References and Notes1. K. W. Behannon, M. H. Acufla, L. F. Burlaga,

R. P. Lepping, N. F. Ness, F. M. Neubauer,Space Sci. Rev. 21, 235 (1977).

2. B. U. 0. Sonnerup and L. J. Cahill, J. Geophys.Res. 72, 171 (1967).

3. A. J. Dessler and V. M. Vasyliunas, Geophys.Res. Lett. 6, 37 (1979).

4. M. H. Acufia and N. F. Ness, J. Geophys. Res.81, 2917 (1976).

5. J. A. Gledhill, Nature (London) 214, 156 (1967);C. K. Goertz,J. Geophys. Res. 81, 3368 (1976).

6. L. J. Gleeson and W. I. Axford, J. Geophys.Res. 81, 3403 (1976).

7. E. J. Smith, L. Davis, Jr., D. E. Jones, P. J.Coleman, Jr., D. S. Colbum, P. Dyal, C. P. So-nett, A. M. A. Frandsen, ibid. 79, 3501 (1974).

8. T. G. Northrop, C. K. Goertz, M. F. Thomsen,ibid., p. 3579; D. E. Jones, E. J. Smith, L. Da-vis, Jr., D. S. Colburn, P. J. Coleman, Jr., P.Dyal, C. P. Sonett, Utah Acad. Sci. Arts Lett.Proc. 51, 153 (1974); A. Eviatar and A. I. Ersh-kovich, J. Geophys. Res. 81, 4027 (1976); M. G.Kivelson, P. J. Coleman, Jr., L. Froidevaux, R.L. Rosenberg, ibid. 83, 4823 (1978).

9. J. H. Piddington and J. F. Drake, Nature (Lon-don) 217, 935 (1968); P. Goldreich and D. Lyn-den-Bell, Astrophys. J. 156, 59(1969).

10. S. D. Shawhan, C. K. Goertz, R. F. Hubbard,D. A. Gumett, G. Joyce, in The Magnet-ospheres of the Earth and Jupiter, V. Formi-sano, Ed. (Reidel, Dordrecht, 1975); J. H. Pid-dington, Moon 17, 373 (1977).

11. S. D. Drell, H. M. Foley, M. A. Ruderman, J.Geophys. Res. 70, 3131 (1965).

12. S. Peale, P. Cassen, R. T. Reynolds, Science213, 892 (1979).

13. See, for example, R. A. Smith, in Jupiter, T.Gehrels, Ed. (Univ. of Arizona Press, Tucson,1976), figure 2 and pp. 1146-1189.

14. We thank our colleagues in the Voyager projectfor discussions of these early results and the en-tire Voyager project team for the success of thisexperiment. We also thank G. Sisk, E. Franz-grote, and J. Tupman of the JPL for their sup-port; C. Moyer, J. Scheifele, J. Seek, and E.Worley for contributions to the design, develop-ment, and testing at GSFC of the experiment in-strumentation; and T. Carleton, P. Harrison, D.Howell, W. Mish, L. Moriarty, A. Silver, andM. Silverstein of the data analysis team for con-tributing to the success of the rapid magneticfields and plasma particles data processing sys-tem. F.M.N. was supported financially by theGerman Ministry of Science and Technology.

20 April 1979

Plasma Observations Near Jupiter:Initial Results from Voyager 1

Abstract. Extensive measurements oflow-energy positive ions and electrons were

made throughout the Jupiter encounter of Voyager 1. The bow shock and magneto-pause were crossed several times at distances consistent with variations in the up-stream solar wind pressure measured on Voyager 2. During the inbound pass, thenumber density increased by six orders of magnitude between the innermost mag-netopause crossing at - 47 Jupiter radii and near closest approach at 5 Jupiterradii; the plasma flow during this period was predominately in the direction ofcoro-tation. Marked increases in number density were observed twice per planetary rota-tion, near the magnetic equator. Jupiterward ofthe Jo plasma torus, a cold, corotat-ing plasma was observed and the energylcharge spectra show well-resolved, heavy-ion peaks at mass-to-charge ratios A/Z* = 8, 16, 32, and 64.

