AE31A-0021
Does Wilson's cloud chamber offer clues on lightning initiation in thunderclouds?
The experimental evidence indicates that the large scale electric field in the cloud at the time of lightning initiation is about 100 kV/m [1], which is an order of magnitude lower than the expected conventional breakdown field. One important problem in atmospheric physics is to understand how lightning flashes are initiated in such low fields. Some scientists suggest that the electric field could reach higher values momentarily in small regions and this combined with the field enhancing action of hydrometeors in the cloud could provide trigger for lightning initiation [2, 3]. Others suggest that energetic electrons produced by cosmic rays could give rise to runaway electron avalanches generating the initial ionization necessary for lightning initiation [4]. Nguyen and Michnowski [2] suggested that in small cloud regions the electric field may exceed 200 to 400 kV/m and in these locations the discharges between hydrometeors could facilitate lightning initiation. This mechanism was further investigated by Cooray et al. [3] who showed that interaction between adjacent hydrometeors cannot produce a streamer discharge, a prerequisite for electric breakdown, unless the field exceeds about 830 kV/m. They also found that long chains of hydrometeors could initiate streamer discharges in relatively low electric fields. For example, in order to generate a streamer discharge in 100 kV/m electric field the length of the chain of hydrometeors of 0.1 mm radius should be about 65 mm with more than 100 particles constituting the chain. However, the question remains on how such long chains of hydrometeors can be produced in the cloud. We suggest the following possibility. Consider an energetic particle passing through the cloud producing ionization in its wake. The passage of such a particle will lead to a stream of positive ions and electrons with the latter being captured within a few tens of nanoseconds by oxygen molecules to form negative ions. If the water vapor in the region under consideration is supersaturated, water molecules will condense on the ions and the resulting droplets can grow to tens of micrometers in a fraction of a second. This is the mechanism utilized in Wilson's cloud chamber to visualize the tracks of ionizing particles. If the track of ionizing particle is aligned with the direction of the electric field in the cloud, the resultant drift of the oppositely charged particles in opposite directions will facilitate collisions among them leading to production of larger droplets. This process can potentially generate long chains of droplets in the cloud which may provide the trigger necessary for the initiation of lightning flashes. [1] Marshall, T. C., M. P. McCarthy and W. D. Rust, Electric field magnitudes and lightning initiation in thunderstorms, J. Geophys. Res., vol. 100, pp. 7097 7103, 1995. [2] Nguyen, M. D. and S. Michnowski, On the initiation of lightning discharges in a cloud, 2. The lightning initiation on precipitation particles, J. Geophys. Res., vol. 101, pp. 26 675 26 680, 1996. [3] Cooray, V., M. Berg, M. Akyuz and A. Larsson, Initiation of ground flashes: some microscopic electrical processes associated with precipitation particles, Proc. International Conference on Lightning Protection, Birmingham, UK, 2002. [4] Gurevich, A. V., G. M. Milikh and J. A. Valdivia, Model of X-ray emission and fast preconditioning during a thunderstorm, Phys. Lett., A 231, pp. 402 408, 1997.
AE31A-0022
Preliminary Results From a Laboratory Study of Positive Streamer Discharges on Simulated Ice Hydrometeors
The initiation of lightning remains an open question, due in large part to a deficit of in-situ observational evidence. Recent theoretical descriptions of lightning initiation have focused on runaway breakdown and related secondary processes, but have not convincingly explained the details of onset of the embryonic lightning leader channel. Among possible mechanisms contributing to the initial leader formation are positive streamer discharges from ice hydrometeors, themselves once favored as the primary explanation of lightning initiation. We present preliminary results from a new laboratory study of positive streamer discharges on simulated ice hydrometeors. Emphasis is given to precisely defining the minimum electric field strength required for onset of positive streamer generation, with variables of interest being ice crystal size, habit and environmental temperature.
AE31A-0023
Ionic Channels in Thunderclouds
We proceed to study the formation and propagation of ionic channels in thunderclouds in the framework of the model of the corona discharge wave propagation (Fomenko A.S., Losseva T.V., Nemtchinov I.V. The corona discharge waves in thunderclouds and formation of ionic channels // 2004 Fall Meeting. EOS Trans. AGU. 2004. V. 85. š 47. Suppl. Abstract AE23A-0835.). In this model we proposed a hypothesis that the structure of a thundercloud becomes nonuniform due to corona discharge on the drops and ice particles and formation of ionic channels with higher conductivity than the surrounding air. When the onset strength of corona discharge becomes smaller than the electric field strength the corona discharge increases concentrations of ions in a small part of the cloud (a hot spot). An additional charge at opposite ends of the hot spot forms due to polarization process. The increased electric field initiates corona discharge in other parts of the cloud on ice particles and water drops with smaller sizes. The corona discharge front moves as a wave with the velocity of the order of ion drift and formes a highly conductive channel. We model this non-stationary problem with Poisson equation which is solved simultaneously with a simplified set of kinetic equations for ions, small charged particles and electrons (at high electric fields), including ionization due to electronic impact, attachment and formation of positive ions. By applying 3D numerical simulations we obtain the parameters of formed ionic channels with respect to onset electric fields both from large particles (in hot spot) and from small particles (surrounding hot spot), microscopic currents from particles with different sizes and the external electric field in the cloud. The interaction of ionic channels is also investigated. This work was supported by Russian Foundation of Basic Research (Project No 07-05-00998-ŕ).
