Short-term Precipitation Forecast Correction in Beijing-Tianjin-Hebei Region Based on Deep Learning
To enhance the quality of short-term precipitation forecast in Beijing-Tianjin-Hebei Region, a U-Net deep learning model is employed to correct short-term precipitation forecasts of 3-12 h based on the forecasting data of INCA (Integrated Nowcasting through Comprehensive Analysis) System and observa...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Editorial Office of Journal of Applied Meteorological Science
2025-05-01
|
| Series: | 应用气象学报 |
| Subjects: | |
| Online Access: | http://qikan.camscma.cn/en/article/doi/10.11898/1001-7313.20250301 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850108564019871744 |
|---|---|
| author | Wang Yuhong Kong Dexuan Zhu Shoupeng Zhang Nan Ji Yan Zhang Shan Xu Huan |
| author_facet | Wang Yuhong Kong Dexuan Zhu Shoupeng Zhang Nan Ji Yan Zhang Shan Xu Huan |
| author_sort | Wang Yuhong |
| collection | DOAJ |
| description | To enhance the quality of short-term precipitation forecast in Beijing-Tianjin-Hebei Region, a U-Net deep learning model is employed to correct short-term precipitation forecasts of 3-12 h based on the forecasting data of INCA (Integrated Nowcasting through Comprehensive Analysis) System and observation data of automatic weather stations during the flood season from 2022 to 2024. The model employs a weighted combination of threat score (TS) and mean square error (MSE) as its loss function. The weight assigned to TS directly influences the model's performance. Through independent experiments, weighting coefficients assigned to TS within loss function are optimized for hourly correction models with lead times ranging from 3 to 12 h, with the following results: 1.2, 1.0, 1.0, 1.0, 1.2, 1.2, 1.2, 1.2, 1.1, and 1.1, respectively. Comparative experiments are conducted to explore the impact of various influencing factors, including temporal and spatial characteristics of precipitation, as well as the thermal and dynamic conditions that affect short-term heavy precipitation. Results indicate that incorporating temporal and spatial characteristics of precipitation, as well as thermal and dynamic circulation conditions influencing precipitation, into the deep learning correction model significantly enhances its correction capability. A single-factor deep learning model only improves the forecast skill for precipitation rates of 5 mm·h-1 and above, while multi-factor deep learning model enhances the forecast skill for precipitation rates of 0.1, 5, 10, and 20 mm·h-1 and above. Multi-factor deep learning model significantly enhances the 3-12-h forecasting capabilities of INCA, with varying degrees of improvement based on precipitation intensity. TS can be increased by up to 0.07, 0.06, 0.06, and 0.03 compared to INCA forecast when the precipitation intensity is at least 0.1 mm·h-1, 5 mm·h-1, 10 mm·h-1, 20 mm·h-1. Multi-factor deep learning model enhances the forecasting accuracy of most stations in Beijing-Tianjin-Hebei Region without increasing precipitation forecast errors. The spatial extent of improvement in TS varies with the intensity of precipitation. For precipitation intensities of at least 0.1 mm·h-1 and 5 mm·h-1, TS can be improved in most areas of Beijing-Tianjin-Hebei Region. For precipitation intensities of at least 10 mm·h-1 and 20 mm·h-1, TS can be improved in the eastern part of Beijing-Tianjin-Hebei Region. |
| format | Article |
| id | doaj-art-7a09fbd8b3aa4d90a07cb856f4038af7 |
| institution | OA Journals |
| issn | 1001-7313 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Editorial Office of Journal of Applied Meteorological Science |
| record_format | Article |
| series | 应用气象学报 |
| spelling | doaj-art-7a09fbd8b3aa4d90a07cb856f4038af72025-08-20T02:38:19ZengEditorial Office of Journal of Applied Meteorological Science应用气象学报1001-73132025-05-0136325726710.11898/1001-7313.20250301yyqxxb-36-3-257Short-term Precipitation Forecast Correction in Beijing-Tianjin-Hebei Region Based on Deep LearningWang Yuhong0Kong Dexuan1Zhu Shoupeng2Zhang Nan3Ji Yan4Zhang Shan5Xu Huan6Xiong'an Key Laboratory of Atmospheric Boundary Layer of CMA, Xiong'an New Area 071800Guizhou Institute of Mountainous Meteorological Sciences, Guiyang 550081Nanjing Innovation Institute for Atmospheric Sciences, Chinese Academy of Meteorological Sciences-Jiangsu Meteorological Service, Nanjing 210041Xiong'an Key Laboratory of Atmospheric Boundary Layer of CMA, Xiong'an New Area 071800School of Atmosphere and Remote Sensing, Wuxi University, Wuxi 