Ground and Space-Based Observations of Lightning

The Geostationary Lightning Mapper (GLM) has many uses operationally, especially in the National Weather Service with situational awareness and severe weather nowcasting and forecasting. The purpose of this project was to look at the efficiency of the GLM on a frame-by-frame basis. The GLM is on the GOES-16 satellite and can detect total in-cloud and cloud-to-ground lightning activity. High-speed video from a Photron camera attached to a trigger can take video at 8 to 12.5 thousand frames per second. The amount of light in the series of frames that are 80-125 microseconds apart are graphed into 2 millisecond increments, which enables the detection of lightning based on peaks in data. The GLM data is read in through the HUDAT program for direct comparison. The relationship between the GLM and the Photron frames are quantified using a contingency table. A set of skill scores of the False Alarm Ratio (FAR), Probability of Detection (POD), and Critical Success Index (CSI) of the frame efficiency were determined. The results when looking on a microscale basis was that the reprocessed 2017 data had better skill scores for all three: FAR, POD and CSI, when compared to the operational data from 2018. Possible reasons for this discrepancy could be different filtering algorithms, frame splitting, and in-cloud lightning out of the field of view. Overall, even with these discrepancies the frame detection efficiency could be improved, but the overall flash detection was 100%.