Selected Abstracts

The dawn–dusk locations of reconnection in the near-earth magnetotail at the time of isolated auroral breakup are studied to clarify whether breakup is always accompanied by reconnection. The near-earth reconnection is identified by tailward plasma flows faster than 200 km/s with southward magnetic field. We first identified 66 breakups in the Polar ultraviolet imager observations of the nightside polar ionosphere. We then studied tailward flows during breakups using Geotail in situ observations of the plasma sheet between 25 and 31 R_E down the tail. It was found that the dawn–dusk (Y) locations of relatively fast (≥400 km/s) tailward flows were associated with breakup magnetic local time (MLT) by a regression line of Y_AGSM = −5.7 × (MLT + 0.6) R_E with a correlation coefficient of 0.8. Most tailward flows were observed within 5 R_E of the modeled Y locations, where tailward flows occurred in 88% of the 26 cases of breakups between 22 and 0 MLT. It is thus inferred that in most cases, breakup is accompanied by tailward flow near the breakup MLT with its dawn–dusk dimension ∼10 R_E. There were only two events without tailward flows in the region where flows have been expected. These two events were an earthward flow event and a traveling compression region event, which are not inconsistent with the initiation of the near-earth reconnection. Auroral breakup is thus likely to always be accompanied by near-earth reconnection near breakup MLT. It is also inferred that reconnection and breakup occur simultaneously within a few minutes, assuming a time delay between reconnection onset and the arrival of tailward flows at satellite locations.

THEMIS was launched on February 17, 2007 to determine the trigger and large-scale evolution of substorms. During the first seven months of the mission the five satellites coasted near their injection orbit to avoid differential precession in anticipation of orbit placement, which started in September 2007 and led to a commencement of the baseline mission in December 2007. During the coast phase the probes were put into a string-of-pearls configuration at 100 s of km to 2 R_E along-track separations, which provided a unique view of the magnetosphere and enabled an unprecedented dataset in anticipation of the first tail season. In this paper we describe the first THEMIS substorm observations, captured during instrument commissioning on March 23, 2007. THEMIS measured the rapid expansion of the plasma sheet at a speed that is commensurate with the simultaneous expansion of the auroras on the ground. These are the first unequivocal observations of the rapid westward expansion process in space and on the ground. Aided by the remote sensing technique at energetic particle boundaries and combined with ancillary measurements and MHD simulations, they allow determination and mapping of space currents. These measurements show the power of the THEMIS instrumentation in the tail and the radiation belts. We also present THEMIS Flux Transfer Events (FTE) observations at the magnetopause, which demonstrate the importance of multi-point observations there and the quality of the THEMIS instrumentation in that region of space.

In this case study we report a substorm, 23 March 2007, which exhibited oscillations with a period of ∼135 s in three substorm phenomena all of which were one-to-one correlated. The in-situ observations are from one THEMIS spacecraft (8.3 R_E geocentric distance) and the geosynchronous LANL-97A spacecraft. The focus here is on the intensification phase during which THEMIS was conjugate to the region of auroral brightening and its foot point was near the high-latitude ground station Kiana. The following results will be demonstrated: (1) THEMIS and LANL-97A (time-delayed) recorded periodic ion injections (>100 keV). (2) Near-conjugate high-latitude ground magnetometer data show very large Pi2 (δH∼150 nT) with a 6-s time delay compared to the THEMIS ion injections. (3) Low-latitude ground magnetometer data also show Pi2 with the same waveform as the high-latitude Pi2 but with longer time delays (20–31 s). (4) Auroral luminosity was periodically modulated during the intensification phase. (5) All three signatures (ion injections, ground Pi2, optical modulation) had the same periodicity of ∼135 s but with various time delays with respect to the THEMIS ion injections. These observations demonstrate that the three substorm phenomena had a common source which controlled the periodicity.

Previously, Chua et al. [2004] computed substorm time scales for over three hundred substorms observed by the Polar Ultraviolet Imager (UVI). They found that statistically the substorm recovery time scales for substorms occurring near winter solstice and equinox (when the nighttime auroral zone was in darkness) was roughly twice as long as the recovery time scale for substorms occurring in the summer (when the nighttime auroral region was sunlit). These results strongly suggest that auroral substorms in the northern and southern hemispheres develop differently during solstice conditions with substorms lasting longer in the dark (winter) hemisphere than in the sunlit (summer) hemisphere. This also implies that more energy is deposited by electron precipitation in the winter hemisphere than in the summer one during substorms. Here, we extend this previous statistical work by analyzing the recovery time scales for simultaneous, conjugate auroral substorms observed by Polar UVI and IMAGE FUV. However, in order to quantitatively compare conjugate substorm time scales, we first inter-calibrate the two instruments by calculating the recovery time scales for both instruments while viewing "same scene" substorms to determine the effect of different temporal and spectral resolutions on the recovery time scales. We will discuss our inter-calibration procedure and present initial results of simultaneous, conjugate auroral substorm recovery time scales.

