Developing Forecast Techniques for Lightning Cessation Using Radar, NLDN, and LMA Analyses
MetadataShow full item record
Dual-polarized radar data are used to develop forecast techniques for total lightning cessation for the Houston, Texas region. A total of 42 isolated thunderstorms observed in the summer months of June, July and August of 2013 through 2015 were analyzed. In-cloud (IC) lightning data were obtained from the Houston Lightning Mapping Array (LMA), and cloud-to-ground (CG) lightning data were obtained from the National Lightning Detection Network (NLDN). Archived radar data from the Houston Next Generation Radar (NEXRAD) radar were obtained from the National Climatic Data Center and converted to constant altitude plan projection indicator (CAPPI) format to examine at the -10, -15, and -20°C levels. NEXRAD radar products of reflectivity and differential reflectivity were used to optimize a radar-based forecast technique for the cessation of lightning, and correlation coefficient was used to determine the quality of data. Results show that the best forecast technique for the cessation of total lightning (IC and CG lightning) was the height of the 30 dBZ reflectivity value decreasing below the -15°C level. This technique produced a Critical Success Index (CSI) of 82%. Results also show that the best forecast technique for the cessation of CG lightning was the 35 dBZ height decreasing below the -10°C level, which produced a CSI of 89%. Four differential reflectivity thresholds of -0.5, 0.5, 1, and 1.5 dB were observed in an area of 30 dBZ or greater to determine the presence of graupel and supercooled water droplets that cause charge separation and lightning within a thunderstorm. Results show that no large percentage of these thresholds decreased at a time before or after the cessation of lightning. Therefore it is concluded that differential reflectivity did not aid in forecasting the cessation of lightning.
Sink, Amanda Marie (2016). Developing Forecast Techniques for Lightning Cessation Using Radar, NLDN, and LMA Analyses. Master's thesis, Texas A & M University. Available electronically from