Using Lightning Data in the Quest for Real-time Detection of Wild Fires Using the ABI Instrument Onboard GOES-16

Abstract:

NOAA’s geostationary GOES-16 satellite is equipped with the Advanced Baseline Imager (ABI) which gives access to near-real-time infrared and visible range images over the western hemisphere. Hall et al. (2019) showed that the current active fire product used with the ABI data has large omission (i.e. missed events) and commission (i.e. false alarm) errors of 84% and 88%, respectively. We present our findings regarding the use of neural networks and lightning data from the National Lightning Detection Network (NLDN) to achieve lower omission and commission errors, while taking full advantage of the near-real-time data supplied by the ABI instrument. In addition, we specifically look at the relationship between forest fires and lightning events in order to better understand the dynamic relationship between the two phenomena. These findings allow us to discuss the igniting sources of wildfires and lead to faster, more accurate, and higher confidence detections.