Ion conics and electron beams associated with … › public › mkivelso › refs › PUBLICATIONS...going energetic electron beams, upward going ion conics, and enhanced broadband

Ion conics and electron beams associated with auroral processes

on Saturn

D. G. Mitchell,1 W. S. Kurth,2 G. B. Hospodarsky,2 N. Krupp,3 J. Saur,4 B. H. Mauk,1

J. F. Carbary,1 S. M. Krimigis,1 M. K. Dougherty,5 and D. C. Hamilton6

Received 21 July 2008; revised 5 December 2008; accepted 15 December 2008; published 14 February 2009.

[1] Ion conics accompanied by electron beams are observed regularly in Saturn’smagnetosphere. The beams and conics are seen throughout the outer magnetosphere, onfield lines that nominally map from well into the polar cap (Ldipole > 50) to well into theclosed field region (Ldipole < 10). The electron beams and ion conics are often observedtogether but also sometimes separately. Typically, the ion conics are prominent at energiesbetween about 30 keV and 200 keV. The electron beams extend from below the �20 keVlower energy threshold for the instrument to sometimes as high as 1MeV. The electrons maybe either unidirectional (upward) or bidirectional; the ions are exclusively unidirectionalupward. The ion conics are usually seen in conjunction with enhanced broadbandelectromagnetic noise in the 10 Hz to few kHz frequency range. Most of the wave energyappears below the local electron cyclotron frequency, hence, is propagating in the whistlermode, although some extension to higher frequencies is sometimes observed, suggesting anelectrostatic mode. Sometimes the particle phenomena and the broadband noise occur inpulses of roughly 5 min duration, separated by tens of minutes. At other times they arerelatively steady over an hour or more. Magnetic signatures associated with some of thepulsed events are consistent with field aligned current structures. The ions are almostexclusively light ions (H, H2, H3, and/or He) with only occasional hints of oxygen or otherheavier species, suggesting an ionospheric source. Taken together, the observations arestrikingly similar to those made at Earth in association with auroral zone downward sheetcurrents, except that in the case of Saturn the particle energies are 20 to 100 times higher.

Citation: Mitchell, D. G., W. S. Kurth, G. B. Hospodarsky, N. Krupp, J. Saur, B. H. Mauk, J. F. Carbary, S. M. Krimigis, M. K.

Dougherty, and D. C. Hamilton (2009), Ion conics and electron beams associated with auroral processes on Saturn, J. Geophys. Res.,

114, A02212, doi:10.1029/2008JA013621.

1. Introduction

[2] As discussed by Saur et al. [2006], field-aligned beamsof energetic electrons associated with auroral processes areubiquitous among the magnetized planets where appropriatemeasurements have been possible (Earth, Jupiter, and Saturn).In particular, electrons often appear to have been acceleratedbetween the spacecraft location and the ionosphere, travelingupward along the magnetic field and consistent with gener-ation within or close to the atmospheric loss cone. Similar

electron events at Jupiter have been discussed in considerabledetail by Mauk and Saur [2007]. These upward going elec-tron beams, as we refer to them, are not peaked at a specificenergy the way downward directed auroral electrons typ-ically are, but rather follow a power law in energy, from thelower threshold of the measurement to hundreds of keV.These electrons may be either unidirectional upward, orbidirectional. Also as discussed by Saur et al. [2006], eventssuch as these were first observed at Earth [e.g., Klumpar,1990; Carlson et al., 1998], and mechanisms for theirgeneration were proposed by Carlson et al. [1998], Klumpar[1990],Marklund et al. [2001], and Ergun et al. [1998]. TheFAST satellite measurements described by Carlson et al.[1998] showed a close correspondence between upwardgoing energetic electron beams, upward going ion conics,and enhanced broadband electromagnetic noise power. Theauthors showed that all of these phenomena were locatedwithin a region of downward field aligned current in theauroral zone Birkeland current system. At Jupiter,Mauk andSaur [2007] found that the beamsmapped to regions of brightaurora, and so concluded that such bright regions, generallyaccepted to be associated with upward field aligned current,must have regions of downward current intermixed.

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 114, A02212, doi:10.1029/2008JA013621, 2009ClickHere

for

FullArticle

1Johns Hopkins University Applied Physics Laboratory, Laurel, Maryland,USA.

2Department of Physics and Astronomy, University of Iowa, Iowa City,Iowa, USA.

