Sylvain Coquillat

Computed conditions of corona emission from raindrops

The altitude of corona emission from charged raindrops located in a vertical ambient electric field is calculated by coupling the corona experimental results obtained by Dawson (1969) with the raindrop model of Coquillat and Chauzy (1993). This model provides the drop shape and electric surface field necessary to calculate the corona occurrence altitude from a fitting of Dawsons data. The original results are presented in the form of vertical profiles of the critical field, which is the ambient field that causes disruption or a corona. These results are directly comparable with in situ measurements of electric field, raindrop size, and net charge. If we make the assumption that positive streamer propagation is of prime importance for lightning initiation, the critical field profiles allow us to determine the minimum net charge of a drop which could initiate a discharge in a given ambient field.

Behavior of precipitating water drops under the influence of electrical and aerodynamical forces

The present work performs a realistic modeling of precipitating charged water drops under the influence of electrical and dynamical forces in the vertical and downward electric field of a thundercloud. The following factors which control the shape of an individual raindrop are taken into account: surface tension, internal hydrostatic pressure, aerodynamic pressure, and electrostatic pressure. Unlike a recent and notable work by Chuang and Beard (1990) in which this problem is approached by adjusting an empirical pressure distribution for the distortion, our model considers simple local pressure balance to determine the drop shape. This computation aims at characterizing drop distortion, falling speed modification, and disruption. The overall present results are similar to those of Chuang and Beards more sophisticated model, and the predicted critical fields are even closer to wind tunnel measurements by Richards and Dawson (1971). The disruption of positively charged drops requires lower ambient fields than that of the negatively charged drops, and for highly charged and large drops they are of the order of those commonly measured within thunderclouds. At last, the terminal velocity is highly affected by net charge and ambient field. These processes are probably important in lightning initiation during drop disruption.

Microdischarges between ice particles

A laboratory experiment is performed in a cold chamber to produce microdischarges between frozen hydrometeors. A charged icicle is placed close to an uncharged one connected to a transient analyser. The electric current signature obtained during each event is then studied and characterized in function of both temperature and net charge. The existence of a transition temperature of about −15°C is detected, whereas net charge clearly governs the microdischarge characteristics. Furthermore, the radiated electric field is calculated and the spectral analysis, by means of fast Fourier transform, shows that this elementary phenomenon radiates within the VHF range. Comparisons are made with previous similar work by Chauzy and Kably (1989) on microdischarges between water drops.

Time and space correlation between sprites and their parent lightning flashes for a thunderstorm observed during the HyMeX campaign

During the night of 22–23 October 2012, together with the Hydrology cycle in the Mediterranean eXperiment (HyMeX) Special Observation Period 1 campaign, optical observations of sprite events were performed above a leading stratiform Mesoscale Convective System in southeastern France. The total lightning activity of the storm was monitored in three dimensions with the HyMeX Lightning Mapping Array. Broadband Extremely Low Frequency/Very Low Frequency records and radar observations allowed characterizing the flashes and the regions of the cloud where they propagated. Twelve sprite events occurred over the stratiform region, during the last third of the lightning activity period, and well after the coldest satellite-based cloud top temperature (−62°C) and the maximum total lightning flash rate (11 min−1). The sprite-producing positive cloud-to-ground (SP + CG) strokes exhibit peak current from 14 to 247 kA, Charge Moment Changes (CMC) from 625 to 3086 C km, and Impulsive CMC (iCMC) between 242 and 1525 C km. The +CG flashes that do not trigger sprites are initiated outside the main convective core, have much lower CMC values, and in average, shorter durations, lower peak currents, and shorter distances of propagation. The CMC appears to be the best sprite predictor. The delay between the parent stroke and the sprite allows classifying the events as short delayed ( 20 ms). All long-delayed sprites, i.e., most of the time carrot sprites, are produced by SP + CG strokes with low iCMC values. All SP + CG flashes initiate close to the convective core and generate leaders in opposite directions. Negative leaders finally propagate toward lower altitudes, within the stratiform region that coincides with the projected location of the sprite elements.

Onset of instability in precipitating water drops submitted to horizontal electric fields

Most intracloud discharges are supposed to be initiated by hydrometeors in a vertical ambient electric field. However, horizontal propagating flashes are commonly observed. The aim of this work is to measure the horizontal onset field of instability for various drop sizes and to compare the results with those previously determined in experiments where the drops did not reach their terminal velocity [Nolan, 1926; Macky, 1931; Ausman and Brook, 1967] and with those recently published by Kamra et al. [1993] for drops suspended in a wind tunnel. In the present experiment, uncharged drops fall about 14 m and are at terminal velocity before entering an interelectrodes region where a horizontal electric field is generated. It appears that the lowest onset field intensities are obtained when the aerodynamical effects are maximum. However, there is a large discrepancy between our results and those by Kamra et al. It is believed that the onset fields of instability are underestimated by the latter authors because of probably unperfect experimental conditions. The present results, which are certainly more representative of the fall of uncharged drops in still air than those previously obtained, seem to indicate that horizontal electric fields in thunderclouds are not strong enough to cause lightning initiation by corona emission in these conditions and to affect the drop-size distributions. This conclusion conflicts with that of Kamra et al. [1993]. However, the case of charged drops falling at higher altitude would certainly lead to lower horizontal field onsets.

Lightning initiation: Strong pulses of VHF radiation accompany preliminary breakdown

We analyze lightning initiation process using magnetic field waveforms of preliminary breakdown (PB) pulses observed at time scales of a few tens of microseconds by a broad-band receiver. We compare these pulses with sources of narrow-band very high frequency (VHF) radiation at 60–66 MHz recorded by two separate Lightning Mapping Arrays (LMAs). We find that almost none of the observed PB pulses correspond to geo-located VHF radiation sources, in agreement with previous results and with the hypothesis that processes generating VHF radiation and PB pulses are only weakly related. However, our detailed analysis discovers that individual peaks of strong VHF radiation seen by separate LMA stations correspond surprisingly well to the PB pulses. This result shows that electromagnetic radiation generated during fast stepwise extension of developing lightning channels is spread over a large interval of frequencies. We also show that intense VHF radiation abruptly starts with the first PB pulse and that it is then continuously present during the entire PB phase of developing discharges.