J. L. Posch

Wave acceleration of electrons in the Van Allen radiation belts

The Van Allen radiation belts are two regions encircling the Earth in which energetic charged particles are trapped inside the Earths magnetic field. Their properties vary according to solar activity and they represent a hazard to satellites and humans in space. An important challenge has been to explain how the charged particles within these belts are accelerated to very high energies of several million electron volts. Here we show, on the basis of the analysis of a rare event where the outer radiation belt was depleted and then re-formed closer to the Earth, that the long established theory of acceleration by radial diffusion is inadequate; the electrons are accelerated more effectively by electromagnetic waves at frequencies of a few kilohertz. Wave acceleration can increase the electron flux by more than three orders of magnitude over the observed timescale of one to two days, more than sufficient to explain the new radiation belt. Wave acceleration could also be important for Jupiter, Saturn and other astrophysical objects with magnetic fields.

Van Allen probes, NOAA, GOES, and ground observations of an intense EMIC wave event extending over 12 h in magnetic local time

Although most studies of the effects of electromagnetic ion cyclotron (EMIC) waves on Earths outer radiation belt have focused on events in the afternoon sector in the outer plasmasphere or plume region, strong magnetospheric compressions provide an additional stimulus for EMIC wave generation across a large range of local times and L shells. We present here observations of the effects of a wave event on 23 February 2014 that extended over 8 h in UT and over 12 h in local time, stimulated by a gradual 4 h rise and subsequent sharp increases in solar wind pressure. Large-amplitude linearly polarized hydrogen band EMIC waves (up to 25 nT p-p) appeared for over 4 h at both Van Allen Probes, from late morning through local noon, when these spacecraft were outside the plasmapause, with densities ~5–20 cm−3. Waves were also observed by ground-based induction magnetometers in Antarctica (near dawn), Finland (near local noon), Russia (in the afternoon), and in Canada (from dusk to midnight). Ten passes of NOAA-POES and METOP satellites near the northern foot point of the Van Allen Probes observed 30–80 keV subauroral proton precipitation, often over extended L shell ranges; other passes identified a narrow L shell region of precipitation over Canada. Observations of relativistic electrons by the Van Allen Probes showed that the fluxes of more field-aligned and more energetic radiation belt electrons were reduced in response to both the emission over Canada and the more spatially extended emission associated with the compression, confirming the effectiveness of EMIC-induced loss processes for this event.

Pi1 magnetic pulsations in space and at high latitudes on the ground

Nightside, high-latitude (68°–80° magnetic) Pi1 waves, measured with a ground array of induction magnetometers, are studied and compared with magnetic field measurements made at synchronous orbit near the meridian of the ground measurements. The objectives of the study are to relate the ground signatures of Pi1B and PiC to the auroral substorm and its manifestation at synchronous orbit in an attempt to understand the origin of the Pi1 waves. Pi1 waves are measured in the equatorial plane by the GOES spacecraft and appear to be initiated by the dipolarization process of the nightside tail magnetic field at the onset of substorms. Across two meridional arrays of ground stations the earliest onset of Pi1B generally occurred at the lowest-latitude station and, in many instances, this burst was superimposed on a ground signature of a sudden onset of the westward electrojet. In one instance, where good coverage of ground optical data was available, this sudden onset Pi1B was time related to the overhead passage of a westward traveling auroral surge. The timing of maximum Pi1B across the array in both longitude and latitude agrees with the westward motion of the local auroral surge and the poleward motion of the aurora after the surge has arrived at a given site, suggesting a local ionospheric source for some of the Pi1 waves. However, across the entire array, extending about 12° in latitude and 20° in longitude, there often was nearly simultaneous Pi1B wave power at all sites which occurred before the maximum signal at a given site (and presumably local aurora), suggesting horizontal ducting of wave power from the onset of PiIB seen earlier at the auroral zone latitude. Prompt turn-on of PiC waves across the array also indicates ducting of these waves. The narrow bandwidth of the PiC waves themselves suggests a resonant cavity source for them which would indicate that some wave power enters the ionosphere from space (consistent with the GOES in situ data) as the trigger wave for this resonant source mechanism. In conclusion, this study finds evidence for local ionospheric currents, magnetospheric waves, and resonant cavity modes as sources for Pi1 ground waves.

