Evaluation of precipitation forecasting base on GraphCast over mainland China
Abstract The accurate cumulative precipitation forecasts are essential for monitoring water resources and natural disasters. The combination of deep learning and big data has become a new direction for precipitation forecasting. However, the current large models are still lacking in-situ data verifi...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-04-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-98944-7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850042101043036160 |
|---|---|
| author | Zihuang Yan Xianghui Lu Lifeng Wu Fa Liu Rangjian Qiu Yaokui Cui Xin Ma |
| author_facet | Zihuang Yan Xianghui Lu Lifeng Wu Fa Liu Rangjian Qiu Yaokui Cui Xin Ma |
| author_sort | Zihuang Yan |
| collection | DOAJ |
| description | Abstract The accurate cumulative precipitation forecasts are essential for monitoring water resources and natural disasters. The combination of deep learning and big data has become a new direction for precipitation forecasting. However, the current large models are still lacking in-situ data verification. To accomplish this goal, the precipitation forecasting performance of a state-of-the-art model GraphCast was evaluated. Using the cumulative precipitation data from 2393 observation stations for the 1–3 day period as a reference, we assessed the cumulative precipitation in mainland China region for the 1–3 day period from 2020 to 2021, utilizing a high-resolution model with 0.25° × 0.25° grid spacing and 13 layers of parameters. The precipitation of European Centre for Medium-Range Weather Forecasts (ECMWF) was also compared. The results show that: (1) During the 2020-2021 period, for the 1-day, 2-day, and 3-day cumulative precipitation forecasts, the Root Mean Square Error (RMSE) values of GraphCast were primarily between 0.46 to 9.38 mm/d, 0.44 to 9.06 mm/d, and 0.44 to 9.06 mm/d, respectively. The Mean Error (ME) values were mainly between − 0.595 to 1.705 (0.01 mm). (2) As the forecast period extends, the forecasting capability of GraphCast declines. (3) In the 1–3 day cumulative precipitation forecasts for various stations in mainland China, GraphCast demonstrates higher predictive accuracy than ECMWF. (4) Compared to ECMWF, GraphCast demonstrated the best forecast performance in the temperate humid and semi-humid regions of Northeast China, with the RMSE being approximately 12 $$\%$$ higher. Our study indicates that GraphCast demonstrates significant potential and higher accuracy in precipitation forecasting. |
| format | Article |
| id | doaj-art-09b0b7accfe644f5b34ed1bed834a2c2 |
| institution | DOAJ |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-09b0b7accfe644f5b34ed1bed834a2c22025-08-20T02:55:38ZengNature PortfolioScientific Reports2045-23222025-04-0115111310.1038/s41598-025-98944-7Evaluation of precipitation forecasting base on GraphCast over mainland ChinaZihuang Yan0Xianghui Lu1Lifeng Wu2Fa Liu3Rangjian Qiu4Yaokui Cui5Xin Ma6School of Soil and Water Conservation, Nanchang Institute of TechnologySchool of Soil and Water Conservation, Nanchang Institute of TechnologySchool of Soil and Water Conservation, Nanchang Institute of TechnologyKey Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of SciencesState Key Laboratory of Water Resources Engineering and Management, School of Water Resources and Hydropower Engineering, Wuhan UniversityInstitute of RS and GIS, School of Earth and Space Sciences, Peking UniversitySchool of Mathematics and Physics, Southwest University of Science and TechnologyAbstract The accurate cumulative precipitation forecasts are essential for monitoring water resources and natural disasters. The combination of deep learning and big data has become a new direction for precipitation forecasting. However, the current large models are still lacking in-situ data verification. To accomplish this goal, the precipitation forecasting performance of a state-of-the-art model GraphCast was evaluated. Using the cumulative precipitation data from 2393 observation stations for the 1–3 day period as a reference, we assessed the cumulative precipitation in mainland China region for the 1–3 day period from 2020 to 2021, utilizing a high-resolution model with 0.25° × 0.25° grid spacing and 13 layers of parameters. The precipitation of European Centre for Medium-Range Weather Forecasts (ECMWF) was also compared. The results show that: (1) During the 2020-2021 period, for the 1-day, 2-day, and 3-day cumulative precipitation forecasts, the Root Mean Square Error (RMSE) values of GraphCast were primarily between 0.46 to 9.38 mm/d, 0.44 to 9.06 mm/d, and 0.44 to 9.06 mm/d, respectively. The Mean Error (ME) values were mainly between − 0.595 to 1.705 (0.01 mm). (2) As the forecast period extends, the forecasting capability of GraphCast declines. (3) In the 1–3 day cumulative precipitation forecasts for various stations in mainland China, GraphCast demonstrates higher predictive accuracy than ECMWF. (4) Compared to ECMWF, GraphCast demonstrated the best forecast performance in the temperate humid and semi-humid regions of Northeast China, with the RMSE being approximately 12 $$\%$$ higher. Our study indicates that GraphCast demonstrates significant potential and higher accuracy in precipitation forecasting.https://doi.org/10.1038/s41598-025-98944-7Cumulative precipitation forecastLarge modelsHigh resolutionECMWF |
| spellingShingle | Zihuang Yan Xianghui Lu Lifeng Wu Fa Liu Rangjian Qiu Yaokui Cui Xin Ma Evaluation of precipitation forecasting base on GraphCast over mainland China Scientific Reports Cumulative precipitation forecast Large models High resolution ECMWF |
| title | Evaluation of precipitation forecasting base on GraphCast over mainland China |
| title_full | Evaluation of precipitation forecasting base on GraphCast over mainland China |
| title_fullStr | Evaluation of precipitation forecasting base on GraphCast over mainland China |
| title_full_unstemmed | Evaluation of precipitation forecasting base on GraphCast over mainland China |
| title_short | Evaluation of precipitation forecasting base on GraphCast over mainland China |
| title_sort | evaluation of precipitation forecasting base on graphcast over mainland china |
| topic | Cumulative precipitation forecast Large models High resolution ECMWF |
| url | https://doi.org/10.1038/s41598-025-98944-7 |
| work_keys_str_mv | AT zihuangyan evaluationofprecipitationforecastingbaseongraphcastovermainlandchina AT xianghuilu evaluationofprecipitationforecastingbaseongraphcastovermainlandchina AT lifengwu evaluationofprecipitationforecastingbaseongraphcastovermainlandchina AT faliu evaluationofprecipitationforecastingbaseongraphcastovermainlandchina AT rangjianqiu evaluationofprecipitationforecastingbaseongraphcastovermainlandchina AT yaokuicui evaluationofprecipitationforecastingbaseongraphcastovermainlandchina AT xinma evaluationofprecipitationforecastingbaseongraphcastovermainlandchina |