3Max-Planck-Institut fur Sonnensystemforschung, Katlenburg-Lindau,Germany.

4Institut fur Geophysik and Meteorologie, Universitat zu Koln, Cologne,Germany.

5Blackett Laboratory, Imperial College, London, UK.6Department of Astronomy, University of Maryland, College Park,

Maryland, USA.

Copyright 2009 by the American Geophysical Union.0148-0227/09/2008JA013621$09.00

A02212 1 of 10

http://dx.doi.org/10.1029/2008JA013621

[3] In this work, we describe observations very similar tothose seen at Earth by Carlson et al. [1998], but in the case ofSaturn the particle energies are higher than those typical atEarth by a factor of up to 100. The composition of the ionconics confirms a likely ionospheric source for the acceler-ated ions. We describe two classes of field-aligned events.The first class is different from the observations at Jupiter,on the basis of their steady appearance over hours of ob-servation time. These events are consistent with extensive,contiguous low-altitude acceleration, presumably in regionsof downward field aligned current. A second class of eventsappears temporally pulsed, with a repetition rate on the orderof 1 h.

2. Observations

[4] We will describe measurements of energetic ions andelectrons from themagnetospheric imaging instrument (MIMI)ion and neutral camera (INCA), the low-energy magneto-spheric measurement system (LEMMS), and the charge,energy and mass sensor (CHEMS); energetic neutral atomsfrom INCA; magnetic fields from the fluxgate magnetometer(MAG); and plasma waves from the radio and plasma wavescience instrument (RPWS). Descriptions of these instru-ments are given by Krimigis et al. [2004], Dougherty et al.[2004], andGurnett et al. [2004]. Because INCAand LEMMSfields of view are nearly orthogonal, and the LEMMS angularcone very limited, it is unusual to have the opportunity tomeasure field-aligned energetic electrons with LEMMS forthe same event that we measure field-aligned ions with INCA.Observations of pitch angle distributions require both a space-craft roll and a favorable magnetic field orientation.[5] The first period we wish to highlight is on day 288 of

2006. Cassini was at 28.3 RS, 42� north latitude, Ldipole �51,and 2220 local time. During the hours 1900 to 2400 UT thespacecraft was rolling about an axis chosen for fields andparticles measurements, which for the measured field orien-tation allowed both the LEMMS and INCA sensors to samplenearly complete pitch angle distributions of energetic ionsand electrons. In Figure 1 (top) we display 100 keVelectroncounting rates from each end of the double-ended LEMMStelescope. Although LEMMS measures electrons over allpitch angles (except between �75� to 105�), the onlyelevated fluxes appear when near 0� pitch angles are sampled.An electron spectrogram (Figure 1, middle) shows that theenergy of these field aligned electrons routinely extends to>100 keV, sometimes reaching nearly 1 MeV. Note also thatexcept when the sensor can view near 0� pitch angle, theintensities are at detector background. In Figure 1 (bottom),we show details of the pitch angle distribution and the energyflux for the highest observed electron beam at 2200UT. Sincethe LEMMS conical collimator has a half-angle of 7.5�, the�10� apparent width of the electron beam is in fact <3�. Themeasured field strength was�4.1 nT. Using 2.5� as the beamangle, the mirror field for these particles would be�1500 nT,corresponding to an altitude under 3 RS. Given that the beamintensity is still rising with decreasing nominal pitch angle (atthe center of the LEMMS cone) when it reaches its minimumvalue of �4.5�, the angular distribution within the 2.5� widebeam is still strongly peaked toward zero degrees; thus theacceleration altitude is likely much closer to 1 RS. The elec-tron energy flux in the beam is nearly flat out to �600 keV,

indicating a process in the source region capable of acceler-ating electrons to at that energy.[6] Figure 2 shows energetic ion data from the INCA sen-

sor for two widely separated energy channels for energeticprotons. In this event, INCAwas in its ion mode, so the colli-mator plates are turned off and INCA measures ions between�10 keV/nucleon and 200 keV/nucleon. As the spacecraftexecutes its roll, the pitch angle contours move through suc-cessive 90� � 120� angular windows. When pitch anglesnear 0� enter the INCA FOV, a proton enhancement ap-pears parallel to the field spanning energies from 30 keV to200 keV, with roughly comparable intensity throughout thatrange. Simultaneously sampled oxygen channels covering60 keV to 500 keV total energy showed no sign of this beam.Note throughout this paper that the INCA sensor capabilityfor mass discrimination is limited. Hydrogen and oxygen arewell separated, but any ‘‘light’’ element (H, H2, H3, He) willbe classified as H, any heavy element (O, N, C, OH, H2O) asoxygen. We will generally refer to the light elements as H orprotons, the heavier elements as oxygen, since we believe onthe basis of published results from the CAPS [Young et al.,2005] andMIMI CHEMS [Krimigis et al., 2004] instrumentsthat these are the most likely candidates.[7] The second period we discuss is 0700–1100UTon day

