Local time asymmetry of energetic ion anisotropies in the Jovian magnetosphere

Local time asymmetry of energetic ion anisotropies in the Jovian magnetosphere

Planetary and Space Science 49 (2001) 283–289 www.elsevier.nl/locate/planspasci Local time asymmetry of energetic ion anisotropies in the Jovian mag...

514KB Sizes 0 Downloads 15 Views

Planetary and Space Science 49 (2001) 283–289


Local time asymmetry of energetic ion anisotropies in the Jovian magnetosphere N. Kruppa; ∗ , J. Wocha , A. Lagga , E.C. Roelof b , D.J. Williamsb , S. Livia , B. Wilkena a MPAE,

D-37191 Katlenburg-Lindau, Germany Laurel, MD 20723-6099, USA


Received 14 December 1999; received in revised form 20 April 2000; accepted 22 May 2000

Abstract Since December 1995 the Galileo spacecraft is in orbit around Jupiter. Up to now the spacecraft performed 25 orbits through the Jovian system within nearly 4 years providing an excellent data coverage in local time speci6cally for distances r ¡ 40RJ . We present 6rst-order anisotropies measured in energetic ion distributions from the Energetic Particles Detector (EPD) onboard Galileo. In this paper, we concentrate on measurements in the inner Jovian magnetosphere at distances between 6 and 40RJ . Results from three di:erent ion species show a pronounced local time asymmetry in the ion distributions at these distances especially between the dawn–prenoon and the dusk–premidnight sector of the magnetosphere. The predominantly 6rst-order anisotropies in the co-rotation direction show larger amplitudes with radial outward components in the dawn sector whereas at dusk smaller anisotropies with small radial inward components are observed. Under the reasonable assumption that the anisotropies are primarily due to
1. Introduction The motion of plasma and energetic particles within the Jovian magnetosphere is very important in understanding the con6guration and dynamics of the magnetosphere. Previous
Corresponding author. Tel.: +49-5556-979-154; fax: +49-5556-6154. E-mail address: [email protected] (N. Krupp).

in a magnetodisk con6guration (VasyliGunas, 1983) as in the Jovian case. Within that distance the plasma and energetic particle should basically co-rotate with the planet and no signi6cant local time dependence is expected. Outside that radius co-rotation is only maintained at a fraction of 30 –70% as observed by previous
c 2001 Elsevier Science Ltd. All rights reserved. 0032-0633/01/$ - see front matter  PII: S 0 0 3 2 - 0 6 3 3 ( 0 0 ) 0 0 1 4 9 - 5


N. Krupp et al. / Planetary and Space Science 49 (2001) 283–289

backwards and is lagging rigid co-rotation. In the dusk sector the same 6eld line changes to a leading con6guration. This transition was observed at 38RJ at about 19:00 LT ◦ at 30 southern magnetic latitude onboard the Ulysses spacecraft (Dougherty et al., 1993). The lengthening 6eld line is stretched towards the magnetotail. Energetic particle instruments onboard Ulysses observed 6eld-aligned distributions during its pass through the dusk sector at southern ◦ magnetic latitudes of up to 45 (i.e. Lanzerotti et al., 1992). Staines et al. (1996) reported larger radial anisotropies on the dusk side than on the prenoon side, indicating a dawn–dusk asymmetry in the particle distribution. Krupp et al. (1997,1999) showed the existence of mono- and bidirectional ion beams along the magnetic 6eld. Laxton et al. 1997 observed anti-co-rotational
Table 1 Selected energy passbands for di:erent species from the time-of-

Energy passbands


Protons Protons Oxygen Oxygen Sulfur Sulfur

0.080 – 0.220 (MeV) 0.220 – 0.540 (MeV) 0.026 – 0.051 (MeV=N) 0.051– 0.112 (MeV=N) 0.016 – 0.030 (MeV=N) 0.030 – 0.062 (MeV=N)

Fig. 1. Top: Sketch of the look directions of the EPD instruments onboard Galileo in the s=c-coordinate system; lower left: projection of the EPD look directions into a plane; lower right: JSE coordinate system. In analogy to the GSE-coordinate system at Earth the x-axis points towards the Sun, the y-axis towards local dusk and z completes a right-handed system. Jupiter is in the center of the coordinate system (original 6gure from Lagg, 1998).

