Preliminary Results from the Inclusion of Lightning Type and Polarity in the Identification of Severe Storms

To date, the literature on the use of lightning information to distinguish severe thunderstorms from ordinary storms has focused primarily on the “lightning jump”, the rapid increase in total flash rate that is generally associated with a strengthening of the storm. The lightning jump, however, ignores other characteristics such as the proportions of cloud and cloud-to-ground (CG) flashes within the total flash rate and the proportions of positive- and negative-polarity flashes, CG strokes, and individual cloud pulses within the flashes. In this work, we present results from an analysis of more than 1000 thunderstorm cells sampled from most of the continental U.S. by the U.S. National Lightning Detection Network (NLDN) on a number of days in 2015 and 2016 when both severe and non-severe thunderstorms occurred. The sample was chosen to represent the real proportion of severe thunderstorms, generally considered to be around 10% or less of all thunderstorms in the U.S. All of these storms were observed by the NLDN in its post-2013 upgraded state, when its cloud flash detection efficiency was between 40-60% and its CG flash detection efficiency was more than 95%, and thus, the NLDN served as a continental-scale total lightning network. In this study, the severe storms were identified on the basis of proximity in space and time to severe storm reports compiled by the NOAA Storm Prediction Center. The objective of the study is a preliminary assessment of whether the inclusion of parameters such as cloud flash fraction and separate rates of positive and negative cloud and CG flashes, strokes, and pulses provide any additional value on top of the total flash rate in terms of distinguishing between severe and non-severe storms in a representative sample of all thunderstorms from around the U.S.