EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Probabilistic Diffusion Models Advance Extreme Flood Forecasting

Zhigang Ou, Congyi Nai, Baoxiang Pan, Yi Zheng, Chaopeng Shen, Peishi Jiang, Xingcai Liu, Qiuhong Tang, Wenqing Li, Ming Pan

Geophysical Research Letters · 16.08.2025 · 52 (15)

DOI: 10.1029/2025gl115705 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Geophysical Research Letters, online erschienen am 31.07.2025. Verfasst von Zhigang Ou und 9 weiteren Autor:innen. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Extreme floods pose escalating risks in a changing climate, yet forecasting remains challenging due to peak flow underestimation and high uncertainty. We introduce diffusion‐based runoff model (DRUM), a probabilistic deep learning (DL) approach that advances extreme flood forecasting across representative basins in the contiguous United States. DRUM outperforms state‐of‐the‐art benchmarks, enhancing nowcasting skill for the top 1‰ of flows in 72.3% of studied basins. Under operational scenarios, DRUM extends reliable lead times by nearly a full day for 20‐ and 50‐year floods. When evaluated with measured precipitation, an ideal condition, recall improves by 0.3–0.4 and the early warning window extends by 2.3 days for 50‐year floods. The enhancement potential varies regionally, with precipitation‐driven flood zones in the eastern and northwestern US benefiting most, gaining 3–7 days in lead time. These findings highlight the transformative potential of diffusion models as a cutting‐edge generative AI technique for advancing hydrology and broader Earth system sciences.

Förderung

  • National Natural Science Foundation of China
  • National Key Research and Development Program of China

Zitieren

APA

Ou, Z., Nai, C., Pan, B., Zheng, Y., Shen, C., Jiang, P., Liu, X., Tang, Q., Li, W., & Pan, M. (2025). Probabilistic Diffusion Models Advance Extreme Flood Forecasting. Geophysical Research Letters, 52(15). https://doi.org/10.1029/2025gl115705

Vancouver

Ou Z, Nai C, Pan B, Zheng Y, Shen C, Jiang P, et al. Probabilistic Diffusion Models Advance Extreme Flood Forecasting. Geophysical Research Letters. 2025;52(15). doi:10.1029/2025gl115705

BibTeX
@article{Ou2025,
  author = {Ou, Zhigang and Nai, Congyi and Pan, Baoxiang and Zheng, Yi and Shen, Chaopeng and Jiang, Peishi and Liu, Xingcai and Tang, Qiuhong and Li, Wenqing and Pan, Ming},
  title = {Probabilistic Diffusion Models Advance Extreme Flood Forecasting},
  journal = {Geophysical Research Letters},
  year = {2025},
  volume = {52},
  number = {15},
  doi = {10.1029/2025gl115705}
}
Kostenlos starten →