Jupiter: A fundamentally different magnetospheric interaction with the solar wind

Transcription

1 Click Here for Full Article GEOPHYSICAL RESEARCH LETTERS, VOL. 34, L20106, doi: /2007gl031078, 2007 Jupiter: A fundamentally different magnetospheric interaction with the solar wind D. J. McComas 1 and F. Bagenal 2 Received 21 June 2007; revised 29 August 2007; accepted 7 September 2007; published 24 October [1] Magnetic reconnection between the solar wind and Earth s magnetic field creates an open magnetosphere with erosion of magnetic flux from the dayside and return on the nightside. At Jupiter, the large source of plasma from Io and fast rotation rate create a magnetosphere whose dynamics is dominated by internal processes. This, along with its sheer physical size, makes the nightside return of flux difficult, if not impossible. However, because magnetic reconnection still occurs on Jupiter s dayside magnetopause, it has been generally assumed that Jupiter s magnetosphere must be similarly open. Here we show how additional reconnection between the IMF and open magnetospheric flux back near flanks can re-close this open flux without having to invoke reconnection in the Jovian magnetotail. Our reconnection cycle solves the problem of closing and returning open magnetic flux in a large, internally dominated magnetosphere that constantly sheds large amounts of plasma down its magnetotail. Citation: McComas, D. J., and F. Bagenal (2007), Jupiter: A fundamentally different magnetospheric interaction with the solar wind, Geophys. Res. Lett., 34, L20106, doi: /2007gl Introduction [2] The solar wind, flowing nearly radially outward from the Sun, pushes the Earth s magnetosphere inward on the dayside and back into an extended magnetotail on the nightside. Dungey [1961] proposed a process whereby the dynamics of the Earth s magnetosphere are directly driven by its interaction with the solar wind and its imbedded Interplanetary Magnetic Field (IMF, see Figure 1). While the microphysics of reconnection is still not well understood, the past 45 years of exploration have demonstrated the primacy of this Dungey cycle of plasma circulation at Earth. [3] Jupiter s magnetosphere is very different from the Earth s, both being dominated by the planet s rapid rotation (10-hour period) and the copious (one ton s 1 ) internal source of plasma provided by ionization of gasses from Io s active volcanoes. Another major difference is that Jupiter s magnetosphere is truly immense, spanning 10 7 km across and reaching at least 10 9 km down its tail as indicated by several apparent encounters of the Voyager 2 (V2) spacecraft with the magnetotail at distances of R J,as 1 Space Science and Engineering Division, Southwest Research Institute, San Antonio, Texas, USA. 2 Laboratory for Astrophysics and Space Physics, University of Colorado, Boulder, Colorado, USA. Copyright 2007 by the American Geophysical Union /07/2007GL031078$05.00 it approached Saturn [Kurth et al., 1982; Lepping et al., 1983; Goldstein et al., 1985]. These deep tail observations are consistent with tail-aligned fields and a hot (b > 1) plasma flowing down the tail. The vastness of Jupiter s magnetosphere on an absolute scale is important for solar wind interactions because the solar wind flows at the same speed (400 km s 1 ) past both planets. This means, for example, that the timescale for the solar wind to traverse the distance from the magnetosphere s nose to its terminator is 100 times longer at Jupiter than at Earth. At Jupiter, the extremely slow magnetosheath flows just outside the magnetopause, where reconnection occurs, take hundreds of hours to move back to the terminator. This long magnetopause transit time is demonstrated in a magnetohydrodynamic (MHD) model of the Jovian magnetosheath presented by Joy et al. [2006]. [4] In the 1970s it was recognized that the electric field of the rotationally dominated Jovian magnetosphere is considerably larger than that applied by the moving solar wind [Brice and Ioannidis, 1970; Vasyliunas, 1975]. Jupiter is clearly a rotation-driven magnetosphere, as demonstrated by numerous in situ observations and studies [Khurana et al., 2004; Krupp et al., 2004]. Iogenic plasma is picked up, accelerated to corotation speeds, and transported radially outwards on times scales of weeks. The plasma cannot continue rotating with the planet indefinitely [Hill et al., 1974] and on average, roughly a half ton s 1 must be lost down the magnetotail. Voyager observations of magnetospheric particles at tens of kev energies showed a distinct break in flow direction at 100 R J down the tail on the dawn side, indicating a magnetospheric wind [Krimigis et al., 1981]. Vasyliunas [1983] proposed that stretched magnetic field lines in the equatorial plasma sheet would reconnect and blobs of