AE31A-0024
Three Dimensional Behaviour Analysis Of Lightning Channel By CFD Modelling
One of the mechanisms assumed to be involved in the ignition of the recoil streamers phase is associated with the development of instability in the lightning channel. Like any other electric arcs, lightning channels become unstable if the provided power is insufficient. A criterion has been derived from Heckman (1) who shows the minimum current to sustain an electrical channel in air as a function of its length. In order to study this phenomenon, ONERA in collaboration with EDF has performed 3D simulations of a part of the lightning channel in order to compute the main characteristic of the arc and understand in the future the physical conditions for the ignition of instability. In a first step, purpose of this paper, the computations have been compared to the experiment of Tanaka et al (2). They have experimentally studied free burning arcs of 1.6 and 3.2 meter long, subjected to continuous current of 100 and 2000A with duration higher than 100ms. The results show that the characteristics of the arc column do not depend on the gap length, which means that the electrodes have no effect on it. Consequently, the electrical arc produced is very similar to a part of the lightning channel. (1) Heckman S., Why does a lightning flash have multiple strokes? Thesis (PH.D.), Massachusetts Institute of Technology, 1992 (2) Tanaka S., Sunabe K. and Goda Y., Three Dimensional behaviour analysis of D.C. free arc column by image processing technique, XIII Intl Conf on Gas Discharges and their applications, Glasgow, 2000
AE31A-0025
Electrical Structure and Preliminary Breakdown of Lightning Flashes in Three New Mexico Thunderstorms
In this presentation, we compare propagation characteristics of lightning flashes from Lightning Mapping Array data to simultaneous electric field (E) measurements at different altitudes for three storms. For normal cloud-to- ground flashes, detailed analysis has shown that when a potential well for negative charge exists between the altitude of the flash initiation and ground, the first return stroke occurs only after a period of horizontal negative polarity branching below the initiation altitude. This period of preliminary breakdown, or the delay between the initiation time and the first return stroke, lasted an average of 117 ms for 14 flashes that initiated when a low-level potential well was present. When no low-level potential well was indicated in the E data, the time between flash initiation and first return stroke averaged 15 ms. Existence of preliminary breakdown for more than about 40 ms before the first return stroke of a cloud-to-ground flash thus suggests that a low-level potential well for negative charge is present at that time in the storm. Examples will be shown that give an indication of the rapid temporal evolution in the electrical structure of the storm as it relates to the lightning activity.
AE31A-0026
Lightning Charge Transport During a Thunderstorm Near Langmuir Laboratory on 18 August 2004
On 18 August 2004, 19:46 UT, a balloon was launched from Langmuir Laboratory, New Mexico, and was carried about 30 km southeast by the wind, traveling at an altitude that varied between 2500 m up to 4300 m above sea level. An electric field sonde or Esonde measured the electric field during the flight, while simultaneously, New Mexico Tech's Lightning Mapping Array recorded the location of VHF pulses generated during lightning. During the 40 minute balloon flight, 182 flashes occurred within 15 km of the balloon; 41 of these flashes were cloud-to-ground flashes. Using the Esonde data and the LMA data, along with recently developed analysis techniques, we estimate the charge transport associated with the lightning occurring near the balloon. This analysis leads to a picture for how charge flows in the 30 km region observed by the balloon-borne E-field change instrument. http://www.math.ufl.edu/~aslan
AE31A-0027
Correlation between lightning flash count and local meteorological parameters
It is interesting to know the likelihood of an electrified storm in the afternoon or evening based on morning measurements of meteorological parameters. Soundings are generally used to obtain such measurements, but not all locations are near to a National Weather Service sounding site. Thus, a statistical study was conducted comparing lightning flash counts with all of the following: mixing ratio (MR) of water to air, temperature, winds, and convective available potential energy (CAPE). Meteorological parameters were compared with lightning flash counts on the ground and at pressure alititudes of 500 hPa at 1200 and 0000 UTC. The study covered an 85-day period in the summer of 2005, with lightning data from the Los Alamos Sferic Array (LASA), and meteorological data from upper-air balloon soundings at NWS stations. In a log-log plot of mixing ratio at ground level and lightning flash count, we found a correlation coefficient r=+0.7 in the southwestern United States. Southwestern summer days with MR greater than 7-8 g/kg at 1200 UTC frequently produce lightning. In Oklahoma, we found r=- 0.5 correlation in a log-log plot of dry-bulb temperature (at 500 hPa) vs. flash count, with a decrease of this temperature corresponding to an increase in the flash count. The strongest correlation for Pennsylvania was r=+0.5 between CAPE and lightning. We also verified at some sites that the correlation was strongest for lightning confined to within 200 km of the sounding site, strengthening our confidence that the measurements of the local air mass were revealing a causal relationship to lightning activity. We will discuss why the best correlating parameters varied from region to region in the context of the theory of colliding ice and riming graupel as the dominant source of electrification in thunderstorms.
