Search

Ions in the F region ionosphere at 150–400 km altitude consist mainly of molecular NO+and O2+${\mathrm{O}}_{2}^{+}$, and atomic O+. Incoherent scatter (IS) radars are sensitive to the molecular‐to‐atomic ion density ratio, but its effect to the observed incoherent scatter spectra is almost identical with that of the ion temperature. It is thus very difficult to fit both the ion temperature and the fraction of O+ions to the observed spectra. In this paper, we introduce a novel combination of Bayesian filtering, smoothness priors, and chemistry modeling to solve for F1region O+ion fraction from EISCAT Svalbard IS radar (75.43° corrected geomagnetic latitude) data during the international polar year (IPY) 2007–2008. We find that the fraction of O+ions in the F1region ionosphere is controlled by ion temperature and electron production. The median value of the molecular‐to‐atomic ion transition altitude during IPY varies from 187 km at 16–17 MLT to 208 km at 04–05 MLT. The ion temperature has maxima at 05–06 MLT and 15–16 MLT, but the transition altitude does not follow the ion temperature, because photoionization lowers the transition altitude. A daytime transition altitude maximum is observed in winter, when lack of photoionization leads to very low daytime electron densities. Both ion temperature and the molecular‐to‐atomic ion transition altitude correlate with the Polar Cap North geomagnetic index. The annual medians of the fitted transition altitudes are 14–32 km lower than those predicted by the International Reference Ionosphere. Ions in the F region ionosphere at 150–400 km altitudes consist mainly of molecular NO+and O2+${\mathrm{O}}_{2}^{+}$, and atomic O+. Incoherent scatter radars are sensitive to the molecular‐to‐atomic ion density ratio, but its effect to the observed incoherent scatter spectra is almost identical with that of the ion temperature. It is thus very difficult to fit both the ion temperature and the fraction of O+ions to the observed spectra. This causes bias to the fitted temperatures and leaves behavior of the F1region ion composition poorly known. We apply a novel combination of inverse mathematics and chemistry modeling to analysis of EISCAT Svalbard incoherent scatter radar data, and solve for both the ion temperature and the fraction of O+ions at the same time. We find that the ion composition is considerably different from a standard model, that it undergoes regular diurnal variations, and it is affected by geomagnetic activity. The typical variations can be qualitatively explained by known diurnal variations in ion temperature, solar photoionization, and ion chemistry. When the ionosphere above Svalbard is in almost complete darkness in mid‐winter, ion composition in the daytime ionosphere is considerably different from that observed during the other seasons. We use novel data analysis techniques and chemistry modeling to fit atomic oxygen ion fractions to EISCAT Svalbard radar data during IPYWe characterize the F1 region ion composition dependence on local time, solar zenith angle, and geomagnetic activityThe molecular‐to‐atomic ion transition altitudes are 14–32 km lower than those predicted by the International Reference Ionosphere We use novel data analysis techniques and chemistry modeling to fit atomic oxygen ion fractions to EISCAT Svalbard radar data during IPY We characterize the F1 region ion composition dependence on local time, solar zenith angle, and geomagnetic activity The molecular‐to‐atomic ion transition altitudes are 14–32 km lower than those predicted by the International Reference Ionosphere