plasma (analogous to plasmoids at Earth) would be ejected down the magnetotail. Cowley et al. [2003] argued for a Dungey cycle at Jupiter with a conduit of return flow on dawn side. Badman and Cowley [2007] estimated parameters for such return flow, including the possibility of a tenuous, slow planetward flow. Observations by the Galileo spacecraft that made 34 orbits of Jupiter are generally consistent with the Vasyliunas picture [Krupp et al., 2004], but other than the very distant V2 encounters, no observations have been reported >150 R J down tail, where these theories can really be tested. [5] Walker and Russell [1985] argued that less frequent and weaker dayside reconnection events observed at Jupiter indicate a lower overall reconnection rate than at Earth. Based on Galileo observations, Joy et al. [2002] estimate that the half-width of Jupiter s magnetotail is R J. Even assuming a reconnection efficiency as large as 10%, as is often assumed for Earth, this would lead to an equatorward drift speed of 40 km s 1 (2 R J per hour) if L of5

2 Figure 1. The Dungey cycle for solar wind driven convection in a magnetosphere taken from Axford [1969]. A time sequence of flux tubes is numbered sequentially. In this cycle, the IMF interconnects with the Earth s magnetic field via magnetic reconnection creating open flux tubes, which are tied to one of the polar caps on one end and extend out into the solar wind on the other. These flux tubes are carried back over the poles and ExB drift down through the tail where they reconnect again, closing and returning flux to the Earth. open flux tubes penetrate the magnetosphere, and it would take 3 4 days for them to reach the equator. In this time, the section of the open magnetic flux tube in the solar wind moving at 400 km s 1 will have moved 1500 to 2000 R J downwind. Thus, we would expect a distant reconnection point or X-line to be R J down the magnetotail. Inside that distance, plasma would be drifting sunward at 40 km s 1 [Kivelson and Southwood, 2005], and would take hours to cover the R J necessary to return back to the planet. In this time Jupiter rotates 100 times and, if one assumes the flux tubes stay connected to the planet, this means 100 complete rotations of the foot points. In contrast, the equivalent drift from a neutral line at 200 R E in the Earth s magnetotail takes only 9 hours, less than half a day. In addition, at Jupiter the circulation speed is much smaller than the estimated plasma outflow sound speed [Joy et al., 2002] and planetward drift would have to occur against a significantly faster tailward flow of a planetary wind and detached plasma blobs that continuously carry a large fraction of a ton of ions per second down the tail. Given these timescales and strong tailward flow of high b plasma, it is almost impossible to imagine that an Earth-like convection can actually work at Jupiter. [6] While Dungey cycle return circulation seems unlikely to be a dominant process in the Jovian magnetosphere, the facts remain that 1) magnetic reconnection does occur, opening flux on the dayside Jovian magnetopause [Walker and Russell, 1985], 2) once opened, that flux must be carried back tailward by the solar wind and magnetosheath flow, and 3) this flux must ultimately close and be returned to the planet somehow. Thus, the real mystery is how and where the flux closes and what effects this process has on the overall magnetosphere compared to its rotationallydominated features. 2. Closing Open Magnetic Flux on the Magnetopause [7] Here we propose that, in contrast to Earth, the Jovian magnetosphere may contain little open magnetic flux and that this open flux remains largely confined to the outer regions of the magnetosphere. This is possible if the magnetic flux is returned via additional reconnection along the magnetopause. This proposal has the huge advantage that the returning flux tubes are already short and near the planet while any extended lobe-like field is shed down the tail. Interestingly, this basic reconnection topology was also suggested by Dungey [1963]. [8] Figure 2 shows one possible geometry for this sort of reconnection and displays how our proposed process works. The opening of magnetic flux is assumed to occur, just as at Earth, when the solar wind brings an IMF with a component opposite to the planetary magnetic field to the sub-solar Figure 2. Schematic diagram of how open magnetic flux can close at the magnetopause. (a) A significantly southward IMF (1) drapes around the dayside magnetopause (2, 3) until it reconnects with an oppositely-directed planetary field (stars). (b) Flux tube 3 reconnects near the magnetopause near both north and south cusps, creating newly closed (3 0 ) and disconnected (3) flux tubes. (Note that Jupiter s intrinsic magnetic field is opposite that of the Earth s so erosion near the nose occurs preferentially for northward IMF instead of southward.) After reconnection, the short, newly closed flux tube is free to work its way back toward a more normal closed shape (4 0 ), while the long, newly disconnected flux tube (4) is lost down the flanks of the tail. 2of5