AE31A-0028
Radiation Patterns of Lightning
New Mexico Tech's Lightning Mapping Array (LMA) uses time-of-arrival measurements at a network of stations to accurately locate VHF radiation sources in three spatial dimensions and time. The peak power values are also measured at each station but have not been fully utilized. Currently the source power calculations assume an isotropic source and an isotropic receiver gain pattern. The event source power is approximated by the median value of the contributing stations. The receiver gain pattern of individual LMA stations were empirically determined, and are in agreement with analytical formulations. Incorporating the receiver gain pattern improved the source power calculation, reducing the standard deviation of the individual station powers to about 1.5 dB. Using the corrected power calculations, three 10-minute time intervals comprising of 7.5 million sources, were used to look for source radiation patterns having 15 dB nulls relative to the median source power. Of the 167,720 candidate events 244 had an identifiable radiation source pattern. These patterns ranged from simple dipole patterns to relativistically beamed patterns with a velocity 99 percent the speed of light. It is expected that many more sources had nulls that were not as strong or that were not directed at a station.
AE31A-0029
Analysis of Charge Transport During Lightning Using Balloon-Born Electric Field Sensors and Lightning Mapping Array
Techniques are developed for processing the wide band measurements of electric field obtained by a balloon-borne electric field sonde (or Esonde), and for estimating the charge transport associated with lightning. The techniques use Lightning Mapping Array measurements of the VHF pulses generated during lightning recorded simultaneously with the Esonde data. A filtering algorithm is developed to separate the background field associated with instrument rotation and cloud charging processes from the lightning-induced electric field change. The charge transport associated with lightning is approximated by constrained monopoles and dipoles. The constraints used to achieve a unique fit include conservation of charge, charge separation constraints, location constraints associated with the observed LMA pulses, and statistical constraints based on estimated errors in instrument calibration. The location constraint is handled using a "pulse graph," a graph whose vertices coincide with the observed LMA pulses. The techniques are illustrated using electric field data measured on 18 August 2004 near Langmuir Laboratory. In our analysis we observe that current flow lags behind the LMA detected channel formation by on the order of 0.1 s, roughly the same time delay observed for lightning optical signals detected by NASA's Lightning Imaging Sensor. http://www.math.ufl.edu/~hager/Lightning
AE31A-0030
High-Speed Electric Field Measurements and Lightning Mapping Observations at Langmuir Laboratory
During the summer of 2007 we recorded broadband RF radiation waveforms, together with ground-based slow and fast electric field waveforms, from lightning in the vicinity of Langmuir Laboratory in central New Mexico. Three-dimensional observations of the lightning flashes from New Mexico Tech's Lightning Mapping Array (LMA) give context for interpreting the RF and electric field waveforms. In addition, high-speed in-situ vector electric field waveforms were recorded by a newly developed balloon-borne instrument. The broadband RF radiation was received using a Rhode & Schwarz model HE010 active rod antenna with a 100~MHz bandwidth, which was sampled at up to 400~MHz. Log-detected RF waveforms for two side-by-side LMA stations, one operating at 63 MHz (TV channel 3) and the other at 183 MHz (TV channel 9), and ground-based fast and slow electric field waveforms, were digitized at 25~MHz. All ground-based waveforms had a depth of 1~second, and were time-tagged with a GPS receiver. These comprehensive datasets are being used to study various processes associated with lightning flashes --- in particular, to detect high-frequency electrical activity at the onset of a flash. The ground based measurements, in conjunction with the in-situ vector field measurements, will be used to study the transfer of charge during a flash. Measurements of the log-detected RF (which is used by the LMA to map lightning flashes), in conjunction with the broadband RF and electric field waveforms, will help us better understand what types of events are detected and located by the LMA. In this paper we will present the results for a few interesting lightning flashes. ~
AE31A-0031
Assessing Patterns in the Surface Electric Field Prior to First CG Flashes and After Last CG Flashes in Air-Mass Thunderstorms
In an effort to elicit patterns in the temporal and spatial evolution of the contours of surface electric field relevant to the occurrence of cloud-to-ground (CG) lightning, we have analyzed data from the network of 31 electric-field mills jointly operated by the John F. Kennedy Space Center (KSC) and Cape Canaveral Air Force Station (CCAFS). To identify cases of interest, we used lightning ground-strike data, maps of in-cloud lightning discharges, rainfall data, and radar data. In particular, we have focused on two critical problems: 1) estimation of when and where the first CG flash in a storm might occur and 2) assessment of the likelihood of CG flashes occurring late in a storm after a long period without a CG flash. Our long-term goal is to understand the evolution of surface contours of electric field for periods of 30 minutes or more before the first flash of any kind and 30 minutes or more before and after the last flash of any kind. For practical reasons, we are reporting here on analysis of data for periods of 30 minutes before the first CG flash and 30 minutes after the last CG flash in each storm of interest. We have analyzed electric-field data from isolated air-mass convective storms that developed over KSC/CCAFS from late May through early September, 2004-2006. To identify thunderstorms that fit the air-mass, or "pop-up" criteria, we started by examining rainfall and CG lightning data, then looked at radar data. Then, for the storms selected, we performed a two-pass Barnes objective analysis on the electric-field data. Each analysis cycle resulted in one contour plot of 20-second averaged data, yielding 90 plots for each 30 minute interval, which we then animated. This resulted in 58 animations of the field contours prior to first CG flashes and 62 animations of the field contours after last CG flashes. Preliminary impressions from examinations of these cases suggest that the electric-field contours before the first flash exhibit a smooth transition from weak to strong gradient, but the strongest field just before the flash may occur up to several kilometers distant from the ground-strike point. Furthermore it appears that in several cases, the field remained strong throughout a broad area for relatively long periods after the last flash.