The primary focus of this study was the motion of auroral patches and the polarization electric field generated therein observed on 9 November 2015 in an experiment using the European incoherent scatter (EISCAT) radars, Kilpisjärvi Atmospheric Imaging Receiver Array (KAIRA), and an all‐sky imager simultaneously. Based on the all‐sky imager data, the drift speed of the auroral patches corresponded to a southward electric field of 14.1( ± 3.7)–17.2( ± 4.5) mV/m. The convective electric field derived from the EISCAT radars and KAIRA observation was approximately 14.6 mV/m in the southward direction. This suggest that the spatial distribution of the auroral patches reflects the distribution of the cold plasma in the magnetosphere. The electron density and the height‐integrated Hall conductance between 80 and 120 km were enhanced by a factor of 2–4 inside the auroral patches. In this situation, a polarization electric field was generated therein. Enhanced ion velocities due to the polarization electric field was observed at up to 200‐km altitude; however, the absolute values of the ion velocities were approximately 40% of what was expected from the polarization electric field. A field‐aligned current (FAC) from 5 to 10 μA/m−2in the edges of the auroral patches could explain the weakening of the polarization electric field. Since a FAC of that order of magnitude corresponded with that observed by the Swarm satellite, it was suggested that the polarization electric field was weakened by the FAC. Furthermore, the polarization electric field propagated upward from the dynamo region to at least 200 km. The drift speed of auroral patches was consistent with the convective electric fieldThe polarization electric fields perpendicular to the convective electric fields were generated inside the auroral patchesThe upward and downward FACs contacting the west and east side edges of the auroral patches weakened the polarization electric field

Solar proton events (SPEs) cause large‐scale ionization in the middle atmosphere leading to ozone loss and changes in the energy budget of the middle atmosphere. The accurate implementation of SPEs and other particle ionization sources in climate models is necessary to understand the role of energetic particle precipitation in climate variability. We use riometer observations from 16 riometer stations and the Whole Atmosphere Community Climate Model with added Dregion ion chemistry (WACCM‐D) to study the spatial and temporal extent of cosmic noise absorption (CNA) during 62 SPEs from 2000 to 2005. We also present a correction method for the nonlinear response of observed CNA during intense absorption events. We find that WACCM‐D can reproduce the observed CNA well with some need for future improvement and testing of the used energetic particle precipitation forcing. The average absolute difference between the model and the observations is found to be less than 0.5 dB poleward of about 66° geomagnetic latitude, and increasing with decreasing latitude to about 1 dB equatorward of about 66° geomagnetic latitude. The differences are largest during twilight conditions where the modeled changes in CNA are more abrupt compared to observations. An overestimation of about 1° to 3° geomagnetic latitude in the extent of the CNA is observed due to the fixed proton cutoff latitude in the model. An unexplained underestimation of CNA by the model during sunlit conditions is observed at stations within the polar cap during 18 of the studied events. CNA during 62 SPEs was studied with WACCM‐D and riometer observationsCNA is modeled well on average and in individual events with a nonlinearity correction for high levels of CNAThe fixed proton cutoff latitude in WACCM‐D at 60 degrees leads to overestimation of the extent of the CNA, especially in small to moderate SPEs

This study investigates the response of the semidiurnal tide (SDT) to the 2013 major sudden stratospheric warming (SSW) event using meteor radar wind observations and mechanistic tidal model simulations. In the model, the background atmosphere is constrained to meteorological fields from the Navy Global Environmental Model—High Altitude analysis system. The solar (thermal) and lunar (gravitational) SDT components are forced by incorporating hourly temperature tendency fields from the ERA5 forecast model, and by specifying the M2and N2lunar gravitational potentials, respectively. The simulated SDT response is compared against meteor wind observations from the CMOR (43.3°N, 80.8°W), Collm (51.3°N, 13.0°E), and Kiruna (67.5°N, 20.1°E) radars, showing close agreement with the observed amplitude and phase variability. Numerical experiments investigate the individual roles of the solar and lunar SDT components in shaping the net SDT response. Further experiments isolate the impact of changing propagation conditions through the zonal mean background atmosphere, non‐linear wave‐wave interactions, and the SSW‐induced stratospheric ozone redistribution. Results indicate that between 80 and 97 km altitude in the northern hemisphere mid‐to‐high latitudes the net SDT response is driven by the solar SDT component, which itself is shaped by changing propagation conditions through the zonal mean background atmosphere and by non‐linear wave‐wave interactions. In addition, it is demonstrated that as a result of the rapidly varying solar SDT during the SSW the contribution of the lunar SDT to the total measured tidal field can be significantly overestimated. Simulations of the semidiurnal tide (SDT) are compared against meteor wind observations in the mid‐to‐high latitude northern hemisphere during the 2013 sudden stratospheric warmingIndividual lunar and solar SDT simulations find that the net tidal response is largely driven by the solar componentThe response of the solar SDT is driven by changing zonal mean propagation conditions and by non‐linear interactions with planetary waves Simulations of the semidiurnal tide (SDT) are compared against meteor wind observations in the mid‐to‐high latitude northern hemisphere during the 2013 sudden stratospheric warming Individual lunar and solar SDT simulations find that the net tidal response is largely driven by the solar component The response of the solar SDT is driven by changing zonal mean propagation conditions and by non‐linear interactions with planetary waves

