observations of thermal plasmas in jupiter’s magnetotail · 2007. 12. 13. · received 8 march...

Observations of thermal plasmas in Jupiter’s magnetotail

L. A. Frank and W. R. PatersonDepartment of Physics and Astronomy, University of Iowa, Iowa City, Iowa, USA

K. K. KhuranaInstitute of Geophysics and Planetary Physics, University of California, Los Angeles, California, USA

Received 8 March 2001; revised 16 August 2001; accepted 16 August 2001; published 3 January 2002.

[1] A survey of thermal plasmas and magnetic fields is presented for the orbit of the Galileospacecraft around Jupiter that occurred during the period May 4 through June 22, 1997. This orbittraversed the magnetotail out to Jovian radial distances of 100.2 RJ in the magnetotail. Perijove waspositioned at 9.3 RJ. Three primary ion populations were detected with the plasma analyzer: coolhydrogen ions with temperatures of 10 eV, hot hydrogen ions with temperatures of �10 keV, and athird population of heavy ions such as O++, O+, S++, and S+++ with temperatures in the range of 500eV. Plasma flows near perijove were in the corotational direction but with speeds �60% of those forrigid corotation with the planet out to radial distances of �18 RJ. In the radial range of 18–26 RJ

there were significant radial components for the bulk flows, and the flow components in thecorotational direction reached values expected for rigid corotation when the current sheet wascrossed. The transient character of the plasma parameters suggests that strong ion plasmaacceleration is occurring in this region. The temperatures of the heavy ions increased from 5 �106 K at 9.3 RJ to �108 K at 26 RJ. At distances <20 RJ there is a strong dependence of iontemperatures on System III longitude. The scalar magnetic field outside of the current sheet in theradial distance range 9.3–20 RJ varied as R�2.78 and similar to that for a dipole field, and atdistances >50 RJ it varied as R�1.19. The thermal plasma pressure in the current sheet is a factor of�10 less than the magnetic pressure at 9.3 RJ at positions above or below the sheet but becomesequal to this magnetic pressure at radial distances >30 RJ. The corresponding values of the ratio ofthe plasma to magnetic pressure, b, are in the range of 10–100 in the current sheet. The numberdensities and temperatures of these plasmas are 0.05–0.1 /cm3 and 0.5–1 � 108 K, respectively. Inthe magnetotail the bulk flows of the thermal plasmas exhibit substantial components in thecorotational and radially outward directions, but the bulk speeds of 50–200 km s�1 are significantlyless than those for rigid corotation. For this single orbit the average bulk flows were �50 km s�1 inthe premidnight sector and 200 km s�1 in the early morning sector at radial distances >�50 RJ. Atapojove of 100 RJ an antisunward flow of �200 km s�1 is found that is supportive of themagnetospheric ‘‘wind’’ reported for Voyager measurements of energetic charged particles. The 10-hour periodicity of the pairs of current sheet crossings at the position of the Galileo spacecraftincludes a variety of dynamical signatures, which are suggested to be due to the changes indirection and pressures in the solar wind and due to the transient acceleration of plasmas in thecurrent sheet. INDEX TERMS: 6220 Planetology: Solar System Objects: Jupiter, 5737Planetology: Fluid Planets: Magnetospheres (2756), 2756 Magnetospheric Physics: Planetarymagnetospheres (5443, 5737, 6030), 2744 Magnetospheric Physics: Magnetotail

1. Introduction

[2] Observations of magnetic fields and charged particles havebeen previously acquired during the flybys of Jupiter with Pioneer10 and 11, Voyager 1 and 2, and the Ulysses spacecraft. TheGalileo spacecraft is currently in orbit around this giant planet andcontinues to return fascinating measurements of plasma phenom-ena in various regions of its magnetosphere and near its moons. Itis the thermal plasmas in the Jovian magnetotail that are theprimary interest of this paper.[3] The first in situ observations of magnetic fields and charged

particles were gained with Pioneer 10 (arrival year: 1973) andPioneer 11 (1974). These measurements showed that Jupiter’smagnetosphere is the largest entity within the heliosphere and is

populated with extremely high intensities of charged particles. Thedynamics of this magnetosphere is dominated by its size and itsrapid 10-hour rotational period. Unlike Earth’s magnetosphere,which is influenced in greatest extent by the solar wind, the Jovianmagnetosphere is distorted into a disk by the forces associated withthe high internal rotation speeds. Overviews of these early findingsare offered by Smith et al. [1976] for the magnetic fields and byVan Allen [1976] and by McDonald and Trainor [1976] for theenergetic charged particle distributions surrounding this planet.[4] The second wave of exploration of the Jovian environment

was provided by Voyager 1 and 2 (both in 1979). Like thetrajectories of the Pioneer spacecraft, the Voyager spacecraftentered the Jovian magnetosphere on the dayside, passed nearJupiter in its dusk sector, and exited during local morning. The insitu plasma measurements during the inbound segment of thetrajectory, together with the remote spectroscopic observations,firmly established that Jupiter was encompassed by a plasma torus,

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 107, NO. A1, 1003, 10.1029/2001JA000077, 2002

Copyright 2002 by the American Geophysical Union.0148-0227/02/2001JA000077$09.00

SIA 1 - 1

which is fed by gases from the volcanoes of Io. These measure-ments of the plasma torus have been reviewed by Bagenal [1994].An overview of the plasma observations during the entire Voyagerencounters is offered by Belcher [1983], and an overview for themore energetic charged particles is offered by Krimigis et al.[1981] and Schardt and Goertz [1983]. A companion survey ofthe magnetic fields along the Voyager trajectories is given byAcuna et al. [1983]. At this point it is important to note that thermalplasma observations in the magnetotail were not acquired by thePioneer and Voyager spacecraft. The fields of view of these plasmaanalyzers were directed toward intercepting the solar wind, theirprincipal objectives during their journeys into the outer solarsystem. For this reason, the plasma instruments were generallyviewing in directions away from that of the corotating plasmaswhen the spacecraft exited through Jupiter’s magnetotail. It is themeasurements of such thermal plasmas that are a primary objectiveof the Galileo Mission.[5] One further flyby of Jupiter occurred before the arrival

of the Galileo spacecraft, that of Ulysses (1992), in order togreatly increase its heliocentric orbital inclination for theensuing passages over the Sun’s poles at lesser distances. Thusthe Ulysses inbound trajectory was similar to that of thePioneer and Voyager spacecraft, but the outbound segmentpassed at high magnetic latitudes through the Jovian dusksidemagnetosphere. Thermal plasma measurements were also takenduring this encounter. A mixture of solar wind ions and heavyions from Io were dispersed throughout the magnetosphere[Geiss et al., 1992], and large intensities of low-energyelectrons were detected at the current sheet crossings [Bameet al., 1992]. Two energetic particle instrumentations, at acombined range of �0.1–1 MeV, were used to infer convec-

tion speeds of the thermal plasmas as generally being near orbelow those expected for rigid corotation for the inbound passand were used to demonstrate the existence of field-alignedbeams during the high-latitude outbound passage [Lanzerotti etal., 1993; Cowley et al., 1996; Seidel et al., 1997; Krupp et al.,1997].[6] The Galileo spacecraft arrived at Jupiter in December of

