Bayesian techniques to analyze and merge lightning locating system data

As more lightning locating systems (LLSs) become available, there is a growing need to assess how each LLS performs and how to best merge data from multiple LLSs. A Bayesian analysis is used to compare the worldwide data of LLSs from three providers for November 2014 to October 2015: Earth Networks Total Lightning Network (ENTLN, Earth Networks), the combined data from the Global Lightning Detection 360 and National Lightning Detection Network (Vaisala), and the World Wide Lightning Location Network (WWLLN, University of Washington). By using the union of the data sets we are able to determine an estimate for the upper limit of the absolute detection efficiency of each system. Globally, ENTLN detected 56.8% of the discharges, the combined Vaisala networks detected 59.8%, and WWLLN detected 7.9%. In addition, there were 2.842 × 10 9 unique discharges detected by these LLSs, an average of 90.1 strokes/s.