High resolution Demeter plasma and wave observations were available during one of the geomagnetic storms of November 2004 when the ionospheric footprint of the plasmasphere was pushed below 64 degrees in the midnight sector. We report here onboard observations of thermal/suprathermal plasma and HF electric field variations with a temporal resolution of 0.4 s, which corresponds to a spatial resolution of 3 km. Local perturbations of the plasma parameters at the altitude of 730 km are analysed with respect to the variation of the field-aligned currents, electron and proton precipitation and large-scale electric fields, measured in-situ by Demeter and by remote optical methods from the IMAGE/Polar satellites.

Flow monitoring in the 21:00 and 24:00 MLT sectors during storm conditions reveals two distinct regions of O⁺ outflow, i.e. the region of the field-aligned currents, which often comprises few layers of opposite currents, and the region of velocity reversal toward dusk at sub-auroral latitudes. Average upward O⁺ velocities are identical in both local time sectors and vary between 200 and 450 m s^-1, with an exception of a few cases of higher speed (~1000 m s^-1) outflow, observed in the midnight sector. Each individual outflow event does not indicate any heating process of the thermal O⁺ population. On the contrary, the temperature of the O⁺, outflowing from auroral latitudes, is found to be even colder than that of the ambient ion plasma. The only ion population which is observed to be involved in the heating is the O⁺ with energies a few times higher than the thermal energy. Such a population was detected at sub-auroral latitudes in the region of duskward flow reversal. Its temperature raises up to a few eV inside the layer of sheared velocity.

A deep decrease in the H⁺ density at heights and latitudes, where, according to the IRI model, these ions are expected to comprise ~50% of the positive charge, indicates that the thermospheric balance between atomic oxygen and hydrogen was re-established in favour of oxygen. As a consequence, the charge exchange between oxygen and hydrogen does not effectively limit the O⁺ production in the regions of the electron precipitation. According to Demeter observations, the O⁺ concentration is doubled inside the layers with upward currents (downward electrons). Such a density excess creates the pressure gradient which drives the plasma away from the overdense regions, i.e. first, from the layers of precipitating electrons and then upward along the layers of downward current.

In addition, the downward currents are identified to be the source regions of hiss emissions, i.e. electron acoustic mode excited via the Landau resonance in the multi-component electron plasma. Such instabilities, which are often observed in the auroral region at 2-5 Earth radii, but rarely at ionospheric altitudes, are believed to be generated by an electron beam which moves through the background plasma with a velocity higher than its thermal velocity.

Mars lacks a global magnetic field, but it does have intense and localized crustal fields yielding a complex magnetic topology. Where the crustal field has a nearly radial orientation, there is a tendency for the field lines to connect with the IMF, forming cusps that provide a conduit for ionospheric plasma to escape and for solar wind plasma to precipitate into the atmosphere. On the nightside one expects ionization due to solar wind electron precipitation in regions of open (radial) field lines at cusps and an absence of ionization in closed (horizontal) field regions. Recently observed accelerated electrons, which appear to be associated with cusps surrounding the strongest custal fields, will also create very localized regions of enhanced ionization.

Using an electron transport model, we calculate the electron density of the nighttime ionosphere of Mars and its spatial structure. As input we use Mars Global Surveyor electron measurements including an interval when accelerated electrons were observed. Precipitating accelerated electrons increase the maximum ionospheric number density by a factor of 3 over that produced by typical tail electrons. These regions of enhanced ionization are localized and occur near magnetic cusps. Horizontal gradients in the ionospheric electron density on the night side of Mars can reach ~4000 per cc over 200 km or ~20 per cc per km. Even sharper gradients occur near plasma voids; the electron density can go from effectively 0 per cc to ~5000 per cc over a few km.

Such strong gradients in the plasma density have several important consequences. These large pressure gradients will lead to localized plasma transport perpendicular to the ambient magnetic field and will generate horizontal currents and electric fields which will in turn lead to localized Joule heating. Additionally, transport of ionospheric plasma by neutral winds, which vary in strength and direction as a funcion of local time, can generate horizontal currents where the ions are collisionally coupled to the neutral atmosphere while electrons are not. Closure of the horizontal currents and electric fields may require the presence of vertical, field-aligned currents and fields which may play a role in high altitude acceleration processes.

