Improving the heavy rainfall forecasting using a weighted deep learning model
Weather forecasting has been playing an important role in socio-economics. However, operational numerical weather prediction (NWP) is insufficiently accurate in terms of precipitation forecasting, especially for heavy rainfalls. Previous works on NWP bias correction utilizing deep learning (DL) meth...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-02-01
|
| Series: | Frontiers in Environmental Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fenvs.2023.1116672/full |
| _version_ | 1811169063573913600 |
|---|---|
| author | Yutong Chen Yutong Chen Yutong Chen Gang Huang Gang Huang Gang Huang Ya Wang Weichen Tao Qun Tian Kai Yang Jiangshan Zheng Hubin He |
| author_facet | Yutong Chen Yutong Chen Yutong Chen Gang Huang Gang Huang Gang Huang Ya Wang Weichen Tao Qun Tian Kai Yang Jiangshan Zheng Hubin He |
| author_sort | Yutong Chen |
| collection | DOAJ |
| description | Weather forecasting has been playing an important role in socio-economics. However, operational numerical weather prediction (NWP) is insufficiently accurate in terms of precipitation forecasting, especially for heavy rainfalls. Previous works on NWP bias correction utilizing deep learning (DL) methods mostly focused on a local region, and the China-wide precipitation forecast correction had not been attempted. Meanwhile, earlier studies imposed no particular focus on strong rainfalls despite their severe catastrophic impacts. In this study, we propose a DL model called weighted U-Net (WU-Net) that incorporates sample weights for various precipitation events to improve the forecasts of intensive precipitation in China. It is found that WU-Net can further improve the forecasting skill of heaviest rainfall comparing with the ordinary U-Net and ECMWF-IFS. Further analysis shows that this improvement increases with growing lead time, and distributes mainly in the eastern parts of China. This study suggests that a DL model considering the imbalance of the meteorological data could further improve the precipitation forecasting generated by numerical weather prediction. |
| first_indexed | 2024-04-10T16:36:54Z |
| format | Article |
| id | doaj.art-906b7a950cd44364ae517268e8dca6e2 |
| institution | Directory Open Access Journal |
| issn | 2296-665X |
| language | English |
| last_indexed | 2024-04-10T16:36:54Z |
| publishDate | 2023-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Environmental Science |
| spelling | doaj.art-906b7a950cd44364ae517268e8dca6e22023-02-08T11:26:59ZengFrontiers Media S.A.Frontiers in Environmental Science2296-665X2023-02-011110.3389/fenvs.2023.11166721116672Improving the heavy rainfall forecasting using a weighted deep learning modelYutong Chen0Yutong Chen1Yutong Chen2Gang Huang3Gang Huang4Gang Huang5Ya Wang6Weichen Tao7Qun Tian8Kai Yang9Jiangshan Zheng10Hubin He11State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, ChinaLaboratory for Regional Oceanography and Numerical Modeling, Qingdao National Laboratory for Marine Science and Technology, Qingdao, ChinaCollege of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing, ChinaState Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, ChinaLaboratory for Regional Oceanography and Numerical Modeling, Qingdao National Laboratory for Marine Science and Technology, Qingdao, ChinaCollege of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing, ChinaState Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, ChinaState Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, ChinaGuangdong Provincial Key Laboratory of Regional Numerical Weather Prediction, Guangzhou Institute of Tropical and Marine Meteorology, CMA, Guangzhou, ChinaState Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, ChinaShanghai Investigation, Design and Research Institute Co., Ltd., Shanghai, ChinaZhejiang Institute of Communications Co., Ltd., Hangzhou, ChinaWeather forecasting has been playing an important role in socio-economics. However, operational numerical weather prediction (NWP) is insufficiently accurate in terms of precipitation forecasting, especially for heavy rainfalls. Previous works on NWP bias correction utilizing deep learning (DL) methods mostly focused on a local region, and the China-wide precipitation forecast correction had not been attempted. Meanwhile, earlier studies imposed no particular focus on strong rainfalls despite their severe catastrophic impacts. In this study, we propose a DL model called weighted U-Net (WU-Net) that incorporates sample weights for various precipitation events to improve the forecasts of intensive precipitation in China. It is found that WU-Net can further improve the forecasting skill of heaviest rainfall comparing with the ordinary U-Net and ECMWF-IFS. Further analysis shows that this improvement increases with growing lead time, and distributes mainly in the eastern parts of China. This study suggests that a DL model considering the imbalance of the meteorological data could further improve the precipitation forecasting generated by numerical weather prediction.https://www.frontiersin.org/articles/10.3389/fenvs.2023.1116672/fullbias correctiondeep learningextremely heavy rainfallimbalanced dataECMWFHenan |
| spellingShingle | Yutong Chen Yutong Chen Yutong Chen Gang Huang Gang Huang Gang Huang Ya Wang Weichen Tao Qun Tian Kai Yang Jiangshan Zheng Hubin He Improving the heavy rainfall forecasting using a weighted deep learning model Frontiers in Environmental Science bias correction deep learning extremely heavy rainfall imbalanced data ECMWF Henan |
| title | Improving the heavy rainfall forecasting using a weighted deep learning model |
| title_full | Improving the heavy rainfall forecasting using a weighted deep learning model |
| title_fullStr | Improving the heavy rainfall forecasting using a weighted deep learning model |
| title_full_unstemmed | Improving the heavy rainfall forecasting using a weighted deep learning model |
| title_short | Improving the heavy rainfall forecasting using a weighted deep learning model |
| title_sort | improving the heavy rainfall forecasting using a weighted deep learning model |
| topic | bias correction deep learning extremely heavy rainfall imbalanced data ECMWF Henan |
| url | https://www.frontiersin.org/articles/10.3389/fenvs.2023.1116672/full |
| work_keys_str_mv | AT yutongchen improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT yutongchen improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT yutongchen improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT ganghuang improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT ganghuang improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT ganghuang improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT yawang improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT weichentao improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT quntian improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT kaiyang improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT jiangshanzheng improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel AT hubinhe improvingtheheavyrainfallforecastingusingaweighteddeeplearningmodel |