AE42A-01 INVITED
Occurrence and distribution of global TLE activities and their effects
During the 3-year survey of ISUAL experiment on the FORMOSAT-2 satellite, thousands of TLEs including sprites, elves, halos and jets have been recorded. The most dominated type (~80%) of TLEs was recognized as elves, and only 20% of the recorded events are sprites and halos. In addition, ISUAL also recorded 13 gigantic jets during this time span. Distributions of the ISUAL TLEs show that sprites congregate mainly over continents as lightning does, whereas elves scatter mostly over oceans. The global occurrence rates of TLEs are also derived. However, a detailed study of the peak CG current distribution and the ISUAL imager detection threshold indicates that only around 1% CG lightning are able to generate ISUAL detectable elves. The true occurrence rates of TLEs could range from a factor of two to an order magnitude higher than detected rates. The strong dependency of elve occurrence on the sea surface temperature, updraft flow of global atmosphere circulation, and precipitation implies that the warm tropical oceans act as the heat reservoirs to drive vertical convection and accelerate charge accumulation by rainfall, thus produce intense oceanic lightnings, which induce a large fraction of observed elves. This process suggests that oceans, atmosphere and ionosphere are coupled. Sprites and jets are able to transport electrons between thunderclouds and lower ionosphere, and elves contribute substantial free electrons and ions at ~90km height from EMP-heated ionization. In this presentation we also discuss the role of TLE in the global circuit and atmospheric chemistry quantitatively.
AE42A-02
Blue And Gigantic Jets From Taiwan 2007 TLE Campaign
Blue and gigantic jets are believed to be blue luminous phenomena. In Taiwan 2007 TLE campaign, a multi- wavelength imaging system was deployed with the aim to elucidate the physical and chemical characteristics of TLEs. On 22 July 2007, twenty blue/gigantic jets and four sprites were observed to occur over a frontal system in Fujian province of China, about 400km away from our observation site at Lulin Observatory, Taiwan. All the observed jets showed little blue band (380-510nm) emissions but had easily recognizable signals in red band (570-2700nm). This result indicates that jets observed from ground are reddish and most of the blue emissions are extinct. One of the jets was observed to propagate upward to ~75km elevation, thus it can be identified as a GJ and is similar to the GJ-event observed by Pasko et al (2001). This GJ was launched 200ms after a small jet from the same cloud top. This implies that the smaller jet could be regarded as the leader of this GJ, similar to the stepped/dart leader in a CG flash. Since the blue luminous events in ISUAL data (Su, et. al. 2005 AGU) have similar features as the jets in this ground observation. Thus, we can conclude that the ISUAL blue luminous events also are blue jets or blue starters. Even though blue jets and blue starter have different ISUAL SP2 (N2 2P, 337.0nm), SP3 (N2 1N, 391.4nm) and SP6 (250-390nm) intensities, but the peak ratio between SP2 and SP6 are ~ 0.5 and the peak ratio between SP3 and SP2 are ~ 0.07. This means blue jets and starters possess the same spectral properties and the same degree of ionization. Finally, the relation between jets and the nearby lightning will also be addressed.