214105Xiong'an Key Laboratory of Atmospheric Boundary Layer of CMA, Xiong'an New Area 071800Lu'an Meteorological Bureau of Anhui, Lu'an 237000To enhance the quality of short-term precipitation forecast in Beijing-Tianjin-Hebei Region, a U-Net deep learning model is employed to correct short-term precipitation forecasts of 3-12 h based on the forecasting data of INCA (Integrated Nowcasting through Comprehensive Analysis) System and observation data of automatic weather stations during the flood season from 2022 to 2024. The model employs a weighted combination of threat score (TS) and mean square error (MSE) as its loss function. The weight assigned to TS directly influences the model's performance. Through independent experiments, weighting coefficients assigned to TS within loss function are optimized for hourly correction models with lead times ranging from 3 to 12 h, with the following results: 1.2, 1.0, 1.0, 1.0, 1.2, 1.2, 1.2, 1.2, 1.1, and 1.1, respectively. Comparative experiments are conducted to explore the impact of various influencing factors, including temporal and spatial characteristics of precipitation, as well as the thermal and dynamic conditions that affect short-term heavy precipitation. Results indicate that incorporating temporal and spatial characteristics of precipitation, as well as thermal and dynamic circulation conditions influencing precipitation, into the deep learning correction model significantly enhances its correction capability. A single-factor deep learning model only improves the forecast skill for precipitation rates of 5 mm·h-1 and above, while multi-factor deep learning model enhances the forecast skill for precipitation rates of 0.1, 5, 10, and 20 mm·h-1 and above. Multi-factor deep learning model significantly enhances the 3-12-h forecasting capabilities of INCA, with varying degrees of improvement based on precipitation intensity. TS can be increased by up to 0.07, 0.06, 0.06, and 0.03 compared to INCA forecast when the precipitation intensity is at least 0.1 mm·h-1, 5 mm·h-1, 10 mm·h-1, 20 mm·h-1. Multi-factor deep learning model enhances the forecasting accuracy of most stations in Beijing-Tianjin-Hebei Region without increasing precipitation forecast errors. The spatial extent of improvement in TS varies with the intensity of precipitation. For precipitation intensities of at least 0.1 mm·h-1 and 5 mm·h-1, TS can be improved in most areas of Beijing-Tianjin-Hebei Region. For precipitation intensities of at least 10 mm·h-1 and 20 mm·h-1, TS can be improved in the eastern part of Beijing-Tianjin-Hebei Region.http://qikan.camscma.cn/en/article/doi/10.11898/1001-7313.20250301deep learningloss functioninfluence factorsforecast correction |
| spellingShingle | Wang Yuhong Kong Dexuan Zhu Shoupeng Zhang Nan Ji Yan Zhang Shan Xu Huan Short-term Precipitation Forecast Correction in Beijing-Tianjin-Hebei Region Based on Deep Learning 应用气象学报 deep learning loss function influence factors forecast correction |
| title | Short-term Precipitation Forecast Correction in Beijing-Tianjin-Hebei Region Based on Deep Learning |
| title_full | Short-term Precipitation Forecast Correction in Beijing-Tianjin-Hebei Region Based on Deep Learning |
| title_fullStr | Short-term Precipitation Forecast Correction in Beijing-Tianjin-Hebei Region Based on Deep Learning |
| title_full_unstemmed | Short-term Precipitation Forecast Correction in Beijing-Tianjin-Hebei Region Based on Deep Learning |
| title_short | Short-term Precipitation Forecast Correction in Beijing-Tianjin-Hebei Region Based on Deep Learning |
| title_sort | short term precipitation forecast correction in beijing tianjin hebei region based on deep learning |
| topic | deep learning loss function influence factors forecast correction |
| url | http://qikan.camscma.cn/en/article/doi/10.11898/1001-7313.20250301 |
| work_keys_str_mv | AT wangyuhong shorttermprecipitationforecastcorrectioninbeijingtianjinhebeiregionbasedondeeplearning AT kongdexuan shorttermprecipitationforecastcorrectioninbeijingtianjinhebeiregionbasedondeeplearning AT zhushoupeng shorttermprecipitationforecastcorrectioninbeijingtianjinhebeiregionbasedondeeplearning AT zhangnan shorttermprecipitationforecastcorrectioninbeijingtianjinhebeiregionbasedondeeplearning AT jiyan shorttermprecipitationforecastcorrectioninbeijingtianjinhebeiregionbasedondeeplearning AT zhangshan shorttermprecipitationforecastcorrectioninbeijingtianjinhebeiregionbasedondeeplearning AT xuhuan shorttermprecipitationforecastcorrectioninbeijingtianjinhebeiregionbasedondeeplearning |