GEOPHYSICAL RESEARCH LETTERS, VOL. 26, NO. 24, PAGES 3605-3608, DECEMBER 15, 1999 Sprites triggered by negative lightning discharges C. P. Barrington-Leigh and U. S. Inan STAR Laboratory, Stanford University, Stanford, CA 94305 M. Stanley Department of Physics, New Mexico Tech, Socorro, NM 87801 S. A. Cummer NASA/Goddard Space Flight Center, Greenbelt, MD 20771 Abstract. High altitude air breakdown, manifested as “red Intensified broad-spectrum CCD video observations were sprites,” is reported in close association with negative cloud- made from Langmuir Laboratory (33.98◦N×107.19◦W) on to-ground lightning (−CG)onatleasttwooccasionsabove the night of August 29 1998, using a frame-strapped timing an unusual storm on August 29, 1998. Data from high speed system [Rairden and Mende, 1995] with 33 ms integration photometry, low-light-level video, and receivers of lightning time and GPS time stamping with 1 ms precision. The electromagnetic signatures in the frequency range 10 Hz to 15◦×20◦ field of view of the camera was aligned with that 20 kHz are used to establish the association and indicate of the Fly’s Eye photometric array [Inan et al, 1997], con- that the causative −CG discharges effected unusually large sisting of nine horizontally aligned photometers, each with ◦ ◦ ◦ ◦ vertical charge moment changes (∆MQv) of up to 1550 C· a ∼2.2 ×1.1 field of view, and a central one with a 6 ×3 km in 5 ms. The existence of sprites caused by −CG’s, field of view. Each photometer bore a red (> 670 nm) filter rather than the regularly associated +CG’s, has immediate and was calibrated using a Hoffman Engineering Corpora- implications for sprite models and observations. tion Spectral Radiance Standard. Data from the array were Introduction recorded digitally in a triggered mode, along with the output of a 350 Hz to 20 kHz Very Low Frequency (VLF) receiver, Sprites are often described as an electric discharge or at a sample rate of 60 kHz per channel. The VLF receiver breakdown at mesospheric altitudes occurring above large was fed by an air-core magnetic loop antenna, oriented in a positive cloud to ground (+CG) lightning. While sprites are vertical plane 23◦ east of north. A GPS clock was used for known to be associated with +CG discharges [Sentman et timing with <1 ms absolute uncertainty. al., 1995; Winckler et al., 1996; Lyons, 1996], not all sprites In addition, a calibrated ELF/VLF (10 Hz to 20 kHz) closely follow such a discharge, or any recorded discharge recording with the same timing system was made at Stan- at all [Franz et al., 1990; Boccippio et al., 1995; Winckler, ford (37.42◦N×122.17◦W). These data were time aligned 1995]. Winckler [1998] reports three sprites each occurring with the Langmuir Laboratory data and allowed a deter- within one second of nearby −CG’s, but provides no spe- ◦ mination of the vertical currents flowing on time scales of 1 cific evidence of an association closer than one second or 2 to 10 ms. (∼11 km) of viewing azimuth. In our observations we reg- ularly recorded sprites associated with a sequence of CG’s Observations spaced by 10 to 50 ms. More often, sprites are closely asso- ◦ ◦ ciated with a large +CG which moves a large positive charge A thunderstorm system centered at 112.7 W×29.8 N(in- (∆MQv of 250 to 3250 C·km in Cummer and Inan [1997]; set, Figure 1) that was part of a major mesoscale convec- 200 to 1100 C·km in Bell et al. [1998]), and in the case of especially impulsive lightning (>60 kA, in Barrington-Leigh 200 km Langmuir Lab and Inan [1999]) the CG is followed by “elves,” the flash due +CG to widespread heating of the lower ionosphere by the elec- 19 CA -CG tromagnetic pulse of lightning. In this paper, we report evi- 15 16 dence of at least two sprites that are closely associated with Mexico AZ negative cloud-to-ground (−CG) lightning strokes. Among 14 our observations, these events are unique. 17 6 18 13710 12 9 3 1 Instrumentation 11 5 8 4 We use three lines of evidence to show that a storm over 2 − northwestern Mexico produced two CG-associated sprites. 20 km Copyright 1999 by the American Geophysical Union. Figure 1. NLDN-recorded flashes from a nighttime MCS on Paper number 1999GL9010692. August 29, 1998. The inset shows the storm which produced 00948276/99/1999GL9010692$05.00 many large −CG’s. Numbered events are detailed in Table 1. 