Lightning-induced Electron Precipitation Research Articles

Numerical simulation of LEP Trimpis observed at Poitiers, France, on signals from VLF transmitters

Transient amplitude and phase perturbations on subionospheric VLF signals, known as the Trimpi effect, are caused by the scattering of VLF radiation from localized ionization enhancements in the nighttime D-region. The patches of ionization are due to precipitation from the radiation belts of keV electrons, that is induced by lightning-generated whistlers.This work is concerned with the numerical simulation of such VLF perturbations, termed LEP (lightning-induced electron precipitation) or classic Trimpis. Two different codes are used to compute the VLF propagation in the Earth-ionosphere waveguide in the presence of a D-region inhomogeneity. The first is based on mode theory, and the second on the FDTD (finite-difference time-domain) method. Both codes are two-dimensional and, therefore, relevant only to LEP events lying on the transmitter-receiver great circle path (TRGCP). A method of simulation is proposed to interpret quantitatively VLF amplitude and phase changes in terms of the approximate location and size of the associated ionospheric perturbation along the TRGCP. The method is applied to LEP Trimpis observed at Poitiers ( L = 2) on signals from the NAA and GQD transmitters. Results are discussed in the light of the information that may be deduced from the high-resolution analysis of VLF temporal signatures. A sequence of simultaneous LEP Trimpis observed at Poitiers on three widely separated transmission paths is studied; two alternative interpretations are proposed.

Trimpi perturbations from large ionisation enhancement patches

A number of increasingly sophisticated and realistic models have been developed in order to investigate the interaction between sub-ionospherically propagating VLF waves and regions of ionisation enhancement (LIE 1 1 The “L” in “LIE” stands for “LEP-produced” ) in the D-region caused by lightning-induced electron precipitation enhancements (LEP). This LEP-produced LIE can result in phase and amplitude perturbations on received VLF radio signals that are referred to as Trimpis or more precisely, classic Trimpis, to distinguish them from “early/fast Trimpis” or “VLF sprites” which are not caused by LEP and are not considered here. It is important, for comparison with experimentally observed Trimpi effects, that the spatial extent of the D-region electron density ( N e) perturbation is modeled accurately. Here, it is argued that most previous modeling has used patch (LIE) sizes that are typically up to 100 km in both latitudinal and longitudinal extent, which are generally smaller than those that actually occur for real lightning induced electron precipitation events. It would also appear that maximum Δ N e values assumed have often been too large, and the patches (LIEs) have been incorrectly modelled as circular rather than elliptical in horizontal extent. Consequently, in the present work, Trimpi perturbations are determined for LIEs with smaller maximum Δ N e, larger spatial extent and elliptical shape. Calculations of VLF Trimpis have been made as a function of the horizontal coordinates of the LIE centre, over the whole rectangular corridor linking transmitter and receiver. The Trimpi modelling program is fully 3D, and takes account of modal mixing at the LIE. The underlying theory assumes weak Born scattering, but the code calculates a non-Born skin depth attenuation function for the LIE in question. The LIE is modelled as an electron density enhancement with a Gaussian profile in all coordinates. Results for a large elliptical LIE ∼ 200 × 600 km show that significant Trimpis, ∼−0.4 dB in amplitude and ∼+4° in phase are predicted, using modest maximum Δ N e values ∼ 1.5 el/cc. Such an electron density enhancement is well within the range that would be expected to result from experimentally observed fluxes of electron precipitation following wave particle interactions with whistler-mode waves.

Recovery signatures and occurrence properties of lightning‐associated subionospheric VLF perturbations

Subionospheric VLF signatures of lightning‐associated ionospheric disturbances observed at multiple sites in the central United States are used to investigate the occurrence properties and recovery signatures of characteristic perturbations exhibiting rapid onset followed by slow recoveries. The two different types of events, so‐called early/fast and lightning‐induced electron precipitation (LEP) [Inan et al., 1996a], are distinguished by the lack of presence of a few hundred millisecond delay between VLF event onsets and causative lightning discharges, respectively. Analysis of recovery signatures of the two types of events indicate subtle but distinct differences in the recovery rates. A majority of early/fast events were found to exhibit a more rapid initial recovery to preevent levels during the first 20 s of recovery, when compared to LEP events. This experimental evidence indicates that the physical nature of the ionospheric disturbance involved in the two classes of events are different, consistent with a recent theoretical suggestion [Inan et al., 1996c]. The occurrence properties of early/fast events observed by the Holographic Array for Ionospheric Lighting (HAIL) receivers indicate that ionospheric disturbances act primarily as forward scatterers.

