A machine learning model for the prediction of hail-affected area in Germany

Hailstorms pose significant risks in Germany, calling for accurate forecasts and warnings. This study explores the application of a convolutional neural network (CNN) to predict daily hail-affected areas using radar-based hail footprints from 2005 to 2019. The ML model utilizes 18 thermodynamic and...

Full description

Saved in:
Bibliographic Details
Main Authors: Siyu Li, Peter Knippertz, Michael Kunz, Jannik Wilhelm, Julian Quinting
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-03-01
Series:Frontiers in Earth Science
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/feart.2025.1527391/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850069129750380544
author Siyu Li
Peter Knippertz
Michael Kunz
Jannik Wilhelm
Julian Quinting
author_facet Siyu Li
Peter Knippertz
Michael Kunz
Jannik Wilhelm
Julian Quinting
author_sort Siyu Li
collection DOAJ
description Hailstorms pose significant risks in Germany, calling for accurate forecasts and warnings. This study explores the application of a convolutional neural network (CNN) to predict daily hail-affected areas using radar-based hail footprints from 2005 to 2019. The ML model utilizes 18 thermodynamic and dynamic convection-related parameters derived from ERA5 reanalysis data. Feature selection identifies seven key predictors, with a particular emphasis on the convective available potential energy and bulk wind shear (CAPESHEAR). Model performance is assessed against climatology- and persistence-based reference forecasts, and sensitivity analyses using gradient-weighted class activation mapping (Grad-CAM) are conducted to interpret the predictions. The CNN model significantly outperforms the reference forecasts, achieving a Heidke Skill Score (HSS) of up to 0.66 for large hail-affected areas. However, lower predictive skill is observed on days with weak CAPESHEAR values or when hailstorms are isolated. Sensitivity analysis highlights CAPESHEAR as the dominant predictor influencing model decisions. These findings demonstrate the potential of ML-based hail prediction using only convective environmental parameters. Given its low computational demand once trained, this approach offers a promising tool for operational forecasting. It would be desirable to extend this approach to a more regional perspective and to include information on severity.
format Article
id doaj-art-7cda91bfbade4d01bbb85881909d8c2e
institution DOAJ
issn 2296-6463
language English
publishDate 2025-03-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Earth Science
spelling doaj-art-7cda91bfbade4d01bbb85881909d8c2e2025-08-20T02:47:52ZengFrontiers Media S.A.Frontiers in Earth Science2296-64632025-03-011310.3389/feart.2025.15273911527391A machine learning model for the prediction of hail-affected area in GermanySiyu LiPeter KnippertzMichael KunzJannik WilhelmJulian QuintingHailstorms pose significant risks in Germany, calling for accurate forecasts and warnings. This study explores the application of a convolutional neural network (CNN) to predict daily hail-affected areas using radar-based hail footprints from 2005 to 2019. The ML model utilizes 18 thermodynamic and dynamic convection-related parameters derived from ERA5 reanalysis data. Feature selection identifies seven key predictors, with a particular emphasis on the convective available potential energy and bulk wind shear (CAPESHEAR). Model performance is assessed against climatology- and persistence-based reference forecasts, and sensitivity analyses using gradient-weighted class activation mapping (Grad-CAM) are conducted to interpret the predictions. The CNN model significantly outperforms the reference forecasts, achieving a Heidke Skill Score (HSS) of up to 0.66 for large hail-affected areas. However, lower predictive skill is observed on days with weak CAPESHEAR values or when hailstorms are isolated. Sensitivity analysis highlights CAPESHEAR as the dominant predictor influencing model decisions. These findings demonstrate the potential of ML-based hail prediction using only convective environmental parameters. Given its low computational demand once trained, this approach offers a promising tool for operational forecasting. It would be desirable to extend this approach to a more regional perspective and to include information on severity.https://www.frontiersin.org/articles/10.3389/feart.2025.1527391/fullhail footprintsmachine learningstatisticsconvective parametersERA5Germany
spellingShingle Siyu Li
Peter Knippertz
Michael Kunz
Jannik Wilhelm
Julian Quinting
A machine learning model for the prediction of hail-affected area in Germany
Frontiers in Earth Science
hail footprints
machine learning
statistics
convective parameters
ERA5
Germany
title A machine learning model for the prediction of hail-affected area in Germany
title_full A machine learning model for the prediction of hail-affected area in Germany
title_fullStr A machine learning model for the prediction of hail-affected area in Germany
title_full_unstemmed A machine learning model for the prediction of hail-affected area in Germany
title_short A machine learning model for the prediction of hail-affected area in Germany
title_sort machine learning model for the prediction of hail affected area in germany
topic hail footprints
machine learning
statistics
convective parameters
ERA5
Germany
url https://www.frontiersin.org/articles/10.3389/feart.2025.1527391/full
work_keys_str_mv AT siyuli amachinelearningmodelforthepredictionofhailaffectedareaingermany
AT peterknippertz amachinelearningmodelforthepredictionofhailaffectedareaingermany
AT michaelkunz amachinelearningmodelforthepredictionofhailaffectedareaingermany
AT jannikwilhelm amachinelearningmodelforthepredictionofhailaffectedareaingermany
AT julianquinting amachinelearningmodelforthepredictionofhailaffectedareaingermany
AT siyuli machinelearningmodelforthepredictionofhailaffectedareaingermany
AT peterknippertz machinelearningmodelforthepredictionofhailaffectedareaingermany
AT michaelkunz machinelearningmodelforthepredictionofhailaffectedareaingermany
AT jannikwilhelm machinelearningmodelforthepredictionofhailaffectedareaingermany
AT julianquinting machinelearningmodelforthepredictionofhailaffectedareaingermany