AE31A-0032
Eyewall Lightning and Hurricane Intensification
Hurricane eyewall lightning is seldom observed during the development and quiescent periods of a hurricane's lifetime. However, observational evidence collected during the 2004 and 2005 hurricane seasons suggests that abundant eyewall lightning is present during periods of rapid intensification. We provide a conceptual model of hurricane electrification during the rapid intensification phase, and analyze the current and historical Atlantic hurricane records to determine frequency and strength of rapid intensification. We correlate hurricane intensification with eyewall lightning frequency using the National Lightning Detection Network (NLDN) and Los Alamos Sferic Array (LASA) lightning databases to assess the potential efficacy of remote eyewall lightning observations for hurricane intensity monitoring.
AE31A-0033
Estimations of Current Densities Above, Below, and Within a Thunderstorm
Thunderstorms are believed to be the main generators in the Global Electric Circuit (GEC). In this presentation we look at the time history of conduction current density above, below, and within a thunderstorm by combining ground and balloon electric field measurements with assumed conductivity profiles. We will compare below- cloud conduction currents to estimates of corona currents. The measurements at the ground allow us to estimate the Maxwell current density produced by the thunderstorm. We find that the ground measurements sometimes provide a good estimation of the current density above the storm. Combining the ground measurements with in- cloud conduction current measurements gives a lower bound on the storm's generator current density.
AE31A-0034
Modeling Transient and Quasi-stationary Electric Fields in a Thundercloud
A numerical model describing the stationary and transient electric fields in thunderclouds is developed. The stationary electric structure of the storm is modeled as a set of horizontal layers of the external current in accordance of the balloon measurements of the electric field profile. The lightning flashes are modeled as a short time currents producing dipole charge structure in the thundercloud in the case of intracloud flash and monopole charge region in the case of cloud to ground flash. The relaxation of the flash charge is described. The model gives the total current flowing from the storm top to the ionosphere and estimates the transient charge transfer to the ionosphere for different flash types. Dependence of the charge transfer on the parameters of the flash and the conductivity modification in the vicinity of the storm is considered. Using the balloon measurements of the electric field through a number of MCS stratiform precipitation regions their electric structure and parameters are modeled.
AE31A-0035
Transient Currents in the Global Electric Circuit due to CG and IC Lightning
Intracloud (IC) and cloud-to-ground (CG) lightning flashes produce transient changes in the electric field (E) above a thundercloud, and this E produces a transient current in the global electric circuit. In a previous presentation, an electodynamic model simulation was developed and used to estimate the charge transferred between cloud top and the ionosphere by the above-cloud transient E resulting from a single CG lightning flash. An alternative model to estimate above-cloud charge transfers has recently been developed, and this new model is applied to a larger group of both CG and IC flashes. For example, two CG flashes moved -11 C and -12 C of charge between cloud and ground and +7 C and +5 C upward from the cloudtop. Two IC flashes that moved 85 C and 62 C inside the cloud caused -7 C and -8 C to move upward from the cloudtop. The results of the two models will be compared to one another. The main goal of this presentation is to show that above-cloud transient currents resulting from lightning may play an important role in the global electric circuit.
AE31A-0036
Thunderstorm Precipitation Current Measurements
Precipitation currents inside and below thunderstorms and electrified shower clouds are thought to be important contributors to the global electric circuit. Using balloon measurements, we find precipitation current densities of both polarities inside four mature stage storms range from 10-40 nA/m2. Measurements in one decaying storm give precipitation currents of only 1-2 nA/m2 (of both polarities) during the End-Of-Storm-Oscillation (EOSO). From these data we estimate the storm-scale precipitation current and compare the results to earlier estimates of this parameter in the global circuit.