1995 to begin its marvelous orbital tour of this giant planet.Among the fascinating discoveries are a ‘‘minimagnetosphere’’at Ganymede [Kivelson et al., 1998], an ionosphere at Callisto[Gurnett et al., 2000], intense electron beams at Io [Williams etal., 1996], and direct detection of ions in Io’s ionosphere [Frankand Paterson, 2001a]. Of present direct interest are the topologyand dynamics of the plasma torus and its extension into theplasma sheet. Energetic particle bursts that are reminiscent ofsubstorm phenomena in Earth’s magnetosphere are detected[Mauk et al., 1997, 1999; Krupp et al., 1998]. Temporalvariations in Jupiter’s magnetosphere that exhibit quasiperiodsof several days are also found [Vasyliunas et al., 1997; Woch etal., 1998], but their cause is not yet identified. Russell et al.[1998] report the existence of patches of strongly northward andsouthward magnetic fields at distances >50 RJ in the magneto-tail, which may be the signatures of magnetic reconnectionwhich are convecting tailward. Frank and Paterson [2001b]report in situ measurements of thermal plasmas in the torus thatshow that there is an ‘‘active sector’’ with greater latitudinalextent that is corotating with Jupiter. These investigationsrepresent only the initial, as yet seemingly disconnected ingre-dients for a comprehensive understanding of Jupiter’s remarkable

Figure 1. The orbit of Galileo in Jupiter-centered solar-eclipticcoordinates during May 4 to June 22, 1997. This orbit is projectedonto the (a) X-Y plane and (b) X-Z plane.

Figure 2. E/Q spectrum of positive ions in the direction 13�relative to the plasma bulk velocity at 1413 UT on May 6, 1997.This spectrum is fitted with the velocity distributions of three M/Qspecies of heavy ions, together with that for hydrogen ions asidentified in the legend.

SIA 1 - 2 FRANK ET AL.: THERMAL PLASMAS IN JUPITER’S MAGNETOTAIL

magnetosphere. The present report provides an overview of thefirst measurements of thermal plasmas in Jupiter’s magnetotail.

2. Instrumentation

[7] The Galileo fluxgate magnetometer (MAG) comprises twoboom-mounted three-axis sensor assemblies. The outboard sensoris positioned at 11.03 m from the spacecraft spin axis, and theinboard sensor is positioned at 6.87 m from this axis. Thismagnetometer is described in detail by Kivelson et al. [1992].The dynamic ranges of the outboard magnetometer are ±32 and±512 nT. These ranges are ±512 and ±16,382 nT for the inboardsensor. Flipper mechanisms are included in the sensor assembliesfor calibration of the offsets. Magnetic vectors averaged over 1-minintervals are used in the present investigation.[8] The plasma instrumentation (PLS) on board the Galileo

spacecraft is composed of spherical-segment electrostatic ana-lyzers, which are capable of measurements of the positive ionand electron velocity distributions over the energy/charge (E/Q)range of 0.9 V to 52 kV in 64 passband steps. Three miniaturemagnetic spectrometers are also positioned at the exit apertures ofthe ion electrostatic analyzers in order to determine the mass/charge (M/Q) of these ions. The PLS has been previously describedby Frank et al. [1992]. The fan-shaped field of view of theelectrostatic analyzers is divided into seven segments with multiplesensors. Electronic sectoring of the responses of the sensorsaccording to the phase of the spacecraft rotation then allows thethree-dimensional determination of the velocity distributions of thepositive ion and electron plasmas, i.e., coverage of 80% of the4p-sr solid angle for arrival directions of charged particles at theanalyzers. Most of the coverage of the orbit shown in Figure 1 wasacquired at the low-rate telemetry rates, which were transmitted inreal time to Earth. These are known as RTS modes. There were twosuch modes used during this orbit. The most frequently used modewas RTS0 for which seven sensors were sampled in four azimuthalspin sectors in 14 E/Q passbands for each of the positive ion andelectron velocity distributions. Every fourth passband in the pass-band step range 11 through 63 was selected for an E/Q range of 7 Vto 53 kV. The total number of samples was thus 392. The timeresolution for these measurements varied according to the availablespacecraft telemetry rate. The slowest time resolution was 12 min.The second RTS mode, RTS4, sampled the 7 sensors in 8azimuthal sectors and 16 E/Q passbands for a total of 896 samplesof the velocity distributions. The total amount of data recorded inthis mode was �24 hours during this orbit. Every third passbandwas sampled in the step range of 18 to 63, corresponding to an E/Qrange of 24 V to 53 kV. A high-rate segment of data was providedby the tape recorder during 1300–1525 UT on May 6, 1997. Forthese high-resolution data the velocity distributions of the positiveions and the electrons were each acquired for 7 sensors, 8azimuthal sectors, and 24 E/Q passbands spanning the E/Q rangeof 8 V to 52 kV, all in 60 s.

3. Observations

[9] The plasma and magnetic field measurements for Galileoorbit G08 during May 4 through June 22, 1997, are presented here.

The G08 identifies the close-targeted flyby of one of Jupiter’smoons, G for Ganymede in this case, during the eighth orbit of thespacecraft around Jupiter. This orbit in Jupiter-centered solar-ecliptic coordinates is shown in Figure 1 and was positioned nearthe ecliptic plane. Apojove was at a Jovian radial distance of 100.2RJ (Jupiter radii) and local time 0121 on June 2 with perijove at 9.3RJ on May 8. Each day is denoted by a dot in Figure 1, and the totalduration of this orbit was �50 days. During an interval of �2 dayson May 4 and 5 there were no measurements of the magneticfields. The rest of the orbit was almost completely covered withreal-time telemetry transmitted to Earth from the spacecraft.[10] In order to compute the ion plasma moments it is necessary

to determine their mass/charge (M/Q). Then these moments can becalculated from the measured three-dimensional velocity distribu-tions of the positive ions [Krall and Trivelpiece, 1973]. The M/Qspectra of the positive ions could not be directly determined withmass spectrometers because (1) their fields of view were not indirections of the ion bulk flows when the ion densities were abovetheir thresholds during a several-day period centered on perijove or(2) the densities were below the spectrometer thresholds for the restof the orbit. However, the responses to hydrogen ions with theenergy/charge (E/Q) analyzers allowed a sufficient determinationof the M/Q for the heavy ions.[11] An example of such a measurement at 1413 UT on May 6

is shown in Figure 2, which was gained with the use of the taperecorder. The ion sensor is P1 in sector 8 with its center of field ofview at 13� with respect to the bulk flow velocity of the plasma.This E/Qspectrum exhibits a strong maximum at �500 V. Theheavy ions are in the E/Q range of �2 to 20 kV with a peak at �8kV. The reader will probably note that the ratio of this peak energyto that of hydrogen is 16, which identifies the M/Q of the heavy

Table 1. Comparison of Ion Plasma Parameters for Tape-

Recorded Data and its Decimation for 1413 UT, May 6, 1997

Tape-Recorded Decimation

Density N, cm�3 0.12 0.12Temperature T, K 1.3 � 107 1.4 � 107

Bulk speed V, km s�1 297 281Latitude for bulk flow, q, deg 83 90Longitude for bulk flow, f, deg 331 332

Figure 3. E/Q spectra of positive ions as observed in the directionof ion bulk flow (sensor P1) and in the direction opposite to thebulk flow (sensor P7) during 1435–1450 UT on May 6, 1997.