Differential energy flux of electrons precipitating into the high‐latitude ionosphere can be estimated from incoherent scatter radar observations of the ionospheric electron density profile. We present a method called ELSPEC for electron spectrum estimation from incoherent scatter radar measurements, which is based on integration of the electron continuity equation and spectrum model selection by means of the Akaike information criterion. This approach allows us to use data with almost arbitrary time resolutions, enables spectrum estimation with dense energy grids, avoids noise amplifications in numerical derivatives, and yields statistical error estimates for all the output parameters, including the number and energy fluxes and upward field‐aligned currents carried by the precipitating electrons. The technique is targeted for auroral energies, 1–100 keV, which ionize the atmosphere mainly between 80 and 150 km altitudes. We validate the technique by means of a simulation study, which shows that Maxwellian, kappa, and mono‐energetic spectra, as well as combinations of those, can be reproduced. Comparison study for two conjugate satellite measurements to the EISCAT UHF radar are shown, for Reimei and Swarm, showing an agreement with the results. Finally, an example of a 2‐hr measurement by the EISCAT radar is shown, during which we observe a variety of precipitation characteristics, from soft background precipitation to mono‐energetic spectra with peak energies up to 60 keV. The upward field‐aligned current varies from 0 to 10 μAm−2and the total energy flux from 0 to 250 mWm−2. Electron energy spectra are estimated from incoherent scatter radar measurements by integrating the electron continuity equationEstimates of upward field‐aligned current and auroral power are integrated from the spectrum estimatesThe results are validated using simulations and comparisons with satellite measurements

Dust impact detection by electric field instruments is a relatively new method. However, the influence of dust impacts on electric field measurements is not completely understood and explained. A better understanding is very important for reliable dust impact identification, especially in environments with low dust impact rate. Using data from Earth‐orbiting Magnetospheric Multiscale mission (MMS) spacecraft, we present a study of various pulses detected simultaneously by multiple electric field antennas in the monopole (probe‐to‐spacecraft potential measurement) and dipole (probe‐to‐probe potential measurement) configurations. The study includes data obtained during an impact of a millimeter‐sized object. We show that the identification of dust impacts by a single antenna is a very challenging issue in environments where solitary waves are commonly present and that some pulses can be easily misinterpreted as dust impacts. We used data from multiple antennas to distinguish between changes in the spacecraft potential (dust impact) and structures in the ambient plasma or electric field. Our results indicate that an impact cloud is in some cases able to influence the potential of the electric field antenna during its expansion. Dust impact detection by multiple electric field antennas in the monopole and dipole configurationsSimilarities in signatures of dust impacts and solitary waves are discussedData obtained during the impact of a millimeter‐sized object are described in detail