3 Figure 3. Schematic views looking from the Sun toward the nose of the magnetopause. By 5 AU the IMF spiral (transverse) component is generally dominant and the sub-solar flow drapes the field around the lower latitude portions of the magnetosphere. (a) For outward sector IMF that are slightly northward, magnetic flux should be opened near the nose and pulled back (dark lines) over the top-right and bottom-left. (b) A change to inward sector IMF would produce reconnection of a single solar wind flux tube on both sides, re-close magnetic flux (dark line), and preferentially release disconnected flux tubes (U-shaped lines) down the top-right and bottom-left flanks for inward IMF sectors. Dashed lines indicate field topologies prior to reconnection. Similar reconnection topologies would be produced for inward sectors over the opposite quadrants. region of the magnetopause. We propose that subsequent reconnection of IMF flux tubes with open flux rooted in both magnetic poles then close the open flux. This type of double reconnection was also proposed to occur poleward of the cusps at Earth during intervals of northward IMF as a possible source of the Low Latitude Boundary Layer (LLBL) [Song and Russell, 1992]. A number of detailed studies have demonstrated that under the right conditions, this process does occur at Earth [Lavraud et al., 2005]. While largely southward IMF is probably rare at Jupiter, Figure 2 displays the general principle of how closed magnetic flux is returned to the planet whenever a single solar wind flux tube reconnects with two open flux tubes anchored to alternate poles of the planet. [9] Reconnection configurations for the more typical, tightly wound IMF spiral at Jupiter are shown looking from the Sun toward Jupiter in Figure 3. For simplicity, the figure shows reconnection and closing of open flux symmetrically on the dayside. Nevertheless, 1) any open flux tubes from the two poles can reconnect with the external field to close flux, 2) the magnetic field is hung up on the magnetopause for extremely long times (hundreds of hours), 3) the two reconnection events do not need to occur simultaneously and can occur at very different times (both local and absolute) over the long magnetopause transit timescale, and 4) this process would work equally well even after the field drapes back further along the flanks. Finally, it is important to remember that these are just simplified schematics showing the basic principle. Other, more complex geometries are possible, particularly once the effects of rotation are included. For example, Isbell et al. [1984] showed how rotation might twist up open flux tubes in the polar regions of Jupiter s and Saturn s magnetospheres. [10] With reconnection at the magnetopause returning flux, the extent of long lobe-like fields along the flanks of the tail and the depth to which they ExB drift into the tail depend on the residence time of open flux tubes before they re-reconnect. Over each 10-hour rotation period, a section of an open flux tube out in the distant magnetosheath moves back 200 R J while the part of the flux tube within the magnetosphere should sink no more than 20 R J into the tail, only about one tenth of the distance from the magnetopause to the equator. The residence time undoubtedly varies with the IMF variability and may change precipitously at sector boundary crossings. Importantly, the process we have described does provide an avenue for the observed escape of particles and entry of solar wind ions (e.g., He ++ ) into the Jovian magnetosphere [Geiss et al., 1992]. 3. Discussion [11] There are a number of independent lines of evidence that support our basic picture. While some analyses of auroral images suggest that there may be a small polar cap-like structure up to 10 across [Pallier and Prange, 2004], much of the time it is hard to identify any clear cap region at all [Clarke et al., 2004]. In either case, a very small polar cap is another direct line of evidence that the magnetosphere cannot be very similar to the terrestrial case. Instead, small amounts of flux that map into the high latitude aurora could be opening and closing all the time as various flux tubes interconnect with the solar wind and then disconnect again along the magnetopause. In fact, these magnetopause interactions could well be related to the frequently observed small patches of brightening in the polar aurora that vary rapidly in location and time [Clarke et al., 2004; Grodent et al., 