Electric Field Detector Research Articles

Electric field measurements during supercharging events on the MAIMIK Rocket Experiment

We report on the observation and analysis of quasi‐DC electric fields measured during active electron emission from the MAIMIK sounding rocket. The MAIMIK mother‐daughter tethered rocket consisted of two separately instrumented payload sections which were detached in‐flight. The daughter payload included a variable current, 8‐keV electron accelerator and instrumentation to measure the vehicle potential relative to the ionospheric plasma. The electric field detector was part of the quiescent mother payload which separated from the daughter at a constant speed. Of particular interest to the present study were the observations of impulsive electric fields measured at the initiation of the beam for emission currents which drove the daughter payload into an overcharged state. At locations well removed from the active payload the electric field signature consisted of a 1‐ to 2‐ms, monopolar impulse field that had a rapid risetime and a more gradual decay. The components of the vector field were examined using a cylindrical coordinate system that placed the active payload at the origin and fixed the azimuthal injection angle of the beam. In general, the radial components of the field were directed out from the positively charged central body and presumed to be of electrostatic origin. The tangential components had consistent positive deflections indicating that these field components were mostly electromagnetic in nature. We suggest that the measured impulse fields were due to a shielded ion blast wave propagating away from the highly charged central body with azimuthal asymmetries created by the distribution of space charge associated with a virtual cathode and escaping beam electrons. The time‐varying electrostatic fields drove transient Hall currents that were carried by cold electrons E×B drifting near the head of the blast wave.

Steady‐state plasma transition in the Venus ionosheath

The results of an extended analysis of the plasma and electric field data of the Pioneer Venus Orbiter (PVO) are presented. We report the persistent presence of a plasma transition embedded in the flanks of the Venus ionosheath between the bow shock and the ionopause. This transition is identified by the repeated presence of characteristic bursts in the 30 kHz channel of the electric field detector of the PVO. The observed electric field signals coincide with the onset of different plasma conditions in the inner ionosheath where more rarified plasma fluxes are measured. The repeated identification of this intermediate ionosheath transition in the PVO data indicates that it is present as a steady state feature of the Venus plasma environment. The distribution of PVO orbits in which the transition is observed suggests that it is more favourably detected in the vicinity of and downstream from the terminator.

Excitation of ELF waves in the Schumann resonance range by modulated HF heating of the polar electrojet

We report the successful generation and detection of ELF waves in the frequency range of 6–76 Hz by polar electrojet modification using the High‐Power Auroral Stimulation (HIPAS) HF Heater Facility near Fairbanks, Alaska. Magnetic field amplitudes of ∼1 pT and vertical electric field amplitudes of ∼0.2 mV have been observed at a receiving site 35 km from the HIPAS site. The wave amplitude does not depend strongly on the ELF frequency, however, the amplitude is closely related to the level of electrojet activity inferred from magnetometer chain data and from ionosonde measurements. The 1‐MW HF heater is modulated at low ELF frequencies using an “array dephasing” technique where the eight‐element antenna array is alternately phased for peak vertical gain and dephased for a spread pattern at the ELF rate. This method can be employed at any modulation frequency without concern for transmitter power supply resonances. Measurements of the wave amplitude are made by conventional analog lock‐in techniques and by continuously digitizing the waveform output of the wideband magnetic search coils and the electric field detector. With the digitized waveform, digital postprocessing can be used to extract the low‐level coherent signal from the Schumann resonance noise background using a software lock‐in routine.

A re‐examination of impulsive VLF signals in the night ionosphere of Venus

Singh and Russell (1986) reported that impulsive electrical signals observed by the Pioneer Venus Orbiter Electric Field Detector at all frequencies clustered about periapsis. However, the impulsive signals consist of both signals entering the instrument through the antenna and artificial pulses associated with telemetry errors as Taylor and Cloutier (1988) have pointed out. Obvious telemetry errors were not included in the original study but records of which signals were thought to be naturally occurring and which were thought to be telemetry noise were not kept. Thus we cannot check the original identifications on a point‐by‐point basis. We can, however, repeat the study and compare statistical results. This study indicates that there is naturally occurring noise in the dark ionosphere of Venus near periapsis as originally reported by Singh and Russell. However, the absolute rates of occurrence of the original study and the present work differ. Sometimes the rates are smaller and sometimes larger than originally reported. These differences may be due to the use of a higher threshold in the earlier study together with the lack of sufficient discrimination against telemetry dropouts. Our rates of “naturally occurring” emissions also differ from the non‐spike noise rates obtained by Taylor and Cloutier although our rates of artificial signals are similar. In any event the Singh and Russell study should not be used to infer the altitude distribution of the noise since the spacecraft spends more time at low altitude than high as it traverses the dark ionosphere.

