-scale convolutional block attention module (MSCBAM) captures local multi-scale features, while the gated attention feature fusion unit (GAFFU) adaptively regulates the fusion intensity, thereby enhancing spatial structure and temporal continuity in a synergistic manner. Experiments were performed on the precipitation dataset from the Royal Netherlands Meteorological Institute (KNMI) under thresholds of 0.5 mm, 5 mm, and 10 mm. The results indicate that SwinNowcast surpasses six state-of-the-art approaches regarding the critical success index (CSI) and the Heidke skill score (HSS), while markedly reducing the false alarm rate (FAR). The proposed model holds substantial practical value in applications such as short-term heavy rainfall monitoring and urban flood early warning, offering effective technological support for meteorological disaster mitigation.https://www.mdpi.com/2072-4292/17/9/1550precipitation nowcastingSwin Transformermulti-scale featuresgated attention mechanismsextreme weather warningdeep learning
spellingShingle Zhuang Li
Zhenyu Lu
Yizhe Li
Xuan Liu
SwinNowcast: A Swin Transformer-Based Model for Radar-Based Precipitation Nowcasting
Remote Sensing
precipitation nowcasting
Swin Transformer
multi-scale features
gated attention mechanisms
extreme weather warning
deep learning
title SwinNowcast: A Swin Transformer-Based Model for Radar-Based Precipitation Nowcasting
title_full SwinNowcast: A Swin Transformer-Based Model for Radar-Based Precipitation Nowcasting
title_fullStr SwinNowcast: A Swin Transformer-Based Model for Radar-Based Precipitation Nowcasting
title_full_unstemmed SwinNowcast: A Swin Transformer-Based Model for Radar-Based Precipitation Nowcasting
title_short SwinNowcast: A Swin Transformer-Based Model for Radar-Based Precipitation Nowcasting
title_sort swinnowcast a swin transformer based model for radar based precipitation nowcasting
topic precipitation nowcasting
Swin Transformer
multi-scale features
gated attention mechanisms
extreme weather warning
deep learning
url https://www.mdpi.com/2072-4292/17/9/1550
work_keys_str_mv AT zhuangli swinnowcastaswintransformerbasedmodelforradarbasedprecipitationnowcasting
AT zhenyulu swinnowcastaswintransformerbasedmodelforradarbasedprecipitationnowcasting
AT yizheli swinnowcastaswintransformerbasedmodelforradarbasedprecipitationnowcasting
AT xuanliu swinnowcastaswintransformerbasedmodelforradarbasedprecipitationnowcasting