Analysis and Automated Detection of Convection-Induced Aviation Weather Hazards in Visible and Infrared Satellite Imagery


Kris Bedka talks about methods helping to detect hazardous weather situations for aviation.

Length: 30 min

Author: Kris Bedka (NASA)


Current generation geostationary satellites are observing convection that is hazardous to aviation at increasingly high spatio-temporal detail. In recent years, commercial and research aircraft have collected automated turbulence and cloud ice water content observations that can be used to better understand exactly where within deep convection the turbulence and icing conditions are typically occurring. Ground-based weather radar and severe weather reports also identify locations of hail, downburst wind, and tornadoes. Research conducted at NASA Langley Research Center (LaRC), in collaboration with a number of U.S. and international partners, has resulted in geostationary-based analyses and automated detection algorithms that can denote where turbulence, icing, and severe weather conditions are likely. These methods are applicable to any geostationary visible and IR imager across the globe and therefore can be used to map these weather hazards in nearreal time, a capability that is especially valuable over regions without weather radars and other conventional observations of aviation hazards.


Go to webcast

Lecture slides