Telemetry interference incorrectly interpreted as evidence for lightning and present‐day volcanism at Venus

Spike‐like responses to interruptions in the data recorded by the Orbiter Electric Field Detector (OEFD) on the Pioneer Venus Orbiter (PVO) have been incorrectly identified as broadband whistler noise and then interpreted as evidence of lightning and volcanism by Singh and Russell [1986; 1987]. We show that these spikes are non‐physical artifacts of the processing of interruptions in telemetry data and are readily distinguished from other non‐artificial noise appearing in the measurements. We independently analyze the spikes and the non‐artificial noise to show that key portions of the results of Singh and Russell are compromised by the inclusion of the artificial data. This conclusion is certified by the fact that (a) the distribution of the spikes‐only data matches the distribution of noise presented by Singh and Russell, while (b) the number and distribution of all non‐spike noise is quite incompatible with their results. This finding is consistent with our prior research, and leads us to reiterate that the PVO electric field data set used by Singh and Russell provides no basis for inferring the presence of either lightning or present‐day volcanism at Venus.

The plasma wave environment of an auroral arc: 3. VLF hiss

Intense auroral hiss was observed by a sounding rocket payload under circumstances that have allowed detailed comparison of the hiss spectrum with the predictions of a convective beam amplification hiss production model. The payload was launched from Poker Flat, Alaska at 0813 UT on March 9, 1978; it carried an electric field detector and electron detectors covering the energy range from 30 eV to 25 keV. A complete pitch angle scan was made every 0.2 s. The trajectory was close to the magnetic meridian and crossed the diffuse aurora, a dark region, and a quiet 40 kR auroral arc. Auroral hiss was first encountered at the poleward edge of the diffuse aurora, and was observed for the rest of the flight. A maximum intensity of 10−8(V/m)²/Hz at 30 kHz was seen in the dark region. Electron distribution functions were seen with δf/δv∥>0 and a large enough slope to have produced appreciable VLF amplification. These distributions were found in a region a few kilometers wide on the equatorward boundary of the arc. A model distribution function has been fitted to one of the observed distributions and used as input into an auroral beam noise generation model. Reasonable assumptions about the distribution of cold plasma and width of the unstable region yields a predicted VLF spectrum in good agreement with observation. Several secondary results were also obtained. The observed VLF hiss does not appear to have been produced by a payload‐plasma interaction. The intensity of the hiss measured in situ, above the ionosphere, was greater equatorward of the arc than within the arc, thus ruling out the ionospheric transmission model as an explanation for this intensity pattern. Quasi‐linear back reaction on the electron distribution was not found to be significant in this case. The correlation between 0.1 – 0.7 keV electron flux and hiss intensity was not better than the correlation with the flux of 5‐keV electrons. The data did not suggest the occurence of a 3 wave parametric decay. Finally, the possibility of a secondary source at ∼100 km altitude is not ruled out by these data.

Partially transmitted lightning signals recorded by pioneer Venus orbiter

Impulsive electric fields appearing on all four frequency channels of the Pioneer Venus electric field detector in the night ionosphere of Venus are characteristic of lightning generated signals. Based on our knowledge of the electron density and magnetic field in the Venus ionosphere, we suggest that lightning waves could be partially transmitted upwards into the ionosphere. The leakage of these lightning waves into the ionosphere on encountering electron density holes may be treated as reradiation into the ionosphere from the hole. Since this radiation pattern is frequency dependent, we should not expect to see all frequency components for every lightning stroke observed.