Lightning induced enhancements of D-region ionisation and whistler ducts

A 3-D magnetospheric ray-tracing program has been constructed in which ray paths can be calculated for propagation in 3-D ducts having a Gaussian enhancement of electron density in both the geomagnetic meridian and geomagnetic longitude directions. Such calculations show a reduction in wave path excursions both in geomagnetic latitude (or L-value) and longitude as rays propagate upwards in a duct between low altitude (∼1000 km) and the equatorial plane. The change in wave energy flux between low altitude and the interaction region for wave–particle interactions and the cross- L extent of related precipitating electron fluxes can then be determined. The size of localised electron density enhancements in the D-region lightning induced enhancements (LIEs) which cause phase and amplitude perturbations on sub-ionosphere VLF propagation (Trimpi effect) can then also be estimated. LIE extents determined in this way appear to be generally about an order of magnitude smaller than estimates from Trimpi observations or low altitude satellite observations of LEP (lightning-induced electron precipitation). This discrepancy has been investigated by ray-tracing in a variety of 3D duct models, including ducts which vary in width and/or enhancement along their length. Calculations made for ducts with maximum enhancement and width in Δ L-value constant along their length imply that precipitation regions and LIEs would typically have smaller dimensions in L-value than the low altitude whistler duct widths measured by the spatial extent of the whistler-mode waves. Therefore, to account for this discrepancy in duct width, ray paths were also determined for ducts which vary in enhancement and/or width in Δ L-value along their length. Although a little better fit with experimental observations is obtained for such ducts, the results can still not explain significantly wider ducts in L-value in the interaction region than at low altitudes. Thus other possible explanations are considered; in particular that the non-linear resonant current in the interaction region near the equatorial plane will radiate wave power into modes with larger wavenormal angles to the geomagnetic field direction than the original trapped modes, and this will then result in a larger region from which precipitating electrons arise and hence larger LIEs than the actual duct width in the interaction region.

Poleward‐displaced electron precipitation from lightning‐generated oblique whistlers

New simulation results show that obliquely‐propagating lightning‐generated whistlers which fill vast volumes of the magnetosphere readily induce significant energetic electron precipitation over large regions of the ionosphere, depositing appreciable energy flux to the upper atmosphere at locations substantially poleward of the lightning source. These results provide a new interpretation for the observed spatial relationships between lightning discharges and subsequently disturbed ionospheric regions inferred from subionospheric signal perturbations (i.e. Trimpi events) reported in several lightning‐induced electron precipitation (LEP) studies.

Satellite observations of lightning‐induced electron precipitation

Lightning‐induced electron precipitation (LEP) from the Earth's radiation belt has been observed on numerous occasions with detectors on the low‐altitude S81‐1/SEEP satellite. A sequence of seven LEP events on September 9, 1982, and eight events on October 20, 1982, are correlated on a one‐to‐one basis with one‐hop whistlers at Palmer, Antarctica. The temporal profile within a LEP burst has a remarkable fine structure. It is shown to be associated with bunches of magnetically guided and focused 100‐to‐200 keV electrons that are repeatedly scattered by the atmosphere and bounce between the northern and southern hemispheres. The delay time between the lightning sferic and the arrival of the first electron bunch increases with increasing L as predicted by the first‐order gyroresonance theory. The global distribution of strong LEP events observed with the SEEP payload correlates with lightning activity and shows a preferred distribution at 2 < L < 3. This L shell range corresponds to the slot region in the electron radiation belt. A single LEP burst (10−3 erg s−1 cm−2) in the slot region is estimated to deplete ∼0.001% of the particles in the region covered by the burst magnetic field lines. The evidence supports the production of structured LEP by ducted rather than nonducted whistlers. It is found that ducted whistlers can be an important pitch angle diffusion mechanism for 100–250 keV electrons in the 2 < L < 3 range although a number of uncertainties in the various parameters remain to be resolved; It is suggested that observations of LEP can be a new tool to measure the presence and transverse dimensions of plasmaspheric whistler mode ducts.

