Articles | Volume 25, issue 1
https://doi.org/10.5194/nhess-25-41-2025
https://doi.org/10.5194/nhess-25-41-2025
Brief communication
 | 
03 Jan 2025
Brief communication |  | 03 Jan 2025

Brief communication: Training of AI-based nowcasting models for rainfall early warning should take into account user requirements

Georgy Ayzel and Maik Heistermann

Related authors

The Potsdam Soil Moisture Observatory: High-coverage reference observations at kilometer scale
Elodie Marret, Peter M. Grosse, Lena Scheiffele, Katya Dimitrova Petrova, Till Francke, Daniel Altdorff, Maik Heistermann, Merlin Schiel, Carsten Neumann, Daniel Scheffler, Mehdi Saberioon, Matthias Kunz, Miroslav Zboril, Jonas Marach, Marcel Reginatto, Anna Balenzano, Daniel Rasche, Christine Stumpp, Benjamin Trost, and Sascha E. Oswald
Earth Syst. Sci. Data Discuss., https://doi.org/10.5194/essd-2025-546,https://doi.org/10.5194/essd-2025-546, 2025
Preprint under review for ESSD
Short summary
Present and future trends of extreme short-term rainfall events in Germany, by downscaling convective environments of ERA5 and a CMIP6 ensemble
Gerd Bürger and Maik Heistermann
EGUsphere, https://doi.org/10.5194/egusphere-2025-3584,https://doi.org/10.5194/egusphere-2025-3584, 2025
This preprint is open for discussion and under review for Natural Hazards and Earth System Sciences (NHESS).
Short summary
The ability of a stochastic regional weather generator to reproduce heavy-precipitation events across scales
Xiaoxiang Guan, Viet Dung Nguyen, Paul Voit, Bruno Merz, Maik Heistermann, and Sergiy Vorogushyn
Nat. Hazards Earth Syst. Sci., 25, 3075–3086, https://doi.org/10.5194/nhess-25-3075-2025,https://doi.org/10.5194/nhess-25-3075-2025, 2025
Short summary
Groundwater recharge in Brandenburg is declining – but why?
Till Francke and Maik Heistermann
Nat. Hazards Earth Syst. Sci., 25, 2783–2802, https://doi.org/10.5194/nhess-25-2783-2025,https://doi.org/10.5194/nhess-25-2783-2025, 2025
Short summary
Virtual Joint Field Campaign: a framework of synthetic landscapes to assess multiscale measurement methods of water storage
Till Francke, Cosimo Brogi, Alby Duarte Rocha, Michael Förster, Maik Heistermann, Markus Köhli, Daniel Rasche, Marvin Reich, Paul Schattan, Lena Scheiffele, and Martin Schrön
Geosci. Model Dev., 18, 819–842, https://doi.org/10.5194/gmd-18-819-2025,https://doi.org/10.5194/gmd-18-819-2025, 2025
Short summary

Cited articles

Agrawal, S., Barrington, L., Bromberg, C., Burge, J., Gazen, C., and Hickey, J.: Machine Learning for Precipitation Nowcasting from Radar Images, arXiv [preprint], https://doi.org/10.48550/arXiv.1912.12132, 2019. a
Ayzel, G.: The RainNet2024 family of models for precipitation nowcasting, Zenodo [code], https://doi.org/10.5281/zenodo.12547127, 2024. a
Ayzel, G., Scheffer, T., and Heistermann, M.: RainNet v1.0: a convolutional neural network for radar-based precipitation nowcasting, Geosci. Model Dev., 13, 2631–2644, https://doi.org/10.5194/gmd-13-2631-2020, 2020. a, b, c, d, e, f
DWD: Warnkriterien für Starkregen, https://www.dwd.de/DE/wetter/warnungen_aktuell/kriterien/warnkriterien.html?nn=508722#doc453962bodyText3 (last access: 2 January 2025), 2024. a
Franch, G., Nerini, D., Pendesini, M., Coviello, L., Jurman, G., and Furlanello, C.: Precipitation Nowcasting with Orographic Enhanced Stacked Generalization: Improving Deep Learning Predictions on Extreme Events, Atmosphere, 11, 267, https://doi.org/10.3390/atmos11030267, 2020. a
Download
Short summary
Forecasting rainfall over the next hour is an essential feature of early warning systems. Deep learning (DL) has emerged as a powerful alternative to conventional nowcasting technologies, but it still struggles to adequately predict impact-relevant heavy rainfall. We think that DL could do much better if the training tasks were defined more specifically and that such specification presents an opportunity to better align the output of nowcasting models with actual user requirements.
Share
Altmetrics
Final-revised paper
Preprint