Previous work [Chua et al., 2004] computed the substorm recovery time scale for over three hundred substorms observed by the Polar Ultraviolet Imager (UVI). When sorted according to season, the substorm recovery times were well ordered by whether or not the nightside auroral region was sunlit: substorms occurring in the winter and equinox periods had similar recovery time scales which were both roughly a factor of two longer than that for summer when the auroral oval was sunlit. These results strongly suggest that simultaneous auroral intensifications in the northern and southern hemispheres develop differently during solstice conditions. We expect the auroral breakup in the dark (winter) hemisphere to be more intense and longer lived than that observed in the sunlit (summer) hemisphere. This also implies that more energy is deposited by electron precipitation in the winter hemisphere than in the summer one during a substorm. Here we extend this previous work by including a similar number of substorms observed by IMAGE Far Ultraviolet Imager (FUV) as well as simultaneous, conjugate auroral substorm observations by Polar UVI and the IMAGE FUV. The observed hemispheric asymmetry and non-conjugacy of auroral substorms is consistent with the suppression of discrete aurora in sunlight and highlights the importance of ionospheric conductivity plays in global-scale dynamics of the aurora and in magnetosphere-ionosphere coupling.

Introduction: Recently, both Mars Global Surveyor and Mars Express have observed downward traveling auroral-like accelerated electrons on the night side of Mars [1, 2]. The spatial distribution of accelerated electrons is patchy and inhomogeneous tending to cluster near magnetic cusps at the periphery of strong crustal magnetic field sources. Upon interacting with the atmosphere, these accelerated electrons will create significant enhancements in the electron number density and total electron content in the Martian ionosphere. These enhancements in ionization will also be patchy and distributed nonuniformly on the night side. Such ionospheric structure has been recently observed by MARSIS onboard Mars Express on both the day and night sides [3-6].

Model: Using a modification of the terrestrial electron transport code of [7], we model the effect of accelerated electrons on the nighttime Martian ionosphere. Appropriate modifications include the incorporation of Martian neutral density profiles given by the MTGCM atmosphere of [8] and the addition of appropriate CO and CO2 cross sections for electron impact compiled by [9] and [10], respectively.

Model Results: We find that a typical accelerated electron spectrum increases the maximum ionospheric electron number density and TEC by a factor of 3 over that produced by a nominal magnetotail electron spectrum. This factor should not be regarded as an upper limit since electron spectra with downward energy fluxes over an order of magnitude larger than that of the example used here have been observed in the MGS data set [1]. The latitudinal width of the region of enhanced ionization is less than 200 km.

Seasonal and Solar Cycle Variations: We also investigate how seasonal and solar cycle changes in the upper atmosphere of Mars affect the maximum electron density, the altitude of the maximum, and the total electron content resulting from the precipitation of accelerated electrons. During northern winter (perihelion) conditions, the ionization peak, as well as the ionosphere in general, moves to higher altitude. Also the peak ionospheric density slightly decreases while the TEC slightly increases due to the larger atmospheric scale height at perihelion versus aphelion. During active solar conditions, the thickness of the ionospheric layer and the TEC increase as the peak ionospheric density decreases. Again this can be understood in terms of a larger upper atmospheric scale height under active solar conditions as compared to solar minimum.

Implications and Conclusions: The spatial and temporal variability of the nighttime Martian ionosphere has important implications for planning and carrying out subsurface radar soundings from orbit. The penetration depth to which soundings can reach is inversely proportional to the signal frequency. To penetrate the ionosphere, this frequency must be above the ionospheric electron plasma frequency which is determined by the maximum electron density. Therefore, increased ionospheric densities lead to decreased penetration depths. Recently [11] showed that MARSIS radar reflections from the surface can completely disappear on the night side during solar energetic particle events due to increased ionospheric electron densities. Finally, most previous work addressing how radio waves propagate through ionospheres assumes horizontal uniformity. This work shows that such an assumption is not valid in the night side ionosphere of Mars.

References: [1] Brain, D. A., et al. (2006) Geophys Res. Lett., 33, L01201. [2] Lundin, R., et al. (2006) Science, 311, 980-983. [3] Gurnett, D. A., et al. (2005) Science, 310, 1929-1933. [4] Duru, F., et al. (2006) J. Geophys. Res., 111, A12204. [5] Kirchner, D. L., et al. (2006) Geophys. Res. Abs., 8, 05224. [6] Kirchner, D. L., et al. (2007) Geophys. Res. Abs., 9, 04627. [7] Lummerzheim, D., and J. Lilensten (1994) Ann. Geophys., 12, 1039-1051. [8] Bougher, S. W., et al. (2000) J. Geophys. Res., 105, 17,669-17,692. [9] Liu, W., and G. A. Victor (1994) Ap. J., 435, 909-919. [10] Itikawa, Y. (2002) J. Phys. Chem. Ref. Data, 31, 749-767. [11] Morgan, D. D., et al. (2006) Geophys. Res. Lett., 33, L13202.