Interferometric phase velocity measurements in the auroral electrojet

A double-probe electric field detector and two spatially separated fixed-bias Langmuir probes were flown on a Taurus-Tomahawk sounding rocket launched from Poker Flat Research Range in March 1982. Interesting wave data have been obtained from about 10s of the downleg portion of the flight during which the rocket passed through the auroral electrojet. Here the electric field receiver and both density fluctuation ( δn/ n) receivers responded to a broad band of turbulence centered at 105 km altitude and at frequencies generally below 4 kHz. Closer examination of the two ( δn/ n) turbulent waveforms reveals that they are correlated, and from the phase difference between the two signals, the phase velocity of the waves in the rocket reference frame is inferred. The magnitude and direction of the observed phase velocity are consistent either with waves which travel at the ion sound speed ( C s ) or with waves which travel at the electron drift velocity. The observed phase velocity varies by about 50% over a 5 km altitude range—an effect which probably results from shear in the zonal neutral wind, although unfortunately no simultaneous neutral wind measurements exist to confirm this.

Plasma and field observations of a Pc 5 wave event

Micropulsation measurements of a Pc 5 wave event on July 14, 1982, in the afternoon magnetosphere are reported as observed by wave and particle instruments on board the Dynamics Explorer 1 (DE 1) spacecraft. The overall structure of the Pc 5 event as noted in the low‐energy particle and quasi‐static electric field data seems to order the event into two distinct halves. The appearance is reminiscent of a wave packet and suggests a temporal or spatial variation of the micropulsation, which has a scale of 20 min or 3000 km. The wave packet structure is also well correlated with a variation in the pitch angle distribution of the low‐energy plasma with single direction field‐aligned flow associated with the maximum amplitude of the wave packet structure and bidirectional field‐aligned flows associated with the nodal point of the wave packet structure. The field‐aligned velocities of the observed H+, He+, O+, N+, and O++ ions combined with the elliptical E × B wave oscillation in the first half of the event produce a helical motion of the plasma along the field line. In the second half of the event although all ion species are still present, strong magnetospheric convection seems to have a significant effect on the field‐aligned motions and Pc 5 oscillations of the low‐energy plasma. The event is characterized in the low‐energy plasma by a left‐hand to near‐linear polarized rotation of the plasma over the first half of the event. Variations of the eccentricity over the first half of the event which are indicated by both the quasi‐static electric field detector and the low‐energy particle detector suggest that the center of resonance lies just radially inside of the DE 1 orbit (L = 4.7) and just outside of the dayside plasmapause, which appears to be located at L = 4 as indicated by experiments on board the DE 2 spacecraft. Comparisons of the measured Pc 5 wave period to the theoretical period derived from in situ evaluation of the plasma mass loading factor yield values of 190 s and 192 s, respectively. This is consistent with the center of resonance being nearly coincident with the DE 1 orbit. The second half of the event occurs in a region of enhanced magnetospheric convection, which makes determination of the eccentricity of the wave difficult and harder to interpret. The wave is found to be right‐hand or linearly polarized during this portion of the event, which suggests DE 1 is just radially inward of the resonance center. However, disagreement between the measured period of 233 s and the derived period of 189 s does not necessarily support this interpretation. Density variations of plasma along the field line which are hemispherically asymmetric may explain the inconsistencies between the measured and derived period in this half of the event.

SSC‐excited pulsations recorded near noon on GEOS 2 and on the ground (CDAW 6)

The SSC occurring on March 22, 1979, at 0826 UT had an unusually sharp onset in Scandinavia, in Middle Europe and in experiments on the geostationary satellite GEOS 2, which was near noon, local magnetic time. The ground magnetometer stations showed a small preimpulse which started ∼5 s before the main impulse. Both impulses needed ∼2 s to “propagate” from ground stations at L = 6.3–4.6. Search coil magnetometers indicate a very small precursor in northern Finland (L ∼ 4.4–6.0) which started ∼15–20 s before the main impulse. This small precursor also occurred close to the time of the SSC onset at GEOS 2. We interpret this precursor as an effect of precipitating electrons changing the ionospheric conductivity in a localized region. The main impulse triggered damped magnetic pulsations (Psc) with periods near 160 s and 50 s visible in northern Scandinavia and the electric field detector on GEOS 2. Furthermore, the magnetic field and the energetic ions at GEOS observed pulsations with periods near 80 s, but these could only be observed at the northernmost ground stations. There are several indications that the first three harmonics of standing hydromagnetic waves are detected. They may correspond to periodic oscillations of the subsolar point or eigenperiods of the SSC‐excited fast mode (compressional cavity resonance). The tentatively identified second harmonic wave (period ∼80 s) is indicative of a bounce resonance of ring current protons. Inside the plasmasphere the dominant period of a superimposed Psc 4 event increased with latitude for the H component indicating several toroidal eigenoscillations.

