[Eoas-seminar] Meteorology MS Defense for Thomas Gard, Monday, October 26, 2020, 9:00 AM, on Zoom https://fsu.zoom.us/j/93507804077?pwd=UmVweThlUHJVQjBwaTR0ZmR2Q1g5QT09

[eoas-seminar at lists.fsu.edu]

NOTE: The zoom link has been updated. Please use this link on Monday.

Meteorology Seminar

Thomas Gard

M.S. Meteorology Candidate



Title:  Examining the utility of ProbSevere For Predicting Pulse Severe Thunderstorms

Major Professor:  Dr. Henry Fuelberg

Date:  Monday, October 26th, 2020                                 Time: 9:00 AM


Location: Zoom Meeting URL: https://fsu.zoom.us/j/93507804077?pwd=UmVweThlUHJVQjBwaTR0ZmR2Q1g5QT09


ABSTRACT
Disorganized, weakly forced convection is a routine summertime occurrence in the Southeast United States. Pulse severe storms are single cell thunderstorms that produce severe wind and/or severe hail for a brief period of time. These thunderstorms pose a major warm season forecasting problem, since forecasters presently do not have sufficient guidance to know which, if any, of the many single cells will become severe. The empirical Probability of Severe (ProbSevere) model, developed by the Cooperative Institute for Meteorological Satellite Studies (CIMSS), fuses real-time data to produce short-term, statistically derived probabilistic forecasts of thunderstorm intensity. Objects corresponding to storms are created from satellite and radar data, tracked in two-minute timesteps, and trained on a composite of all modes of severe thunderstorms. ProbSevere has been found to increase forecaster confidence and forecast lead time. However, it is unclear how well ProbSevere performs specifically for pulse severe storms. This study evaluates the ability of ProbSevere to represent and predict pulse severe storms.
Pulse severe environments are often classified as being unstable (CAPE * 2000 J kg-1) and weakly sheared (0-6 km AGL shear < 10 m s-1). ProbSevere objects fitting the accepted definition of a pulse severe environment were matched with severe events from Storm Data to create a dataset of ProbSevere objects that corresponded to pulse severe thunderstorms. A null dataset comprised of objects that corresponded to pulse severe environments but did not match with a severe event was also created.
Pulse severe objects were evaluated at two times - the time at object initialization and the time of the severe event. This way, the evolution of objects could be analyzed as would be done by a forecaster in real time. Objects were examined in four ways - 1) their temporal evolution, 2) their spatial evolution, 3) how well the severe hail and cellular wind models perform for pulse severe objects, and 4) how well individual predictors discriminate between severe and nonsevere objects.
Results reveal that ProbSevere's ability to represent pulse severe thunderstorms is limited, both temporally and spatially. There is little difference between severe hail-producing objects and severe wind-producing objects. At object initialization, ProbSevere exhibits insignificant probabilities of severe weather. At the time of severe events, probabilities are greater but still small. No predictor, with the exception of lightning flash rate, shows a strong correlation with the increasing probability of pulse severe events. Furthermore, no discernable difference in the distribution of probabilities as a function of predictor value is found between severe and nonsevere objects. This study demonstrates that ProbSevere's ability to represent pulse severe thunderstorms is limited, and that ProbSevere is not optimized to give forecasters sufficient guidance for pulse severe storm events.



Shel McGuire
Florida State University
Academic Program Specialist
Department of Earth, Ocean, & Atmospheric Science
1011 Academic Way, 2019 EOA Building
Tallahassee, FL 32306
850-644-8582
To make an appointment please login to my.fsu.edu and choose the Campus Connect (CC) icon

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.fsu.edu/pipermail/eoas-seminar/attachments/20201023/e234da5e/attachment.html>