FRANK ET AL.: THERMAL PLASMAS IN JUPITER’S MAGNETOTAIL SIA 1 - 3

ions in the range of 16. The bulk flow velocity is computed withthe assumption that the M/Q of the heavy ions is 16. Thecomponents of this bulk flow velocity in spacecraft-centeredsolar-ecliptic coordinates are (Vx, Vy, Vz) = (273, �143, 15 kms�1). A good fit to the E/Q spectrum is shown in Figure 2 with ahydrogen density of 0.05 cm�3 and a temperature of 20 eV.Because the angular distributions of the convecting plasma arefrequently similar to or smaller than the analyzer fields of view, thefits to the E/Q spectra are accomplished by dividing the analyzerfields of view into sets of smaller elements [Frank and Paterson,2001a]. The number of such elements varied from 77 to 143,dependent upon the ion sensor. The fit in Figure 2 exhibits threeheavy ion components with M/Q = 8 with 0.02 cm�3, 10.7 with0.02 cm�3, and 16 with 0.05 cm�3, all with a temperature of 500eV. These three ion groups presumably are O++ at 8, S+++ at 10.7,and O+ and S++ at 16. The above E/Q spectrum was obtained at aJovian radial distance of 25.3 RJ and a local time of 0558. The

composition is not grossly dissimilar to that reported previously forVoyager 1 observations at a radial distance of 19.8 RJ in thedayside magnetosphere [McNutt et al., 1981].[12] For the survey of measurements reported here a M/Q= 16

is used in the computation of the plasma moments. This valueintroduces a tolerable uncertainty in the ion densities and bulkflow velocities. Hydrogen ions are excluded from the momentscalculation by setting a lower E/Q threshold of 2 kV. The hydrogenions are excluded from the moments calculation because their M/Qis much less than that of the heavy ions. The bulk flow velocitycan be computed for the hydrogen ions and is in agreement withthat independently determined for the heavy ions. If the hydrogenion energies are lower, then the effects of spacecraft charging mustbe considered [Frank and Paterson, 1999]. The ion densities varyas (M/Q)1/2, and the bulk flow speed varies as (M/Q)�1/2. Theaverage M/Q for the E/Q spectrum shown in Figure 2 is 13.0. Thusthe errors due to the assumed value of 16 are �0.01 cm�3at a

Figure 4. Survey plot for plasma parameters measured during the period May 5 to June 22, 1997: (a) ion densities,(b) ion temperatures, (c–e) ion bulk flows in solar-ecliptic coordinates, (f ) ion bulk flow component in the radialdirection, (g) ion bulk flow component in the corotational direction, and (h) the bulk flow speed. These plasmamoments are given for the crossings of the current sheet as determined with the simultaneous measurements of themagnetic fields. The jovicentric radial distance and local time of the spacecraft position are given along the abscissa.


density of 0.09 cm�3 and +35 km s�1 at a bulk flow speed of 306km s�1.[13] The high-resolution measurements taken with the tape

recorder for Figure 2 also provide the opportunity to investigatethe effects of the decimation of the sampling of the velocitydistributions during the real-time telemetry periods. As notedpreviously, this real-time coverage provides the major fraction ofour survey. For the real-time data in mode RTS0 the sampling ofthe velocity distributions is reduced from eight to four azimuthalsectors around the spin axis and the number of E/Q passbands isdecimated such that alternate interleaving samples are eliminated.All of the ion sensors are retained. The effects of this decimationare shown in Table 1for the tape-recorded velocity distributiontaken at 1413 UT (Figure 2). The ion densities N, temperatures T,flow speed V, longitude f, and latitude q of the flow vector inecliptic coordinates are shown in Table 1. It is seen that theseparameters are robust to the decimation.[14] This is also an appropriate place in the text to discuss the

errors in the computation of temperatures due to errors in thedistribution of ions with different M/Q values. It has beenestablished from the analyses of Voyager 1 and Galileo plasmadata that the most abundant heavy ions are characterized with M/Qvalues of 8, 10.7, 16, and 32 [Bagenal, 1994; Crary et al., 1998;Frank and Paterson, 2000, 2001b]. The maximum error intemperature occurs when the densities of the two major M/Qspecies are about equal. Then the error is comparable to thedifference in their kinetic energies, i.e., �T ’ 1/2 (M2 � M1)V2/Q. As an example, for M2/Q = 16 and M1/Q = 8, with V = 200km s�1in the magnetotail �T = 2 � 107K (1.7 keV). Beyond aradial distance of �30 RJ the temperatures are >5 � 107 K, and theabove effect leads to an overestimate of the order of �50% or less.On the other hand, the maximum uncertainty in temperature isyielded by distributions, such as that in Figure 2, which have lowertemperatures. The moments computation gives a temperature of1.3 � 107 K (1.1 keV). The fit shown in Figure 2 for a flow speedof 300 km s�1 indicates that the temperatures of the three ions withM/Q = 8, 10.7, and 16 are �6 � 106 K (500 eV). Thus the possibleerror in the determination of the temperature is about a factor of 2if only a single ion is present. At lesser radial distances in theplasma torus the temperature errors due to a distribution of M/Qvalues are significantly less because ions with M/Q = 16 aredominant [Frank and Paterson, 2001b].[15] In addition to the thermal plasmas such as those shown in

Figure 2, there is also a hotter component that is detected in theE/Q range of tens of kV near the limit of the plasma analyzer.The presence of this hotter component is exhibited in Figure 3.

These measurements were also obtained with the tape recorderduring 1435–1450 UT on May 6 at a Jovian radial distance of25.2 RJ. Sensor P1 in its azimuthal sector 8 is viewing into theplasma bulk flow direction. The ion peak centered at �10 kV isassociated with the thermal plasma previously shown in Figure 2.On the other hand, sensor P7 in its sector 8 in Figure 3 isviewing hot ions in the direction opposite to that of the plasmabulk flow. The angular distributions of these ions are consistentwith those expected for hydrogen ions at a temperature of �20keV and bulk speeds and densities of 300 km s�1 and 0.05 cm�3,respectively. Hot hydrogen ions with similar characteristics havebeen previously detected with medium-energy charged particleinstrumentation on Voyager 1 [Krimigis et al., 1981]. These hotions contribute a substantial energy density in this region of themagnetosphere.[16] The survey of plasma parameters is given in Figure 4. It is

important to note that, for example, one ion density is plotted for asingle crossing of the current sheet in Figure 4a. This density is theaverage of the three determinations of densities that are nearest tothe current sheet crossing as determined by the reversal ofdirection of the radial component of the magnetic field. Thetemporal resolution of individual plasma measurements is deter-mined by the rate of the real-time telemetry stream from thespacecraft. This temporal resolution is typically in the range of 5–12 min. Inspection of Figure 4 shows that the telemetry coverageduring May 5 through June 22 was remarkably good. Perijove isevident in the density profile as the number densities reach amaximum of �100 cm�3. Beyond a radial distance of 50 RJ thedensities in the current sheet are typically in the range of 0.05–0.1cm�3. The plasma temperatures are at a minimum of �5 � 106 Kat perijove and increase to values in the range of 5 � 107to 108 Kbeyond 50 RJ.[17] Inspection of the temperature profiles in Figure 4b during