space. Nearly the whole unit sphere is covered. A complete description of the instrument can be found in Williams et al. (1992). Due to the main antenna failure of Galileo the operation modes of EPD had to be changed. In the so-called record-mode EPD data are stored on the onboard tape recorder with the original high resolution in space and time (448 di:erent directions within 140 s). The record-mode is reserved only for a few minutes mainly during the close
N. Krupp et al. / Planetary and Space Science 49 (2001) 283–289

measurements of three di:erent ion species by using spherical harmonics and derived the direction and amplitude of the 6rst-order anisotropy. The particle intensity I (E; ; ) can be described as I (E; ; ) =

∞  m=n 

Anm Ynm


n=0 m=−n

and the functions Ynm are de6ned as  Pnm (cos #) cos(m’) if m ¿ 0; Ynm = Pn|m| (cos #) sin(|m|’) if m ¡ 0 with Pnm being the normalized Legendre polynomials. Details on this technique can be found, e.g. in Sanderson and Page (1974), and Alevizos et al. (1999). The measured 6rst-order anisotropies A1 are normally a combination of various components like particle
3. Observations We present the observations and results in the Jovian Solar Ecliptic coordinate system (JSE) which is shown in the lower right corner of Fig. 1. We focus on observations made in the inner part of the magnetosphere between 6 and 40RJ . Fig. 2 shows maps of intensities on the left and maps of the energy spectral index  on the right. The colour-coded intensities and -values for protons, oxygen ions, and sulfur ions are measured between June 1996 and November 1999 (orbits G1 to I25). The data are sorted into circular bins (2RJ in diameter) along the trajectory of Galileo and projected into the x–y-plane of the JSE system. For the distance range under study we 6nd that the intensity levels are slightly asymmetric in local time. By comparing the intensities along the orbit of Callisto (indicated by the solid circle labelled with C) at di:erent local times


it is obvious that the values are higher in the local time sector between dusk and midnight compared to dawn for all three ion species. The spectral index  does not show a pronounced local time asymmetry. The values of  at di:erent local times appear nearly the same at comparable distances. If the values of  are large the spectrum of the intensity vs. energy is steep indicative for a soft spectrum, i.e. comparatively few high-energy particles are present; whereas a small  points to a hard spectrum with a major fraction of high-energy particles. The values are higher for the heavy ions oxygen and sulfur compared to protons. A change in the slope for all three species is clearly visible at around 20 –25RJ . The spectra become harder inside that distance where the transition between dipolar-like 6eldlines and radially stretched magnetic 6eldlines occurs. Fig. 3 shows the colour-coded 6rst-order anisotropy amplitude and direction as derived from energetic ion distributions. Anisotropy values are only calculated for those time intervals where reliable spin axis information of the spacecraft are available. For the other time periods where EPD data exist only the trajectory is shown by black lines. On the left-hand side the 6rst-order anisotropy amplitude A1 relative to the isotropic component A0 is presented for protons, oxygen ions and sulfur ions. For all three species the same colour-coding is applied. Proton anisotropy amplitudes are smaller than the heavy-ion anisotropy amplitude as expected from the ion velocity dependence of convected anisotropies and already found in Voyager data (Krimigis and Roelof, 1983). For all three investigated species we 6nd a pronounced local time asymmetry in the anisotropy amplitudes. They are larger in the dawn and pre-noon region than in the dusk and pre-midnight region at comparable distances beyond 15 –20RJ . Inside that distances much smaller 6rst-order amplitude values are observed, and the pronounced local time asymmetry seem to disappear (all blue). The directional information about the particle motion is presented on the right-hand side of Fig. 3. The angle between the co-rotation direction and the direction of the measured 6rst-order anisotropy vector is shown in colour. Yellow to red indicate radial outward anisotropies, green and blue radial inward components. The local time asymmetry is very pronounced in the direction of the 6rst-order anisotropy vector inside 20RJ . In the dawn sector we 6nd radial outward ◦ deviations from the co-rotation direction of more than 30 at distances between perijove and the orbit of Callisto. This is true for all the three ion species. On the other hand, near local dusk the deviation from the co-rotation direction is smaller and slightly radially inward (green and blue colours in Fig. 3). Beyond about 20RJ on the dusk and premidnight sectors the anisotropies deviate towards the radial outward direction (yellow and red colours). Fig. 4 shows the anisotropy vectors for oxygen ions (26 –51 keV=nucleon) projected into Jupiter’s equatorial plane.