269, 2006. The spacecraft was outbound at dusk, movingfrom 8 to 10 RS, and from 11 < L < 15. In this case, thespacecraft attitude was fixed such that INCA (which was inneutral mode, and so only sensitive to ions with energiesabove the collimator ion rejection cutoff of roughly 150 keV)could sample field aligned ions, whereas LEMMS could notsample field aligned electrons.[8] In Figure 3 we present the data from this event. The

first panel shows electrons with energies between 200 keVand 1 MeV. LEMMS is oriented perpendicular to the mag-netic field, so the pulsed enhancements are all at 90� pitchangle (not beams). Below that hydrogen intensities measuredby INCA are shown. The intensities rise at 0730 UT as aspacecraft maneuver moves the Sun out of the INCA field ofview, and the INCA voltages are turned up. Four times overthe next 3 h the INCA high-energy proton intensities rise(lower energy protons are electrostatically excluded by theion rejection plates). Below each proton intensification, angle-angle plots of the highest-energy channel are displayed. Ineach case, a distinctly (upward) field-aligned intensificationis evident. Below this plot, the magnetic field angles andmagnitude appear. The field angles exhibit abrupt changes by�2� to 3�, then return to their prepulse values. This behavioris consistent with the repeated passage of field aligned currentstructures. The fourth panel in Figure 3 shows broadbandelectromagnetic wave enhancements, again well correlatedwith the field aligned currents, upward ion beams, and elec-tron enhancements. Most of the wave energy in these broad-band bursts lies below the electron cyclotron frequency fcesignified by the white trace and is electromagnetic, propa-gating in the whistler mode. Awave normal analysis of thesedata shows the Poynting flux to be upward directed; thiswould be characteristic of the driving electrons also movingupward, indicative of a downward directed field alignedcurrent. The whistler mode cannot propagate above fce,however, so the emissions above fce in this example cannotbe whistler modes. In fact, no magnetic component is

A02212 MITCHELL ET AL.: SATURN AURORAL ION AND ELECTRON BEAMS

2 of 10

A02212

measured above the cyclotron frequency, suggesting anelectrostatic mode.[9] In this event, we can verify the existence of field-

aligned ions, but we have no direct measurement of field-aligned electrons. However, a survey of all of the missiondata to date reveals many instances where there is correspon-dence between electron and ion beams, in similar regions ofspace (high latitude, relatively low altitude, L > 10). In thiscase, we suggest that the (very weak) electron enhancements,

whose timing is slightly later than the onset of the ion beams,are backscattered from field irregularities much farther outthe flux tube, from electron beams similar to those describedin Figure 1.[10] In the data presented to this point, ion conics have not

been obvious. Whereas conics created at low altitudes willappear as beams in weak, high-altitude magnetic fields, untilvery recently in the mission we have not had many opportu-nities to measure the ion angular distributions at lower alti-

Figure 1. Energetic electron beam characteristics. (top) The time history of 100 keVelectrons measuredin opposite facing telescopes on the LEMMS sensor as the spacecraft spin sweeps the sensor throughvarious angles with respect to the magnetic field. The sensors are counting near background except whenthey sample near 0� pitch angle. (middle) Electron intensity as a function of time and energy. Spikes appearwhen the sensor samples 0� pitch angle. (bottom left) A quantitative plot of the electron pitch angles. Sincethe detector cone has a 7.5� half angle, the data are consistent with electrons confined to <3� from themagnetic field. (bottom right) The electron energy flux in the highest spike shows significant energy up to700 keV.