AE31A-0037 INVITED
ELF Q-bursts from African Squall Lines
A number of large amplitude ELF transient signals (Q-bursts) are documented at multiple sites around the world (Japan, Hungary, Israel and USA) in association with westward moving mesoscale convective systems (MCSs) in West Africa during the African Monsoon and Multidisciplinary Analysis campaign in 2006. Some of these bursts are associated with red sprites first observed from ground-based measurement in Africa. Using MIT Doppler radar and electric field measurements locally installed in Niamey, we investigate quantitatively the meteorological conditions responsible for generating these exceptionally large Q-bursts. Detailed meteorological information is provided from radar such as the spatio-temporal evolution of radar echo, while the electrical properties oflarge Q-bursts (e.g. charge moment charge (CMC)) are experimentally derived by using the remote-sensing method taking into account the theory for the earth-ionoshere wave guide. Furthermore, detailed propagation characteristics of Q-bursts and their effect on the location and remotely derived accuracy of the CMC will be discussed by comparing the results from multi ELF stations under different ionospheric conditions such as day-night asymmetry and the ionospheric terminator lines. Preliminary results from similar campaign in 2007 in Niger will also be presented to be compared with those in 2006.
AE31A-0038
Atmospheric Electricity Measurements on the Proposed European Venus Explorer Mission
The European Venus Explorer (EVE) mission has been proposed to the European Space Agency as part of their Cosmic Vision programme. It comprises an orbiter, balloon and descent probe to provide detailed measurements of the atmosphere's physical and chemical properties. The proposed combination of in situ measurements, with an orbiter for remote sensing, will constrain cloud microphysics and contribute to the study of Venus' climate and radiative balance, in addition to providing detailed corroboration of existing observations. The electrical instrumentation is motivated by the lack of in situ measurements at Venus. It aims to elucidate the detailed nature of any discharges, and determine the role of atmospheric electricity in chemical and transport processes (e.g. through the formation of trace species by lightning). Measurements of non-lightning atmospheric electrical processes on Venus will also contribute to understanding of trace gas chemistry, transport processes, and even cloud formation. The EVE orbiter will carry an electromagnetic instrumentation suite comprising a wave analyser, a camera to identify upper atmosphere discharges, thought to be likely on Venus, and a detector searching for gamma-ray bursts associated with electrical activity. The EVE balloon will carry optical lightning detectors and an electromagnetic event counter. This unique instrumentation will permit synchronous detection of atmospheric electrical events both optically, electrical and via gamma-ray bursts. Ionization by cosmic rays, which can penetrate to the surface of Venus, is a lightning-independent source of atmospheric charge, which will produce molecular cluster-ions, charged aerosols, and consequently atmospheric electric fields and a finite air conductivity. In addition to the lightning detectors, the EVE balloon will carry a permittivity instrument, which will detect the "fair weather" electrical properties (conductivity, electric field) from which ion and aerosol characteristics can be inferred.
AE31A-0039
Laboratory Measurements of Charging of Apollo 17 Lunar Dust Grains by Low Energy Electrons
It is well recognized that the charging properties of individual micron/sub-micron size dust grains by various processes are expected to be substantially different from the currently available measurements made on bulk materials. Solar UV radiation and the solar wind plasma charge micron size dust grains on the lunar surface with virtually no atmosphere. The electrostatically charged dust grains are believed to be levitated and transported long distances over the lunar terminator from the day to the night side. The current models do not fully explain the lunar dust phenomena and laboratory measurements are needed to experimentally determine the charging propertied of lunar dust grains. An experimental facility has been developed in the Dusty Plasma Laboratory at NASA- Marshall Space Flight Center for investigating the charging properties of individual micron/sub-micron size positively or negatively charged dust grains by levitating them in an electrodynamic balance in simulated space environments. In this paper, we present laboratory measurements on charging of Apollo 17 individual lunar dust grains by low energy electron beams in the 5-100 eV energy range. The measurements are made by levitating Apollo 17 dust grains of 0.2 to 10 um diameters, in an electrodynamic balance and exposing them to mono- energetic electron beams. The charging rates and the equilibrium potentials produced by direct electron impact and by secondary electron emission processes are discussed.
AE31A-0040
Electrical activity and dust lifting on Earth and Mars
Dusty phenomena, such as dust devils and dust storms, are the primary sources of atmospheric dust on both Earth and Mars, and are the most important currently active geological processes on Mars. Electric fields larger than 100 kV/m have been measured in terrestrial dusty phenomena. Such electric fields can substantially reduce the critical wind speed necessary for wind to move sand grains and can even directly lift them from the surface (Kok and Renno, 2006). On Mars, these electric fields have potentially important implications for atmospheric chemistry. Indeed, they dissociate water vapor and produce large amounts of hydrogen peroxide (Atreya et al., 2006). The emission of dust is driven by saltation, the process by which sand grains bounce on the surface and eject the smaller, harder to lift, dust particles into the air. We developed the first physically based numerical model for saltation that has been tested with data from measurements. Our model includes the generation of electric fields and the effects of electric forces on saltation. Preliminary results confirm that electric fields strong enough to directly lift particles from the surface can develop in terrestrial dust storms and dust devils. Our model results suggest that on Mars electric fields frequently exceed the 25 kV/m for the thin Martian air to break down. We thus expect electric discharges to commonly occur in Martian dust storms, which could have important implications for atmospheric chemistry and dust lifting.