The Voyager plasma experiment is acooperative effort by experimentersfrom the Massachusetts Institute ofTechnology, the Goddard Space FlightCenter, the Jet Propulsion Laboratory,the High Altitude Observatory of the Na-tional Center for Atmospheric Research,the University of California at Los Ange-les, and the Max-Planck-Institut furAeronomie. The instrument and the ex-perimental objectives have been de-scribed in detail (1) and only a brief sum-mary is given here. The instrument con-sists of four Faraday cup sensors. Threeof these (the A, B, and C cups) are ar-ranged in a symmetric cluster whose axisusually points toward Earth. The axis ofthe fourth sensor (the D cup) is at rightangles to the axis of the cluster andpoints roughly into the direction of co-rotational flow on the inbound leg of thetrajectory at Jupiter; see Fig. 1. The en-ergy range for protons and electrons is 10to 5950 eV. The L and M modes are posi-tive ion modes spanning this energy/charge range in 16 contiguous steps (29percent nominal resolution in energy)and 128 contiguous steps (3.6 percentSCIENCE, VOL. 204, 1 JUNE 1979

nominal resolution in energy), respec-tively. Positive ion measurements aremade with all four sensors. Electronmeasurements are made only with the Dsensor. The El mode measures electronswith energies in the range 10 to 140 eV,using 16 contiguous steps at 3.6 percentnominal resolution in energy. The E2mode measures electrons with energiesin the range 10 to 5950 eV, using 16 con-tiguous steps at 29 percent nominal reso-lution in energy. During encounter, thetime required for a complete measure-ment cycle offour modes is 96 seconds.

In this report we describe (i) the ob-served crossings of the bow shock andmagnetopause and the changes in theirpositions with external conditions, (ii)plasma properties in the dayside outermagnetosphere, (iii) properties of theplasma in the inner magnetosphere, (iv)plasma properties in the nightside outermagnetosphere, and (v) radiation effectson the performance of the instrument.The reader should bear in mind that re-sults given here are based on a very pre-liminary state of the analysis. For ex-ample, the positive ion densities quoted

for the inbound pass for the outer mag-netosphere are based on assumptionsthat are only partially true, such as thatthe flow is corotational and supersonicand that all of the ions are protons. Thefirst two assumptions will have to bemodified and the third is certainly wrong.For this reason, no precise estimate canbe given of the accuracy of the densitiesshown in Fig. 3. They are probably goodto a factor of 5, but a final determinationcan only be made on the basis of a moredetailed analysis. Densities quoted forthe inner magnetosphere are much moreaccurate, for reasons which will becomeapparent below.Bow shock and magnetopause cross-

ings seen by Voyager 1 on the inboundand outbound trajectories are listed inTable 1 and shown on the trajectory plotin Fig. 1. Figure 2 shows the upstreampressure measured by Voyager 2 and ex-trapolated to Voyager 1, taking into ac-count corotation delay and the differentradial distances of the two spacecraftfrom the sun. The latter effect was themajor source of delay; a typical delaytime between observation at Voyager 2and arrival at Voyager 1 was - 35 hours.The first bow shock crossing occurred at85.5 Jupiter radii (RJ) at a dynamic pres-sure of 8 x 10-10 dyne/cm2. Using thesevalues of pressure and distance, PO andRO, the five shock crossings observed onthe inbound pass fit extremely well therelationship P = PO(RdR)8 with 8 = 3.This characterization agrees with that ofSmith et al. (2) and confirms the con-clusion of the Pioneer experimenters thatJupiter's magnetosphere is much morecompressible than that of Earth. Thedashed curve of Fig. 2 represents theequilibrium position of the bow shockversus pressure given by the relationabove; similarly, the solid curve showsthe expected position of the magneto-pause, using 8 = 3 and a value of Ro ap-propriate for the initial position of themagnetopause rather than the bowshock. The agreement with actual cross-ings is good.The six magnetopause crossings ob-

served on the inbound and outboundpasses occurred at subspacecraft SystemIII (1965) longitudes ranging from 100 to173°; the prediction by Dessler andVasyliunas (3) of a tendency for mag-netopause crossings to cluster in therange 290° ± 65', based on the presenceof such a tendency in the Pioneer 10 and11 observations, was not confirmed.In the dayside outer magnetosphere,

plasma ions exhibit a strong corotationalsignature on the inbound pass, as evi-denced by a consistently enhanced signalin the D cup as compared to ion currents