Study of long-wave radio signal disturbances due to lightning-induced energetic electron precipitation

Observing long-wave radio signals at short distances simultaneously with the broadband reception of VLF radiations, we obtained an abrupt (by more than 10%) amplitude variation of signals from the Moscow long-wave transmitter which can be identified as the reflection of long-wave signals from lightning-induced energetic electron precipitation regions in the ionosphere. Some amplitude variation events were accompanied by single-hop whistlers which arrived from the southern hemisphere and multihop whistlers originating from the northern hemisphere.

Characteristics of localized ionospheric disturbances inferred from VLF measurements at two closely spaced receivers

The very low frequency (VLF) NPM signal from Hawaii was recorded at two closely spaced (∼50 km) receivers (located at Palmer and Faraday stations) on the Antarctic Peninsula. Measurements of characteristic amplitude and phase signatures of lightning‐induced electron precipitation (LEP) events were made on three different days in March 1992. Both amplitude and, for the first time, phase measurements are quantitatively interpreted using a three‐dimensional model of VLF propagation in the Earth‐ionosphere waveguide in the presence of lower ionospheric disturbances. This is the first time such a study has been undertaken with mirrored precipitation. Differences between the amplitude and phase changes at the two sites are accounted for by the location of the LEP ionospheric disturbance transverse to the VLF signal propagation paths. The change in these differences is explained by the horizontal movement of the disturbance region and, therefore, the causative whistler duct footprint across the transmitter‐receiver paths. Trends in the amplitude and phase changes on a timescale of order 1 hour are found to be encompassed by the modeling of the passage of the day‐night terminator along the paths.

VLF strip holographie imaging of lightning‐associated ionospheric disturbances

A linear array of very low frequency (VLF) receivers, deployed along a line roughly perpendicular to the direction of signal propagation, enables the determination of the size and location of D region disturbances produced by lightning‐induced electron precipitation (LEP) bursts or by lightning‐induced heating in the vicinity of the great circle VLF propagation paths. The configuration essentially constitutes a strip hologram so that the width of both the amplitude and phase perturbation of the signal pattern recorded along the strip are simply related to the size and location of the disturbance for the range of altitude profiles of disturbed ionization expected for LEP events. The validity of the approach is demonstrated both analytically for single waveguide‐mode propagation and also for a realistic propagation path (Annapolis‐Stanford) by using a three‐dimensional numerical model of VLF propagation and scattering in the Earth‐ionosphere waveguide. We also discuss the criteria by which the spacing of discrete elements along the strip can be optimally choosen.

Aspects of wave-particle interactions at mid-latitudes

Arguably the most significant source of particle precipitation into the ionosphere at mid-latitudes ( L ∼2–3), at least for electrons in the tens to hundreds of keV energy range, is that which arises from the interaction between trapped particles and whistler-mode waves in the magnetosphere. Quasisteady plasmaspheric hiss has long been postulated as a major class of waves contributing to the loss of radiation belt particles but recent studies have suggested that under some conditions, impulsive precipitation caused by interactions with lightning-generated whistlers may be equally important as a loss process. Such lightning-induced electron precipitation (LEP) and its ionospheric signature is the subject of this paper. Although LEP may be detected and studied by a variety of ground-based, balloon-borne and satellite-borne sensors, through optical emissions, X-ray production, enhanced conductivity, or the direct measurement of the precipitating particles themselves, a technique using ground-based narrow-band VLF receivers to measure the Trimpi effect (the transient perturbations in amplitude and/or phase of received narrow-band VLF transmissions) caused by LEP-associated ionisation enhancements has become increasingly popular due to its simple instrumentation and wide field of view. Most work has concentrated on the 2 < L < 3 region where typicalwhistler spectra, trapped electron energy distributions and magnetospheric plasma densities and magnetic field strengths are most favourable for the Trimpi effect. In order to use the technique to study in detail the characteristics and distribution of LEP (and its importance as a trapped-particle loss mechanism), using a network of intersecting transmitter-receiver great-circle paths (TRGCPs), a consensus on how to interpret the observational data is crucial, though this has recently been the subject of controversy. Whilst most studies suggest that only LEP within ∼200 km of the TRGCP gives rise to an observable Trimpi event, some have suggested that non-gaussian perturbations well off the TRGCP can be important. In this paper, the current state of LEP research is summarised and some recent results are presented, using data from a network of both widely and closely spaced observing sites in Antarctica.