2003]. [12] Encounters of the very deep Jovian magnetotail by V2 occurred with remarkable periodicity at very nearly the solar rotation rate. Further, they correlated with stream rarefaction regions generally, and a clear sector boundary crossing in at least one example [Kurth et al., 1982; Lepping et al., 1983; Goldstein et al., 1985]. While no definitive physical process was identified for this correlation at the time, the enhanced closing of open magnetic flux and release of tail-like field is a natural consequence of our proposed magnetopause reconnection whenever the IMF reverses. Such reversals occur at solar wind sector boundaries. Interestingly, cometary tail disconnection events are also well correlated with sector boundaries and thought to be reconnection between roughly oppositely directed flux tubes slowed by cometary pickup ions [Brandt et al., 1999], 3of5

4 very similar to what we have described here (except of course with no connection to an internal magnetic field of the obstacle). [13] Another aspect of the V2 deep tail observation that supports our premise is the lack of any significantly twisted magnetic flux. A detailed analysis of the magnetic field observations during these intervals [Goldstein et al., 1985] showed only a very slight twist (pitch angle of 2 3 )tothe field. While the field should start untwisting after becoming disconnected in the tail, we would still expect a significantly twisted field if it had been connected to Jupiter for 100 days (rotations) prior to disconnection. [14] The Ulysses spacecraft flew through the Jovian magnetosphere reaching magnetic latitudes poleward of ±45. Several instruments observations support the interpretation of some open, polar cap or cusp-like region for two high latitude excursions of 34 and 47 at 15 and 8.7 R J, respectively [Bame et al., 1992; Smith and Wenzel, 1993]. This interpretation is consistent with previous Pioneer 10 and 11 observations that the Jovian magnetospheric cusps were at much lower latitude than at Earth [Thomas and Jones, 1984]. In fact, the combined observations from these three spacecraft locate this low latitude open field region near dawn, but not near local noon [Smith and Wenzel, 1993]. These results were surprising from the perspective of an Earth-like Dungey cycle where flux would be pulled back over the poles, but seem natural if flux is opened along the flanks and then closed and returned by low-latitude reconnection as suggested here. [15] Combining the idea of closing reconnection on the magnetopause with the slow penetration of open flux tubes into the lobes leads us to conclude that the open region of Jupiter s magnetosphere is likely confined to a boundary layer within a few 10s of R J of the magnetopause. If magnetopause reconnection is the primary method for both opening and closing magnetic flux, then the Jovian magnetotail should contain essentially entirely tailward flow of largely planetary plasma. Along the flanks, inside the magnetopause, we would expect alternately open flux that is still connected to the planet and disconnected flux tubes, which have been recently released by magnetopause reconnection back nearer the planet, depending on the time history of the IMF conditions. [16] While described for Jupiter, the process developed here may also be important for magnetospheres of other outer planets, where the IMF spiral is tightly wound, and other astrophysical objects where reconnection occurs between a magnetized body and a largely transverse magnetic field. Thus, we have proposed a new reconnection cycle that can explain the open aspects of the Jovian magnetosphere while simultaneously allowing flux to return without the need for a magnetotail reconnection point and its requisite (but unlikely) planetward flow against a strong downtail flow. References Axford, W. I. (1969), Magnetospheric convection, Rev. Geophys., 7, Badman, S. V., and S. W. H. Cowley (2007), Significance of Dungey-cycle flows in Jupiter s and Saturn s magnetospheres, and their identification on closed equatorial field, Ann. Geophys., 25, 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 (1992), Jupiter s magnetosphere: Plasma description from the Ulysses flyby, Science, 257, Brandt, J. C., F. M. Caputo, J. T. Hoeksema, M. B. Niedner, Y. Yi, and M. Snow (1999), Disconnection events (DEs) in Halley s comet : The correlation with crossings of the heliospheric current sheet, Icarus, 137, Brice, N. M., and G. A. Ioannidis (1970), The magnetospheres of Jupiter and Earth, Icarus, 13, Clarke, J. T., D. Grodent, S. W. H. Cowley, E. J. Bunce, P. Zarka, J. E. P. Connerney, and T. Satoh (2004), Jupiter s aurora, in Jupiter: The Planet, Satellites and Magnetosphere, vol. 1, edited by F. Bagenal, T. E. Dowling, and W. B. McKinnon, pp , Cambridge Univ. Press, New