[Eoas-seminar] Meteorology MS Defense for Jerry Kung, Monday, March 23, 2020, 11:00 AM, Love 353

eoas-seminar at lists.fsu.edu eoas-seminar at lists.fsu.edu
Mon Mar 9 16:20:13 EDT 2020

Meteorology Seminar

Jerry Kung

M.S. Meteorology Candidate

Title:  Three modeling approaches to predicting pyrocumulus formation during prescribed burns at eglin air force base, florida

Major Professor:  Dr. Jeffrey Chagnon

Date: Monday, March 23, 2020                        Time: 11:00 AM

Location: Werner A. Baum Seminar Room (353 Love Building)


Prescribed burning is a land management technique conducted at Eglin Air Force Base (AFB), Florida, by Jackson Guard.  Sometimes a fire-induced cumulus cloud, called a pyrocumulus (pyroCu), forms over a surface fire.  A towering pyrocumulus can lead to downbursts that challenge Jackson Guard teams' capacity to keep fire behavior under control.  There has been scant effort devoted to forecasting pyrocumuli accurately.  This thesis presents three different methods for predicting pyrocumulus formation prior to scheduled burns at Eglin AFB: numerical modeling, statistical regression, and qualitative modeling.  The numerical modeling approach entailed simulating 14 burn dates using the Weather Research and Forecasting (WRF) model's coupled fire-atmosphere module, WRF-Fire.  A Rate-of-Spread (ROS) parameter was developed for simulating the outcomes of future burns, with the prescribed burn model matching observed pyrocumuli at a 71% success rate.  The statistical regression approach used a sample of 31 burn dates to build, train, and test a statistical model using logistic regression.  The statistical model performed with 74% accuracy in hindcasting the sample size's observed outcomes.  In the third approach, a qualitative model was built and tested on the same 31 burn dates' thermodynamic soundings and burn parameters.  The categorization algorithm was converted into a flowchart, with decision points trained to optimally predict the sample's observed outcomes, hindcasting pyrocumulus formation at a 74% success rate.  These approaches to prediction yielded three distinct operational tools for Jackson Guard.  While higher success rates were not obtained this time, a fruitful proof of concept was accomplished.  Furthermore, the three methodological frameworks can all be tailored for fixed-site prescribed burn operations outside northwestern Florida.

Shel McGuire
Florida State University
Academic Program Specialist
Department of Earth, Ocean, & Atmospheric Science
1011 Academic Way, 2019 EOA Building
Tallahassee, FL 32306

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.fsu.edu/pipermail/eoas-seminar/attachments/20200309/0842c148/attachment.html>

More information about the Eoas-seminar mailing list