AE42A-03 INVITED
Very Active Sprite-Producing Thunderstorms Observed Over Argentina and Brazil
To investigate sprite activity over Argentina and Brazil, a field campaign was conducted from the Brazilian Southern Space Observatory near Santa Maria, Rio Grande do Sul. We observed more than 500 sprites and halos during two large thunderstorm systems on February 23 and March 4, 2006, which included the third most active sprite-producing storm ever reported. We present an overview of the campaign, which includes sprite video images and extremely low to very low frequency (ELF/VLF) electric and magnetic field data recorded at the observatory, along with coincident lightning location data from the World Wide Lightning Location Network (WWLLN). We also present meteorological conditions inferred from GOES Infrared imagery, TRMM rainfall data, WWLLN lightning locations, local radar data, and the BRAMS mesoscale model. Within the high sprite activity, we identified several sprites driven by negative cloud-to-ground (CG) lightning. Downward directed electric fields should be absent following negative CG lightning, so the observation of negative CG sprites challenges runaway breakdown theories of sprite production. We show that the negative CG sprites were less bright and had limited vertical extent compared with positive CG sprites from the same storm. A second finding is that these storms were large mesoscale convective systems with total cloud shields of 200,000 to 500,000 km2 and that most sprites occurred in stratiform regions where cloud-top temperatures were about 15°C warmer than spriting regions of US High Plains storms. http://earthweb.ess.washington.edu/lnk/jnt/
AE42A-04
IR Temperature Characteristics of a Prolific TLE Producing Storm in South America Observed During a Sprite Campaign in Brazil
A total of 445 Transient Luminous Events, TLEs, were observed above a Mesoscale Convective System, MCS, over Argentina on the night of 22-23 February, 2006, during the third sprite campaign ever conduced in Brazil. We investigated the temporal-spatial evolution of the cloudtop temperature of the MCS based on satellite infrared imagery and their relationship with cloud-to-ground lightning and TLE activity. We used the same methodology São Sabbas and Sentman [2003] applied to a US sprite producing MCS and compared the results. The Argentinean thunderstorm was a mesoscale system with very different characteristics than regular TLE producing MCSs. It originated from the merge of several convective storms that remained as collection of "isolated" convective regions (Tc < -52o C) under a common cloud cover (Tc < - 32o C) throughout its lifetime. We started to record TLEs at least 3 hr before the system could be classified as an MCS and the temperature of the cloud cover, ~ -52o C, was at ~15 o C higher than of the US MCS. Most of the lightning tended to occur at the borders of the convective cores instead of occurring at the cores themselves. Another difference was that while the US MCS produced sprites only in the back region, the Argentina system produced sprites/TLEs both in the back and front regions. The sprites occurred above regions with ~ -40o C in the beginning of the observations and with ~ -58o C at the end. This also differs from the US case study, in which all sprites happened above regions with ~ -70o C and ~ -64o C. The main similarity was that in the growth phase the lightning occurrence rate increased in association with regions with decreasing cloudtop temperatures (~ -50o C to ~ -58o C ), until a maximum rate was reached in association with minimum temperatures, however in the US MCS the temperatures involved were lower (~ -64o C to ~ -70o C). Another interesting similarly between the US and Argentina MCSs is that both moved towards Northeast at ~85 km/h and ~45 km/h, respectively. The overall development of this system was very complex and highly structured, which may have contribute in unpredicted ways to the exceptional TLE activity documented.
AE42A-05 INVITED
Sprite Processes Quantified With Submillisecond Video and Electromagnetic Measurements
During a summer 2005 observation campaign, high speed imaging and ground-based electric and magnetic field measurements were combined in an effort to probe the details of the connection between the low altitude lightning processes and high altitude transient luminous events (TLEs). The low frequency sensitivity of the magnetic field sensors enabled measurement of the continuing lightning current that is involved in many complex sprites but that can be difficult to detect by other means, while the high speed video gave precise timing of TLE onset and features relative to the driving lightning processes below. We report results from an ongoing detailed and quantitative analysis of this combined data set, including the following elements: measurements of streamer dynamics, including velocity profiles and observations of nearly constant deceleration of many downward streamers below about 60 km altitude; observations of the lightning processes involved in the initiation of delayed sprites; the quantitative agreement between conventional breakdown theory and measurements of sprite initiation time and altitude combined with the integrated lightning charge moment change; and measurements of high altitude sprite current and an analysis of possible sources of this mesospheric charge transfer.
