Evaluation of a long-range lightning detection network with receivers in Europe and Africa

Abstract

This study focuses on the performance evaluation of the experimental long-range lightning detection network Zeus, with receivers located in Europe and Africa. The evaluation is carried out in terms of locating error and detection performance. Zeus' receivers record the radio noise (sferics) emitted by cloud-to-ground lightning discharges in the very low frequency (bandwidth between 5 and 15 kHz). The lightning location is retrieved by means of the arrival time difference (ATD) triangulation technique. The presented study includes evaluation of the system's locating error and detection performance over the African continent, equatorial/mid-Atlantic, the Caribbean, and northern South America. This dataset is validated over a period extending from July through October 2004. The Lightning Imaging Sensor onboard NASA's Tropical Rainfall Measuring Mission satellite is used as the validation source. The analysis shows that the number of receivers as well as their relative location (relative position of the flash-event to the network's receivers) plays a key role in the retrieval error magnitude. Median error values within the network's periphery are found to be approximately 20 km, while outside the periphery significantly larger (/spl sim/130 km). Simulation of chi-square values, implemented from the ATD algorithm, shows an adequate agreement between theoretical error-related computations and the network's retrievals. In terms of detection performance, the Zeus system is also validated over the same area/period. The chief observation is that the detection performance is likewise dependent on the number of receivers used by the system rather than their relative location.