0036-8075/79/0601-0987$00.50/0 Copyright © 1979 AAAS 987

on

June

16,

201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

in the main sensor [see Fig.1 for the ge-ometry of the inbound pass; see also fig-ure 3 in (1)]. If one assumes that theplasma is corotating and supersonic andthat the positive ions are all protons,then one can derive a "density" from themeasured positive ion currents. This

~10

-20

-110

axis ofbeginnlook dindicat

solar vthe posolar v

um,mtime. Idyne/c

Cc

LuJ

If)

a-

C-,

11

Fig. 3.ter. ReVerticfrom Jvertica

overall variation in positive ion numberdensity is shown schematically in Fig. 3.At times (but not all of the time) there

are two resolved peaks in the D cupspectra, which we interpret as a narrowproton peak plus a broad heavy-ion peakconsisting of a number of unresolved

w

Stntn

e

90 80 70 60 50 40R,

r~~~~

S ~~~~~~~~~~~~~~~~~IfZs ~~ ~ ~ ~~~~~20'-osre nte ue antshr

species. Figure 3 shows examples ofsuch L mode spectra where the normal-ized curves have been obtained by divid-ing the measured current in a given ener-gy channel by the energy width of thechannel. Note the extreme variability ofthe apparent abundance ratio of heavyions to protons. Such variations some-times occur from one measurement tothe next. The apparent proton peak doesnot usually appear at the energy ex-pected for strict corotation but occursmost often at lower energies. This effectseems too large to be accounted for by apositive spacecraft charge (see below),but could be caused by a deviation of theflow from strict corotation. At othertimes, the positive ions are much hotterand no resolved peaks appear in thespectra (see also Fig. 3).The population of plasma electrons

15 J~ ~~ observed in the outer magnetosphere

59 60 61 62 63 during the inbound pass is characterizedNt8 USOLAR WIN O PREDICTED VOYRCERI SCET(DAYS) 1979 by a mean energy ranging from few

L~MAGNETOSHEATH Fig. I (left). Voyager I trajectory during Jupi- hundred to a few thousand electron

ter encounter. The trajectory of the spacecraft volts. In this respect it is similar to ther84 and the orbits of the four Galilean satellites electron population of the plasma sheet

-, . IS are seen as projected onto Jupiter's equatorial in the terrestrial magnetosphere. The in-SM-LINE "ITEREoTlAL(M,) 150 plane. Tick marks are shown every 12 hours.

The projections of the look directions of the ensity of electrons Increases with de-symmetry of the mainsenlsor and D cup (labeled S and D, respectively) are shown at the creasing radial distance in a manner that

iing of every third day, starting with day 57 (spacecraft event time), 26 March 1979. The is consistent with the variation of the ionlirection of the former lies almost in the plane; the foreshortening of the D cup line is density described above. In the out-

tive of its orientation out of the plane. Distances are all in units of Jovian radii (1 ermost regions of the magnetosphere the71,372 km). The sections of the trajectory along which the plasma experiment measured

wind or magnetosheath plasma are indicated. Unkeyed trajectory sections correspond to itensity of electrons is relatively smalltsitions at which the spacecraft was inside the magnetosphere. Fig. 2 (right). Predicted and in this region relatively strong steadyvind pressure at Voyager1 based on Voyager 2 measurements. The interplanetary medi- currents are observed in the lowest ener-

iagnetosheath, and magnetosphere are indicated as in Fig. 1; SCET is spacecraft event gy windowsof E1. We interpret theserhe solid and dashed smooth curves are discussed in the text. Pressure is in units of 10-10

sgnoel.cWe irpretothese:m2. signals as photoelectrons or secondary

electrons (or both) that are trapped by a

positive spacecraft potential. Similar ob-

ENERGY (KEV) servations have been reported in near-

s.1 3 s Earth space (4) and in the magneto-

8RJ § 12 RJ l15 RJ 17RJ 25RJ 35RJ sphere of Mercury (5). In agreement withthis interpretation, there is a detailed in-verse correlation between the observed

intensity of plasma electrons and the po-

tential of the spacecraft derived from theI observed high-energy cutoff of the as-

sumed photoelectron component. Nearthe magnetopause the potential of the

2\\\spacecraft occasionally rose to 30 to 40

V, but at distances less than 40 RJ it wasusually below the minimum value detect-

