[Eoas-seminar] PhD Defense - William Curtis

[eoas-seminar at lists.fsu.edu]

Hello all,

Please join us for William Curtis' PhD Defense on Thursday, November 6 @ 10:00 am EST, in EOAS 5067 and on zoom.


Title: NON-THUNDERSTORM CUMULUS ELECTRIFICATION ASSESSMENT AND PREDICTIVE ANALYSIS USING MACHINE LEARNING METHODS

Name: William Curtis
Date: Thursday, November 6, 2025, at 10:00 am EST
Major Professor: Henry Fuelberg
Location: EOAS 5067
Zoom: https://fsu.zoom.us/j/99741098539<https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Ffsu.zoom.us%2Fj%2F99741098539&data=05%7C02%7Ceoas-seminar%40lists.fsu.edu%7C58554249b22945d5080008de1af2b767%7Ca36450ebdb0642a78d1b026719f701e3%7C0%7C0%7C638977826907148860%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=k9kcdTGNjn6G%2BIAcetoTXRphIcWAAD7zlr1MK%2BpqXTs%3D&reserved=0>

Abstract: The electrification of developing convective, non-thunderstorm cumulus clouds is investigated utilizing a combination of a rapidly scanning dual-pol radar and a dense network of surface field mills that measure the vertical electric potential gradient. Current launch safety criteria employ a combination of cloud top temperature and potential gradient threshold values to assess the potential for natural and triggered lightning. A novel dataset of clouds within the field mill network is constructed and investigated based on the behavior of the surrounding field mill sites to assess the electrification. Field mill data is analyzed using gridded 1-minute averaged potential gradients. Previous research utilized various reflectivity thresholds for different temperature levels to differentiate between electrified clouds, lightning producing clouds, and different electrification thresholds. Reflectivity data at different temperature levels and hydrometeor classification algorithm output are utilized in concert with cloud top temperature to represent cloud characteristics and explain surface electric field behavior.
            This study utilizes supervised machine learning models and explainable AI diagnostic tools to provide a comparison between radar parameters and their association with surface potential gradients. Hydrometeor classification algorithm output provides a substantial benefit to cloud classification based on surface potential gradients and predicting the magnitude of negative potential gradient values through the vertical ice class of hydrometeors. Cloud top temperatures colder than -5°C indicate steadily greater likelihood for strong negative potential gradients, and increasing vertical ice further contributes to stronger negative potential gradients. The application of machine learning and hydrometeor classification produces novel insight into the dependencies between radar parameters and their diagnostic capabilities for assessing electrification in developing convective clouds.

Best,
Adea

Adea Arrison
Sr. Academic Program Specialist
Department of Earth, Ocean & Atmospheric Science
[cid:image001.png at 01DC4CB1.608D2F20]<https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.eoas.fsu.edu%2F&data=05%7C02%7Ceoas-seminar%40lists.fsu.edu%7C58554249b22945d5080008de1af2b767%7Ca36450ebdb0642a78d1b026719f701e3%7C0%7C0%7C638977826907166582%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=SlmnQoY9F8ZqQicMr25GHNJ3SN7%2BBXQb2w33%2F6sXtG0%3D&reserved=0>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.fsu.edu/pipermail/eoas-seminar/attachments/20251103/1010931d/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 3433 bytes
Desc: image001.png
URL: <http://lists.fsu.edu/pipermail/eoas-seminar/attachments/20251103/1010931d/attachment-0001.png>