Using data from the Sodankylä Geophysical Observatory (67°22′N, 26°38′E, Finland) meteor camera from the whole year 2015, we identified and investigated 28 optical meteors with accompanying ionization trails unambiguously detected by the Sodankylä Geophysical Observatory ionosonde, which sounded the ionosphere once per minute with frequency rising from 0.5 to 16 MHz. These ionosonde reflections were obtained from heights around 90 km. The electron line densities of the trails were found to be between 1014and 1016m−1, which characterize the trails as nonunderdense (i.e., transitional and overdense). The ionosonde reflections were observed for a few minutes, with decreasing maximal frequency of the return. During the first 250 s, for the trails with initial line density about (2–3) · 1015m−1the return frequency decreased with time corresponding to the diffusional expansion of cylindrical meteor trails, that is, f∝ t−γ, where the exponent γ= 0.5, whereas less dense trails decayed slower (γ≈ 0.2) and more dense trails decayed faster (γ≈ 1). In many cases the meteor events were accompanied by nonspecular long‐lived detections using a colocated all‐sky interferometric meteor radar with operating frequency 36.9 MHz. As a rule the meteor radar echo durations were longer than expected from diffusional expansion of cylindrical meteor trails and their amplitudes were highly variable. We suggest that the slower frequency decrease of the ionosonde echoes and the nonspecular long‐lived meteor radar echoes might be associated with the presence of meteoric dust. Optical, ionosonde, and meteor radar observations of non‐underdense (transitional and overdense) meteor trails were made at high latitudeDuring the first 4 min the decay rate of ionosonde echo depends on initial electron line density of trails (empirical dependence obtained)Meteor events were accompanied by nonspecular long‐lived detections using a colocated meteor radar with operating frequency 36.9 MHz

We present the first study of dust impact events on one of the Earth‐orbiting Cluster satellites. The events were identified in the measurements of the wide band data (WBD) instrument on board the satellite operating in monopole configuration. Since 2009 the instrument is operating in this configuration due to the loss of three electric probes and is therefore measuring the potential between the only operating antenna and the spacecraft body. Our study shows that the WBD instrument on Cluster 1 is able to detect pulses generated by dust impacts and discusses four such events. The presence of instrumental effects, intensive natural waves, noncontinuous sampling modes, and the automatic gain control complicates this detection. Due to all these features, we conclude that the Cluster spacecraft are not ideal for dust impact studies. We show that the duration and amplitudes of the pulses recorded by Cluster are similar to pulses detected by STEREO, and the shape of the pulses can be described with the model of the recollection of impact cloud electrons by the positively charged spacecraft. We estimate that the detected impacts were generated by micron‐sized grains with velocities in the order of tens of km/s. The first hypervelocity dust impact detection by the Cluster spacecraftThe concept of dust impact detection with electric field instruments is discussedThe size and velocity of impinging dust grains are estimated

It is well known that auroral patterns at the substorm recovery phase are characterized by diffuse or patch structures with intensity pulsation. According to satellite measurements and simulation studies, the precipitating electrons associated with these aurorae can reach or exceed energies of a few hundreds of keV through resonant wave‐particle interactions in the magnetosphere. However, because of difficulty of simultaneous measurements, the dependency of energetic electron precipitation (EEP) on auroral morphological changes in the mesoscale has not been investigated to date. In order to study this dependency, we have analyzed data from the European Incoherent Scatter (EISCAT) radar, the Kilpisjärvi Atmospheric Imaging Receiver Array (KAIRA) riometer, collocated cameras, ground‐based magnetometers, the Van Allen Probe satellites, Polar Operational Environmental Satellites (POES), and the Antarctic‐Arctic Radiation‐belt (Dynamic) Deposition‐VLF Atmospheric Research Konsortium (AARDDVARK). Here we undertake a detailed examination of two case studies. The selected two events suggest that the highest energy of EEP on those days occurred with auroral patch formation from postmidnight to dawn, coinciding with the substorm onset at local midnight. Measurements of the EISCAT radar showed ionization as low as 65 km altitude, corresponding to EEP with energies of about 500 keV. Aurora is emission of the atmospheric particles excited by electrons coming from the magnetosphere. The electrons have energies of 1–10 keV or higher. In particular, it is known that the energy can increase more than 100 keV in association with the pulsating aurora and that morphology of the pulsating aurora changes with time. However, relationships between the energy increase and the morphological change have not been studied well. This study analyzed the ionospheric density and auroral images and found that significant increases of the energy coincides with evolution of the patch structures in the pulsating aurora. Ground‐based and satellite measurements showed that auroral EEP (>100 keV) coincided with the patch appearance in the late morningMeasurements of the EISCAT radar showed ionization as low as 65 km altitude by ~500 keV electron precipitationEEP activity highly depended on MLT rather than geomagnetic activity