Altitude profiles of localized D region density disturbances produced in lightning‐induced electron precipitation events

A three‐dimensional model of very low frequency (VLF) radio wave propagation in the Earth‐ionosphere waveguide in the presence of lower ionospheric disturbances is used to quantitatively interpret VLF signatures of lightning‐induced electron precipitation (LEP) events observed in two previously reported cases. One case is that of a 28.5‐kHz signal originating in Puerto Rico and propagating to a receiver in Lake Mistissini, Quebec. The other case involves a 24.0‐kHz signal originating in Cutler, Maine, and received at Stanford, California. In both cases, high‐resolution measurements of the VLF signals were made to accurately document characteristic signatures of LEP events (Inan et al., 1988b, 1990). The comparison of the model calculations with the data yields information about the altitude profiles of electron density of both the extra ionization produced by the LEP events and of the ambient ionospheric D region. The comparisons are carried out using generally accepted values of the spatial extent of the disturbed regions and the intensity of the particle flux constituting the LEP burst.

Recovery signatures of lightning‐associated VLF perturbations as a measure of the lower ionosphere

A new model of the physical processes associated with subionospheric VLF signal perturbations caused by lightning‐induced electron precipitation (LEP) bursts is developed to diagnose the state of the lower ionosphere (e.g., electron number density and rate coefficients for various chemical reactions) on the basis of measurements of VLF recovery signatures. The model accounts for the energy spectrum of the electron bursts precipitated by lightning‐generated whistlers, the chemical relaxation of enhanced secondary ionization in the nighttime D region due to LEP bursts, and quantitatively treats the resultant effects on propagation of the VLF signal in the Earth‐ionosphere waveguide. Application of the model to experimental data obtained for the VLF propagation path from NPM station (Hawaii) to Palmer station (Antarctica) indicates that effective electron detachment rate γ, enhanced secondary ionization profile (e.g., energy content of LEP bursts), as well as the ambient electron density distribution, may be estimated using observed subionospheric VLF recovery signatures. The effective detachment rate was identified as ∼10−18 N s−1, where N is total number density of neutrals. Model indicates in particular that the attachment‐detachment processes play the dominant role in recovery of subionospheric VLF signal perturbations on timescales ∼ 100 s, and that the observed perturbations of the NPM‐Palmer signal correspond to the LEP bursts consisting of relatively soft (&lt; 250 keV) electrons.

On the spatial relationship between lightning discharges and propagation paths of perturbed subionospheric VLF/LF signals

A study has been made of the spatial relationship between propagation paths of subionospheric VLF/LF signals exhibiting sudden amplitude perturbations (Trimpi events) and time correlated cloud‐to‐ground lightning flashes. On each of the 4 days examined the storm centers were located close to the signal path from the NAU transmitter (28.5‐kHz) in Puerto Rico to Stanford (SU) and were at large distances from the propagation path of the 48.5‐kHz transmitter signal from Nebraska to SU. Nevertheless, no Trimpi events were observed on the former path, while many were seen on the latter. Furthermore, the detected Trimpi perturbations of the 48.5‐kHz signal received at Stanford were found to be associated with the lightning activity in the distant storm centers. Since the NAU‐SU path lies entirely at L &lt; 2 and the 48.5‐SU path is located mostly at 2 &lt; L &lt; 3, the L dependent magnetospheric conditions which determine the level of lightning‐induced electron precipitation are different along the two paths. Thus we postulate that the observed difference in Trimpi occurrence on the two paths was due to the different magnetospheric conditions. Hence the occurrence of Trimpi events over the geographical region corresponding to L &lt; 3 may be more dominantly controlled by magnetospheric conditions than the source lightning distribution.