Characteristics of broadband ULF magnetic pulsations at conjugate cusp latitude stations

Although cusp latitude pulsation studies have for the most part focused on narrowband waves, analysis of magnetometer data from the Arctic has shown that the most common type of dayside long-period ULF wave activity at very high latitudes is broadband noise (Pi1-2), and that its occurrence and intensity is largely controlled by solar wind velocity [Engebretson et al., 1995]. However, the origin of temporal variations in the intensity of these waves is not understood. In order to further investigate these broadband waves and their origins, we present a similar data set from another season, data from a roughly conjugate site, and multi-instrument data. Comparison of conjugate station data revealed that there was a substantial fraction of days during which there was significant temporal disagreement between hemispheres, but the solar wind velocity still appears to control overall daily intensity in broadband power. The coincidence of increased riometer absorption from conjugate locations with strong broadband ULF wave power suggests that precipitating energetic particles are responsible for much of the broadband ULF noise, and further suggests that high solar wind velocity plays a role in precipitation of significant fluxes of energetic particles. Quantitative estimates based on riometer and photometer observations also indicate that modulated electron precipitation is sufficient to drive the broadband pulsations. We review possible source mechanisms for these broadband waves and the precipitating electrons associated with them. Finally, the clear temporal association between these waves and Pc5 waves on closed field lines may suggest a causal connection via modulations of a three-dimensional current system.

Latitudinal and seasonal variations of quasiperiodic and periodic VLF emissions in the outer magnetosphere

We have analyzed ELF-VLF receiver and search coil magnetometer data from five Antarctic stations from 1998 and 1999 to study quasiperiodic emissions (QPs) and periodic emissions (PEs), which occur as ULF-range modulations of ELF-VLF signals between 0.5 kHz and similar to4 kHz. QPs are modulated at frequencies of similar to20-50 mHz, and PEs are modulated at frequencies of similar to200-500 mHz. The stations used covered a range of magnetic latitudes from -62degrees (Halley) to -74degrees (South Pole Station); three automated geophysical observatories (AGOs) were located at intermediate latitudes. Consistent with earlier studies, most QPs were observed with magnetic pulsations of identical period in the Pc3 range ( type I QPs). Of those QPs not observed with simultaneous magnetic pulsations ( type II QPs), nearly all were accompanied by PEs. Type I QPs, PEs, and events during which both appeared together (QPPEs) were found to have different latitudinal, seasonal, and diurnal occurrence patterns: QPs of both types were more likely to occur between -65degrees and -70degrees magnetic latitude, while PEs occurred more often around -60degrees magnetic latitude. QPs were more common during the months of October though March, while PEs were more common during the months of May through September. QPs, whether with or without simultaneous PEs or magnetic pulsations, were predominantly a dayside phenomenon, with a broad maximum near local noon. The occurrence of QPs unaccompanied by PEs was restricted to the dayside, however, while a small number of QPPEs appeared even during nighttime hours. PEs, on the other hand, could be seen at all local times, but with latitudinally dependent diurnal patterns. Most higher-latitude QPs were type I events (observed with magnetic pulsations), while type II QP events (without simultaneous magnetic pulsations) occurred relatively more often at lower latitudes. A case study from 1 August 1999 using wideband data from South Pole and Halley provides evidence of a transition from echoing whistler activity to PE activity and then to QP activity and suggests a causal relationship.