Venus nightside ionospheric troughs: Implications for evidence of lightning and volcanism

Ionization troughs are frequently observed in the Venus nightside ionosphere by the orbiter ion mass spectrometer (OIMS) on the Pioneer Venus Orbiter (PVO). These events typically exhibit density depletions which may range from a factor of 2 to an order of magnitude or more, often feature sharp density gradients, and frequently are associated with superthermal ion fluxes. Although their full spatial extent cannot be determined, the ion trough dimensions as perceived along the PVO orbit track extend from 1° to more than 10° in latitude and longitude. Troughs have been encountered at altitudes ranging from more than 1000 to less than 250 kilometers. The nightside trough boundaries are similar to both the ionization gradients which mark the Venus dayside ionopause and the ionization gradients which define the high‐latitude ionization trough in the earth ionosphere. In each of these Venus and earth phenomena there is a close association between sharp gradients in thermal ion distributions and energetic and dynamic processes stimulated by the interaction between the solar wind/interplanetary magnetic field (IMF) and the thermal ionosphere. A close correlation is identified between ion trough events and prominent published examples of 100‐Hz plasma waves detected by the orbiter electric field detector (OEFD), which have been interpreted by Scarf and Russell (1983) and by Ksanfomaliti et al. (1983) as whistler mode propagation stimulated by lightning in the lower atmosphere and frequently occurring over mountains. While the occasional coincidence between the detection of whistler waves and the overflight of Venus mountain ranges is also observed for the ion troughs, neither phenomenon is seen to exhibit convincing evidence of an association with the mountains. Overall, these findings lead us to believe that many of the 100‐Hz noise bursts previously attributed to lightning are, like the ion troughs, actually the result of the interaction between the ionosphere and the combined effects of the solar wind and the draped IMF. If this interpretation is correct, there may be little evidence for lightning from the PVO measurements. Similarly, a supporting argument for active volcanism as a possible source for the deduced lightning events would also be discounted.

Plasma‐depleted holes, waves, and energized particles from high‐altitude explosive plasma perturbation experiments

High explosive shaped charge experiments King Crab and Bubble Machines I and II, designed to perturb the ambient plasma and magnetic field, were flown above 460 km on Taurus Tomahawk rockets from Poker Flat in March 1980, 1981, and 1982, respectively. The last flights were a mother‐daughter combination with the instrumentation section remaining attached to the rocket. The detectors consisted of a single‐axis dipole electric field detector, a fixed bias cylindrical Langmuir probe, a three‐axis attitude magnetometer, and curved plated energetic ion and electron electrostatic analyzer. The first experiment, King Crab, revealed the existence of a barium plasma depleted region or dark hole of about 5 km diameter centered on the burst. All electric power to the instruments failed on the lift‐off of the 1981 Bubble Machine I, but useful optical data were obtained. The third flight a year later produced useful data from all but the electron detector on the electrostatic analyzer. Ground‐based optical and telluric field instrumentation recorded evidence for injection‐induced waves of about 5‐s period. Delay times indicate a slow (175–385 km/s) propagation from the burst point. Promptly following the release, auroral intensity ion beams were observed in the field‐aligned ion electrostatic analyzer, with energies up to 6.8 keV. Evidence suggests that these particles were not energized by the explosion but that they represent an existing ion conic population pitch angle scattered by the released barium into the view of the detector. Concerning locally produced electrostatic instabilities, the wave measurements of Bubble Machine II indicate that finite Larmor radius stabilization occurs for the weakly driven low (barium) density situation, but that shorter‐wavelength waves can occur early in the expansion event when the barium density is higher. These results are in reasonable agreement with a theory described by Sperling and Krall (1981).