)0 > F . able (10 V).During the first magnetospheric pas-

sage, from 59 to 67 RJ, sporadic in-creases in the intensity of kilovolt elec-

ki; w lnLt ll 1 trons were noted; at least ten events

1 t t t t t t t were recorded during the first - 12, , .. 1,I., . , i,,,,l,, ..... ......... hours spent inside the magnetosphere.

0 10 20 30 40R so 60 70 80 90 The timing of these events was not re-

RJ ~ ~~~~~~~~latedin any way to predicted crossings

Positive ion densities during the Voyager 1 inbound pass, in number per cubic centime- Of the magnetic equator. In the sub-hgions of solar wind and magnetosheath flow are indicated in the same manner as in Fig. 1.al arrows indicate crossings of the magnetic equator. Also shown at different distances sequent inbound magnetospheric pas-lupiter are representative L mode spectra (see text) from the D cup, plotted on a linear sage, sporadic events were also ob-ai scale. The various spectra are normalized to the same maximum value. served at distances less than 47 Rj, but

9 8 8 ~~~~~~~~~~~~~~~~~~~~~~~~S ENCE, VOL. 204

63

rZ 60SN

0s 57

S

69

7S

72

1(

988 SCIENCE, VOL. 204

on

June

16,

201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

as the distance decreased the enhancedelectron fluxes occurred more oftenclose to expected crossings of the mag-netic equator.

Figure 3 clearly shows the regions ofenhanced plasma density encountered atregular intervals on the inbound pass. Atthe times when the density is high, themagnetic field strength is low (6). Fur-thermore, the density increases occurwhen the spacecraft is close to the mag-netic equator. Thus, it seems natural tointerpret the density enhancements seeninside 35 RJ as crossings of the space-craft through a thin plasma sheet or cur-rent sheet. The sheet rotates with theplanet and moves up and down relativeto the Zenographic equator. Such a cur-rent sheet was inferred by Smith et al. (7)from the Pioneer 10 and 11 measure-ments. In Fig. 4 we show a detailed viewof one of the inbound current sheetcrossings (other crossings on both the in-bound and the outbound pass are similarin character). The magnetic field strength(6) decreases from 30 to 12 y when theelectron density increases from 0.2 to 0.9cm-3. We also show the variation of thenominal ion density during this plasmasheet crossing. Particle densities peak atthe time of minimum magnetic fieldstrength and it seems reasonable that themagnetic field depression is due to thediamagnetic effect of plasma in theplasma sheet. In that case the decreaseof magnetic field pressure B2/87r shouldbe balanced by a corresponding increasein particle pressure. Based on a prelimi-nary analysis, the observed low-energy

Table 1. Bow shock (S) and magnetosphere(M) boundaries observed by the Voyagerplasma experiment.

Spacecraft Sys-event time Dis- tem

Bound- tance IIIary Hours, R ogi

DOY* mi- (RJ) longi-utes tude

Inbound passS 59 1434 85.7 870S 59 1952 82.2 2860S 60 1226 71.7 1640M 60 1943 67.1 660M 61 0755 59.1 1490S 61 0950 57.8 2220S 61 1308 55.7 3340M 62 0235 46.7 980

Outbound passM 74 0930 158.3 980M 74 1657 162.8 100,M 74 2128 165.4 1730S 77 0704 199.2 980S 79 0810 227 750S 79 0827 227 750

[ >80 0607 20 13<80 0816 240 1530Data 80 1024gap 81 0657S 81 0943 256 500S 81 1302 258 2000

*Day of year (1 January = 1).