N. Krupp et al. / Planetary and Space Science 49 (2001) 283–289

Fig. 2. Colour-coded maps of ion intensities and energy spectral indices as measured by EPD onboard Galileo for distances r ¡ 40RJ . A dynamic averaging of 1 min=RJ distance from the planet is used. If the orbits overlap within 2RJ the circular bins are averaged. The labels of the dashed circles indicate the distances in Jovian radii from the planet and solid circles mark the orbits of the Galilean moons Io (I), Europa (E), Ganymede (G) and Callisto (C) for reference. Left column: Intensity maps for protons (80 –220 keV), oxygen ions (26 –51 keV=nucleon), and sulfur ions (16 – 30 keV=nucleon) (from top to bottom). Right column: Maps of energy spectral indices  for protons, oxygen and sulfur as derived from power laws for the intensity I = I0 (E=E0 )− taken at the boundary between from two adjacent energy channels for each species (see Table 1).

It is very clear from that 6gure that the anisotropy=
4. Discussion The derived angular distributions of energetic protons, oxygen ions, and sulfur ions were used to derive global anisotropy maps of ion motion in the equatorial plane of Jupiter inside 40RJ . From the fact that the derived
N. Krupp et al. / Planetary and Space Science 49 (2001) 283–289


Fig. 3. First-order anisotropy maps (dynamic averaging of 1min=RJ distance from the planet is used) for protons (80 –220 keV), oxygen ions (26 – 51 keV=nucleon), and sulfur ions (16 –30 keV=nucleon) in the Jovian magnetosphere for distances r ¡ 40RJ . Dashed circles indicate the distances from the planet and solid circles mark the orbits of the Galilean moons Io (I), Europa (E), Ganymede (G) and Callisto (C) for reference. Left: Colour-coded maps of the 6rst-order anisotropy amplitude A1 relative to the isotropic component A0 . Right: Colour-coded maps of the angle between the 6rst-order ◦ ◦ anisotropy vector and the co-rotation direction. An angle of 45 (red=grey) is indicative of a positive radially outward directed anisotropy and −45 (dark blue) describes the opposite case of radially inward directed 6rst-order anisotropies.

velocities are nearly the same for all three species we conclude that to a 6rst approximation the 6rst-order anisotropy is convection dominated. Otherwise, a mass and or energy dependence would have appeared. Independent studies from Kane et al. (1999) found that gradient e:ects could be neglected in the center of the plasma sheet and therefore also

support the assumption. If the ion velocity vion is much larger than the


N. Krupp et al. / Planetary and Space Science 49 (2001) 283–289

Fig. 4. First-order anisotropy vectors (3-h averages) for oxygen ions (26 –51 keV/nucleon) as derived from Galileo=EPD measurements and projected into the equatorial plane of the Jovian magnetosphere for distances r ¡ 40RJ . The lengths of the vectors are de6ned by the 6rst-order anisotropy amplitude. The vector scaling is given in the upper right corner.

where  is the measured energy spectral index (see i.e., Krimigis and Roelof, 1983). The
troduce an asymmetry is the fact that co-rotation
N. Krupp et al. / Planetary and Space Science 49 (2001) 283–289 Brice, N., Ioannidis, G., 1970. The magnetospheres of Jupiter and Earth. Icarus 13, 173–183. Carbary, J.F., Krimigis, S.M., Keath, E.P., Gloeckler, G., Axford, W.I., Armstrong, T.P., 1981. Ion anisotropies in the outer Jovian magnetosphere. J. Geophys. Res. 86, 8285–8299. Cheng, A.F., Krimigis, S.M., 1989. A model of global convection in Jupiter’s magnetosphere. J. Geophys. Res. 94, 12,003–12,008. Cowley, S.W.H., Balogh, A., Dougherty, M.K., Dunlop, M.W., Edwards, T.M., Forsyth, R.J., Hynds, R.J., Laxton, N.F., Staines, K., 1996. Plasma

1997. Field-aligned particle streaming in the duskside high latitude Jovian magnetosphere. Adv. Space Res. 20, (2)225–(2)228. Krupp, N., Roelof, E., Woch, J., Kane, M., Williams, D., Lagg, A., Wilken, B., Livi, S., 2000. Global