A Fuzzy Logic Method for Lightning Prediction Using Thermodynamic and Kinematic Parameters from Radio Sounding Observations in South Korea

Abstract

Abstract Lightning is one of the most troubling weather phenomena for weather forecasters at space centers. In this study, proximity sounding and lightning data were used to evaluate the utility of thermodynamic and kinematic parameters for forecasting lightning prior to launch operations. Various parameters from 4138 radio sounding observations at five sites and cloud-to-ground (CG) stroke data from the Korea Meteorological Administration’s Lightning Detection Network (KLDN) over South Korea during 2004–09 were used. To support launch operations, forecasts of the total membership function for lightning (TMF) were derived from the combination of membership functions of selected thermodynamic and kinematic parameters with each objective weight using a fuzzy logic algorithm. The forecast skill of TMF was evaluated by computing several skill statistics, which include probability of detection (POD), false alarm rate (FAR), percent correct (PC), critical success index (CSI), and the true skill statistic (TSS). The lightning forecasting method for Gwangju, South Korea (site nearest to the Naro Space Center), was found to have a POD of 0.68, an FAR of 0.45, a PC of 0.76, a CSI of 0.44, and a TSS of 0.47.