ion plasma pressure at the center of thesheet contributes substantially to thispressure balance but the electron pres-sure does not.Heavy ions are found throughout the

inner magnetosphere apparently corotat-ing with Jupiter. Within the Io plasmatorus these corotating ions are quite hot;Jupiterward of the torus the plasma cools

extremely rapidly. The differences in theplasma between these two regions, hotand cold, are shown in Fig. 5. In bothregions the experimental data show thatthe bulk motion of the plasma is strictlycorotational; that is, vr - 0 and vt = or,where v, and vt are the radial and tan-gential components of the plasma veloc-ity, r is the radial distance from Jupiter,and o is the angular velocity of Jupiter.Under these conditions, all ions at a giv-en radius move with the same speed andthe corotation acts as a velocity selector.The energy/charge scan of the plasma in-strument can then separate different ion-ic species by their mass-to-charge ratios,A/Z* (A is atomic mass number and Z* iseffective charge number). Thus in Fig. 5two of the energy/charge spectra showclearly resolved peaks. The changing ge-ometry is such that the component of thecorotation vector into the A sensor is al-most constant for the three spectrashown (see Table 2); consequently,peaks at the same energy per charge cor-respond to the same value of A/Z*. Foreach spectrum the various peaks, aftercorrection for overlap, were separatelyfit to a convected Maxwellian distribu-tion function; in all cases the convectionspeed obtained from the fit was within0.5 km/sec of the geometrically expectedspeed and typically within 0.2 km/sec(thus the spacecraft potential in this re-gion must be small). Consequently, theaccuracy in A/Z* is better than ± 2 per-cent for these spectra. The density andtemperature found from the fits are tabu-lated in Table 3; absolute densities are

zED

LUJ

zED

zC1

0

a)

J

z

L1J

CDU.-

CE(a)

al:

Uy-

30 -01 0: ::o o~ 0 03~~~~~~~~~~~~~~~~~~~~~~~ i0 = ° - . _, *1. ° |

2C2 1,> 0-00 200 300 400 500 600

10-C _ t _ENERGY/CHRBGE (VOLTS)0 Fig. 4 (left). Plasma and magnetic field parameters for the current6.0 7.0 n.0 9o0

n nn11. 0 12n sheet crossing near 25 RJ inbound. Shown are electron densities (num-6. 7r.0u o.0 9.0u 10.0 1. 12. ber per cubic centimeter) derived from a fit to an isotropic Max-

DOY 63 wellian, the nominal ion densities, and the field strength F (nano-teslas). Fig. 5 (right). Comparison of the distribution function of positive ions within the lo plasma torus (0931) and at two points in the coldplasma region Jupiterward of the torus. Each curve is labeled with the spacecraft event time in hours and minutes UTC (coordinated universaltime). The distribution function is plotted on a linear scale and is computed from the A sensor data of selected M mode spectra, assuming all ionsare protons. The energy/charge scale is linear to 600 V, although the instrument scans to 6 kV.

I JUNE 1979 989

on

June

16,

201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

Table 2. Input data for composition analysis.

Time, DOY 64 0931 1015 1121Distance, RJ 5.63 5.30 4.96Corotation speed 46.5 40.5 34.3

relative to space-craft, km/sec

Speed into A sen- 32.8 33.7 33.2sor, km/sec

good to 10 percent, relative densities to 3percent, and temperatures to 25 percent.The spectral peaks in Fig. 5 and Table

3 cannot be identified in an unambiguousway from the plasma data alone. How-ever, some possibilities seem more likelyon physical grounds. Since no peak isobserved for values of A/Z* between 8and 16, the A/Z* = 8 peak is almost cer-tainly 02+ or a higher charge state of aheavier species. A possibility is S4+, butthat implies the presence of S3+, which isnot observed; similar arguments hold forMg3+. The A/Z* = 32 peak is probablyS+. The AIZ* = 16 peak is either 0+ orS2+; however, in the cool region the ki-netic temperatures of 02+, 0+, and S+are the same, whereas S2+ is about twiceas hot. Consequently, assuming thermalequilibrium, the A/Z* = 16 peak in thecool region is probably mainly O+. Thepeak at A/Z* = 64 is unidentified butcertainly real. There is no spectral evi-dence for N, Ne, Na, Mg, or Si, a roughlimit on the density in each case beingbetween about 1/30 and 1/10 of the ap-propriate neighboring peak.The relative abundances, densities,

