All Physics Faculty Publications

Title

Statistical Analysis of Space-time Relationships Between Sprites and Lightning

Document Type

Article

Journal/Book Title/Conference

Journal of Atmospheric and Solar-Terrestrial Physics

Volume

65

Issue

5

Publisher

Elsevier

Publication Date

3-2003

First Page

525

Last Page

535

DOI

10.1016/S1364-6826(02)00326-7

Abstract

We present a detailed statistical analysis of the association of 40 sprite events with lightning from the parent thunderstorm. Both temporal and spatial criteria were used to identify the parent cloud-to-ground (CG) lightning. Sprite images were GPS time stamped and their locations triangulated. In contrast to previous reports of nearly one-to-one association of sprites with positive cloud-to-ground (+CG) lightning, 11 events (27%) did not have a +CG recorded by the National Lightning Detection Network (NLDN), and 7 events (17%) had neither NLDN nor very low frequency (VLF) signatures associated with them. A negative cloud-to-ground (−CG) preceded one of these events by 9 ms. As expected for ∼16:7 ms integrated images, none of the sprites without a +CG had any discernible visual characteristic that would distinguish them from “regular positive sprites”. We have calculated the distribution of time intervals (At = tsprite − tlightning) for the sprites that had a parent +CG Cash registered by the NLDN or VLF systems, and the distribution of distances between the sprite nadir positions and the Cash locations registered by the NLDN. The time interval (At) distribution had a peak around 10–20 ms and a mean of 30 ms (total). This distribution is broadly consistent with the characteristic single electron avalanche time scale associated with streamer growth between ∼70 and 85 km. The distribution of the distances (As) between the nadir point of sprites and the parent +CGs showed that approximately two-thirds of the sprites occurred within 50 km lateral displacement from the parent +CG. The parent +CG peak current distribution had a maximum at 40–50 kA and mean of 60 kA, suggesting that high peak currents (I ¿75 kA) are not a necessary prerequisite for sprites. The peak current distribution for all +CGs of the storm, with a maximum around 10–20 kA and mean of 27 kA, exhibits a qualitatively diEerent form from the peak current distribution of the parent +CGs producing sprites.

Comments

Published by Elsevier in Journal of Atmospheric and Solar-Terrestrial Physics.

http://www.sciencedirect.com/science/article/pii/S1364682602003267

Publisher PDF is available for download through the link above.

http://www.sciencedirect.com/science/article/pii/S1364682602003267