Search for lightning‐induced electron precipitation with rocket‐borne photometers

Photometers at 3914Å and 5577Å and an optical imager were part of an experimental package launched on a sounding rocket in the 1987 Wave Induced Particle Precipitation (WIPP) campaign at Wallops Island, Virginia. The objective was to measure lightning‐induced electron precipitation (LEP) by means of its optical signature. This was the first attempt to measure LEP using rocketborne optical instrumentation. Launch criteria included nearby thunderstorm activity and ground‐based observations of Trimpi events. Lightning flashes are clearly discernible in the data. The photometer data was also characterized by large spin and precession modulations in the photon count rate, consistent with elevated steady particle fluxes in the northern portion of the instrument field of view. No evidence of LEP was observed by the photometers or onboard particle detectors [Arnoldy and Kintner, 1989]. Analysis of the data has enabled us to place an upper limit of 8 × 10−4 ergs‐cm−2‐sec−1 on any burst precipitation energy flux that may have occurred during the rocket flight in the regions explored by the photometers.

Lightning-induced perturbations on VLF subionospheric transmissions

Phase and amplitude perturbations on VLF subionospheric transmissions from transmitter NWC to Dunedin have been studied on both MSK frequencies and at spaced receivers, 9 km apart. In any one event (a ‘Trimpi’) the phase and amplitude perturbation can be expressed in terms of a perturbation phasor. This is generally believed to be the result of lightning-induced electron precipitation (LEP) producing a localized increase in ionization near the normal reflection height for subionospheric (waveguide) VLF waves. Most of the Trimpis received on the NWC-Dunedin path can be best explained if the LEP ionization is sufficiently localized so that it acts as a scattering centre for the subionospheric VLF wave from the transmitter. It is then this scattered wave or echo at the receiver which makes the perturbation phasor. We call these ‘echo Trimpis’. The phase of the echo relative to the direct signal will differ on spaced antennae if the angle of arrival of the two signals differ. Similarly, this relative phase will vary with frequency if the group delay of the signals differ. Thus measurement of these differences allows location of the scattering centres, and so too the LEP. Locations made show a significant grouping in a region where the lightning intensity is high. This and other features strongly suggest that these echo Trimpis originate from local (southern hemisphere) lightning. This and other reasons are suggested to explain the high proportion of echo Trimpis on this path.

Location of lightning‐induced electron precipitation from measurement of VLF phase and amplitude perturbations on spaced antennas and on two frequencies

The phase and amplitude perturbations on a VLF subionospheric transmission can always be expressed in terms of a perturbation or echo phasor. It is believed that lightning‐induced electron precipitation (LEP) can produce a localized increase in ionization which acts as a scattering center and so echo source for subionospheric VLF waves. The phase of the echo relative to the direct signal will differ on spaced antennas if the angle of arrival of the two signals differ. Similarly, this relative phase will vary with frequency if the group delay of the signals differ. Thus measurement of these differences allows location of the scattering centers and so the LEP. Locations made show a significant grouping in a region where both the lightning intensity is high and the field line passes through the center of the inner radiation belt.

Energy spectra and pitch angle distributions of lightning‐induced electron Precipitation: Analysis of an event observed on the S81‐1(SEEP) satellite

Temporal and spectral signatures of a lightning‐induced electron precipitation (LEP) burst observed on the S81‐1 (SEEP) satellite are analyzed and compared with the predictions of a test particle model of the gyroresonant whistler‐particle interaction in the magnetosphere. The flux to be detected by specific detectors on the low altitude (∼220 km) satellite at L ≃ 2.24 is calculated in terms of the integral counting rate as a function of time and in terms of the dynamic energy spectra during the initial ∼300‐ms precipitation pulse. For a whistler wave packet with frequency range 500 Hz to 6 kHz the dynamic energy spectra are found to depend sensitively on the electron angular distribution in the vicinity of the loss cone. In the case of a whistler wave originating in northern hemisphere lightning the maximum whistler‐induced pitch angle scattering of electrons occurs near ∼10°S geomagnetic latitude. However, scattering occurring over the latitude range of ∼20°N to ∼20°S is found to be significant and contributes to the observed LEP pulse. The dynamic energy spectra of the LEP pulse and the temporal profile of the integral counting rate are consistent with the predictions of a test particle model of the gyroresonant scattering of the electrons by a whistler wave having an equatorial intensity at 6 kHz of ∼200 pT. The measured LEP pulse pitch angle distribution is wider than that estimated on the basis of the test particle model.