Ground observations and possible source regions of two types of Pc 1‐2 micropulsations at very high latitudes

We have used 1-years data from the recently installed Magnetometer Array for Cusp and Cleft Studies (MACCS) in Arctic Canada and from two stations of the developing “conjugate” array of Automated Geophysical Observatories (AGOs) in Antarctica to study ULF waves in the Pc 1–2 (100–600 mHz) frequency band at cusp and polar cap latitudes (Λ ∼ 74° – 80°). In this paper we focus on the spectral properties and latitudinal and local time distributions of Pc 1–2 events observed during 1994 and use these along with several case studies to infer the source locations of the two major wave types we have observed. We found little variation in center-band frequency of the Pc 1–2 waves we observed, but the average event bandwidth was distinctly wider at stations near 80° MLAT than at stations near 75° MLAT. Broadband waves, with diffuse spectral character, dominated at the higher latitudes, but their occurrence was confined at most stations to within 4 hours of local magnetic noon. Waves with narrower bandwidth were much more common in our data set, and were the statistically dominant wave type at the lower-latitude MACCS stations. Their occurrence was also limited to the dayside but extended both later and more widely in local time than the more broadband waves. These multistation observations, combined with data from the DMSP, IMP 8, and Geotail satellites, suggest the possibility that these two wave types originate in quite different regions near the magnetospheric boundary; the more narrowband waves in the subsolar and postnoon equatorial region, and the more broadband waves in the high-latitude plasma mantle (and possibly at the poleward edge of the cusp). The cusp itself appears to not be a significant source of Pc 1–2 wave activity that can be detected by ground observatories.

A multipoint determination of the propagation velocity of a sudden commencement across the polar ionosphere

We use magnetic field and riometer data from ground observatories in both the Arctic and Antarctic regions to characterize the high-latitude propagation of a sudden storm commencement (SC) that occurred at 0901 UT February 21, 1994. (1) High time resolution magnetic field data from both hemispheres indicate extremely rapid propagation of the initial part of the SC signal at high latitudes. An initial inflection point was observed in the data from dawn sector stations in both polar caps nearly simultaneously (Δt <2 s) but ∼3 s earlier in the southern hemisphere. (2) Data from the Magnetometer Array for Cusp and Cleft Studies (MACCS) in Arctic Canada, with stations from 0130 to 0600 magnetic local time, indicate dispersive propagation of the preliminary impulse (PI) at speeds decreasing from ∼150 to ∼50 km/s, directly away from a source near or slightly poleward of the cusp, rather than along the auroral oval. (3) Plots of two-dimensional, 20-s resolution equivalent convection vectors in the Northern Hemisphere reveal the imposition and rapid propagation and decay of a short-lived large-scale flow pattern opposite to the normal dawn sector flow, followed by an intensification of the original pattern. However, the perturbed flows were not dominated by the localized, tailward propagating vortices predicted in some models of SC events. In both hemispheres, observations of the PI are consistent with the temporary imposition of a field-aligned current pair antisymmetric about local noon in the polar ionosphere and with horizontal fast mode propagation of a pulse through the ionosphere. (4) Riometer signatures do not match the magnetic variations in either time or intensity, and they propagate with a much lower velocity (∼1 km/s). We infer that the initial magnetic signatures are generated by the arrival of Alfven waves and associated Birkeland currents at the near-cusp ionosphere, while the riometer signatures at these high latitudes are generated predominantly by interplanetary particles associated with the causative coronal mass ejection.

Probing the relationship between electromagnetic ion cyclotron waves and plasmaspheric plumes near geosynchronous orbit