3 of 10

A02212

tude where their conic nature can be more easily measured. InFigure 4, we present data from day 269, 2008 between 1330and 1540 UT, when Cassini was at 7 to 8 Rs altitude, between60 and 70 degrees south latitude, at 37 < Ldipole < 52. Themagnetic field strength ranged from 110 to 80 nT. The ionangular distributions obtained over the course of each space-craft roll show very clear conic structures, especially at ener-gies above 100 keV (where the intensity peaks at about 155�to 160� pitch angle, which for this L and altitude correspondsto a 3.3 to 3.75 Rs mirror altitude). They are unidirectionalupward, and appear primarily in hydrogen; oxygen is presentin this event (as it is in a small fraction of these events) atintensities approximately 3 orders of magnitude lower thanhydrogen at equal total energy. While the proton conics arestrictly antiplanetward, the electron beam is bidirectionaland field aligned. Given that these observations are madein�100 nT magnetic field, it is not reasonable to assume thata mirror point is to be found on anything but a closed fieldline. The electrons, then, show that despite the large Ldipole

value this is a topologically closed field line, with foot pointsin both southern (below the spacecraft) and northern (oppo-site hemisphere) ionospheres.[11] On day 291, 2008, Cassini was outbound near mid-

night, at about�64� latitude and 5 Rs, when INCAmeasuredone of the most intense upward ion beams seen in the missionto date (Figure 5). Although we show only one energy band,this beam like others in this paper extends in energy fromless than 10 keV to over 200 keV. The beam appears and

disappears twice, a characteristic of a subset of these events,which we have referred to as pulsed events. The beamintensity was higher than the typical ion intensities foundnear the equatorial plane, but as in most of the other suchbeams contained no signature in oxygen. In this case, at thetime of its second pulse the angle of the beam in spacecraftcoordinates allowed it to enter one of the three CHEMSsensor telescopes, so we were able to determine its compo-sition and charge state. In Figure 6 (top), we show data thatrepresents the composition found in the magnetosphericplasma just prior to the appearance of the beam at theCHEMS sensor, which, because of its position on the space-craft and the spacecraft attitude, only detects the beam at itssecond appearance between 0853 and 0900 UT. This data istaken from the full energy sweep of the CHEMS electrostaticanalyzer, which covers the range 3 keV/q < E/q < 230 keV/q.The composition and charge states are typical of the energeticmagnetospheric plasma, with protons and oxygen dominat-ing, and significant H2

+, He+ and He++ included. In Figure 6(bottom), He and O are only present in trace quantities, andthe beam is dominated by protons and H3

+, typical of iono-spheric composition.[12] Next we investigate an event on day 284, 2006, that

unlike the one on day 269, 2006, but similar to that on day269, 2008, remains steady over periods of an hour or more,and continues to reappear for over 7 h. Figure 7 shows theion beams associated with this event. Again, as in mostevents, there is no evidence for heavier ions (e.g., oxygen

Figure 2. Energetic ions for the same period as Figure 1. The INCA sensor measures proton intensity in16 � 16 pixel maps covering its 90� � 120� field of view (known as angle-angle plots). Two of its eightenergy channels are shown, with pitch angle contours overlain. Each time pitch angles close to zero degreesare covered, high-intensity field-aligned ions appear at energies between 30 and 200 keV.


4 of 10

A02212

Figure 3. Pulsed particle acceleration events, day 269, 2006. First panel shows energetic electronenhancements, 200 keV to 1MeV fromMIMI LEMMS. Second panel shows time versus average intensityover the INCA FOV for eight hydrogen energy channels. Lower energies are neutrals, but sufficientlyenergetic ions enter through the INCA ion rejection plates. Angular plots of ion intensity with pitch anglecontours appear beneath each peak in intensity. The magnetic field (third panel) shows sharp angularvariations associated with each electron and ion enhancement. Broad band electromagnetic (below whiteline, fce) and electrostatic noise (above fce) (from RPWS, fourth panel) also aligns with these events.


5 of 10

A02212

or water products) in these beams. They are consistent withhydrogen, or possibly H2 or H3. As before, they appear asfield aligned, upward moving beams.[13] Just prior to the event shown in Figure 7, INCAwas in

neutral imaging mode, and at relatively high (southern)latitude (�34�) and low altitude (9.3 RS). From this vantage,a localized region of strong ENA emission appears just belowthe south polar cap. The image in Figure 8a shows thisemission, which like the proton beams has no correspondingemission in oxygen. In Figure 8b when INCA returned to

ENA imaging mode after the ion measurements shown inFigure 7, this locus of bright hydrogen emission from thesouth polar region once again appears, again absent inoxygen. In both Figures 8a and 8b, the usual emission fromthe ring current beyond the outer edge of the E ring can beseen (cut off by the edges of the INCA FOV).We suggest thatthe hydrogen emission is generated as protons are acceleratedat low altitude above the auroral zone by wave particle inter-action, generating ion conics (the same mechanism invokedby Carlson et al. [1998]). INCA images this emissionintermittently, but only when Cassini is located at latitudesconsistent with the pitch angles of the protons as they chargeexchange in Saturn’s exospheric hydrogen and convert toneutral atoms. The emission is sufficiently weak and local-ized that we can only detect it when Cassini is at relativelylow altitude, as in these cases.