York. Cowley, S. W. H., E. J. Bunce, T. S. Stallard, and S. Miller (2003), Jupiter s polar ionospheric flows: Theoretical interpretation, Geophys. Res. Lett., 30(5), 1220, doi: /2002gl Dungey, J. W. (1961), Interplanetary field and auroral zones, Phys. Rev. Lett., 6, Dungey, J. W. (1963), The structure of the exosphere, or adventures in velocity space, in Geophysics: The Earth s Environment, edited by C. DeWitt, J. Hieblot, and A. Lebeau, pp , Gordon Breach, New York. Geiss, J., G. Gloeckler, H. Balsiger, L. A. Fisk, A. B. Galvin, F. Gliem, D. C. Hamilton, F. M. Ipavich, S. Livi, and U. Mall (1992), Plasma composition in Jupiter s magnetosphere: Initial results from the solar wind ion composition spectrometer, Science, 257, Goldstein, M. L., R. P. Lepping, and E. C. Sittler Jr. (1985), Magnetic field properties of Jupiter s tail at distances from 80 to 7500 Jovian radii, J. Geophys. Res., 90, Grodent, D., J. T. Clarke, J. H. Waite Jr., S. W. H. Cowley, J.-C. Gérard, and J. Kim (2003), Jupiter s polar auroral emissions, J. Geophys. Res., 108(A10), 1366, doi: /2003ja Hill, T. W., A. J. Dessler, and F. C. Michel (1974), Configuration of the Jovian magnetosphere, Geophys. Res. Lett., 1, 3 6. Isbell, J., A. J. Dessler, and J. H. Waite (1984), Magnetospheric energization by interaction between planetary spin and the solar wind, J. Geophys. Res., 89, 10,716 10,722. Joy, S. P., M. G. Kivelson, R. J. Walker, K. K. Khurana, C. T. Russell, and T. Ogino (2002), Probabilistic models of the Jovian magnetopause and bow shock locations, J. Geophys. Res., 107(A10), 1309, doi: / 2001JA Joy, S. P., M. G. Kivelson, R. J. Walker, K. K. Khurana, C. T. Russell, and W. R. Paterson (2006), Mirror mode structures in the Jovian magnetosheath, J. Geophys. Res., 111, A12212, doi: /2006ja Khurana, K. K., M. G. Kivelson, V. M. Vasyliunas, N. Krupp, J. Woch, A. Lagg, B. H. Mauk, and W. S. Kurth (2004), The configuration of Jupiter s magnetosphere, in Jupiter: The Planet, Satellites and Magnetosphere, vol. 1, edited by F. Bagenal, T. E. Dowling, and W. B. McKinnon, pp , Cambridge Univ. Press, New York. Kivelson, M. G., and D. J. Southwood (2005), Dynamical consequences of two modes of centrifugal instability in Jupiter s outer magnetosphere, J. Geophys. Res., 110, A12209, doi: /2005ja Krimigis, S. M., J. F. Carbary, E. P. Keath, C. O. Bostrom, W. I. Axford, G. Gloeckler, L. J. Lanzerotti, and T. P. Armstrong (1981), Characteristics of hot plasma in the Jovian magnetosphere: Results from the Voyager spacecraft, J. Geophys. Res., 86, Krupp, N., et al. (2004), Dynamics of the Jovian magnetosphere, in Jupiter: The Planet, Satellites and Magnetosphere, vol. 1, edited by F. Bagenal, T. E. Dowling, and W. B. McKinnon, pp , Cambridge Univ. Press, New York. Kurth, W. S., D. A. Gurnett, J. D. Sullivan, H. S. Bridge, F. L. Scarf, and E. C. Sittler Jr. (1982), Observations of Jupiter s distant magnetotail and wake, J. Geophys. Res., 87, 10,373 10,383. Lavraud, B., M. F. Thomsen, M. G. G. T. Taylor, Y. L. Wang, T. D. Phan, S. J. Schwartz, R. C. Elphic, A. Fazakerley, H. Rème, and A. Balogh (2005), Characteristics of the magnetosheath electron boundary layer under northward interplanetary magnetic field: Implications for highlatitude reconnection, J. Geophys. Res., 110, A06209, doi: / 2004JA Lepping, R. P., M. D. Desch, E. C. Sittler Jr., K. W. Behannon, L. W. Klein, J. D. Sullivan, and W. S. Kurth (1983), Structure and other properties of Jupiter s distant magnetotail, J. Geophys. Res., 88, Pallier, L., and R. Prange (2004), Detection of the southern counterpart of the Jovian northern polar cusp: Shared properties, Geophys. Res. Lett., 31, L06701, doi: /2003gl Smith, E. J., and K.-P. Wenzel (1993), Introduction to the Ulysses encounter with Jupiter, J. Geophys. Res., 98, 21,111 21,127. Song, P., and C. T. Russell (1992), Model of the formation of the lowlatitude boundary layer for strongly northward interplanetary magnetic field, J. Geophys. Res., 97, Thomas, B. T., and D. E. Jones (1984), Modeling Jupiter s magnetospheric currents using Pioneer data: Evidence for a low-latitude cusp, J. Geophys. Res., 89, of5

5 Vasyliunas, V. M. (1975), Concepts of magnetospheric convection, in The Magnetospheres of the Earth and Jupiter, vol. 52, edited by V. Formisano, pp , D. Reidel, Dordrecht, Netherlands. Vasyliunas, V. (1983), Plasma distribution and flow, in Physics of the Jovian Magnetosphere, edited by A. J. Dessler, pp , Cambridge Univ. Press, New York. Walker, R. J., and C. T. Russell (1985), Flux transfer events at the Jovian magnetopause, J. Geophys. Res., 90, F. Bagenal, Laboratory for Astrophysics and Space Physics, University of Colorado, Duane Physics UCB 392, Boulder, CO , USA. D. J. McComas, Space Science and Engineering Division, Southwest Research Institute, P.O. Drawer 28510, San Antonio, TX , USA. 5of5