May 6–9 finds that the ion temperatures exhibit alternate high

Table 2. Ion Temperature and Density at Current Sheet Crossings Near Perijove

Date 1997 UniversalTime, UT

N, cm�3 T, K Radial Distance,RJ

Latitude, deg System IIILongitude, deg

Ion Temperature

May 6 0918 0.09 1.1 � 108 27.1 �0.9 291.1 highMay 6 1513 0.05 1.0 � 107 25.0 �0.8 141.5 lowMay 6 1925 0.13 8.6 � 107 23.4 �0.7 290.4 highMay 7 0034 0.51 2.0 � 107 21.4 �0.6 112.2 lowMay 7 0546 0.35 4.1 � 107 19.3 �0.4 294.8 highMay 7 1058 2.4 1.3 � 107 17.1 �0.2 115.8 lowMay 7 1521 2.1 1.9 � 107 15.2 0.0 266.7 highMay 7 2148 11 7.4 � 106 12.6 0.2 123.8 lowMay 8 0251 15 1.5 � 107 10.8 0.3 288.2 highMay 8 0830 120 4.3 � 106 9.5 0.4 105.1 lowMay 8 1423 47 8.5 � 106 9.4 0.4 284.7 highMay 8 2018 27 5.6 � 106 10.7 0.3 109.5 lowMay 9 0148 4.3 2.0 � 107 12.7 0.2 288.6 highMay 9 0712 2.8 6.8 � 106 14.9 0.1 110.1 lowMay 9 1224 0.77 3.7 � 107 17.1 0.0 288.4 highMay 9 1743 0.42 3.0 � 107 19.3 �0.1 113.2 lowMay 9 2248 0.13 8.9 � 107 21.5 �0.1 291.5 high

Table 3. Ion Plasma Parameters and Uncertainties Due to

Counting Statistics

May 8, 19970830 UT

May 30, 19971330 UT

Density N, cm�3 116 (±1) 0.044 (±0.005)Temperature T, K 4.3 (±0.07) � 106 4.6 (±0.8) � 107

Flow X component, Vx, km s�1 �11 (±0.5) �36 (±14)Flow Y component, Vy, km s�1 76 (±0.5) �90 (±14)Flow Z component, Vz, km s�1 �11 (±0.3) �15 (±12)


and low values for the crossings of the current sheet. An activesector for sulfur ion emissions was previously reported for theSystem III longitude range of �180� to 230� [Pilcher andMorgan, 1985]. In order to investigate the possibility that theion temperatures are correlated with System III longitude, the iontemperatures and this longitude are given in Table 2. A strongcorrelation of temperature with longitude is observed. The iontemperatures at longitudes �290� are considerably greater thanthose at �110�. The near-equatorial plane of the Galileo orbit andthe tilt of the magnetic dipole axis limit the ranges of longitudethat can be sampled even though the local time of the orbitchanges during the mission.[18] Figures 4c–4e display the components of ion bulk flow in

solar-ecliptic coordinates. The errors in densities and bulk flowsdue to counting statistics are given in Table 3 for two representativetimes. The Z components of the bulk flows in the current sheet aresmall and are representative of the statistical errors in the deter-mination of flow components. During the period of May 10–15 at

radial distances in the range of 20–30 RJ in the premidnight sector,there are large, unsteady flows of plasmas up to 400 km s�1 withtailward directions. As the spacecraft subsequently moves outwardfrom Jupiter, these flows become notably more quiescent. The Xcomponent generally is in the range of 0–100 km s�1 with twoexceptions. The first is the tailward flow of plasma at apojove atradial distances of �100 RJ. The second exception is the persistentcomponent of flow at 50–100 km s�1 as the spacecraft returns todistances nearer the planet during June 19–24. In Figure 4d it isapparent that the plasmas in the magnetotail exhibit a substantialnegative Y component of bulk flow. This is directed toward localmorning and is in the sense of corotation with the planet. This flowcomponent is in the range of �50 to �100 km s�1 during May 15–25 and increases in magnitude to �150 to �200 km s�1 by the endof the orbit. The scalar magnitude of bulk flows in the current sheetis given in Figure 4h.[19] The components of ion bulk flows radially outward from

Jupiter, Vr, and in the corotation direction, Vf, are shown in

Figure 5. Continuation of Figure 4: (a) ion density, (b) ion bulk flow in the X direction, (c) ion bulk flow in the Ydirection, (d–f ) components of the magnetic field in solar-ecliptic coordinates, (g) magnetic field component in theradial direction, (h) magnetic field component in the corotational direction, (i) scalar magnitude of the magnetic field,( j) ion and magnetic field pressure in units of pascals, and (k) ratio of ion and magnetic field pressures, b. Themagnetic fields are given for positions above or below the current sheet for which b < 0.1. Note that the componentsof the magnetic fields are given as scalar magnitudes on logarithmic scales.


Figures 4f and 4g, respectively. It is of considerable interest to notethat the bulk flow is directed along the corotational direction atperijove but that a substantial radial component appears at largerdistances of >15 RJ. This topic will be addressed further in a laterplot. The plasma flows during May 18 to June 1 are significantlyless than those for rigid corotation. During this period the radialcomponent averages to �0 km s�1, and the component in thedirection of corotation is �50 km s�1. In other words, the flow isquite stagnant in the premidnight sector. This condition changesafter the apojove tailward flows. During the June period when thespacecraft is in the early morning sector, the corotational compo-nent gradually increases to �200 km s�1 with a measurable radialcomponent of �50 km s�1 directed outward from Jupiter. Thefuture analysis of other orbits is necessary in order to establish thisremarkable local time effect as a durable feature of the magnetotail.[20] The magnetic field measurements with the magnetometer

(MAG) are summarized in Figure 5. For reference the ion

densities and bulk flows in the x and y directions are replicatedin Figures 5a–5c. The magnetic fields are given for positionsoutside of the current sheet. All magnetic field measurements forwhich the ratio of the plasma pressure to magnetic field pressure,b, is <0.1 are shown. These field measurements are averaged over1-min intervals. In order to show the full range of the magneticfields the absolute magnitudes are plotted on a logarithmic scale.At perijove the field is �400 nT and is characteristically 5–10 nTin the magnetotail. The magnetic field components are given insolar-ecliptic coordinates (Figures 5d–5f ) and in the radial andcorotational directions (Figures 5g–5h). Figure 5i for the scalarmagnetic fields shows that the magnitudes exhibit relatively smallfluctuations in the magnetotail. Such fluctuations are often thesignatures of dynamic variations in the solar wind pressures. Themagnetic and plasma pressures are shown in Figure 4j in units ofpascals. In the vicinity of perijove the magnetic pressure domi-nates over the plasma pressure by factors of �5, but for the

Figure 6. Continuation of Figure 4 with the components of the magnetic fields given in linear scales: (a) iondensity, (b) ion bulk flow in the X direction, (c) ion bulk flow in the Y direction, (d–f ) components of the magneticfields in solar-ecliptic coordinates, (g) magnetic field component in the radial direction, (h) magnetic field componentin the corotational direction, and (i) scalar value of the magnetic field. Note that values for the magnetic fields inside aradial distance of 30 RJ are off-scale and thus are not shown.


magnetotail these pressures are closely equal. This equality ofmagnetic pressures in the lobe outside of the current sheet withthe plasma pressures in the current sheet provides confidence thatthe bulk of the plasma pressure is being measured. The plasma bis displayed in Figure 5k and ranges from 1 to an impressive100. This b is determined from simultaneous magnetic fieldmeasurements and the plasma observations at the current sheetcrossing. Some of the variations in b can be due to rapidcrossings of the current sheet relative to the time resolution ofthe measurements.[21] In order to complete the record of plasmas and magnetic

fields the vector components of the magnetotail fields are shownin Figure 6. Again, the components are shown in solar-eclipticcoordinates in Figures 6d–6f, and the radial and azimuthalcomponents are shown in Figures 6g–6h. The primary newfeature to note is that the crossings consecutively alternate aboveand below the current sheet at the closer distances to Jupiter,and at the greater radial distances the crossings are confined toone lobe or the other for significant numbers of planetaryrotations. This effect is probably due to a combination ofthickening of the plasma sheet in which the current sheet isembedded and due to the larger lever arm for solar winddeflection of the magnetotail.[22] The radial dependence of the thermal ion densities during