3. Discussion

[14] The events presented in Figures 1–7 are similar inmost respects with the events analyzed by Carlson et al.[1998], which they attributed to perpendicular wave-particleacceleration and electrostatic confinement of ions and up-ward field aligned acceleration of electrons in downwardfield-aligned auroral currents. The whistler mode emissionsshown are consistent with VLF saucers found in the FASTobservations in downward field aligned current regimes eventhough the Cassini observations to date do not have sufficientresolution to show the characteristic saucer-shaped dynamicspectrum. Although the magnetic field strength in the auroralzone is similar for Earth and Saturn, the distance scale as

Figure 4. Antiplanetward field-aligned ion conic and bidirectional electron beam. Scale for the ionintensities is identical for all energies except for first and sixth rows. Ion conics are clearly indicated athigher energies by the intensity depression about the field direction (180� pitch angle). Energetic electronsare clearly bidirectional field aligned beams, with about a factor of 100 variation in intensity with pitchangle.

Figure 5. Very intense upward field-aligned ion beam,day 291, 2008.


6 of 10

A02212

measured by planetary radius at Saturn is roughly 10 timesthat of Earth. The magnitude of the total potential drop alongan auroral field line apparently depends on something like thesquare of the distance scale, as the observed peak energies ofthe accelerated ions and electrons produced by this mecha-nism are roughly 100 times higher at Saturn than at Earth.At Earth, the upgoing electron beams have a broad peak inenergy flux with an upper limit in the vicinity of 5 keV, atSaturn that upper limit is as high as >500 keV. At Earth, theion conics have peak energies around 100 eV to 2 keV, atSaturn the ion beams and conics have peak energies from 20to >200 keV. And while the electron differential energy fluxin the events studied by Carlson et al. [1998] is an order ofmagnitude higher than that in the events at Saturn (consistentwith higher electron densities in the events at Earth), the totalenergy flux integrated over energy for the electron beams atSaturn exceeds that at Earth by about one order of magnitude.

[15] A suggestion for understanding the apparent depen-dence on distance scale might be the scale of the magneto-sphere as characterized by the standoff distance of themagnetopause, which for Saturn relative to Earth is about afactor of 20. With the similar values of the magnetic field atthe surface and at the equatorial plane for a given L value, themapping of potential structures driven by convective motionin the equatorial plane at Saturn will be similar to Earth(although the latitude or L value at which the magnetopauseboundary maps to the ionosphere is generally higher atSaturn). The convective velocities are likely to be similar atboth planets (ranging from solar wind velocities to corota-tional velocities, and down to much weaker radial velocitiesof 10 or less km/s in the inner and middle magnetosphere),and the resulting electric field magnitudes in the equatorialplane will be similar to those at Earth. If the scale size of aconvected structure scales with the size of the system, thenthe total potential drop across such a structure (a Kelvin-Helmholtz eddy, a reconnection flow at the magnetopauseor in the tail, a flux tube interchange motion) will also scaleas the flux tube cross-sectional area, or roughly as the ratioof the standoff distance squared. This would yield a factor of400 rather than the factor of �100 observed in the energy ofthe particles, but the additional factor of 4 could easily beaccommodated in considering the differences in the magneticfield strength at the two planets, the differences in the mag-nitudes of the shear flows that generate the electric fields, thedifferences in the coupling with the ionosphere, etc.[16] Because the MIMI instrument depends on spacecraft