Plasmaspheric plumes created during disturbed geomagnetic conditions have been suggested as a major cause of increased occurrences of electromagnetic ion cyclotron (EMIC) waves at these times. We have catalogued occurrences of strong Pc1 EMIC waves from 1996 through 2003 at three automated geophysical observatories operated by the British Antarctic Survey at auroral zone latitudes in Antarctica (L = 6.28, 7.68, and 8.07) and have compared them to the occurrence of plasmaspheric plumes in space, using simultaneous data from the Magnetospheric Plasma Analyzer on the Los Alamos National Laboratory 1990-095 spacecraft, in geosynchronous orbit at the same magnetic longitude. A superposed epoch analysis of these data was conducted for several categories of disturbed geomagnetic conditions, including magnetic storms, high-speed streams, and storm sudden commencements. We found only a weak correspondence between the occurrence of strong Pc1 waves observed on the ground and either plasmaspheric plumes or intervals of extended plasmasphere at geosynchronous orbit before, during, or after the onset of any of these categories. Strong Pc1 activity peaked near or slightly after local noon during all storm phases, consistent with equatorial observations by the Active Magnetospheric Particle Tracer Explorers/Charge Composition Explorer satellite at these L shells. The highest Pc1 occurrence probability was at or 1-2 days before storm onset and during the late recovery phase. Occurrence was lowest during the early recovery phase, consistent with the decrease in solar wind pressure often seen at this time. The peak at onset is consistent with earlier observations of waves in the outer magnetosphere stimulated by sudden impulses and magnetospheric compressions.

Low-harmonic magnetosonic waves observed by the Van Allen Probes†

Purely compressional electromagnetic waves (fast magnetosonic waves), generated at multiple harmonics of the local proton gyrofrequency, have been observed by various types of satellite instruments (fluxgate and search coil magnetometers and electric field sensors), but most recent studies have used data from search coil sensors, and many have been restricted to high harmonics. We report here on a survey of low-harmonic waves, based on electric and magnetic field data from the Electric Fields and Waves double probe and Electric and Magnetic Field Instrument Suite and Integrated Science fluxgate magnetometer instruments, respectively, on the Van Allen Probes spacecraft during its first full precession through all local times, from 1 October 2012 to 13 July 2014. These waves were observed both inside and outside the plasmapause (PP), at L shells from 2.4 to ~6 (the spacecraft apogee), and in regions with plasma number densities ranging from 10 to >1000 cm−3. Consistent with earlier studies, wave occurrence was sharply peaked near the magnetic equator. Waves appeared at all local times but were more common from noon to dusk, and often occurred within 3 h after substorm injections. Outside the PP occurrence maximized broadly across noon, and inside the PP occurrence maximized in the dusk sector, in an extended plasmasphere. We confirm recent ray-tracing studies showing wave refraction and/or reflection at PP-like boundaries. Comparison with waveform receiver data indicates that in some cases these low-harmonic magnetosonic wave events occurred independently of higher-harmonic waves; this indicates the importance of including this population in future studies of radiation belt dynamics.

ULF Waves at Very High Latitudes

Observations in the ULF frequency range (from mHz to Hz) at very high latitudes, at the poleward auroral boundary and beyond, make it possible to characterize the turbulent energy transfer from the solar wind into dayside regions of the magnetosphere and the magnetotail. The regions near the cusp are characterized by elevated levels of ULF wave activity. There was hope that several categories of these waves (Pc1-2, Pc3-4, Pc5) could be used as ground-based indicators of location of the cusp or related regions. However, recent studies have revealed wave sources distinct from the cusp proper, often in the low-latitude boundary layer (LLBL) or slightly deeper in the magnetosphere, and for unstructured Pc1-2 activity, in the plasma mantle. The polar cap, on the other hand, has often been assumed to be a quiet region, with any wave power observed entering only from neighboring regions. Studies using arrays of stations in Antarctica have now shown the presence of ULF phenomena unique to the polar cap in Pc3-4 and Pc5-6 bands. A class of long-period variations (4-20 min) that are coherent throughout the polar cap and are decoupled from auroral and cusp ULF activity has been revealed. New types of waves and transients in all nominal ULF bands have been found in the magnetotail, but their ground images have not been thoroughly examined yet. The high-latitude regions of Earths magnetosphere, both the cusp/LLBL/mantle and polar cap, are thus even more complex than we had thought, generating ULF waves in regions and by processes that we had not anticipated and do not yet fully understand.