May 4 to June 22 at the current sheet is shown in Figure 7a. Thesedensities have been fitted for two radial distance ranges, <20 RJ and

>50 RJ. Ion densities inside 20 RJ vary steeply as R�6.90 whereas the

radial dependence is quite shallow beyond 50 RJ, R�1.28. As shown

in Figure 7b for these two radial distance ranges, the magnetic fieldamplitude outside of the current sheet for b < 0.1 varies as R�2.78

and R�1.19 for R< 20 RJ and >50 RJ, respectively. The variationinside the radial distance of 20 RJ is not greatly dissimilar to that fora dipole field. The corresponding plasma and magnetic fieldpressures are given in Figures 8a and 8b, respectively, in units ofpascals. The plasma pressure varies as R�4.71 and R�1.87 for R< 20RJ and >50 RJ, respectively. The corresponding magnetic pressure isshown in Figure 8b and varies as R�5.56 and R�2.37 for R< 20 RJ and>50 RJ, respectively. The variations of plasma and magnetic fieldpressures are similar for R > 50 RJ, when the relatively largefluctuations by factors of 3 among the individual samples areconsidered.[23] It is of interest to further examine the ion parameters and

magnetic fields in the vicinity of perijove. Figure 9 exhibits thesedata for the period centered on perijove at 9.3 RJ at 1200 UT onMay 8. For this plot all plasma and magnetic field measurementsare shown regardless of their position within the current sheet.The vertical dashed lines indicate crossings of the current sheet.The radial distance extends to �26 RJ on the inbound andoutbound legs of the trajectory. The ion number densities andthe temperatures are displayed in Figures 9a and 9b, respectively.The number densities decrease by a factor of �100 in the currentsheet, and the temperatures increase by a factor of �10 withincreasing radial distance. Figures 9c–9e show the radial andcorotational components of the ion bulk flows, together with theion bulk speed, respectively. Rigid corotational speed is given bythe dashed lines in Figures 9d and 9e. The ion bulk flow is in thecorotational direction, with small or no radial component insideof a radial distance of �16 RJ. The corotational flow is sub-stantial but only �60% of that corresponding to rigid corotationalflow. At distances beyond 16 RJ there is a substantial, fluctuating

Figure 7. (a) Ion number densities as a function of Jovian radialdistance for the current sheet crossings. Inbound observations areidentified with circles, and the outbound observations are identifiedwith dots. Power law fits are given for two radial distance ranges,<20 RJ and >50 RJ. (b) Scalar magnetic fields as a function ofJovian radial distance above and below the current sheet for b <0.1. Again, inbound measurements are shown as circles, andoutbound observations are shown as dots. Fits to the radialdependences are also shown.

Figure 8. Continuation of Figure 7 for (a) ion energy densitiesand (b) magnetic field energy densities.


radial outflow from the planet in the range of 50–200 km s�1. Itis also important to note that the values for the corotationalcomponent are nearly equal to those expected for rigid corotationin the range of 150–200 km s�1 when the current sheet iscrossed. Values for the corotational components are in the rangeof 50–150 km s�1 at positions above and below the currentsheet. The components and scalar magnitude of the magneticfields are shown in Figures 9f–9i. On the linear scale used in thisplot the field components vary smoothly with each crossing ofthe current sheet. Examination of the magnitude in Figure 9i findsthat the decreases are similar and in the range of 50 nT, and thusthat the fractional decrease increases rapidly with increasingradial distance. The magnetic field magnitudes at the currentsheet crossings are small at radial distances > 20 RJ for which thecorotational components in the current sheet increase to rigidcorotational speeds.

[24] The plasma parameters and magnetic fields for a crossingof the current sheet and the adjacent regions are shown in Figure10. The crossing of the plasma sheet that contains the current sheetis centered at �1900 UT on May 10. Radial distance from Jupiterand local time of the spacecraft position are shown along theabscissa. The crossing of the current sheet is clearly seen in theprofile of the magnetic field component Br in Figure 10f. As shownfor the plasma parameters in Figures 10a–10e, the plasma sheet iscrossed during 1800–2000 UT with an increase of densities by afactor of �2 to 0.2 cm�3, with no change in temperature and withno readily identifiable changes in the bulk flow components.[25] The 3-day snapshots of plasmas and magnetic fields shown

in Figures 11–13 record several types of temporal variability in themagnetotail. The density threshold for determination of the plasmaparameters is 0.005 cm�3. A dramatic change in the geometry of theplasma sheet as seen at the Galileo spacecraft is seen at�1200UTon

Figure 9. Plasma moments and magnetic field parameters covering the period when the Galileo spacecraft was atperijove on May 8, 1997: (a) ion density, (b) ion temperature, (c) radial component of ion bulk flow, (d) the azimuthalcomponent of flow, (e) ion bulk speed, (f ) radial component of the magnetic field, (g) azimuthal component of thefield, (h) Z component of the field, and (i) scalar magnetic field. Vertical dotted lines identify the current sheetcrossings.


June 1 in Figure 11. During this crossing of the plasma sheet thexcomponent of bulk flow increased to �200 km s�1 as shown inFigure 11c. The dramatic change in the signature of Bx is due to thefact that the spacecraft does not penetrate into both the southern andnorthern lobes surrounding the plasma sheet. This change ingeometry at the spacecraft could be simply due to a change in thedirection of the solar wind. For an assumed plasma sheet thickness of5 RJ at a radial distance of 100 RJ it would take only a 1� change inthe elevation of the solar wind to sufficiently deflect Jupiter’smagnetotail. The reader should note that such crossings are respon-sible for missing north-south cycles in the magnetometer record forthe lobe in Figure 6d, for example, during June 1–6. Figure 12presents a situation for which the crossings are symmetric but forwhich the magnitude of the lobe magnetic fields decreases from Bx =10 nT to 6 nT. Such a variation could be accounted for by a decreasein plasma and/or magnetic field pressures in the solar wind.[26] Figure 13 provides an example for which the temporal

variation is likely to be due to a plasma instability in Jupiter’s

magnetotail. Inspection of Figure 13f shows that the spacecraft ispositioned more or less symmetrically with respect to the currentsheet. The magnetic field is increasing owing to motion of thespacecraft toward Jupiter. At the two crossings centered at �0900UT on June 20 the Y component of bulk flows is substantial atabout �200 km s�1. For the first crossing there is a large impulsivedeflection of the magnetic field in the �Z direction as to be seen inFigure 13h. Large magnetic field fluctuations occur during thesetwo current sheet crossings. These transient events may be indica-tive of large-scale instabilities in the current sheet. The currentsheet becomes stable by the time of the next current sheet crossing,�6 hours later.