attitude for pitch angle coverage with the LEMMS sensor,and because only the INCA sensor has sufficient sensitivityto measure the associated energetic ion beams and conics (thebackground intensities in the LEMMS ion solid state detec-tors are too high in all but the most intense events), it issomewhat unusual that both populations can be measuredduring the same event. The LEMMS telescopes are mountedperpendicular to the INCA sensor, so they view differentpositions in the sky at all times. Auroral particle events forwhich we have measurements of both species are either likethat in Figures 1 and 2, where the spacecraft attitude varies soas to alternately place field-aligned particles in LEMMS andINCA, or like that in Figure 3, where LEMMS does notmeasure the field-aligned electrons, but rather electrons thatare apparently scattered out of the primary electron beam intolarger pitch angles. The latter, being weak and more difficultto identify, are less likely to be recognized as auroral events.[17] These complications in the identification of auroral

particle events make a statistical analysis of their frequencyof occurrence difficult. Nevertheless, we have been able todraw some general observations regarding their locations.[18] 1. Neither the ion nor the electron beams have been

observed for Ldipole < 9. Whether this lower limit in L isbecause the isotropic ion intensities exceed the beam andconic intensities, or the beams and conics are truly absent,remains to be determined.[19] 2. The electron beams may be either unidirectional

upward, or bidirectional (as in the work by Saur et al. [2006]),even for large L (as in Figure 4). The existence of the ionconics and electron beams at large L suggests active auro-ral processes at those locations. Given that some of theseevents (such as that shown in Figure 4) are accompanied bybidirectional electrons, we conclude that this subset at least

Figure 6. (top) Composition data from the MIMI CHEMSsensor measured prior to the ion beam shown in Figure 5.This is typical of Saturn’s magnetosphere, a mix of protons,He+, He2+, H2

+, O+, and little or no H3+. (bottom) Similar

data measured during the intense beam in Figure 5. The beamis primarily protons and H3

+, with only traces of He2+ and O+

and relatively little H2+.


7 of 10

A02212

remain on topologically closed magnetic fields. We havenever observed a unidirectional downward electron beam,nor have we observed an electron conic that was identifiablewith this class of events (if, as we conclude, these are similarto the events at Earth, electron conics are not anticipated).

[20] 3. The ion beams and conics are exclusively upward.[21] To date, these observations have taken place at alti-

tudes of�5 RS and higher. The altitude at which the particlesare accelerated is likely to be much closer to the ionosphere,as the reason for the development of the field aligned

Figure 7. Long-duration field-aligned ion event. Peak field-aligned intensity is mostly above 50 keVandoften absent below 20 keV. This event continued, waxing and waning, for at least 9 h as Cassini’s inboundorbit remained within 13 < L < 14.

Figure 8. (a) Energetic neutral atom image of intense hydrogen emission from Saturn’s south polarregion, just prior to the event shown in Figure 7. (b) ENA image of south polar hydrogen emissionimmediately following the end of the event in Figure 7. The orbits of Dione, Rhea, and Titan are includedas a reference. Axes corresponding to a fixed solar frame are shown in white (Z parallel to Saturn’s spin axis,X in the spin axis/Sun plane, pointed generally sunward, andY completing the system). The violet axes alsohave Z aligned with the spin axis, but they rotate with the SLS3 coordinate system [Kurth et al., 2008].


8 of 10

A02212

potentials responsible for the acceleration of the electronbeams is that themagnetospheric circuit is demanding currentthrough what would otherwise be a vacuum region. Theionospheric electrons are the only large reservoir available,but they cannot supply the current without an accompanyingion population that can maintain charge quasi-neutrality, andthe ionospheric ions are cold and gravitationally bound. Inthe model discussed by many of the investigators of similarevents at Earth [e.g., Ergun et al., 1998;Carlson et al., 1998],a few ions at sufficiently high altitude are acceleratedperpendicular to the field by the observed perpendicularstochastic electric field fluctuations and move upward alongthe field aided by the mirror force. A similar density of elec-trons can accompany those ions, and the electrostatic shockstructure responsible for the field aligned potential energizesthose electrons, until their motion can supply the currentrequired by the system. Of course, that same structure con-fines the ion conics, so that the ions cannot escape theiracceleration region until their parallel energy exceeds theparallel potential that confines them. This confinement of theions by the field aligned potential while they are acceleratedperpendicular to the field by the stochastic wavefield hasbeen called the ‘‘pressure cooker’’ mechanism at Earth[Gorney et al., 1985]. As discussed by Carlson et al. [1998],this results in a characteristic lower limit to the ion conicenergy, and should also result in a larger cone opening anglewith increasing energy. In Saturn’s case, the low atmosphericand ionospheric scale heights presumably mean that the ionacceleration process takes place in regions of lower iondensity than the equivalent regions at Earth; so higherpotentials are required to provide enough current from theconsequently lower density of electrons permitted by quasi-neutrality, hence the much higher energies developed in thesepopulations for downward current, ‘‘black auroral’’ regionsat Saturn. At the altitudes where most of the ion events havebeen measured, the conic opening angle is usually too narrowto resolve, but in the event shown in Figure 4 where it isresolved, the energy dependence of the ion conic openingangle does follow this trend.[22] The fact that some of the events are pulsed (coming