AE42A-06
Photoionization Models Based on Radiative Transfer and Helmholtz Equations for Sprite Streamer Modeling
Sprite streamers are driven by highly nonlinear space charge waves [Raizer, 1991, p. 327], and develop in a self- consistent manner. The photoionization produced by UV photons originating from a region of high electric field in the streamer head is responsible for creation of seed electrons in front of the head of a propagating streamer, and is believed to play a critical role in the spatial advancement of both positive and negative streamers. The accurate and efficient evaluation of the effects of photoionization remains one of the most challenging tasks in streamer modeling. In the current literature, it is a common practice to evaluate the photoionization effects using an integral model proposed by Zheleznyak et al. [High Temp., 20, 357, 1982] for non-thermal discharges in air. Recently, two different approaches to account for the photoionization effects have been proposed to avoid the calculation of the global quadrature over the simulation domain. The first approach is based on the direct numerical solution of radiative transfer equation [Ségur et al., Plasma Source Sci. Technol. 15, 648, 2006]. The second approach is built on the solution of a set of Helmholtz equations [Luque et al., Appl. Phys. Lett., 90, 081501, 2007]. In this talk, we discuss the photoionization models based on differential equation approaches, and develop improved models based on the same principles by more accurately accounting for the spectral dependence of the photoionization [Bourdon et al., Plasma Sources Sci. Technol. 16, 656, 2007; http://www.iop.org/EJ/abstract/0963-0252/16/3/026/]. We report modeling results on streamers obtained by using the developed photoionization models and compare them with those obtained by using the Zheleznyak integral model. We emphasize that the actual advantage of differential models advanced in our study in comparison with the integral model lies in the simplicity of implementation of this type of models, and in unquestionable simplicity of extension of these models to complex two- and three-dimensional simulation geometries, involving, for example, branching of sprite streamers, and the presence of obstacles on the streamer path (i.e., ionospheric inhomogeneities, dust particles, aerosols, etc).
AE42A-07 INVITED
High time-resolution sprite observations
Imaging sprites at 10,000 fps have revealed new details about their temporal development. TV observations show a highly structured central body with downward tendrils and upward branches. But rather than being leaders, as suggested by the long streaks in the TV recordings, tendrils and branches are actually formed by spatially compact streamer heads moving at velocities up to 0.3 c. In an individual sprite event the downward moving streamer heads start first forming the tendrils; later, and from a lower altitude and from existing luminous sprite structures, upward moving streamer heads may appear to form the branches. If there are no upward moving streamer heads the event would be classified as a C-sprite, otherwise it would be a carrot sprite. Following the streamer head activity we see afterglow in which little or no temporal and spatial activity is present. The streamer heads are very bright and they appear to be point sources, i.e. their spatial dimensions are less than our 100-200 m image resolution. Streamer head modeling indicates a scale size of ~25 m in which case the brightness would be in the range 1-100 GR. Other models predict volume emission rates leading to a streamer head spatial scale size in the 10 to 100 m range. Our observations conclusively show the downward and upward propagating streamer heads to be separated in time and space. This is in contrast to a number of models in which both down and up going streamer heads emanates from the origin of the process. We frequently see old sprites re-appear in response to new activity suggesting that sprite activity leaves some imprint on the background atmosphere. Given the very large brightness of the streamer heads it would not be surprising if sprite activity initiates chemical processes that could locally affect the composition of the atmosphere, but whether this affects the mesosphere on a larger scale remains uncertain.
AE42A-08
Plasma Chemistry of Sprite Streamers
The expected chemical effects of passage of a sprite streamer through the mesosphere at an altitude of 70 km are described using a nonlinear coupled chemistry model of 80+ species and 500+ reactions. The transient ionization and chemical response of the atmosphere are computed for a streamer descending at high speed ~107 m s-1. The streamer head impulsively (τ~10 μs) ionizes the gas (fractional ionization density ~10-9) and creates large populations of dissociated and excited neutral byproducts. Electrons created by ionization within the head persist for about 1 s, with losses occurring approximately equally by dissociative attachment with ambient O3, and by dissociative recombination with the positive ion dimer N2O2+ created in the sprite channel by a two-step process involving O2+ creation via N2+ charge exchange. Large amounts (fractional densities ~10-9) of metastables are created, with most of them persisting for less than 1 second. These include a possibly detectable (~1 kR) enhancement of O2 Atmospheric airglow emissions from metastable O2(b1Σg+) populations. Electron impact- dissociated N(2D) interacts with O2 to create a long lived (>100 s) increase (fractional enhancement >50%) of the ambient NO density within the streamer channel. Other long lived (>100 s) effects are weak O2 Infrared Atmospheric emissions from metastable O2(a1Δg), and OH Meinel emissions produced by O(3P) perturbations to the catalytic cycle of hydroxyl chemistry.