4. Summary and Discussion

[27] We have presented a survey of the thermal plasmas andmagnetic fields during the Galileo orbit G08, which covered the

Figure 10. Plasma parameters during a crossing of the plasma sheet during 1800–2000 UT on May 10, 1997: (a)ion density, (b) ion temperature, (c) radial component of bulk flow, (d) component of bulk flow in the corotationaldirection, (e) ion bulk speed, and (f– i) radial, azimuthal, and Z components of the magnetic field and its magnitude.The radial distance and local time for the spacecraft are given along the abscissa.


period May 4 through June 22, 1997. This orbit traversed themagnetotail out to the apojove radial distance of 100 RJ at a localtime of 0121. Perijove was positioned in the outer region of thetorus at a radial distance of 9.3 RJ. There was marvelous, almostcomplete coverage of this orbit with real-time receipt of thespacecraft telemetry at Earth.[28] The fields of view of the mass spectrometers in the

plasma analyzer (PLS) did not view into the ion beams whentheir densities were sufficiently high to obtain a statisticallyreliable determination of the M/Q of the ions. However, thehydrogen ion beam was clearly detected in the E/Q spectra of theions, which allowed determination of the constituents of theheavy ion portion of the flowing plasmas. An exemplary fit tothese E/Q spectra at a Jovian radial distance of 25.3 RJ revealsthe M/Q of the primary heavy ion as 16 with lesser contributionsfrom ions with M/Q = 8 and 10.7. Ions with M/Q = 16 areprobably O+ and S++, and those with 8 and 10.7 are O++ and

S+++, respectively. The presence of these constituents is inagreement with the previous analyses of in situ and remotemeasurements of the torus at lesser radial distances with Voyager1 [Bagenal, 1994] and with the findings in the dayside of themiddle magnetosphere with the Voyager plasma instrumentationat 19.8 RJ [McNutt et al., 1981]. Although the fiducial hydrogendistribution was not always detected along the magnetotailtrajectory with the Galileo plasma analyzers, it is reasonable toassume that the average M/Q of the heavy ion distributions was16 in the computation of the plasma moments, because theircharacteristic E/Q spectra were similar during the orbit with theexpected large variations in number densities. This assumptionintroduced typical errors of �10% in the values of the numberdensities and bulk velocities when these quantities are determinedfrom the observed three-dimensional velocity distributions withthe standard computation of plasma moments [Krall and Trivel-piece, 1973].

Figure 11. Plasma and magnetic field parameters during the 3-day interval May 31 through June 2, 1997: (a) iondensity, (b) ion temperature, (c, d) components of ion bulk flows in solar-ecliptic coordinates X and Y, respectively, (e)ion bulk flow speed, and (f– i) components of the magnetic field in solar-ecliptic coordinates and its magnitude. Notethe change in the geometry of the current sheet crossings at �1200 UT on June 1, which was accompanied by a strongtailward flow of plasmas of �200 km s�1.


[29] It is noteworthy that there were three ion componentsdetected with the plasma analyzer. The first of these componentswere the heavy ions, which were used to determine the iondensities and bulk velocities. The temperature of these ions wastypically 500 eV. The second component was cool hydrogen ionswith a temperature in the range of tens of eV. At a radial distanceof 25 RJ, for example, the number density of these hydrogen ionswas in the range of 0.05 cm�3, and the total density of the heavyions was 0.09 cm�3. Of course, the mass density of thesehydrogen ions was only �3% of the total mass density. Inaddition, there was a hot hydrogen ion distribution frequentlydetected at radial distances <50 RJ. The temperature of these ionswas in the range of 10 keV. These are probably part of theenergetic ions reported by Krimigis et al. [1981] with themedium-energy charged particle detector on the Voyager space-craft at energies above those of the Galileo plasma instrumenta-tion. It is reassuring to note that computations of the bulk flowsof each of the above three thermal plasmas detected with theGalileo plasma analyzer were mutually equal to within theaccuracies of their determinations.

[30] The plasma bulk flows in the vicinity of perijoverevealed several interesting dynamical features. The reader maydesire to inspect Figure 9 for this summary. In the radialdistance range of perijove at 9.0 RJ to �18 RJ the plasma iscorotating but with a speed that is �60% of that expected forrigidly corotating plasmas with Jupiter’s rotation. Various esti-mates of the radial distance for which coupling with the iono-sphere is sufficient to maintain rigid corotation are available inthe literature [Hill, 1979, 1980; Hill et al., 1981; Vasyliunas,1994]. If the limiting distance is interpreted observationally asthe distance for which significant radial flows of plasmacommence, then the above analytical estimates of �20 RJ arein reasonable agreement with the present measurements. Beyond�18 RJ, significant radially outward flows in the range of 50–200 km s�1 are detected with the Galileo plasma analyzer. Infact, in the radial distance of 18–26 RJ shown in Figure 9, theplasma bulk flows at the current sheet crossing are equal tothose expected for rigid corotation. The increase in plasma bulkflows and the large radial components, together with heating ofthe plasmas to temperatures of 108 K, strongly suggest that

Figure 12. (a– i) Continuation of Figure 11 for the period June 9 through June 11, 1997. Note that the magnetotailfield decreased at �1200 UT on June 10, concurrent with a sporadic increase of plasma flows in the Y direction.


acceleration of these plasmas is occurring at these radialdistances both along the inbound and outbound segments ofthe orbit. This plasma acceleration may be associated with thesystem of X and O lines due to magnetic merging in themagnetotail as predicted by Vasyliunas [1983]. On the basis ofmagnetic field observations with the Galileo magnetometer,Russell et al. [1998] suggest that impulsive north and southwardtransients of magnetic fields in the current sheet at these radialdistances are the signatures of localized reconnection in theseregions. Indeed, energetic charged particle acceleration eventsthat are qualitatively similar to those observed in Earth’smagnetosphere are found at radial distances of �9 to 27 RJ inJupiter’s magnetosphere [Mauk et al., 1997, 1999]. Energeticparticle bursts observed at distances farther into Jupiter’s mag-netotail are also reported for Galileo mission [Krupp et al.,1998; Woch et al., 1999].[31] A convincing signature of an active sector in the Jovian

magnetosphere at radial distances in the range of 9.3–27 RJ is

reported for the present plasma observations with Galileo. Thetemperatures in this active sector of System III longitudes can behigher by factors of 2–10 relative to those outside this sector (seeTable 2). However, the present observations are limited by thegeometry of current sheet crossings to two values of longitude,�110� and 290�, with maximum temperatures at 290�. Pilcher andMorgan [1985] have previously reported an active sector for sulfurion emissions as located in the System III longitude range of�180�–230�. This bright sector for sulfur ion emissions has beenextensively studied [Schneider and Trauger, 1995; Schneider et al.,1997]. These latter authors suggest that the brighter emissions areassociated with a low Tk in the active sector. Such an active sectorhas been recently confirmed with in situ thermal plasma measure-ments at lesser radial distances of 6–8 RJ in the torus, whichexhibit higher ion densities off the equator in the active sector[Frank and Paterson, 2001b].[32] The scalar magnetic field in the radial distance range of 9–

20 RJ is found to vary as R�2.78 and is similar to that expected for a

Figure 13. (a– i) Continuation of Figure 11 for the period June 19 through June 21, 1997. At these closer distancesin the range of 60 RJ the plasma sheet is considerably thinner than at the spacecraft apojove distances of �100 RJ.Strong ion bulk flows of �200 km s�1 occur during the two current sheet crossings centered at 0900 UT on June 20,along with a transient magnetic deflection in the Z direction during the first crossing.


