Montgomery Flora, Ph.D. (@MontePhD) header
Montgomery Flora, Ph.D. (@MontePhD) profile picture

Self-motivated individual actively pursuing knowledge. Passions include being a loving husband/father, studying severe weather, and playing music.

Cumming, GA
Joined January 2019
Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)
🚨 Exciting Announcement! 🚨 Introducing WoFSCast: our AI-powered system for predicting thunderstorm evolution at watch-to-warning scales. 🌩️⚡ essopenarchive.org/users/829…
5
11
4
66
8,987
Can we use AI to predict thunderstorm evolution? Recently published in @theAGU GRL, we refactored @GoogleDeepMind's GraphCast for regional modeling and trained it on @CIWRO_ / @NOAANSSL's archived Warn-on-Forecast System (WoFS) data.
Meteorologists are turning to AI as a quick, efficient alternative to traditional weather prediction algorithms. Now researchers including @MontePhD & @corey_potvin have extended this tech to cover localized extreme weather as well as global forecasts. eos.org/research-spotlights/…
1
1
1
183
Pretty animations comparing WoFS and WoFSCast 😀
55
🚨 New paper! 🚨 Available in EOR is our paper exploring training AI models on MRMS MESH vs. reported hail sizes: journals.ametsoc.org/view/jo…
1
1
4
237
The craziest result: Although the training inputs did not include time of day or any social data, the local storm report (LSR)-trained model produced lower probs at night. This suggests that the biases were coming from the physical environment
58
Super proud of this work! More to come this upcoming year with @corey_potvin and @DopplerChase as we transition to an ensemble-based WoFSCast.
WoFSCast is the first AI system trained and developed to emulate a sub-hourly convection-allowing weather model (CAM). WoFSCast speeds up forecast generation versus WoFS by at least a factor of ten, from minutes to seconds. #AMS2025
8
297
Super proud to see that forecasters find WoFS-ML-Severe useful. #AMS31SLS Always keep your end user in mind when developing new products!
1
6
217
“Concepts of a plan” = “Manuscript in preparation”
1
144
We can also create ensemble predictions using the WoFS members as ICs
1
219
Best of all? WoFSCast achieves this in 30-40 seconds on 1 GPU, compared to the 8-12 minutes and 60+ CPUs WoFS needs.
1
1
224
While development is ongoing, WoFSCast matches WoFS's critical success index (CSI) with fewer false alarms and less spurious convection.
213
Trained on archived Warn-on-Forecast System (WoFS) data and a refactored GraphCast for limited area domains, WoFSCast can predict thunderstorm evolution for up to 6 hours.
384
Montgomery Flora, Ph.D. retweeted
Our paper combining AI/ML, storm-scale ensemble prediction, and "forecasting forecast skill" is now in final form in @AMS_AIES: doi.org/10.1175/AIES-D-23-01… Co-authors: @MontePhD, Patrick Skinner, @tonyreinhart, @BrianWxFLA
1
4
371