and temperatures of the ionic species ob-served in the cool plasma region varysystematically with time. For example,the S/O ratio is clearly different at 1015and 1121.Within the torus the plasma is too hot

to permit separation of closely spacedionic species; equal amounts of S+ andS2+ would, however, be resolved. Thissituation is illustrated by the spectrum at0931. The peak atAIZ* = 16 could be ei-ther 0+ or S2+. Independent of that iden-tification, the most probable thermal

speed is - 16 km/sec, significantly lessthan that expected from the pickupspeed for ions of - 64 km/sec. Thislower speed is consistent with the cool-ing inferred from the large intensity of ul-traviolet radiation observed by the ul-traviolet spectrometer (8). Furthermore,the temperature is high enough so thatS2+ is favored over S+ and no emissioncharacteristic of S+ should be observablefrom the torus.The inner boundary of the hot torus is

well defined and the exit and entry loca-tions are consistent with L shell controlof the boundary between the hot andcold regions. A noteworthy feature is thechange from S2+ in the torus to S+ in thecool region.An important unanswered question

concerns the total positive ion and elec-tron densities. Because of the small co-rotation velocity near 5 RJ, protons andalpha particles fall outside the range ofthe instrument. Thus, although an elec-tron number density can be inferred fromthe observations of ions (see Table 3), itdoes not include contributions from ionicspecies with A/Z* 4 6. However, esti-mates of the total electron density in thisregion are available from the planetaryradio astronomy and the plasma waveexperiments. Using these data, it shouldbe possible to estimate the number den-sity of light ions that were not measuredby the plasma experiment.

In the nightside outer magnetosphere,the low-energy ion and electron in-tensities varied regularly with distance.From day 65 until day 69-that is, fromabout 20 RJ to 85 RJ-the intensitiespeaked twice per planetary rotation. Fig-ure 6 shows the System III (1965) longi-tudes at which the E2 response peaks asa function of distance from Jupiter. If theplasma were confined to the rigid equa-torial plane of a magnetic dipole, thepeaks should lie along the lines denotedby R. Clearly, there is a considerable lagbetween these predicted peaks and theobserved locations. Such a lag was alsoobserved on the Pioneer 10 outbound

Table 3. Summary of composition analysis.

0931 1015 1121Possible

A/Z* ionic Den- Temp- Den- Temp- Den- Temp-species sity erature sity erature sity erature

(cm-3) (103 K) (cm-3) (103 K) (cm-3) (103 K)8 02+ 30 35 47 9.716 S2+ 1023 490 247 53 126 17

0+ 2046 250 494 27 254 8.532 S+ 1150 25 134 9.564 S2+, S02+, Zn+ 23 77>8 (Ne) 2046 1730 435

990

901

180'

2700

Fig. 6. System III (1965) longitudes at whichthe E2 response peaks on the outbound trajec-tory, as a function of distance from Jupiter.

pass and has been explained by North-rop et al. (9) and Kivelson et al. (10) as aresult of the finite propagation time ofthe wave that transmits the informationabout the rocking magnetic dipole to thedistant parts of the magnetosphere.CurvesN andK in Fig. 6 indicate the ex-pected positions according to the modelsof Northrop et al. and Kivelson et al.,respectively. The observed peaks liecloser to the Kivelson et al. predictions,although there is a systematic deviationof observed positions from the curves la-beled K, in that the two peaks per rota-tion period occur closer together thananticipated.

In the models of Kivelson et al. andNorthrop et al. the current sheet reachesa maximum latitude of 9.60 once everyplanetary rotation; that is, the currentsheet follows the motion of the rigidmagnetic equatorial plane with a certaintime delay, which increases with radialdistance. Since the spacecraft latitude isabout 5°, the current sheet in these mod-els should sweep across it twice per rota-tion period. If the maximum latitude Xmof the current sheet is smaller than 9.60,the two crossings should occur closer to-gether; in the limit Xm = 50 the currentsheet should only graze the spacecraftonce per rotation period, and if km < 5°the current sheet never reaches thespacecraft. The peaks in "plasma den-sity" should then be ill-defined and oc-cur between the two curves labeled K.Beyond 85 RJ this appears to be the case.We would expect, on the basis of thismodel, that the peaks inside 85 RJ corre-spond to actual crossings of the currentsheet (that is, Xm > 50) and that the mag-netic field should change direction fromaway to toward the planet or vice versawhen the magnitude is low. Most of thepeaks outside 85 RJ should coincide withtimes when the magnetic field magnitude