dipole magnetic field. At the larger radial distances this depend-ence decreases to R�1.19. This value is not dissimilar to thosereported for Voyager 1 and 2 and Pioneer 10, which ranged fromR�1.36 to R�1.70. These values are dependent upon the trajectoryand the pressure in the nearby solar wind. Whereas the thermalplasma pressure is a factor of �10 less than the magnetic pressureat 9 RJ, the thermal plasma pressure in the current sheet becomesequal to the magnetic pressure outside of the sheet at radialdistances >30 RJ. The observational equality of these two pressuresis one of the major findings of the present investigation. The ratiosof the plasma to magnetic pressures, b, within the current sheet arelarge and range from 1 to 100. Beyond 30 RJ the number densitiesand the temperatures of the thermal plasmas are 0.05–0.1 cm�3

and 0.5–1 � 108 K, respectively. These new findings for thethermal plasmas are significant in evaluating the stress balance inthe Jovian current sheet, which lies beyond the torus. Estimationsof these stresses from thermal plasmas and charged particles athigher energies have been given for the Voyager measurements[Mauk and Krimigis, 1987; Sands and McNutt, 1988; Paranicas etal., 1991]. The observational difficulty is that terms in the forcebalance equation require an accurate evaluation of the magneticfield geometry [Vasyliunas, 1983]. With the present work we havegained an understanding of the pressure balance in the magnetotail,which should be expected to be followed by future carefulmodeling of the magnetic fields in order to quantitatively assessthe force balance.[33] In the magnetotail the bulk flow speeds of the thermal

plasmas range from �50 to 200 km s�1 and are significantly lessthan those expected from rigid corotation with Jupiter. Suchspeeds are consistent with the analysis of bulk flows frommeasurements of more energetic charged particles with theVoyager spacecraft [Kane et al., 1995]. The bulk flows of thethermal plasmas exhibit significant components in the corota-tional and outward radial directions. Inspection of the radial andcorotational components of bulk flow in Figures 4f–4g finds thatthe flow is relatively stagnant in the premidnight sector whencompared to those in the early morning sector of the magnetotail.A similar asymmetry in bulk flows has been reported by Krupp etal. [2001] with the assumption that the anisotropies of energeticions observed with the Galileo energetic particle detector (EPD)are due to bulk flow. At apogee at 100 RJ the flow changes to anantisunward flow of �200 km s�1, which is indicative of themagnetospheric ‘‘wind’’ reported by Krimigis et al. [1981]. Thepersistence of this magnetospheric wind will be tested withobservations during other Galileo orbits. The mixing of solarwind with the hot magnetotail plasmas is found with Voyager 2observations at radial distances in the range of 5000–9000 RJ

[Sittler et al., 1987].[34] The current sheet crosses the position of the Galileo

spacecraft in the magnetotail with the 10-hour periodicity ofJupiter’s rotational period. The actual signature of this relativemotion of the sheet with respect to spacecraft depends upon thegeometry and dynamical motions of the sheet and its responsesto the solar wind. Internal instabilities should also affect theapparent motion of the sheet. Considerations of these effectshave been previously published for Voyager and Pioneer obser-vations [Carbary, 1980; Behannon et al., 1981; Khurana, 1997].The possibility of periodicities with longer periods of days hasbeen suggested by Vasyliunas et al. [1997] and Woch et al.[1998]. Our purpose in this initial paper is to show threeexamples of these dynamical responses in the 10-hour rotationalperiodicity with suggested interpretations in terms of change indirection of the solar wind, a change in the solar wind pressure,and a transient instability with acceleration of plasmas in theplasma sheet.

[35] Acknowledgments. This research was supported in part bycontract JPL-958778 at the University of Iowa and by contract JPL-

958694 at the University of California, Los Angeles, both contracts withthe Jet Propulsion Laboratory.[36] Janet G. Luhmann thanks Joachim Woch and another referee for

their assistance in evaluating this paper.

ReferencesAcuna, M. H., K. W. Behannon, and J. E. P. Connerney, Jupiter’s magneticfield and magnetosphere, in Physics of the Jovian Magnetosphere, editedby Dessler, A. J., pp. 1–50, Cambridge Univ. Press, New York, 1983.

Bagenal, F., Empirical model of the Io plasma torus: Voyager measure-ments, J. Geophys. Res., 99, 11,043–11,062, 1994.

Bame, S. J., B. L. Barraclough, W. C. Feldman, G. R. Gisler, J. T. Gosling,D. J. McComas, J. L. Phillips, M. F. Thomsen, B. E. Goldstein, and M.Neugebauer, Jupiter’s magnetosphere: Plasma description from theUlysses flyby, Science, 257, 1539–1543, 1992.

Behannon, K. W., L. F. Burlaga, and N. F. Ness, The Jovian magnetotailand its current sheet, J. Geophys. Res., 86, 8385–8401, 1981.

Belcher, J. W., The low-energy plasma in the Jovian magnetosphere, inPhysics of the Jovian Magnetosphere, edited by Dessler, A. J., pp.68–105, Cambridge Univ. Press, New York, 1983.

Carbary, J. F., Periodicities in the Jovian magnetosphere: Magnetodiscmodels after Voyager, Geophys. Res. Lett., 7, 29–32, 1980.

Cowley, S. W. H., A. Balogh, M. K. Dougherty, M. W. Dunlop, T. M.Edwards, R. J. Forsyth, R. J. Hynds, N. F. Laxton, and K. Staines, Plasmaflow in the Jovian magnetosphere and related magnetic effects: Ulyssesobservations, J. Geophys. Res., 101, 15,197–15,210, 1996.

Crary, F. J., F. Bagenal, L. A. Frank, and W. R. Paterson, Galileo plasmaspectrometer measurements of the composition and temperature in the Ioplasma torus, J. Geophys. Res., 103, 29,359–29,370, 1998.

Frank, L. A., and W. R. Paterson, Production of hydrogen ions at Io, J.Geophys. Res., 104, 10,345–10,354, 1999.

Frank, L. A., and W. R. Paterson, Observations of plasmas in the Io toruswith the Galileo spacecraft, J. Geophys. Res., 105, 16,017–16,034,2000.

Frank, L. A., and W. R. Paterson, Passage through Io’s ionospheric plasmasby the Galileo spacecraft, J. Geophys. Res.,106, 26,209–26,224, 2001a.

Frank, L. A., and W. R. Paterson, Survey of thermal ions in the Io plasmatorus with the Galileo spacecraft, J. Geophys. Res., 106, 6131–6149,2001b.

Frank, L. A., K. L. Ackerson, J. A. Lee, M. R. English, and G. L. Pickett,The plasma instrumentation for the Galileo Mission, Space Sci. Rev., 60,283–307, 1992.

Geiss, J., et al., Plasma composition in Jupiter’s magnetosphere: Initialresults from the solar wind ion composition spectrometer, Science, 257,1535–1539, 1992.

Gurnett, D. A., A. M. Persoon, W. S. Kurth, A. Roux, and S. J. Bolton,Plasma densities in the vicinity of Callisto from Galileo plasma waveobservations, Geophys. Res. Lett., 27, 1867–1870, 2000.

Hill, T. W., Inertial limit on corotation, J. Geophys. Res., 84, 6554–6558,1979.

Hill, T. W., Corotation lag in Jupiter’s magnetosphere: Comparison of ob-servation and theory, Science, 207, 301–302, 1980.

Hill, T. W., A. J. Dessler, and L. J. Maher, Corotating magnetosphericconvection, J. Geophys. Res., 86, 9020–9028, 1981.