and going with a repetition period of roughly 1 h) whileothers are relatively steady may point to two somewhatdifferent mechanisms (although the basic ‘‘pressure cooker’’mechanism remains the same). In one case it may be verymuch like those at Earth, where relatively steady downwardfield aligned current structures continually accelerate par-ticles for hours, while in the pulsed case, these emissionsmay be linked to the ubiquitous flux tube interchange that hasbeen reported throughout the magnetosphere from roughly6 < L < 12 [e.g., Hill et al., 2005; Burch et al., 2005; Leisneret al., 2005].[23] The ENA images of the ion conic generation region

further confirm that the process is taking place relativelyclose to the auroral zone. The characteristic ENA emissionfrom this region (Figures 8a and 8b) cannot constrain thelocation precisely, because the instrument angular resolutionis insufficient at the distances obtained to date, because theneutral emission decreases with altitude owing to decreasingSaturn exospheric gas density that serves as a charge ex-change medium, and because as the conic angle collapsesabout the field with increasing altitude, the trajectories of anyENA produced would no longer intersect the spacecraft.

However, it is clear from these images that the light ionsreach energies of at least 100 keV in a region not more than1 RS above the surface (there is no indication of a heavy ioncomponent, such as water products, methane or nitrogen ineither the ion conics or the ENA emission).

4. Conclusions

[24] We have found strong evidence for ion conic andelectron beam acceleration in Saturn’s auroral zone, associ-ated with downward field aligned electric currents (regionsidentified as black auroras at Earth, and almost always inclose proximity to bright, upward current regions in the auro-ral zone).[25] These come in two distinct types: the first are �1 h

cadence pulsed events clearly associated with transientdownward field aligned current structures and pulses ofwhistler mode noise. The second are relatively steady events,lasting an hour to several hours, also associated with whistlermode noise. The sense of the current associated with thesesteady events in those cases where we have been able toidentify it is downward, and their characteristics are nearlyidentical to those found by various groups at Earth where aclear association with downward field aligned current regionsis shown. Both types of events seem to occur at any locationwith Ldipole > 9 or 10, up to at least L = 70. The region insideL = 10 is typically characterized by fairly intense nearlyisotropic energetic ion and electron intensities, so the cutoff atabout L = 10 may simply reflect masking of the ion events bythe higher ambient particle populations, rather than cessationof the generation mechanism. A separate statistical study oftheir occurrence locations is under way, to identify any trendswith local time or SLS3 longitude.[26] While similar in most respects with the events iden-

tified with ‘‘black’’ aurora at Earth, the characteristic energiesof the particles in the Saturnian version of these events is�100 times larger than those at Earth, and the energy flux isroughly 10 times higher at Saturn than at Earth.[27] Given the range in Ldipole over which these events are

seen, and accepting both their identification with downwardcurrent associated with auroral processes and the tenet thatdownward current regions are likely to lie closely adjacentto upward current, bright aurora regions, we conclude thatSaturn’s auroral zone subtends a region from well equator-ward of the open-closed field boundary to possibly somewhatpoleward of that boundary (although events on demonstrablyclosed field lines are see for Ldipole as large as 52).

[28] Acknowledgments. We thank one of the reviewers for introduc-ing the idea that the higher energies of particles at Saturn may be related tothe scale of the cross-sectional area of magnetic flux tubes in regions ofshear flows and both reviewers for constructive reviews. This research wassupported in part by the NASA Office of Space Science under task order003 of contract NAS5-97271 between NASA Goddard Space Flight Centerand the Johns Hopkins University. The research at the University of Iowa issupported by NASA through contract 1279973 with the Jet PropulsionLaboratory.[29] Wolfgang Baumjohann thanks David Klumpar and another reviewer

for their assistance in evaluating this paper.

