Application of Machine Learning to Severe Weather Prediction from Storm Top Indicators

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F25%3A10513144" target="_blank" >RIV/00216208:11320/25:10513144 - isvavai.cz</a>
Result on the web
<a href="https://physics.mff.cuni.cz/wds/proc/pdf25/WDS25_11_f8_Dolezalova.pdf" target="_blank" >https://physics.mff.cuni.cz/wds/proc/pdf25/WDS25_11_f8_Dolezalova.pdf</a>
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Application of Machine Learning to Severe Weather Prediction from Storm Top Indicators
Original language description
This case study explores the practical application of a machine learning modelfor detecting overshooting tops (OTs) from high-resolution visible satellite imagery,without relying on infrared data. The model was applied to a severe convective weatherevent, and its OT detections were compared with ground-based reports of hazardousweather, including hail, wind gusts, and intense precipitation. The results demonstratethat the model successfully identified several regions of intense convection, showing goodspatial agreement with areas of reported severe weather. However, the correlation variedbetween individual events - while some hazardous occurrences aligned closely withpredicted OTs, others showed weaker correspondence. These findings highlight boththe potential and the limitations of OT-based storm severity assessment and support theuse of such models as complementary tools in real-time meteorological monitoring andnowcasting.
Czech name
—
Czech description
—

Project
—
Continuities
S - Specificky vyzkum na vysokych skolach<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year
2025
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
WDS'25 Proceedings of Contributed Papers - Physics
ISBN
978-80-7378-532-1
ISSN
—
e-ISSN
—
Number of pages
8
Pages from-to
94-101
Publisher name
Matfyzpress
Place of publication
Prague
Event location
Praha
Event date
Jun 3, 2025
Type of event by nationality
EUR - Evropská akce
UT code for WoS article
—

Similar results(10)