AE31A-0041
Electron Distribution Functions in Gas Mixtures Relevant to Planetary Atmospheres
A kinetic solution to the Boltzmann equation for electrons will be presented. Electron distribution functions will be calculated over a range of 0.1 eV 50 MeV for gas mixtures containing molecular nitrogen, molecular oxygen, carbon dioxide, hydrogen, helium, and methane subject to an applied electric field. Results for each gas individually, and gas mixtures where available, will be benchmarked against Swarm' experiments published in the literature. Drift velocity and characteristic energy as well as attachment and ionization rates will be considered. Additional results for gas mixtures relevant to planetary atmospheres will be presented. In light of these distribution functions the characteristics of runaway and conventional breakdown in various planetary atmospheres will be discussed as well as their implications for the possible emission of gamma rays and optical photons from these events.
AE31A-0042
Sprites and lightning in Venus: constraints for observations by the Planet-C mission
Lightning activity in Venus has been mystery for long period, although many studies based on observations both by spacecrafts and by ground-based telescope have been carried out. This situation may be attributed to the ambiguity of these evidential measurements. In order to conclude this controversial subject, we are developing a new type of lightning detector, LAC (Lightning and Airglow Camera), which will be onboard Planet-C (Venus Climate Orbiter: VCO). PLanet-C will be launched in 2010 by JAXA. To distinguish optical lightning flash from other pulsing noises, high-speed sampling at 50kHz for each pixel, that enables us to investigate the time variation of each lightning flash phenomenon, is adopted. On the other hand, spatial resolution is not first priority. For this purpose we developed new type of APD (avalanche photo diode) array with a format of 8 x 8. Narrow band interference filter at wavelength of 777.4 nm (OI), which is expected lightning color based on laboratory discharge experiment, is chosen for lightning measurement. LAC detects lightning flash with an optical intensity of average of Earthfs lightning or less at a distance of 3 Rv. We also present results of theoretical calculations of the expected occurrance heights and emissions of sprites above thunderstorms in the CO2 atmosphere of Venus and the Hydrogen-Helium atmospheres of Jupiter and Saturn. General detection methodology of sprites/lightning in planetary atmospheres by orbiting spacecraft will be discussed.
AE31A-0043
Lightning in the Saturnian system
The Cassini/RPWS (Radio and Plasma Wave Science) instrument has measured the so-called SEDs (Saturn Electrostatic Discharges) which are the radio signatures of lightning in Saturn's atmosphere. During the Cassini mission SEDs have occurred infrequently, but a typical SED storm lasts several weeks with a highly variable flash rate, and in most cases there has been also a correlated cloud feature observed by the Cassini cameras. We will discuss the main physical characteristics of these SEDs like flash duration and rates, spectral radio power, and their polarization, and we will make a comparison with terrestrial lightning flashes. We will also report on our search for bursty radio signals indicative of lightning from Titan. Although lightning seems possible in Titan's thick nitrogen and methane dominated atmosphere, no clear signature has been detected at any of the first 35 close Titan flybys.
AE31A-0044
Generation of Runaway Electrons Induced by Radon Decay Products in Thundercloud Electric Fields
Gamma ray dose-rate increases associated with winter thunderstorm activities have been observed in the coastal areas facing the Sea of Japan. In order to investigate the generation of energetic photons which originate in thunderstorm electric fields, we have calculated the behavior of electrons and photons in electric fields by the Monte Carlo method. We have carried out the calculations of the beta and gamma rays emitted by radon decay products, as a source of energetic electrons, in thundercloud electric fields. From the calculated results, it is confirmed that the energy deposition increased greatly and an enormous amount of the electron-ion pairs will be generated in the high field region. Since the radon decay products form a large part of the energetic electrons in the atmosphere, they can serve as the source of a considerable amount of electrons. Furthermore, we have developed a transport model of radon and its decay products for the Asian region based on a regional meteorological model and a material transport model to determine the mechanism for occurrences of lightning during the winter season in the coastal areas of the Sea of Japan. The analyzed concentration distribution clarifies that seasonal winds carry a large amount of radon from the Asian continent to the coastal area of the Sea of Japan. A large amount of radon and its decay products from the continent stays in that area between the ground and the altitude of 3 km msl. This result suggests that regions with strong electric fields in winter thunderclouds may be filled with a large concentration of radon. From these results, it suggests that beta and gamma rays emitted by radon progenies cause electromagnetic showers and the production of copious runaway electrons in the thundercloud electric fields.
