Lightning warnings with NLDN cloud and cloud-to-ground lightning data

Previous studies [1,2] have evaluated the performance of the U.S. National Lightning Detection Network (NLDN) with respect to warnings of cloud-to-ground lightning (CG) at specific locations such as airports. The statistics used to evaluate warning performance have included probability of detection (POD), false alarm ratio (FAR), duration of warnings, and number of storms reported. In 2013, the NLDN was upgraded to Vaisala´s LS7002 sensors replacing the older generation LS7001 and IMPACT sensors. As a result of this upgrade, the NLDN is expected to have a 50% cloud flash detection efficiency and a 95% cloud-to-ground flash detection efficiency [3]. In this study, we examine the warning performance using cloud and cloud-to-ground data from the NLDN after the 2013 upgrade. Data from several regions of the continental U.S. are considered. The warning statistics are evaluated with and without cloud lightning data. The impact of varying outer radii, inner verification areas, and end-of-storm wait times are examined. Upon conclusion of the study, it was found that addition of NLDN cloud lightning added about 10% to the POD, while the FAR was unchanged, and the percent time under valid warning and false alarms increased somewhat. As found previously, a 15-minute warning expiration time was better than 10 minutes, and POD decreased steadily for longer lead times for CG lightning. POD increased steadily for larger outer radii, but FAR also increased. Finally, a short exploration of a smaller warning area showed better warning statistics, but there was significant sensitivity to the relative size of the inner versus outer areas.