The Illumination of Thunderclouds by Lightning: 3. Retrieving Optical Source Altitude

Area of Science:

• Atmospheric Science and Meteorology
• Remote Sensing
• Lightning Physics

Background:

• Geostationary Lightning Mapper (GLM) detects lightning via cloud top illumination but lacks native altitude reporting.
• Previous research indicates a correlation between optical source altitude and thundercloud illumination patterns.
• Accurate lightning altitude is crucial for understanding thunderstorm dynamics and charge structure.

Purpose of the Study:

• To develop a method for extracting lightning altitude information from GLM optical data.
• To utilize machine learning to predict lightning source altitude using spatial energy distribution metrics.
• To enable three-dimensional (3D) lightning geolocation using GLM data.

Main Methods:

• Matched GLM 'groups' with Lightning Mapping Array (LMA) 3D source location data.
• Employed machine learning algorithms to predict mean LMA source altitudes from GLM optical data features (amplitude, breadth, texture).
• Validated model performance against LMA-derived altitudes.

Main Results:

• Developed a model predicting LMA mean source altitude from GLM group data with a median absolute error <1.5 km.
• The model successfully identified the altitude of the optical energy's origin within thunderstorms.
• Predicted altitudes captured variations linked to convective processes and revealed vertical flash structure.

Conclusions:

• The developed model enables 3D lightning geolocation using GLM data for the first time.
• This technique provides insights into the vertical structure of thunderstorms and lightning flashes.
• Future research will refine altitude estimation by considering thunderstorm variability and sensor viewing angles.