Discovery of co-rotating magnetic reconnection in Saturn's magnetosphere Z. H. Yao1, 2*, A. J. Coates1,3, L. C. Ray1, 4, I. J. Rae1, D. Grodent2, G. H. Jones1,3, M. K. Dougherty5, C. J. Owen1, R. L. Guo6, W. Dunn1,3, A. Radioti2, Z. Y. Pu7, G. R. Lewis1, J. H. Waite8 and J.-C. Gérard2 1
UCL Mullard Space Science Laboratory, Dorking, RH5 6NT, UK.
2
Laboratoire de Physique Atmosphérique et Planétaire, STAR institute, Université de Liège, Liège, Belgium. 3
Centre for Planetary Sciences at UCL/Birkbeck, Gower Street, London WC1E 6BT, UK
4
Department of Physics, Lancaster University, Lancaster, UK.
5
Imperial College of Science, Technology and Medicine, Space and Atmospheric Physics Group, Department of Physics, London SW7 2BW, UK 6
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, China.
7
School of Earth and Space Sciences, Peking University, Beijing, China.
8
Southwest Research Institute, San Antonio, TX, United States.
*Correspondence to:
[email protected]. Magnetic reconnection, as a fundamental energy conversion mechanism, can explosively convert energy contained in a magnetic field into ionized particles in solar system, astrophysical plasmas and laboratory plasmas1,2. Planetary magnetic reconnection can be driven by solar wind energy or planetary internal energy3-5. Externally driven processes, from the solar wind, have been extensively investigated at Earth and Mercury6,7. Internally driven processes dictate important dynamics in giant planetary magnetospheres3,5,8,9; However, it is not yet established how this process drives magnetospheric energy release and re-loading. Here, using measurements from the Cassini spacecraft in Saturn’s magnetosphere, we propose a creative 3D co-rotating reconnection model, which is driven by an internal source. Our results demonstrate that co-rotating magnetic reconnection can drive an expansion of the current sheet in Saturn’s magnetosphere, and consequently produce field-aligned electron acceleration. We reveal the nature of 3D reconnection in giant planets, and show the existence of a co-rotating reconnection process in the universe. Our co-rotating reconnection picture can be widely applied to the fast rotating magnetized plasma environments in the solar system and beyond. For example, magnetic reconnection is a potential heating source for the interstellar medium and halo gas10. The terrestrial magnetospheres receive energy from the Sun via magnetopause reconnection, and occasionally explosively release this energy, causing strong perturbations within their magnetospheres, ionospheres and atmospheres. Electrical currents that flow along the magnetic field couple the magnetosphere and ionosphere, disturbing the polar geomagnetic field and powering the aurorae11,12. Magnetospheric perturbations and auroral intensifications also exist at other planets, including the giant planets (i.e., Saturn and Jupiter)13. Unlike at Earth, internally
generated plasma sources (moons and rings) at the giant planets play an important role in driving dynamics in planetary magnetotails4,5. The centrifugal forces caused by the rotation of the giant planets are usually considered to be the main mechanism in driving internal plasma dynamics in a process referred to as the Vasyliūnas cycle14. The solar wind controlled, externally driven large-scale magnetospheric circulation of energy defined for Earth’s magnetosphere is well known as the Dungey cycle15. Although the Dungey cycle plays a major role in transporting energy to Earth’s nightside magnetotail, it is well accepted that the Dungey cycle does not directly drive the major explosive energy release (EEC) process at the Earth, termed terrestrial substorms. Instead, the disruption of near-Earth currents is the major reason for the substorm EECs6,16. A near-Earth current disruption results in a re-configuration of the global magnetic field (also known as magnetic dipolarization), and consequently releases the majority of free energy of the magnetotail. On the other hand, it is widely accepted that the Vasyliūnas cycle is important in driving the dynamics of giant magnetospheres. The Vasyliūnas cycle is often described in models3,4,14, or shown in simulations17. Evidence from in-situ measurements is rare, although the quasi-periodicity of the particle and magnetic field perturbations at Jupiter implies that the planet’s rotation may participate in planetary energy re-loading18. So far, it is still unclear how the Vasyliūnas cycle drives the dynamics of giant magnetospheres. Energetic particle injections in the inner Saturnian magnetosphere have been shown to be related to radial transport, driven by the centrifugal interchange instability8. Here we report the first direct observation of an internally driven substorm-like EEC and the subsequent re-loading of magnetic energy at Saturn. Regarding the different environments between Saturn and Earth, we hereby define the substorm-like EEC as a process that includes current sheet expansion, particle acceleration and magnetic dipolarization. Figure 1(a-c) shows the magnetic field data from the Cassini magnetometer19 in Kronographic Radial-Theta-Phi coordinates during 20 and 21 September 2006. During this period, Cassini was located post-midnight between 0.9 and 1.5 Saturn local time, at latitudes from ~ 9.4oN to 5.8oN north of Saturn’s equatorial plane, and radius R ~ 30 to 