ISEE 1 and 2 observations of an oscillating outward moving current sheet near midnight

An outward moving current sheet is examined with the ISEE 1 and 2 magnetometers. It is found that the current sheet is moving outward with a velocity of about 17 km/s, a thickness of about 1300 km, and with the current flowing into the ionosphere. Oscillations observed in the magnetic field profile indicate the presence of a wave traveling along the current sheet from midnight toward the east with a velocity of 400 km/s. The oscillations of the plasma normal to the current sheet associated with this wave are sufficient to explain the amplitude of the electric field oscillations in the plane of the current sheet observed by the University of California‐Berkeley (UCB) electric field detector. In addition a strong electric field is observed normal to the sheet, corresponding to a flow along the sheet of 400 km/s, which is consistent with the velocity of the wave deduced from the intersatellite timing of the oscillations. Thus the wave is stationary in the frame of the current sheet plasma. This oscillating current sheet model explains the observed behavior of both the magnetic and electric field observations. The potential drop across the current sheet was very large at this time, 41±26 kV, and is comparable to the total normal potential drop across the polar cap.

Plasma and electric field measurements of the PVO in the Venus ionosheath

A study of the plasma and electric field measurements of the PVO in the Venus ionosheath near the terminator region is presented. A drastic decrease of the particle flux intensity measured with the plasma instrument is encountered in the inner ionosheath before the inbound crossing of the ionopause. The outer boundary of the region of weak plasma fluxes is consistent with the presence of a rarefaction wave as reported previously from the Mariner 5, Venera and PVO plasma measurements in the Venus wake. Near that boundary there are, in addition, appreciable changes in the electric field signals detected with the PVO electric field detector. These observations suggest the existence of local plasma wave activity associated with turbulent flow conditions in that region.

A sounding rocket observation of an apparent wake generated parallel electric field

Sounding rocket Terrier‐Malemute 29.007 was launched over a 40 kR quiet auroral arc at 0813 UT on March 9, 1978. Payload instrumentation included a single‐axis double probe electric field detector. The reduced data from this detector are modulated by a square wave which is coherent with the diagnostic float‐bias cycle of the detector for the last ∼80 s of the flight when the analysis is carried out under the assumption that E · B = 0. This modulation may be removed by assuming that a time‐varying, vehicle‐generated apparent parallel field was present during the periods when the probes were floating. Three possible models for the effect were considered: a gradient in the nonthermal probe current, a gradient in plasma temperature near the rocket, and a variable dipole moment of the wake. There are significant problems with all these models. On balance, the third one appears to be the most likely. The observed effect may account for some previously reported measurements of low altitude parallel fields. It must be stressed that it cannot account for all such reports.

Dynamic variations observed in thermal and superthermal ion distributions in the dayside ionosphere of venus

In-situ measurements of the ion composition and concentration of the ionosphere of Venus are obtained with the Bennett rf ion mass spectrometer (OIMS) on the Pioneer Venus Orbiter (PVO). Dayside ion profiles exhibit considerable variability in the height of the ionopause as well as the scale heights of the ion constituents, which reflect the compression and expansion of the ionosphere in response to solar wind variations. Near the dayside upper boundary of the thermal O + distribution, superthermal (E ⋍ 10–90 ev) ions are detected by the OIMS, presenting a complication for identifying the ion signature of the ionopause. Correlated with the presence of the superthermal ions, the ac electric field detector (OEFD) detects regions of intensified signals, with peak response in the 100 Hz frequency channel. A limited set of comparisons indicates that both the OIMS and the OEFD detect more pronounced enhancements in the superthermal ions and 100 Hz fields, respectively, at midlatitudes (∼ 30°N), relative to low latitudes (∼ 5°N), and that wave like variations are sometimes present. These characteristics of the superthermal ion-plasma wave results suggests that these phenomena are generated in the vicinity of the ionopause, possibly by the turbulent acceleration of planetary ions at the lower boundary of the ionosheath. It is expected that further analysis of the superthermal ion-electric field signatures will contribute to a clearer understanding of the physical processes underlying the formation of the ionopause.

Generation and propagation of an electromagnetic pulse in the trigger experiment and its possible role in electron acceleration

The dc electric field detector onboard the instrumented portion of the Trigger payload detected a large‐amplitude (∼200 mV/m) electric field pulse which was observed with a time delay consistent only with an electromagnetic wave. A model for this perturbation is constructed in this paper and the associated field‐aligned current calculated as a function of altitude. Near the explosion point, downward currents of 2500 µA/m² and upward currents of 90 µA/m² are indicated by the model. The lower upward current has the same direction as the field‐aligned current in discrete auroral arcs. Taking into account the attenuation of the wave and the changes in Alfven speed and plasma density, the associated differential drift velocity between electrons and ions is found to peak at 750‐km altitude at a value one‐third of the electron thermal speed. This is well above the threshold for ion cyclotron wave turbulence and approaches the value necessary for production of electrostatic shocks or double layers. The differential velocity associated with the downward current well exceeds the electron thermal speed. The time delay for observation of accelerated electrons is consistent with the Alfven propagation time to 750‐km altitude. The initial explosion pulse had internal temporal variations just above the oxygen ion cyclotron frequency in agreement with other experiments of this type. The Poynting flux in the electromagnetic pulse was sufficiently intense to create the observed particle precipitation.