SCIENCE, VOL. 204

on

June

16,

201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

decreases without a large change in di-rection. Comparison with the magnetom-eter data is necessary to confirm this in-terpretation. The fact that Xm < 9.60 in-dicates that the current sheet is distortedaway from the magnetic equator towardthe rotational equator in the way sug-gested by Smith et al. (11). However, in-side 85 RJ the distortion must be smallbecause the peaks are well separated intime. Beyond 85 RJ the current sheet ap-parently seldom moves up to latitudes inexcess of 5°. This cannot be explainedsimply by the effects of centrifugalforces. For a centrifugally dominatedmagnetosphere the current sheet shouldreach at least a latitude of Xm = 6.40 (12)and thus cross the spacecraft twice perrotation. A more plausible explanationfor the confinement of the current sheetto small latitude excursions might in-volve control by the solar wind; that is,formation of a tail-like structure, whichcould prevent the current sheet fromreaching latitudes in excess of 5°.

Finally, although the current diskmodel is strongly supported by the ob-servations, the existence of a once-per-rotation asymmetry in the observed re-gion of the disk (R < 85 RJ), whichwould be indicative of significant effects caused by a magnetic anomaly as sug-gested by Dessler and Vasyliunas (3),cannot yet be ruled out without a closerexamination of the data. However, thefact that beyond 85 RJ the peak in-tensities do not occur in what Desslerand Vasyliunas called the active hemi-sphere (the region in which the plasmasource associated with the anomaliesshould lie) is inconsistent with the sim-plest form of the magnetic anomaly mod-el unless large time delays are involved.The high intensities of penetrating ra-

diation belt particles found in the innerJovian magnetosphere are potentiallytroublesome for a plasma experiment(13). The modulated grid Faraday cup,however, measures an a-c signal fromwhich the background (which is a d-c sig-nal) is automatically excluded. There re-mains a small a-c background, which isproportional to the square root of the in-tensity of penetrating particles. We esti-mated this effect to be - 10-13 A at 5.0RJ (1, p. 267); the background actuallymeasured agrees with the prediction. Asa result, clean ion signals were obtaineddirectly from the Voyager plasma in-strument, and the ion measurementscould be immediately processed withoutrequiring any background subtraction.

It was also anticipated by some (14)that the high intensities of radiation beltelectrons could result in a high negativeSCIENCE, VOL. 204, 1 JUNE 1979

spacecraft potential, which would shiftthe positive ion signals to apparent highenergies and possibly completely out ofthe range of the plasma instrument. Thisdid not occur. Within the inner magneto-sphere, significant ion currents were ob-served down to the lowest-energy win-dows of the instrument, indicating thatthe magnitude of the negative spacecraftpotential (if any) did not exceed aboutyOV.

H. S. BRIDGE, J. W. BELCHERA. J. LAZARUS, J. D. SULLIVAN

R. L. MCNUTT, F. BAGENALCenterfor Space Research andDepartment ofPhysics,Massachusetts Institute ofTechnology,Cambridge 02139

J. D. SCUDDERE. C. SITTLER

Laboratoryfor Extraterrestrial Physics,Goddard Space Flight Center,Greenbelt, Maryland 20771

G. L. SISCOEDepartment ofAtmospheric Sciences,University ofCalifornia, Los Angeles

V. M. VASYLIUNASC. K. GOERTZ

Max-Planck-Institutfur Aeronomie,Katlenburg-Lindau, West Germany

C. M. YEATESJet Propulsion Laboratory,California Institute ofTechnology,Pasadena 91103

Rebrences and NotesI. H. S. Bridge, J. W. Belcher, R. J. Butler, A. J.

Lazarus, A. M. Mavretic, J. D. Sullivan, G. L.Siscoe, V. M. Vasyliunas, Space Sci. Rev. 21,259 (1977).