AE31A-0045
Monte Carlo simulation of the temporal behavior of Terrestrial Gamma ray Flashes
Based on TGF-measurements from BATSE it has been found that high energy photons (>300 keV) are observed earlier than low energy photons (25-50 keV), with an average of 240 μs. The explanation for this time dispersion has not yet been identified. Using a Monte Carlo simulation where the Compton scattering, pair production and absorption of X-rays are included, we find that the time dispersion, for most of the TGF, can be explained as a pure Compton effect, assuming a discrete TGF-production altitude. This is because the majority of low energy photons escaping the atmosphere are originally high energy photons that have experienced multiple Compton scattering and energy reduction. Each Compton scattering will make the photon's traveling path longer, which result in a longer traveling time. Our simulations show that the time delay increases as the TGF production altitude decreases, and as a satellite's nadir angle of measurements increases. The initial TGF spatial distribution at the producing altitude is also important for the time delay. If the TGF distribution is isotropic within a solid angle, a satellite measuring at angels inside this cone will hardly observe any time delay. These findings are supported by the BATSE measurements.
AE31A-0046
X-ray Production in Laboratory Sparks in air
Dwyer et al. [2005] recently observed X-ray bursts (in the 30 to 150 keV range) from high-voltage laboratory sparks (positive and negative; 5 cm - 2 m) in air. In this paper, we present results of an independent experiment conducted at a different high-voltage facility and using different instrumentation to confirm (or refute) the production of X-rays by laboratory discharges. The experiment was conducted at the high-voltage laboratory of Uppsala University, Sweden. A spark was created in air at atmospheric pressure applying a standard lightning impulse voltage (1.2/50 impulse; front time: 1.2 μs, time to half-value: 50 μs) to an 80-cm long rod-to-hemisphere air gap. In this experiment the generator was charged to 1 MV with negative polarity. The voltage across the gap at breakdown, the current in the spark, the visible optical radiation from the spark, and the emission of X-rays produced by the spark were measured. The emission of X-rays was detected by a barium fluoride (BaF2) crystal scintillator which had a shape of frustum of a right cone (front diameter: 2.2 cm, rear diameter: 4 cm, and length: 4 cm). The scintillator was attached to a 51 mm PMT (Photonis XP2020/URQ, rise time 1.4 ns, spectral sensitivity 200-550 nm). The ungrounded, light-tight, and electromagnetically sealed metal cabinet having the complete X-ray detector, optical measuring system, oscilloscope and power supplies, was placed about 1 m away from the spark gap so that the center of the crystal was a few centimetres below the high voltage rod electrode. A total of 83 negative voltage impulses were applied across the gap and in 49 cases (59% of the events) the X-ray signals were detected. In 23 cases, the X-ray signal appeared before the collapse of the applied voltage across the gap and about 1 μs prior to onset of the main discharge current. In 26 cases, the X-ray signal appeared at around the main discharge current peak. A rough estimation of total deposited energy in the crystal was carried out based on measured signals from three X-ray sources, which are 137Cs, 60Co, and 241Am. The lowest total deposited energy was calculated to be around 30 keV. The amplitudes and the width of the detected X-ray signals were different for different events and some signals were saturated. The average total deposited energy of the unsaturated signals was calculated to be around 170 keV. The total deposited energy in the case of saturated signals is in the order of few MeV. The detected X-ray signal is consistent with the response of the BaF2 detector to a signal from a 137Cs radioactive source placed on the aluminum-covered-window of the metal cabinet. Moreover, a comparison to the X-ray and optical signals shows clearly that the first recorded X-ray signal, which appears before the collapse of the applied voltage across the gap, is associated with the pre-discharge activity in the gap. The second recorded X-ray signal appears during the collapse of the applied voltage across the gap. There were 4 common cases out of 49 when the signals appeared simultaneously both before and during the collapse of the gap. No X-rays were detected from sparks with positive voltage impulse applied to the rod for the same gap geometry, although in this case the breakdown voltage was lower than that corresponding to the negative voltages.
AE31A-0047
Search for Thunderstorm Associated Nuclear Gamma Rays
We present results from further analysis of thunderstorm data taken with a liquid nitrogen-cooled germanium spectrometer with energy range from 13 keV-2.6 MeV that was setup on South Baldy Peak at Langmuir Laboratory in New Mexico during June through August of 2005. The purpose of the experiment was to investigate the suggestion made by Greenfield et al. (2003) that delayed gamma ray emissions associated with thunderstorms may be attributed to Chlorine-39 and Chlorine-38. We improve upon results of Boggs et al. (2005) by doing a careful spectral calibration to improve sensitivity.Two methods to determine the presence of chlorine emission lines were used; we created storm length background-subtracted spectra, and also examined the change in count rates in energy bins that correspond to the Chlorine emission energies. Both these methods confirm the conclusions of Boggs et al. that chlorine emission was not detected and any signature of Chlorine production was below the detectability of the detector. These results lead to an upper limit on photon flux in chlorine line emission that can be used to place an upper limit on chlorine production during thunderstorms.