We report interannual variations of the correlation between the reflectivity of polar mesospheric summer echoes (PMSEs) and solar wind parameters (speed and dynamic pressure), and AEindex as a proxy of geomagnetic disturbances, and cosmic noise absorption (CNA) in the declining phase (2001–2008) of solar cycle 23. PMSEs are observed by 52 MHz VHF radar measurements at Esrange (67.8°N, 20.4°E), Sweden. In approaching the solar minimum years, high‐speed solar wind streams emanate from frequently emerging coronal holes, leading to 7, 9, and 13.5 day periodicities in their arrival at Earth. Periodicities of 7 and/or 9 days are found in PMSE reflectivity in 2005–2006 and 2008. Periodicity‐resolved correlations at 7 and 9 days of both Dregion ionization observed by cosmic noise absorption (CNA) and PMSE with solar wind speed and AEindex vary from year to year but generally increase as solar minimum is approached. PMSEs have a higher periodicity‐resolved correlation with AEindex than the solar wind speed. In addition, cross correlation of PMSE reflectivity with AEindex is mostly higher than with CNA in solar minimum years (2005–2008). This can signify that high‐speed solar wind stream‐induced high‐energy particles possibly have strong influence on CNA, but not as much as on PMSE, especially for the years of significant periodicities occurring. PMSE periodicities of 7 and 9 days characterize solar minimum yearsCIR/HSS‐driven geomagnetic disturbances induce strong Dregion ionizationPMSE is sensitive to the impact of CIR/HSS‐driven geomagnetic disturbances

During disturbed geomagnetic conditions, the energetic particles in the inner magnetosphere are known to undergo precipitation loss due to interaction with various plasma waves. This study, investigates the energetic particle precipitation events statistically using coordinate observations from the ground riometer network and the inner‐magnetospheric satellite mission, Arase. We have compared cosmic noise absorption (CNA) data obtained from the Finnish ground riometer network located in the auroral/sub‐auroral latitudes with the comprehensive data set of omnidirectional electron/proton flux and plasma waves in ELF/VLF frequency range from the Arase satellite during the overpass intervals. The study period includes one and a half years of data between March 2017 and September 2018 covering Arase conjunctions with the riometer stations from all magnetic local time sectors. The relation between the plasma flux/waves observed at the satellite with the riometer absorptions are investigated statistically for CNA (absorption >0.5 dB) and non‐CNA (absorption <0.5 dB) cases separately. During CNA events, Arase observed elevated electron flux in the medium energy range (2–100 keV), and plasma wave activity in the whistler‐mode frequency range (0.5–3 kHz) of the spectra. Our study provides an estimate of the statistical dependence of the electron flux and plasma wave observations at Arase with the ground reality of actual precipitation. Arase omnidirectional electron flux and plasma wave data are investigated statistically in conjunction with the Finnish riometer networkCNA events associate with Arase electron flux in the energy range >a few keV and plasma waves in 0.5–3 kHz at the ELF/VLF frequency rangeStudy suggests that whistler mode chorus waves are the major driver for EEP responsible for the observed CNA Arase omnidirectional electron flux and plasma wave data are investigated statistically in conjunction with the Finnish riometer network CNA events associate with Arase electron flux in the energy range >a few keV and plasma waves in 0.5–3 kHz at the ELF/VLF frequency range Study suggests that whistler mode chorus waves are the major driver for EEP responsible for the observed CNA

