Using Deep Learning to Identify Circulation Patterns of Intense Rainfall in the Beijing–Tianjing–Hebei Region
The properties and distributions of precipitation are often determined by specific synoptic patterns. Hence, the objective identification of corresponding impact patterns is an important field of research for improving rain forecasting. However, the identification of the weather patterns producing i...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/14/6/930 |
| _version_ | 1797596148470382592 |
|---|---|
| author | Linguo Jing Qi Zhong Xiaojie Li Xiuming Wang Lili Shen Yong Cao |
| author_facet | Linguo Jing Qi Zhong Xiaojie Li Xiuming Wang Lili Shen Yong Cao |
| author_sort | Linguo Jing |
| collection | DOAJ |
| description | The properties and distributions of precipitation are often determined by specific synoptic patterns. Hence, the objective identification of corresponding impact patterns is an important field of research for improving rain forecasting. However, the identification of the weather patterns producing intense rainfall is much more challenging. Since they are violent and local, impact patterns tend to be meso- or smaller-scale systems and are often incompletely presented or only presented in limited regions. In this paper, a deep learning network with a feature cross-fusion module, FConvNeXt, was proposed to address this difficulty and showed great potential. Four major patterns corresponding to intense rainfall in the Beijing–Tianjing–Hebei Region were studied. Statistical testing showed that FConvNeXt performed better than ConvNeXt and ResNet and that the model could identify the weak synoptic forcing type, the subtropical high-pressure type, and the low-vortex pattern with high accuracy. Furthermore, a strictly independent 2021 dataset was tested, and FConvNeXt maintained an equal if not even slightly better performance in spite of a decrease in the subtropical high-pressure type. Meanwhile, the study showed that the accuracy in identifying the upper-level trough type is the lowest for the three deep learning methods, which may be because the northeast vortex was intercepted in the limited region, making it difficult to distinguish from the shallow upper-level trough type. This study is useful for improving the fine objective of forecasting intense rainfall. |
| first_indexed | 2024-03-11T02:47:26Z |
| format | Article |
| id | doaj.art-17eff488cd2e4e819dfb41a56dddbe21 |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-03-11T02:47:26Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-17eff488cd2e4e819dfb41a56dddbe212023-11-18T09:13:53ZengMDPI AGAtmosphere2073-44332023-05-0114693010.3390/atmos14060930Using Deep Learning to Identify Circulation Patterns of Intense Rainfall in the Beijing–Tianjing–Hebei RegionLinguo Jing0Qi Zhong1Xiaojie Li2Xiuming Wang3Lili Shen4Yong Cao5The College of Computer Science, Chengdu University of Information Technology, Chengdu 610225, ChinaChina Meteorological Administration Training Center, Beijing 100081, ChinaThe College of Computer Science, Chengdu University of Information Technology, Chengdu 610225, ChinaChina Meteorological Administration Training Center, Beijing 100081, ChinaHebei Province Meteorological Disaster Prevention and Environment Meteorology Center, Shijiazhuang 050021, ChinaNational Meteorological Center, Beijing 100081, ChinaThe properties and distributions of precipitation are often determined by specific synoptic patterns. Hence, the objective identification of corresponding impact patterns is an important field of research for improving rain forecasting. However, the identification of the weather patterns producing intense rainfall is much more challenging. Since they are violent and local, impact patterns tend to be meso- or smaller-scale systems and are often incompletely presented or only presented in limited regions. In this paper, a deep learning network with a feature cross-fusion module, FConvNeXt, was proposed to address this difficulty and showed great potential. Four major patterns corresponding to intense rainfall in the Beijing–Tianjing–Hebei Region were studied. Statistical testing showed that FConvNeXt performed better than ConvNeXt and ResNet and that the model could identify the weak synoptic forcing type, the subtropical high-pressure type, and the low-vortex pattern with high accuracy. Furthermore, a strictly independent 2021 dataset was tested, and FConvNeXt maintained an equal if not even slightly better performance in spite of a decrease in the subtropical high-pressure type. Meanwhile, the study showed that the accuracy in identifying the upper-level trough type is the lowest for the three deep learning methods, which may be because the northeast vortex was intercepted in the limited region, making it difficult to distinguish from the shallow upper-level trough type. This study is useful for improving the fine objective of forecasting intense rainfall.https://www.mdpi.com/2073-4433/14/6/930intense rainfallcirculation patternobjective classificationdeep learningBeijing–Tianjing–Hebei Region |
| spellingShingle | Linguo Jing Qi Zhong Xiaojie Li Xiuming Wang Lili Shen Yong Cao Using Deep Learning to Identify Circulation Patterns of Intense Rainfall in the Beijing–Tianjing–Hebei Region Atmosphere intense rainfall circulation pattern objective classification deep learning Beijing–Tianjing–Hebei Region |
| title | Using Deep Learning to Identify Circulation Patterns of Intense Rainfall in the Beijing–Tianjing–Hebei Region |
| title_full | Using Deep Learning to Identify Circulation Patterns of Intense Rainfall in the Beijing–Tianjing–Hebei Region |
| title_fullStr | Using Deep Learning to Identify Circulation Patterns of Intense Rainfall in the Beijing–Tianjing–Hebei Region |
| title_full_unstemmed | Using Deep Learning to Identify Circulation Patterns of Intense Rainfall in the Beijing–Tianjing–Hebei Region |
| title_short | Using Deep Learning to Identify Circulation Patterns of Intense Rainfall in the Beijing–Tianjing–Hebei Region |
| title_sort | using deep learning to identify circulation patterns of intense rainfall in the beijing tianjing hebei region |
| topic | intense rainfall circulation pattern objective classification deep learning Beijing–Tianjing–Hebei Region |
| url | https://www.mdpi.com/2073-4433/14/6/930 |
| work_keys_str_mv | AT linguojing usingdeeplearningtoidentifycirculationpatternsofintenserainfallinthebeijingtianjinghebeiregion AT qizhong usingdeeplearningtoidentifycirculationpatternsofintenserainfallinthebeijingtianjinghebeiregion AT xiaojieli usingdeeplearningtoidentifycirculationpatternsofintenserainfallinthebeijingtianjinghebeiregion AT xiumingwang usingdeeplearningtoidentifycirculationpatternsofintenserainfallinthebeijingtianjinghebeiregion AT lilishen usingdeeplearningtoidentifycirculationpatternsofintenserainfallinthebeijingtianjinghebeiregion AT yongcao usingdeeplearningtoidentifycirculationpatternsofintenserainfallinthebeijingtianjinghebeiregion |