Kane, M., B. H. Mauk, E. P. Keath, and S. M. Krimigis, Hot ions inJupiter’s magnetodisc: A model for Voyager 2 low-energy charged par-ticle measurements, J. Geophys. Res., 100, 19,473–19,486, 1995.

Khurana, K. K., Euler potential models of Jupiter’s magnetospheric field, J.Geophys. Res., 102, 11,295–11,306, 1997.

Kivelson, M. G., K. K. Khurana, J. D. Means, C. T. Russell, and R. C.Snare, The Galileo magnetic field investigation, Space Sci. Rev., 60,357–383, 1992.

Kivelson, M. G., J. Warnecke, L. Bennett, S. Joy, K. K. Khurana, J. A.Linker, C. T. Russell, R. J. Walker, and C. Polanskey, Ganymede’s mag-netosphere: Magnetometer overview, J. Geophys. Res., 103, 19,963–19,972, 1998.

Krall, N. A., and A. W. Trivelpiece, Principles of Plasma Physics,McGraw-Hill, New York, 1973.

Krimigis, S. M., J. F. Carbary, E. P. Keath, C. O. Bostrom, W. I. Axford, G.Gloeckler, L. J. Lanzerotti, and T. P. Armstrong, Characteristics of hotplasma in the Jovian magnetosphere: Results from the Voyager space-craft, J. Geophys. Res., 86, 8227–8257, 1981.

Krupp, N., E. Keppler, R. Seidel, J. Woch, A. Korth, A. F. Cheng, S. E.Hawkins, III, L. J. Lanzerotti, C. G. Maclennan, and M. K. Dougherty,Field-aligned particle streaming in the duskside high latitude Jovian mag-netosphere, Adv. Space Res., 20(2), 225–228, 1997.

Krupp, N., J. Woch, A. Lagg, B. Wilken, S. Livi, and D. J. Williams,Energetic particle bursts in the predawn Jovian magnetotail, Geophys.Res. Lett., 25, 1249–1252, 1998.

Krupp, N., J. Woch, A. Lagg, E. C. Roelof, D. J. Williams, S. Livi, and B.


Wilken, Local time asymmetry of energetic ion anisotropies in the Jovianmagnetosphere, Planet. Space Sci., 49, 283–289, 2001.

Lanzerotti, L. J., et al., Measurements of hot plasmas in the magnetosphereof Jupiter, Planet. Space Sci., 41, 893–917, 1993.

Mauk, B. H., and S. M. Krimigis, Radial force balance within Jupiter’sdayside magnetosphere, J. Geophys. Res., 92, 9931–9941, 1987.

Mauk, B. H., D. J. Williams, and R. W. McEntire, Energy-time dispersedcharged particle signatures of dynamic injections in Jupiter’s inner mag-netosphere, Geophys. Res. Lett., 24, 2949–2952, 1997.

Mauk, B. H., D. J. Williams, R. W. McEntire, K. K. Khurana, and J. G.Roederer, Storm-like dynamics of Jupiter’s inner and middle magneto-sphere, J. Geophys. Res., 104, 22,759–22,778, 1999.

McDonald, F. B., and J. H. Trainor, Observations of energetic Jovian elec-trons and protons, in Jupiter, edited by T. Gehrels, pp. 961–987, Univ.of Ariz. Press, Tucson, 1976.

McNutt, R. L. Jr., J. W. Belcher, and H. S. Bridge, Positive ion observationsin the middle magnetosphere of Jupiter, J. Geophys. Res., 86, 8319–8342, 1981.

Paranicas, C. P., B. H. Mauk, and S. M. Krimigis, Pressure anisotropy andradial stress balance in the Jovian neutral sheet, J. Geophys. Res., 96,21,135–21,140, 1991.

Pilcher, C. B., and J. S. Morgan, Magnetic longitude variations in the Iotorus, Adv. Space Res., 5(4), 337–345, 1985.

Russell, C. T., K. K. Khurana, D. E. Huddleston, and M. G. Kivelson,Localized reconnection in the near Jovian magnetotail, Science, 280,1061–1064, 1998.

Sands, M. R., and R. L. McNutt, Jr., Plasma bulk flow in Jupiter’s daysidemiddle magnetosphere, J. Geophys. Res., 93, 8502–8518, 1988.

Schardt, A. W., and C. K. Goertz, High-energy particles, in Physics of theJovian Magnetosphere, edited by A. J. Dessler, pp. 157–196, CambridgeUniv. Press, New York, 1983.

Schneider, N. M., and J. T. Trauger, The structure of the Io torus, Astrophys.J., 450, 450–462, 1995.

Schneider, N. M., M. H. Taylor, F. J. Crary, and J. T. Trauger, On the natureof the lIII brightness asymmetry in the Io torus, J. Geophys. Res., 102,19,823–19,833, 1997.

Seidel, R., E. Keppler, N. Krupp, J. Woch, S. E. Hawkins, L. J. Lanzerotti,and M. K. Dougherty, Energetic electron beams in the duskside Jovian

magnetosphere: Ulysses EPAC and HI-SCALE measurements, Planet.Space Sci., 45, 1473–1480, 1997.

Sittler, E. C. Jr., R. P. Lepping, B. H. Mauk, and S. M. Krimigis, Detectionof a hot plasma component within the core regions of Jupiter’s distantmagnetotail, J. Geophys. Res., 92, 9943–9948, 1987.

Smith, E. J., L. Davis, Jr., and D. E. Jones, Jupiter’s magnetic field andmagnetosphere, in Jupiter, edited by T. Gehrels, pp. 788–829, Univ. ofAriz. Press, Tucson, 1976.

Van Allen, J. A., High-energy particles in the Jovian magnetosphere, inJupiter, edited by T. Gehrels, pp. 928–960, Univ. of Ariz. Press, Tucson,1976.

Vasyliunas, V. M., Plasma distribution and flow, in Physics of the JovianMagnetosphere, edited by A. J. Dessler, pp. 395–453, Cambridge Uni-versity Press, New York, 1983.

Vasyliunas, V. M., Role of the plasma acceleration time in the dynamicsof the Jovian magnetosphere, Geophys. Res. Lett., 21, 401–404,1994.

Vasyliunas, V. M., L. A. Frank, K. L. Ackerson, and W. R. Paterson,Geometry of the plasma sheet in the midnight-to-dawn sector of theJovian magnetosphere: Plasma observations with the Galileo spacecraft,Geophys. Res. Lett., 24, 869–872, 1997.

Williams, D. J., B. H. Mauk, R. E. McEntire, E. C. Roelof, T. P. Armstrong,B. Wilken, J. G. Roederer, S. M. Krimigis, T. A. Fritz, and L. J. Lanzer-otti, Electron beams and ion composition measured at Io and its torus,Science, 274, 401–403, 1996.

Woch, J., N. Krupp, A. Lagg, B. Wilken, S. Livi, and D. J. Williams, Quasi-periodic modulations of the Jovian magnetotail, Geophys. Res. Lett., 25,1253–1256, 1998.

Woch, J., et al., Plasma sheet dynamics in the Jovian magnetotail: Signaturesfor substorm-like processes, Geophys. Res. Lett., 26, 2137–2140, 1999.

��

L. A. Frank and W. R. Paterson, Department of Physics and Astronomy,University of Iowa, Iowa City, IA 52242, USA. ([email protected])K. K. Khurana, Institute of Geophysics and Planetary Physics, University

of California, Los Angeles, CA 90095, USA.


observations of thermal plasmas in jupiter’s magnetotail · 2007. 12. 13. · received 8 march...

Documents