Skip to main content

Using AI to Predict Convective Weather Hazards

The American Meteorological Society (AMS) Committee on Satellite Meteorology, Oceanography, and Climatology (SatMOC)  is pleased to offer a short course titled “Monitoring the El Niño – Southern Oscillation and its Impacts on the Weather'' during June – July 2024. This short course will consist of four 3-hour virtual training sessions described below. The training sessions are scheduled for 1:00 PM to 4:00 PM ET on June 18, 25, 27, and July 16. This series of four training sessions will demonstrate the use of environmental satellite data to monitor the El Niño – Southern Oscillation (ENSO) cycle and three weather impacts commonly associated with the ENSO cycle: flooding, drought, and severe weather.

These training sessions will provide hands-on experience for selecting and applying environmental satellite data products for monitoring ENSO and assessing several weather impacts associated with ENSO. Certificates of completion will be issued to students who participate in a minimum of 3 training sessions and the participation in each session must exceed one hour. The short course is primarily designed for undergraduate and graduate college students but others who are changing careers or moving to a position requiring increased environmental satellite knowledge will benefit from the course.

Sponsored by the AMS Committee on Satellite Meteorology, Oceanography, and Climatology 

Session Description

Satellite, radar, lightning and numerical model data all contain information that can be used to predict the onset of convective hazards.  Machine learning trains a model to learn linear and non-linear relationships between the best predictors and the predictand, or the target (“truth”) that the model is designed to predict. Historical or archive data are used to train algorithms in order to predict future events. This Training session will incorporate three different machine-learning tools have been developed at CIMSS to predict different aspects of convective weather: NOAA/CIMSS ProbSevere (that uses ABI, radar, NWP model, and lightning data), IntenseStormNet (that relies on ABI and GLM data only) and LightningCast (that uses only ABI data); the products can be used operationally to monitor the development and evolution of severe weather, strong convection, and lightning, and the hands-on activities of this session will help the user understand how the products were created, and how the different data sources drive the probabilistic output.

 

John Cintineo, Research Meteorologist
NOAA/OAR/NSSL, Warning Research and Development Division

John Cintineo smiling in front of a white background.

John Cintineo received his bachelor's degree in atmospheric science from Cornell University and his master's degree in meteorology from the University of Oklahoma. As a research meteorologist for the Cooperative Institute for Meteorological Satellite Studies (CIMSS) at the University of Wisconsin -- Madison, he served as the primary developer of the ProbSevere portfolio of machine-learning models, some of which are in widespread use by NOAA's National Weather Service. ProbSevere products use remotely sensed observations from satellite, radar, and lightning platforms, as well as environmental data, for the short-term forecast of convective hazards. At CIMSS, he has also been involved in research projects supporting volcanic ash cloud detection and diagnosis, as well as NOAA's Next-Generation Fire System. John recently moved to NOAA's National Severe Storms Laboratory, where he continues to research and develop methods to improve the operational prediction of convective hazards.

Contact Information
Email: john.cintineo@noaa.gov

 

Scott Lindstrom, Scientist/Trainer
CIMSS, University of Wisconsin-Madison

Scott Lindstrom smiling in a car.

Scott Lindstrom has a BS in Meteorology (and one in Comp Sci) from Penn State, and a Masters and Phd from the UW-Madison.  For the last 20 years, much of his work has centered on advocating for the use of Satellite Data and products in the forecast process.  He creates various training products for the National Weather Service (Quick Guides, Quick Briefs, Blog Posts), for the Hazardous Weather Testbed, and for JPSS (with an especial focus on NUCAPS).  He has done in-person training internationally on Satellite imagery and Tropical and Synoptic Meteorology and has also helped with earlier SATMOC short course presentations.

Contact Information
Email:  scott.lindstrom@noaa.gov