2. E. J. Smith, R. W. Fillius, J. H. Wolfe, J.Geophys. Res. 83, 4733 (1978).

3. A. J. Dessler and V. M. Vasyliunas, Geophys.Res. Lett. 6, 37 (1979).

4. M. D. Montgomery, J. R. Asbridge, S. J. Bame,F. W. Hones, in Photon and Particle Inter-actions with Surfaces in Space, R. J. L. Grard,Ed. (Reidel, Dordrecht, 1973).

5. K. W. Ogilvie, J. D. Scudder, V. M. Vasyliunas,R. C. Hartle, G. L. Siscoe,J. Geophys. Res. 82,1807 (1977).

6. N. F. Ness, personal communication.7. E. J. Smith, L. R. Davis, D. E. Jones, inJupiter,

T. Gehrels, Ed. (Univ. of Arizona Press, Tuc-son, 1976), p. 788.

8. L. Broadfoot, private communication.9. T. G. Northrop, C. K. Goertz, M. F. Thomsen,

J. Geophys. Res. 79, 3579 (1974).10. M. G. Kivelson, P. J. Coleman, L. Froidevaux,

R. L. Rosenberg, ibid. 83, 4823 (1978).11. E. J. Smith, L. Davis, Jr., D. E. Jones, P. J.

Coleman Jr., D. S. Colburn, P. Dyal, C. P. So-nett, A. iM. A Frandsen, ibid. 79, 3501 (1974).

12. C. K. Goertz, ibid. 81, 3368 (1976).13. L. A. Frank, K. L. Ackerson, J. H. Wolfe, J. D.

Mihalov, ibid., p. 457.14. D. A. Mendis and W. I. Axford, Annu. Rev.

Earth Planet. Sci. 2, 419 (1974).15. It is impossible to make proper acknowledge-

ment to all individuals whose efforts were essen-tial to the success of this experiment. However,we extend special thanks to the Jet PropulsionLaboratory (JPL) project staff; to J. Binsack, R.J. Butler, C. Goodrich, G. Gordon, A. Mavretic,and Dale Ross at Massachusetts Institute ofTechnology; to K. Ogilvie at Goddard SpaceFlight Center (GSFC) and to the GSFC pro-gramming staff under W. Mish; and to the JPLdata records group under J. Tupman. Finally,we express our gratitude to E. Franzgrote, ourexperiment representative at JPL, for his majorcontributions to the experiment. This researchwas sponsored by the National Aeronautics andSpace Administration and was performed undercontract 953733 to the Jet Propulsion Labora-tory.

23 April 1979

Jupiter Plasma Wave Observations:An Initial Voyager 1 Overview

Abstract. The Voyager I plasma wave instrument detected low-frequency radioemissions, ion acoustic waves, and electron plasma oscillations for a period ofmonths before encountering Jupiter's bow shock. In the outer magnetosphere, mea-

surements of trapped radio waves were used to derive an electron density profile.Near and within the Jo plasma torus the instrument detected high-frequency elec-trostatic waves, strong whistler mode turbulence, and discrete whistlers, apparentlyassociated with lightning. Some strong emissions in the tail region and some impul-sive signals have not yet been positively identified.

The Voyager 1 mission provided thefirst opportunity to examine directlywave-particle interaction phenomenawithin the magnetosphere of Jupiter andin the extensive region of disturbanceupstream from the planet. This reportcontains an overview of encounter mea-surements from the Voyager 1 plasmawave investigation for the period startingwith initial detection of Jupiter phenome-na and ending a few days after closestapproach. During this time, the magneto-sphere was strongly affected by a denseplasma torus associated with lo (1) andperturbed by changing interplanetary

conditions (2). Some interactions in thelo plasma torus resembled those found inEarth's plasmasphere. This region con-tained strong whistler mode turbulence(chorus, hiss, and impulsive signals thatappear to be associated with lightning).Electrostatic emissions related to theelectron gyrofrequency harmonics andupper hybrid resonance were also de-tected beyond the boundary of the torusand near the magnetic equator crossings.Radio waves trapped between the outertorus region and the dayside magneto-pause (continuum radiation) allowed usto determine electron density profiles, as

0036-8075/79/0601-0991$00.S0/0 Copyright X 1979 AAAS 991

on

June

16,

201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from