Both high‐power large aperture radars and smaller meteor radars readily observe the dense head plasma produced as a meteoroid ablates. However, determining the mass of such meteors based on the information returned by the radar is challenging. We present a new method for deriving meteor masses from single‐frequency radar measurements, using a physics‐based plasma model and finite‐difference time‐domain (FDTD) simulations. The head plasma model derived in Dimant and Oppenheim (2017), https://doi.org/10.1002/2017ja023963depends on the meteoroids altitude, speed, and size. We use FDTD simulations of a radar pulse interacting with such head plasmas to determine the radar cross section (RCS) that a radar system would observe for a meteor with a given set of physical properties. By performing simulations over the observed parameter space, we construct tables relating meteor size, velocity, and altitude to RCS. We then use these tables to map a set of observations from the MAARSY radar (53.5 MHz) to fully defined plasma distributions, from which masses are calculated. To validate these results, we repeat the analysis using observations of the same meteors by the EISCAT radar (929 MHz). The resulting masses are strongly linearly correlated; however, the masses derived from EISCAT measurements are on average 1.33 times larger than those derived from MAARSY measurements. Since this method does not require dual‐frequency measurements for mass determination, only validation, it can be applied in the future to observations made by many single‐frequency radar systems. The material left behind as meteoroids burn up in the upper atmosphere has significant effects on atmospheric chemistry and dynamics. However, the amount of mass deposited by any single meteoroid, and therefore the overall input rate, is difficult to calculate. We present a new method for determining individual meteor masses using radar observations and numerical simulations. We use a physics‐based model of the meteor plasma distribution to simulate the interaction between a radar pulse and a meteor, and calculate observable quantities. Using these simulations, we relate the radar observations to physical characteristics of the meteor, which we then use to estimate the mass. Since this method only requires a single radar observation to calculate a meteor's mass, we apply it to a set of meteors observed at the same time by two radar systems, and compare the results. A finite difference time domain (FDTD) model is used to simulate radar observations of meteorsMeteor mass estimations are made by combining observed radar cross sections with FDTD simulation resultsA data set of coincident observations by two radar systems is used to verify the mass estimation procedure A finite difference time domain (FDTD) model is used to simulate radar observations of meteors Meteor mass estimations are made by combining observed radar cross sections with FDTD simulation results A data set of coincident observations by two radar systems is used to verify the mass estimation procedure

Pulsating auroras (PsAs) are low‐intensity diffuse aurora, which switch on and off with a quasiperiodic oscillation period from a few seconds to ∼10 s. They are predominantly observed after magnetic midnight, during the recovery phase of substorms and at the equatorward boundary of the auroral oval. PsAs are caused by precipitating energetic electrons, which span a wide range of energies between tens and hundreds of keV. Such energetic PsA electrons will deposit their energy at mesospheric altitudes and induce atmospheric chemical changes. To examine the effects of energetic PsA electrons on the atmosphere, we first collect electron flux and energy measurements from low‐latitude spacecraft to construct a typical energy spectrum of precipitating electrons during PsA. Among the 840 PsA events identified using ground‐based auroral all‐sky camera (ASC) network over the Fennoscandian region, 253 events were observed by DMSP, POES, and FAST spacecraft over the common field of view of five ASCs. The combined measurements from these spacecraft enable us to obtain an energy spectrum consisting of nonrelativistic and relativistic (30 eV to 1,000 keV) electrons during PsA. The median spectrum was found to be in good agreement with earlier estimates of the PsA spectra. We then use the Sodankylä Ion‐neutral Chemistry (SIC) model to assess the chemical effect of PsA electrons. The observed extreme and median spectra of PsA produce a significant depletion in the mesospheric odd oxygen concentration up to 78%. Spacecraft particle measurements were used to construct a statistical precipitation spectrum for 253 pulsating aurora eventsLarge flux variations of energetic electron precipitation associated with pulsating aurora between the energies of 10 and 200 keVModel results showed mesospheric Oxdestruction up to 78% for the median duration of pulsating aurora