Modal effects on amplitude perturbations on subionospheric signals (Trimpis) deduced from two‐frequency measurements

Interference between the first two modes of Earth‐ionosphere waveguide propagation at the high end of the VLF band (&gt;18kHz) increases with distance from the transmitter out to very large distances and can add amplitude perturbations to the phase perturbations (trimpis) produced by lightning‐induced electron precipitation (LEP) on the great circle path. Since the two modes have slightly different phase velocities, an interference pattern or standing wave is formed which is shifted slightly along the propagation path by the LEP‐induced change in differential phase velocity. The modal effect at the receiver depends on the local gradient (along the great circle path) of amplitude with respect to the differential phase. Since this differential or mode beat phase varies with frequency, measurement of the resultant amplitude at two close frequencies enables an estimation of the modal effects. In this study, measurements were made at Dunedin at the two MSK frequencies, 22250 Hz and 22350 Hz, of the transmitter NWC, during a night of frequent “one‐dimensional” trimpis (i.e., those produced by large‐area LEP occurring close to the great circle path) and of strong and varying modal interference. Modal generation or modification of trimpi amplitude was related to the local gradient of amplitude as expected. From these results it was deduced that modal modification of “echo trimpis” (those produced by small area LEP occurring well off the great circle path), even under extreme conditions, is insignificant.

Geographic distribution of lightning‐induced electron precipitation observed as VLF/LF perturbation events

Expected occurrence characteristics of lightning‐induced electron precipitation (LEP) events at longitudes of the western (110° W) versus eastern (71° W) United States are considered from the point of view of available trapped particle flux at the edge of the loss cone. Considering published data on nighttime fluxes of &gt;68 keV electrons observed at L ≃ 2.5, and for “direct” precipitation into the northern hemisphere induced by northern hemisphere lightning, the occurrence rate and flux levels are expected to be a factor of 20–200 higher in the west than in the east, assuming no significant variation in lightning source activity with longitude. Again assuming lightning sources in the north, it is predicted that at 71° W, “mirrored” precipitation into the southern hemisphere would involve precipitation fluxes 30–300 times higher than “direct” precipitation into the northern hemisphere. However, at 110° W and again assuming lightning in the north, southern hemisphere precipitation would tend to be limited to that small fraction of particles that were initially scattered into the northern loss cone and that were then backscattered from the northern atmosphere so as to reach the south. Preliminary experimental investigation of these predictions is based on observations of lightning‐associated perturbations of two geographically separate subionospheric VLF/LF signal paths, one (48.5 kHz) originating at Silver Creek, Nebraska, and observed at Stanford, California, and the other (28.5 kHz) originating at Aguadilla, Peurto Rico, and observed at Lake Mistissini, Quebec. The association of the characteristic VLF signal perturbations with lightning is generally evidenced by simultaneous (within ∼1 s) observation of single or multiple radio atmospherics. In most cases, high‐resolution measurements of event signatures reveal a ∼0.5–1 s delay between the atmospheric and event onset, as well as an ∼1‐s onset duration, consistent with theoretical predictions of a test particle model of the gyroresonant whistler‐particle interaction. The data, considered in the light of previous observations in the southern hemisphere, provide qualitative support for several of the predictions based on considerations of the trapped flux level near the loss cone, in particular the prediction of comparable rates in the north at 110° W and the south at 71° W, and the prediction of substantially larger rates in the south than in the north near 71° W.

Reply [to “Comment on ‘Are fast atmospheric pulsations optical signatures of lightning‐induced electron precipitation?’”

Reply [to “Comment on ‘Are fast atmospheric pulsations optical signatures of lightning‐induced electron precipitation?’”