Observations of ULF electric field fluctuations in the dayside auroral oval

Electric field oscillations at frequencies centered below 10 Hz were measured by a payload equipped with double‐probe electric field detectors which was launched into the dayside auroral oval during the Greenland rocket campaign on January 11, 1975. These oscillations exceeded an amplitude of 3 mV/m broadband and occurred over a period of about 90 s. These oscillations occurred during a proton injection event when the quasi‐static electric field was less than 20 mV/m on the average. However, during a brief 2‐s interval the electric field exceeded 50 mV/m. Analysis of the amplitude as a function of the spin frequency showed that the electric field was confined to a plane perpendicular to the magnetic field. The power spectrum was of the form f−6.5±1.5, where f is the frequency over the range from 8 to 18 Hz. The existence of a large electric field fluctuation is a spatially or temporally confined region of low‐amplitude turbulence suggests that this may be an observation of the low‐altitude projection of a turbulent region associated with electrostatic shocks which have recently been observed at higher altitudes.

Electrodynamic structure of the late evening sector of the auroral zone

Detailed observations from a late evening sector pass of S3‐2 over the northern auroral zone are presented and compared with nearly simultaneous DMSP imagery. High time resolution measurements are presented from electric field, magnetic field, energetic electron (80 eV ≤E ≤17 keV), and thermal plasma detectors. The region of diffuse aurorae was characterized by nearly isotropic electron fluxes and slowly varying (≤20 mV/m) electric fields. The Birkeland currents in this region were predominantly into the ionosphere with embedded, narrow sheets of upward currents. Simultaneous magnetometer and satellite potential fluctuations show that the upward current sheets are real and that these upward currents are probably carried by precipitating, low‐energy plasma sheet electrons. Poleward of the diffuse aurorae, the covective electric field varied rapidly in magnitude, anticorrelating with the intensity of precipitating electrons. An inverted‐V structure was observed above a visible arc. An intense downward current sheet (>10 µA/m²) was found near the equatorward edge of the arc. At this place the ionospheric thermal plasma was near marginal stability for the onset of O+ ion cyclotron turbulence. An electric field component parallel to B of ∼10 mV/m was measured in this region of strong downward current. The plasma and electric field distributions in the vicinity of the arc suggest that the auroral electrojet was distributed over the region of diffuse auroral but was most intense near the equatorward boundary of the auroral arc.

Electric fields, electron precipitation, and VLF radiation during a simultaneous magnetospheric substorm and atmospheric thunderstorm

A balloon payload instrumented with a double‐probe electric field detector and an X ray scintillation counter was launched from Roberval, Quebec, Canada (L =4.1) at 0828 UT (0328 LT) on July 9, 1975. A magnetospheric substorm was observed locally between 0815 and 1100 UT, which produced a maximum ΔB of ∼500 nT at ∼0930 UT. A single‐cell atmospheric thunderstorm developed northeast of Roberval beginning around 0925 UT which was most intense from ∼1000 to 1035 UT. Detailed study of the electrical properties of the thunderstorm, the X ray precipitation data, and VLF spheric data leads to three conclusions. First, the electrical coupling from the thunderstorm to the magnetosphere increases with frequency from dc to the VLF; for the observed storm the amplitude at the ionosphere of thunderstorm produced electric fields was not significant at frequencies below 0.1 Hz. Second, the atmospheric conductivity above the thunderstorm was observed to be about one‐half the fair weather value prior to 1000 UT; decreased to about one‐quarter the fair weather value at about 1000 UT; and remained depressed after the end of the thunderstorm. This result was contrary to that expected on the basis of previous work and is one which merits considerably more investigation. Third, the data show a high probability that half‐hop whistlers initiated by sferics from the thunderstorm triggered energetic electron precipitation from the magnetosphere.