AE31A-0048
Spatial distribution of X-ray bursts in triggered lightning experiments
During rocket-triggered lightning experiments, it has been observed that the dart leader phase is accompanied by bursts of X-rays. These experiments were carried out at the University of Florida/Florida Tech International Center for Lightning Research and Testing (ICLRT) at Camp Blanding, FL. We report here the analysis of a sequence of X-ray bursts that occurred during the dart leader of a lightning flash triggered on July 31, 2007. The measurement of the penetrating X-ray photons was done with the Thunderstorm Energetic Radiation Array (TERA), which consists of more than 40 NaI detectors systematically distributed at various distances within 500 m from the launching tower. The integrated observed X-ray intensity at ground level over a short time prior to the return stroke shows a strong radial decay. On the other hand, the integrated radial X-ray distribution is rather flat at earlier times. The measurements are compared with simulations that include the physics of photon propagation through air. The results are consistent with a source that travels from high altitude to near ground level (the top of the rocket launching tower).
AE31A-0049
Energetic electron beams injected into the magnetosphere by terrestrial gamma-ray flashes
We shall present evidence that at least 6 of the 36 BATSE Terrestrial Gamma-ray Flashes (TGFs) that are publicly available for analysis were produced by electron beams directly striking the spacecraft and not by gamma-rays. These electron beams, which can have energies extending above 30 MeV, form a previously unidentified population of energetic electrons in the inner magnetosphere. Recent work modeling the emission of terrestrial gamma-ray flashes as measured by the RHESSI satellite has shown that the source of TGFs is likely to be deep within the atmosphere, at altitudes ranging from about 21 km down to 15 km. As a result, the runaway electrons that create the gamma-rays via bremsstrahlung interactions with air do not escape to space from these depths. On the other hand, electrons and positions created by Compton scattering and pair-production by the gamma- rays near the top of the atmosphere can escape along the geomagnetic field lines and can be detected by orbiting spacecraft, either near the TGF or near the far conjugate point. Indeed, two of the BATSE events are found to be conjugate events originating from thousands of kilometers away from the far geomagnetic foot point. In this presentation, we will present the results of detailed Monte Carlo simulations that show that the intensities of the electron beams and the gamma-rays associated with the TGFs are similar at spacecraft altitudes and that the time-intensity profiles and anisotropy data of several BATSE events can be well fit by these simulations.
AE31A-0050
Meteorological Context of RHESSI Terrestrial Gamma-ray Flashes
The Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI) has detected over 750 Terrestrial Gamma-ray Flashes (TGFs). The RHESSI data provide a large database for statistical studies of TGF characteristics and production environments, such as location and storm type and evolution. We present a statistical comparison of the spectral characteristics and brightness of RHESSI TGFs detected over land and sea. We also examine the timing of TGF events with respect to the storm evolution. The development stage of the storm is estimated from the time sequence of cloud to ground sferics measured by the World Wide Lightning Location Network (WWLLN). Finally we attempt to localize a few TGFs by identifying isolated thunderstorm events that occur in the RHESSI footprint.
AE31A-0051
Unusual RHESSI TGFs: Electron Beams and Others
We will summarize observations and conclusions about two unusual Terrestrial Gamma-ray Flashes (TGFs) observed by the RHESSI satellite. Both events are unusually bright and occur in the Northern hemisphere during the winter -- unusual in itself, since TGFs are overall a phenomenon associated with summer storms. The first event, on January 17, 2004, can be modeled as the signature of the satellite passing through an electron beam ejected from the atmosphere at the magnetic conjugate point. The mechanism behind this process is now understood, and will be presented here and, in more depth, in the accompanying paper by Dwyer, Grefenstette, and Smith in this session. The second event occurred in the Mediterranean -- the only TGF ever reported in that region -- on November 7, 2004. This event seems to be associated with a nearby thunderstorm, not a conjugate one, but is still extremely unusual. We will examine the question of whether its unusual location and brightness is due to its being another electron beam event (caught on the way out instead of the way back in), or whether the storm itself was unique. Finally, we will briefly show the other RHESSI TGFs with the most extreme behavior in duration, average energy per photon, and geographical location.
AE31A-0052
Observations of X-Rays and Electric Field in Thunderstorms: Revisited
In 1994 and 1995 we launched balloon-borne x-ray spectrometers and electric-field meters into thunderstorms to test the hypothesis that strong electric-fields could accelerate cosmic-ray secondary electrons and produce bremsstrahlung x-rays. There were four flights into the stratiform regions of mesoscale convective systems. In three of the flights, increases in x-ray intensity were observed in conjunction with strong electric fields, while on flight four, large x-ray pulses were observed above the MCS. Both negative and positive polarity electric fields (as referenced to the vertical field) produced x rays. Each of these events lasted for times on the order of ten of seconds, demonstrating that it was the strong quasi-electrostatic field, not lightning, that caused the x-ray emissions. It was also observed that the increased x-ray intensity would return to near background levels after lightning reduced the local electric field. The observations indicate that x-rays observed above background are produced by strong electric field present in thunderstorms, most likely through a runaway electron process. In addition, they also show that the runaway electron process can occur over long periods of time without initiating a large-scale lightning discharge.