Radar Thresholds and Lightning Prediction at Wallops Flight Facility

A study was conducted at Wallops Flight Facility (WFF) to determine a radar reflectivity threshold associated with lightning to provide greater lead time for WFF lightning alerts and to improve lightning prediction during launch operations. Earth Networks Total Lightning Network and WSR-88D Level II radar data were used from March 2013 through the summer of 2017, amassing over 200 lightning events within 25nm of WFF during both warm and cold seasons. Radar reflectivity values were recorded preceding and after the first lightning strike occurred within 25nm of WFF at mixed phase isothermal heights (-10C and -15C) based on RAP model sounding data. A subset of 100 of the 200 total events were divided based on whether a storm had active lightning moving into 25nm of WFF or developed its first lightning strike within 25nm of WFF. An additional real-time test case was conducted during the spring and summer of 2017 to refine the radar thresholds preceding lightning within 25nm of WFF to determine the best combination of probability of detection (POD), false alarm rate (FAR), and critical success index (CSI). The results of this study and the real-time test case show that the best indicator of lightning was a radar reflectivity of greater than 42 dBZ at -10C or 38 dBZ at -15C, with a POD of 0.88. Vertically integrated liquid (VIL) values and VIL density (VILD) values were recorded for the subset of 100 events with the best indicators of lightning for VIL at 8.03 kg/m² and VILD at 1.27 kg/m³. VIL and VILD POD were slightly lower at 0.74 and 0.70, respectively. The lead time computed for cells that developed lightning with 25nm of WFF on average ranged from 11 to 13 minutes for radar reflectivity, VIL, and VILD. Quantitative and qualitative analysis of this study shows that best practice for identifying cells that will produce lightning within 25nm of WFF is by using the radar reflectivity values as a baseline identifier and the VIL and VILD values as supplementary data. The reflectivity, VIL, and VILD values found in this study could be used in an alerting based system at WFF to quickly identify storms capable of producing lightning based on model or sounding thermal profiles and radar data. This study at WFF will provide improved lightning prediction using radar data to provide timely lightning alerts for launch build up activities, as well as during launch operations.