Using an electron transport model, we investigate the effect of electron precipitation on the electron density and total electron content in the nightside ionosphere of Mars. As input we use Mars Global Surveyor observations of a typical tail electron spectrum and an auroral-like electron spectrum. The accelerated electron spectrum increases the maximum number density and total electron content by a factor of 3 over that produced by the typical tail spectrum. Our calculations show a secondary electron density peak due to precipitation of several keV electrons not seen in previous modeling efforts. Regions of enhanced ionization are expected to be localized in space, corresponding to magnetic cusps formed by the interaction of crustal sources with the interplanetary magnetic field. Radio and radar measurements from both Mars Global Surveyor and Mars Express agree with this expectation. The horizontally inhomogeneous regions of ionization can affect signals used for subsurface sounding from orbit.

The hypothesis that ionospheric conductivity plays a major role in the global-scale dynamics of the aurora is further evaluated in this study. The substorm recovery time scale during auroral intensifications are computed for over three hundred substorms observed by the Polar Ultraviolet Imager (UVI) and the IMAGE Far Ultraviolet Imager (FUV) in both hemispheres and then sorted according to season. The substorm recovery times are well ordered by whether or not the nightside auroral region is sunlit: substorms occurring in the winter and equinox periods have similar recovery time scales which are both roughly a factor of two longer than that for summer when the auroral oval is sunlit. These results strongly suggest that simultaneous auroral intensifications in the northern and southern hemispheres develop differently during solstice conditions. We expect the auroral breakup in the dark (winter) hemisphere to be more intense and longer lived than that observed in the sunlit (summer) hemisphere. This also implies that more energy is deposited by electron precipitation in the winter hemisphere than in the summer one during a substorm. Simultaneous, conjugate auroral substorm observations by Polar UVI and the IMAGE FUV instrument are used to confirm this behavior. The observed hemispheric asymmetry and non-conjugacy of auroral substorms is consistent with the suppression of discrete aurora in sunlight and highlights the importance of this effect in magnetosphere-ionosphere coupling.

Using an electron transport model, we investigate the effect of ionization due to electron precipitation on the electron density and total electron content in the nightside ionosphere of Mars. As input we use typical tail electron spectra and recently reported auroral-like peaked electron spectra that appear to have undergone an acceleration process. The accelerated electron spectra increase the maximum number density and total electron content by nearly a factor of 3 over that produced by typical tail spectra. The regions of enhanced ionization are localized in space and correspond to magnetic cusps formed by the interaction of the Martian crustal sources with the interplanetary magnetic field. We find that our modeled peak ion production rates are several times less than previous calculations and that our calculated maximum ionospheric electron densities are less than recent observations. The main source of the discrepancy between our results and previous work appears to arise from differences in the neutral atmosphere profile. The upper atmosphere of Mars changes significantly with season and solar cycle. Likewise, peak ionization rates and maximum ionospheric electron number densities can also change significantly, with these quantities being minimum for conditions of large atmospheric scale heights; however, the thickness of the ionospheric layer is greatest under these conditions leading to a small net change in total electron content.

Correlations between global auroral images and in-situ measurements in the plasma sheet and magnetotail have led to many new insights into how the magnetosphere and ionospheric auroral region are coupled. Several studies have shown that intense auroral emission in the ionosphere is well correlated with plasma sheet disturbances characterized by one or more of the following: large ion velocity moments, enhancements in the energetic ion and electron fluxes, increases in plasma temperature, and high frequency fluctuations in the magnetic field. In addition global images provide unambiguous timing of auroral brightenings and direction and speed of propagation of auroral forms. The timing and propagation determined from images, in turn, can be used to put the plasma sheet observations in context. We show several examples of auroral substorms and other auroral activity observed by global auroral imagers and simultaneous, in-situ near-Earth plasma sheet plasma and magnetic field observations. We conclude that when a spacecraft in the near-Earth plasma sheet (X < 20 RE) detects plasma sheet activity, it is in a region magnetically connected to intense auroral emission in the ionosphere. This implies a near Earth source (near 10 RE) and that plasma sheet disturbances propagate tailward as regions of intense auroral emission migrate poleward. This result is mostly inconsistent with the NENL model of substorm onset. Several case studies which were originally interpreted to support the NENL model are re-interpreted to be inconsistent with this model. Future refinements to the NENL model must address these inconsistencies.