ReferencesBurch, J. L., J. Goldstein, T. W. Hill, D. T. Young, F. J. Crary, A. J. Coates,N. Andre, W. S. Kurth, and E. C. Sittler Jr. (2005), Properties of localplasma injections in Saturn’s magnetosphere, Geophys. Res. Lett., 32,L14S02, doi:10.1029/2005GL022611.


9 of 10

A02212

Carlson, C. W., et al. (1998), FAST observations in the downward auroralcurrent regions: Energetic upgoing electron beams, parallel potential drops,and ion heating, Geophys. Res. Lett., 25, 2017–2020, doi:10.1029/98GL00851.

Dougherty, M. K., et al. (2004), The Cassini magnetic field investigation,Space Sci. Rev., 114, 331–383, doi:10.1007/s11214-004-1432-2.

Ergun, R. E., et al. (1998), FAST satellite observations of electric fieldstructures in the auroral zone, Geophys. Res. Lett., 25, 2025–2028,doi:10.1029/98GL00635.

Gorney, D. J., Y. T. Chiu, and D. R. Croley (1985), Trapping of ion conicsby downward parallel electric fields, J. Geophys. Res., 90, 4205–4210,doi:10.1029/JA090iA05p04205.

Gurnett, D. A., et al. (2004), The Cassini radio and plasma wave investiga-tion, Space Sci. Rev., 114, 395–463, doi:10.1007/s11214-004-1434-0.

Hill, T. W., A. M. Rymer, J. L. Burch, F. J. Crary, D. T. Young, M. F.Thomsen, D. Delapp, N. Andre, A. J. Coates, and G. R. Lewis (2005),Evidence for rotationally driven plasma transport in Saturn’s magneto-sphere, Geophys. Res. Lett., 32, L14S10, doi:10.1029/2005GL022620.

Klumpar, D. M. (1990), Near equatorial signatures of dynamic auroral pro-cesses, in Physics of Space Plasmas, pp. 265–276, Sci. Publ., Cambridge,Mass.

Krimigis, S. M., et al. (2004), Magnetospheric imaging instrument (MIMI)on the Cassini mission to Saturn/Titan, Space Sci. Rev., 114, 233–329,doi:10.1007/s11214-004-1410-8.

Kurth, W. S., T. F. Averkamp, D. A. Gurnett, J. B. Groene, and A. Lecacheux(2008), An update to a Saturnian longitude system based on kilometricradio emissions, J. Geophys. Res., 113, A05222, doi:10.1029/2007JA012861.

Leisner, J. S., C. T. Russell, K. K. Khurana, M. K. Dougherty, andM. Andre (2005), Warm flux tubes in the E-ring plasma torus: InitialCassini magnetometer observations, Geophys. Res. Lett., 32, L14S08,doi:10.1029/2005GL022652.

Marklund, G. T., et al. (2001), Temporal evolution of the electric field accel-erating electrons away from the auroral ionosphere,Nature, 414, 724–727,doi:10.1038/414724a.

Mauk, B. H., and J. Saur (2007), Equatorial electron beams and auroralstructuring at Jupiter, J. Geophys. Res., 112, A10221, doi:10.1029/2007JA012370.

Saur, J., et al. (2006), Anti-planetward auroral electron beams at Saturn,Nature, 439, 699–702, doi:10.1038/nature04401.

Young, D. T., et al. (2005), Composition and dynamics of plasma in Saturn’smagnetosphere, Science, 307(5713), 1262–1266.

��J. F. Carbary, S. M. Krimigis, B. H. Mauk, and D. G. Mitchell, Johns

Hopkins University Applied Physics Laboratory, Laurel, MD 20723, USA.([email protected])M. K. Dougherty, Blackett Laboratory, Imperial College, London SW7

2AZ, UK.D. C. Hamilton, Department of Astronomy, University of Maryland,

College Park, MD 20742, USA.G. B. Hospodarsky and W. S. Kurth, Department of Physics and

Astronomy, University of Iowa, Iowa City, IA 52242, USA.N. Krupp, Max-Planck-Institut fur Sonnensystemforschung, D-37191

Katlenburg-Lindau, Germany.J. Saur, Institut fur Geophysik and Meteorologie, Universitat zu Koln,

D-50937 Cologne, Germany.


10 of 10

A02212

Documents

Ion conics and electron beams associated with … › public › mkivelso › refs › PUBLICATIONS...going energetic electron beams, upward going ion conics, and enhanced broadband