3605 3606 BARRINGTON-LEIGH ET AL.: SPRITES TRIGGERED BY NEGATIVE LIGHTNING Table 1. NLDN-recorded large CG events (>60 kA) sprites with vertical (columnar) structure, despite interven- clustered around the −CG sprites, between the times ing cloud cover and the large distance (694 km) of the storm 05:49:00 UT and 06:49:00 UT. Charge moment refers to from Langmuir Laboratory. The full vertical extent of the the change in the first 5 ms after the onset of the sferic. sprites is difficult to ascertain, as their apparent lower limit Events 2, 15, 17, and possibly 18 were associated with may be due to a foreground cloud. Figure 2 also shows observed sprites. the vertical current moment and cumulative vertical charge moment change ∆MQv extracted from the calibrated sferic Time NLDN current Charge moment receiver at Stanford with the method described in Cummer − − · 1 05:49:13 73 kA 230 C km and Inan [1997]. By 5 ms after the arrival of the sferic, · 2 05:49:23 +69 kA +480 C km ∆M ≈ 1380 C·km, indicating an abnormally high con- − − · Qv 3 05:49:31 64 kA 110 C km tinuing current for a −CG [Uman, 1987, p. 172 and 341]. 4 05:52:39 −62 kA −120 C·km ∆MQv before the onset of the second optical peak, or by 5 05:52:51 −71 kA −140 C·km · − − · about 1.38 ms after the onset of the sferic, is 750 C km, well 6 05:53:07 79 kA 370 C km above the 250 C·km threshold observed for the production of 7 05:53:30 −69 kA −280 C·km sprites associated with +CG’s in Cummer and Inan [1997]. 8 05:54:23 −78 kA −80 C·km Figure 3 shows another similar event, corresponding to 9 05:56:47 −64 kA −140 C·km a −CG recorded by NLDN at 06:11:14.808 UT with peak 10 05:58:28 −69 kA −120 C·km current of −93 kA. This discharge (event 17 in Figure 1) 11 05:59:22 −71 kA −130 C·km produced similar unambiguous video recording of columnar 12 06:01:00 −91 kA −270 C·km sprite luminosity between 70 and 80 km through an opening 13 06:09:36 −73 kA −310 C·km in the foreground clouds. The photometric channels and 14 06:10:02 −79 kA −200 C·km 15 06:11:14 −93 kA −1550 C·km ELF sferic also exhibit evidence of elves and a high current 16 06:13:39 −64 kA −300 C·km moment, respectively. None of the photometers are pointed 17 06:15:16 −97 kA −1380 C·km directly at this sprite, however, so none of the photometers 18 06:18:14 −110 kA −1340 C·km shows an obvious second pulse in luminosity for the event. 19 06:48:04 +120 kA +1000 C·km Event 18, recorded at 06:18:14.239 UT, has very similar properties as the two others, but because the region below tive system (MCS) over the northern Gulf of California in Mexico produced mainly −CG lightning on August 29. Fig- ure 1 shows the +CG (+) and −CG (◦) discharges recorded P3 by the National Lightning Detection Network (NLDN, see P2 P4 P5 Cummins et al., [1998]) with peak current >10 kA between P6 P7 P8 05:49 and 06:49 UT. The largest events (peak currents over 90 60 kA) in the storm region of interest are numbered and o listed in Table 1. According to NLDN, no +CG’s (>10 2 60 kA) occurred in this region during the 22 minutes prior to P11 each of the two unusual events recorded at 06:11:14.808, and 40 horizontal o 06:15:16.305. Indeed, over the entire duration of the storm 229.1 E NLDN recorded only five +CG flashes from this active re- 1998-241 06:15:16 289-339 ms (b) ∼ P1 350 kR gion, constituting 1.5% of the total flashes. In contrast, P2 310 90 km the larger MCS surrounding this region exhibited a +CG P3 210 occurrence of ∼ 6%. P4 Figure 2(a) shows one of these unusual events, each of P5 1190 P6 700 which consists of a closely associated −CG flash, an elves P7 310 70 km − P8 370 event, and accompanying sprites. NLDN recorded a 97 (c) P11 P9 500 kA stroke (event 17 in Figure 1) at 06:15:16.305 UT. The 4 kR P11 (log scale) polarity of the sferics recorded at Langmuir and at Stanford 2nd pulse (Figure 2) is unambiguous at this range and confirms the Sferic polarity of the lightning. The bearing to this stroke and 1.5 nT Current moment Charge moment ⋅ the altitudes overlying its location are shown on the video -660 kA⋅km -1900 Ckm image. The video’s pointing direction was determined with (a) 305 306 307 308 309 310 311 312 313 Time (ms) after 06:15:16 UT star field alignment.