EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Dynamic Hourly Correction for 2024 Guangdong Flood Season Precipitation Forecasts: Frequency Matching With CMA‐GD(R3) Model

Wang Zhang, Donglong Li, Bo Zhou, Jiawen Zheng, Hongke Cai, Qiang Li

Advances in Meteorology · 2026-01 · 2026 (1)

DOI: 10.1155/adme/8247065 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advances in Meteorology, online erschienen am 07.08.2026. Verfasst von Wang Zhang und 5 weiteren Autor:innen. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Basic and Applied Basic Research Foundation of Guangdong Province, Sichuan Province Science and Technology Support Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study focuses on the hourly precipitation forecasts based on the CMA‐GD(R3) model. By combining with the actual precipitation dataset of meteorological stations in Guangdong, different training durations are adopted, and the frequency‐matching method (FMM) is utilized to correct the hourly precipitation. Moreover, through case analysis and statistical tests, a comparative analysis of the forecasting effects before and after the correction is carried out. From the perspective of case analysis, the hourly frequency‐matching correction method has a certain corrective effect on both the precipitation area and intensity. The model shows a negative bias in predicting light rain during the flood season in 2024. As the precipitation threshold and the number of lead forecast hours increase, the model gradually exhibits the characteristic of “overestimating light rain and underestimating heavy rain.” Based on the analysis of the correction coefficients (CCs), the model has poor forecasting ability for precipitation that has not occurred upstream, resulting in underestimated precipitation forecasts. For precipitation that has already occurred, the model has an overestimated forecasting error. However, as the precipitation system becomes more distinct, the model’s grasp of the situation gradually improves. When comparing the threat score (TS) before and after the correction, it is found that when using the FMM to correct the hourly precipitation forecasts of the CMA‐GD(R3) model in the Guangdong region, it is not always the case that a longer training duration leads to a better correction effect. On average during the flood season, the correction effect is optimal when the training duration is set to 6 h, and from the perspective of horizontal distribution, the correction effect is the most stable. For the correction of short‐term precipitation forecasts in the northern part of Guangdong and the western part of western Guangdong, the correction effect is the best when a training duration of 24 h is adopted.

Förderung

  • National Natural Science Foundation of China
  • Basic and Applied Basic Research Foundation of Guangdong Province
  • Sichuan Province Science and Technology Support Program

Zitieren

APA

Zhang, W., Li, D., Zhou, B., Zheng, J., Cai, H., & Li, Q. (2026). Dynamic Hourly Correction for 2024 Guangdong Flood Season Precipitation Forecasts: Frequency Matching With CMA‐GD(R3) Model. Advances in Meteorology, 2026(1). https://doi.org/10.1155/adme/8247065

Vancouver

Zhang W, Li D, Zhou B, Zheng J, Cai H, Li Q. Dynamic Hourly Correction for 2024 Guangdong Flood Season Precipitation Forecasts: Frequency Matching With CMA‐GD(R3) Model. Advances in Meteorology. 2026;2026(1). doi:10.1155/adme/8247065

BibTeX
@article{Zhang2026,
  author = {Zhang, Wang and Li, Donglong and Zhou, Bo and Zheng, Jiawen and Cai, Hongke and Li, Qiang},
  title = {Dynamic Hourly Correction for 2024 Guangdong Flood Season Precipitation Forecasts: Frequency Matching With CMA‐GD(R3) Model},
  journal = {Advances in Meteorology},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1155/adme/8247065}
}
Kostenlos starten →