We analyzed a series of solar energetic particle events in late April and early May of 1998, during which lunar surface potentials reached values as large as 4.5 kV (the largest recorded by Lunar Prospector). The two largest surface charging events during this time period correspond to energetic particle injections, when the electron flux between 50 keV and 5 MeV exceeded the proton flux over the same energy range. We searched the entire Lunar Prospector data set for other large negative surface charging events, and found that they occur almost exclusively during magnetotail crossings (when the Moon encounters the plasmasheet) and solar energetic particle events. Lunar surface charging (and its effect on the lunar dust environment) during inherently unpredictable space weather events represents a significant hazard for exploration.

In the high latitude magnetotail, the solid body of the Moon acts as a particle absorber of interplanetary electrons traveling Earthward creating a lunar shadow in the electron flux. Electrons that are just outside the lunar shadow mirror nearer Earth and return to the vicinity of the Moon deflected by the cross tail convection electric field. By measuring this displacement from lunar orbiting spacecraft, the convection electric field can be determined. The convection electric field is an important parameter for magnetospheric dynamics but is difficult to accurately measure by other means. Pioneering work measuring the convection electric field from lunar orbit done during the Apollo Era will be reviewed. We revisit this problem using data from Lunar Prospector which had better energy, angular, and temporal resolutions. Out initial results show that at times the high latitude convection velocity is stable and on the order of 10 km/s in the equatorward direction corresponding to a dawn-dusk electric field of 0.1 mV/m, close to expectations. At other times no consistent electric field can be determined from the data either due to noise, spatial or temporal variability, or another breakdown in the assumptions that go into this method. We will present more detailed analyses of several events with consistent and non-consistent electric fields. Reasons for discrepant cases will be discussed. Finally, we discuss more ideal detector designs and orbit configurations to address this problem for future lunar missions.

Correlations between global auroral images and in-situ measurements in the plasma sheet and magnetotail have led to many new insights into how the magnetosphere and ionospheric auroral region are coupled. Several studies have shown that intense auroral emission in the ionosphere is well correlated with large ion velocity moments, enhancements in the energetic ion and electron fluxes, increases in plasma temperature, and high frequency fluctuations in the magnetic field measured in the plasma sheet. In addition global images provide unambiguous timing of auroral brightenings and direction and speed of propagation of auroral forms. The timing and propagation determined from images, in turn, can be used to put the plasma sheet observations in context. In this presentation, we focus on one particular conjunction of global auroral images and in-situ plasma sheet plasma and magnetic field measurements. Global auroral images from both the Polar UVI and IMAGE FUV instruments show multiple intensifications at different latitudes and local times over the course of a few hours following a solar wind pressure pulse and southward turning of the interplanetary magnetic field. The Cluster spacecraft were located in the plasma sheet at ~ 20 R_E. Cluster observations show complex and dynamic plasma behavior and current sheet structure developing a few minutes after the onset of the auroral intensifications. Initial analyses indicate that strong spatial gradients and plasma boundaries are present and that multiple current sheets or boundary waves develop. A kinetic interpretation will be presented to accurately describe the plasma dynamics of the complex nature of plasma sheet observations during auroral activity.

We present an analysis of simultaneously observed electromagnetic waveforms and ion and electron distribution functions measured by the WAVES and the 3-D Plasma experiments onboard the Wind spacecraft during a perigee pass through the near-Earth magnetotail. The WAVES experiment is composed of state-of-the-art high-time resolution spectral receivers and waveform analyzers sampling electric and magnetic signals up to 120000 samples/sec. The 3-D Plasma experiment provides measurements of the full three-dimensional ion and electron distribution functions covering energies from 10 eV to 20 keV. The dataset analyzed here comes from one Wind perigee pass through the magnetotail at radial distances between 5 and 8 Earth radii during a magnetically quiet period. Analysis of the data show several types of coherent waveforms between 50 Hz and a few hundred kHz such as quasi-monochromatic waves with frequencies close to the electron cyclotron frequency, solitary-like structures, low frequency whistlers, electron Bernstein waves, as well as coherent burts of Langmuir waves or upper-hybrid waves. In addition very peculiar particle distribution functions are observed at times of wave activity. We discuss the nature and detailed properties of the observed wavemodes. We also look for correlations between the waves and particular features in the ion and electron distribution functions in an attempt to identify the source(s) of the waves and understand the plasma micro-instability process(es) leading to these coherent fluctuations. Correlations between wave activity and auroral and magnetospheric activity are also investigated.

Return to Matt's Publications and Presentations page

Return to Matt's Home Page