On 26 March 2008, simultaneous measurements of a large substorm were made using the Poker Flat Incoherent Scatter Radar, Time History of Events and Macroscale Interactions during Substorm (THEMIS) spacecraft, and all sky cameras. After the onset, electron precipitation reached energies ≳100keV leading to intense Dregion ionization. Identifying the source of energetic precipitation has been a challenge because of lack of quantitative and magnetically conjugate measurements of loss cone electrons. In this study, we use the maximum entropy inversion technique to invert altitude profiles of ionization measured by the radar to estimate the loss cone energy spectra of primary electrons. By comparing them with magnetically conjugate measurements from THEMIS‐D spacecraft in the nightside plasma sheet, we constrain the source location and acceleration mechanism of precipitating electrons of different energy ranges. Our analysis suggests that the observed electrons ≳100keV are a result of pitch angle scattering of electrons originating from or tailward of the inner plasma sheet at ~9RE, possibly through interaction with electromagnetic ion cyclotron waves. The electrons of energy 10–100 keV are produced by pitch angle scattering due to a potential drop of ≲10kV in the auroral acceleration region (AAR) as well as wave–particle interactions in and tailward of the AAR. This work demonstrates the utility of magnetically conjugate ground‐ and space‐based measurements in constraining the source of energetic electron precipitation. Unlike in situ spacecraft measurements, ground‐based incoherent scatter radars combined with an appropriate inversion technique can be used to provide remote and continuous‐time estimates of loss cone electrons in the plasma sheet. On the night of 26 March 2008 in Alaska, the sky lit up with intense auroral activity. A ground‐based electronically steerable radar and a spacecraft were simultaneously monitoring this activity. The radar observed intense signatures of charged particle density at unusually low altitudes (<70 km), suggesting that unusually high‐energy electrons were raining down from the magnetosphere. Although numerous studies in the past have estimated the source population of electrons, it has been difficult to determine and separate contributions from different physical processes. In this study, we deduced the energy and number of these electrons using radar measurements of charged particle density as a function of altitude. By comparing this estimate with that measured by the spacecraft along the same magnetic field line, we identified the source location and mechanism of acceleration of electrons of different energies. Our analysis suggests that the observed high‐energy electrons are accelerated at distances beyond ~60,000 km from the Earth by interacting with electromagnetic waves, whereas lower energy electrons are mostly accelerated by electric fields much closer to Earth. Our study demonstrates the utility of combining radar and spacecraft measurements in determining the source region and acceleration process of these high‐energy electrons. Subrelativistic electron precipitation ≳100 keV observed using incoherent scatter radar during a substormSource of the precipitating subrelativistic electrons lies at or tailward of the inner plasma sheetIncoherent scatter radar can measure energy spectra of precipitating magnetospheric plasma

Intensity scintillations of cosmic radio sources are used to study astrophysical plasmas like the ionosphere, the solar wind, and the interstellar medium. Normally, these observations are relatively narrow band. With Low‐Frequency Array (LOFAR) technology at the Kilpisjärvi Atmospheric Imaging Receiver Array (KAIRA) station in northern Finland we have observed scintillations over a three‐octave bandwidth. “Parabolic arcs,” which were discovered in interstellar scintillations of pulsars, can provide precise estimates of the distance and velocity of the scattering plasma. Here we report the first observations of such arcs in the ionosphere and the first broadband observations of arcs anywhere, raising hopes that study of the phenomenon may similarly improve the analysis of ionospheric scintillations. These observations were made of the strong natural radio source Cygnus‐A and covered the entire 30–250 MHz band of KAIRA. Well‐defined parabolic arcs were seen early in the observations, before transit, and disappeared after transit although scintillations continued to be obvious during the entire observation. We show that this can be attributed to the structure of Cygnus‐A. Initial results from modeling these scintillation arcs are consistent with simultaneous ionospheric soundings taken with other instruments and indicate that scattering is most likely to be associated more with the topside ionosphere than the Fregion peak altitude. Further modeling and possible extension to interferometric observations, using international LOFAR stations, are discussed. First multioctave low‐frequency observations of ionospheric scintillationProbe ionospheric scintillation velocities and scattering altitudes