Terrestrial Gamma-Ray Flash Intensity Distribution Andrew B. Collier 1,2 , Thomas Gjesteland 3 , Nikolai Østgaard 3 1 Hermanus Magnetic Observatory, Hermanus, P.O. Box 32, Hermanus, 7200, South Africa (collierab@gmail.com ) 2 School of Physics, University of KwaZulu-Natal, Westville Campus, Durban, 4000, South Africa 3 Physics and Technology, University of Bergen, Postboks 7803, Bergen, 5020, Norway Abstract The most probable source lightning discharges associated with TGFs detected by the RHESSI satellite are determined from WWLLN data. Of the 972 TGFs considered, matches were found for 93 events. For these the causative lightning was found to occur at distances between 51 and 769 km, with a mean distance of 314 km. The average delay between the matched WWLLN events and the corresponding TGFs was found to be Δt = -0.773 s, indicating that the TGFs preceded the lightning peak power. The distribution of matched TGFs indicates that WWLLN is more likely to identify lightning strokes associated with weaker TGFs. 1. Introduction Terrestrial Gamma-ray Flashes (TGFs) are brief bursts of highly energetic photons observed by Low Earth Orbit (LEO) satellites [1] and are associated in both location and time with nearby thunderstorm activity [1,2,3]. Electrons are thought to be accelerated to relativistic speeds by thunderstorm electric fields and subsequently produce a deluge of bremsstrahlung photons with energies extending up to 40 MeV [3,4,5]. A catalogue of TGFs has been compiled using data from the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI) satellite [6]. The catalogue includes 972 events distributed in latitude between 29.4° S and 38.0° N, and localised over the three major zones of lightning activity in the Americas, Africa and the Maritime Continent. The TGFs are identified with temporal precision of 1 ms. Some uncertainty exists with regard to the absolute accuracy of the RHESSI timing and all times have been adjusted by 1.8 ms [6]. The World Wide Lightning Location Network (WWLLN) provides continuous coverage of global lightning activity [7]. WWLLN is biased towards the most intense lightning, with a threshold peak current of around 50 kA [7,8]. The detection efficiency of WWLLN is currently around 10.3% for CG and 4.8% for IC discharges [9]. The spatial and temporal accuracy of the WWLLN events are 10 km and 30 μs respectively. Locations identified by WWLLN are only considered to be valid if the timing residual is less than 30 μs and data from 5 or more stations are involved. These restrictions eliminate the majority of the data from the first few years of WWLLN operation. The results thus only become highly reliable after around August 2003. 2. Results and Discussion The identification of TGF source strokes in the WWLLN was a two stage process. For each of the TGFs in the catalogue, first all WWLLN events within a 20 minute window centred on the TGF time were identified. The distance between each of these events and RHESSI was determined at the time of the TGF and the WWLLN time was then corrected for the light propagation time to the satellite. Those WWLLN events with a corrected time in a 20 ms window centred on the TGF and located less than 2400 km from the RHESSI sub-satellite point were considered to be matches. If multiple candidates existed for a given TGF then the one that minimised the product of delay and distance was chosen. Figure 1 illustrates the results of this analysis, showing cases where a match was found in the WWLLN data and one where no match was present despite there being a significant amount of lightning activity in the vicinity of the sub-satellite point. Comparison of the WWLLN events to data from the Lightning Imaging Sensor (LIS) satellite detector indicate that the locations of the WWLLN events are broadly consistent with those from LIS and thus validating the WWLLN data. Using this procedure matches were found for 93 of the TGFs. Figure 1: Location of WWLLN lightning activity (blue dots) within a 20 minute period centred on the TGF event for cases where (left) a matching source lightning stroke is found and (right) no match is found. The plots are centred on the RHESSI sub-satellite point. Lightning data from LIS are included (red circles) for comparison. The approximate LIS field of view at the time of the TGF is indicated by the green shaded circle. The distances from the RHESSI sub-satellite point to the corresponding WWLLN match ranged from 51 to 769 km, with a mean of 314 km. This range should be compared with the results of [10] who identified causative lightning for RHESSI TGFs at distances exceeding 960 km, while [11] found that all 15 of the matches for TGFs observed by Fermi GBM occurred within 300 km of the sub-satellite point. It is probable that these differences relate to the sensitivities of RHESSI and Fermi GBM. The distribution of WWLLN lightning activity at the times of TGFs as a function of the distance from the RHESSI sub-satellite point is displayed in Figure 2. It is apparent that regardless of whether or not a match was found in WWLLN, the level of lightning activity in the vicinity of the satellite is enhanced. However, for the matched TGFs the activity is significantly higher. At large distances both the matched and unmatched curves converge on the background level of WWLLN lightning activity. 978-1-4244-5118-0/11/$26.00 ©2011 IEEE