35 RS (1 RS~60268 km). Periodic current sheet crossings caused by Saturn’s rotation are identified by the reversals of Br and BΦ (Br, BΦ ~ 0 are marked by the vertical dashed lines). These current sheet crossings are accompanied by enhancements of electron flux (Figure 1d) with energies at tens to hundreds of eV. The red rectangle indicates a significant Bθ component magnetic field spike at around 21:30 UT, accompanied by an electron flux enhancement, which represents magnetic dipolarization20. We define the measurements made one Saturn rotation period before the event as pre-event (a baseline), and from one Saturn rotation period after the events as post-event. Saturn’s magnetospheric rotation period is time variable21. We here determine a period from the largescale current sheet crossing, which can better represent the magnetospheric rotation. As indicated by the magenta arrows in Figure 1a, the time difference between the pre-event and dipolarization event is 11 hours. Moreover, the major change of magnetic field components Bθ and BΦ between the dipolarization event and post-event is highly consistent. It is thus reasonable to adopt 11 hours as the magnetospheric rotation period during our event. In Figure 2, we compare the measurements amongst the dipolarization event between 20:30 UT and 22:30 UT, pre-event (between 20:30 UT – 11 hours and 22:30 UT – 11 hours), and postevent (between 20:30 UT + 11 hours and 22:30 UT + 11 hours). The magnetic field variations shown in Figure 2 are generally consistent among the three periods before 21:30 UT, suggesting that the three periods of measurements were made in the same region of the rotating magnetosphere as the planet rotated twice. In the pre-event period, the magnetic field was not
significantly perturbed, with a dominant component Br ~ 3 nT, and relatively low electron flux, suggesting Cassini was in the outer plasma sheet or lobe region22. One rotation later, a significant magnetic dipolarization signature (Br decrease and Bθ increase) was observed around 21:30 UT, followed by Bθ decreasing to negative. Meanwhile, the electron flux was enhanced during the magnetic dipolarization. The Br decrease, Bθ increase and electron flux enhancements are typical features of current sheet expansion and magnetic dipolarization in Earth’s magnetotail16,23, so this event satisfies our definition of a substorm-like EEC event. Electrons are accelerated both parallel and antiparallel to the magnetic field direction, which is further confirmed as Fermi acceleration, a consequence of current sheet expansion (Extended Data Fig. 1 and Methods). The acceleration is efficient for electrons with energies at few hundred eV to several keV. For example, the differential energy flux in the dipolarization region for 1 keV electrons is about three times higher than the current sheet population. As a consequence of current sheet expansion, the Fermi acceleration differs from the reconnection acceleration in Saturn’s magnetotail, which is efficient in producing tens of keV electrons24. In addition to the previous reconnection evidence, the clear BΦ decrease that accompanies the Bθ increase is likely driven by the magnetic tension force when reconnection suddenly reduces the length of the magnetic field line. The BΦ decrease may also relate to Saturn’s supercorotational return flow phenomenon25. In the post-event measurements, a Bθ enhancement and BΦ decrease are also observed. Obviously, the variations in post-event measurements are similar to the dipolarization event measurements, although with a smaller amplitude. We thus conclude that the post-event status is developed from the dipolarization condition after a significant magnetic energy re-loading process. We use Figure 2g to illustrate the evolution of the magnetic field configuration from pre-event to postevent. Pre-event, the magnetic field was highly stretched (blue), and Cassini was in the lobe (or outer plasma sheet) region. After the substorm-like EEC event, the magnetic field is dipolarized (black), and Cassini, relatively, moves into the plasma sheet. Thereafter, the re-loading process stretches the magnetic field to approach the pre-event configuration (red). To explain the Bθ positive to negative feature in the dipolarization event and the post-event, we propose a co-rotating magnetic reconnection model, as presented in Figure 3. As the magnetic reconnection region co-rotates with Saturn, Cassini’s relative position moves from Saturnward (Figure 3a) to tailward (Figure 3b), consequently observing the reversal of Bθ (Figure 3c), together with the decrease of BΦ (Figure 3d). In previous literature, the Bθ reversal is usually described as a tailward plasmoid26, which is based on a 2D model, i.e., not including azimuthal variation. However, the retreat of a 2D plasmoid structure cannot well explain a similar Bθ reversal signature during post-event (Figure 2b), which can be well explained by our co-rotating magnetic reconnection model. The azimuthal scale of the reconnection line in our event is about 2 MLT (or ~ 1.2×106 km), corresponding to the ~ 1 hour duration between 21:30:00 UT and 22:30:00 UT. We would like to point out that the co-rotating aurora breakup with a 3 MLT azimuthal scale is presented in previous literature, although observations there have not been discussed in the context of a 3D model27. The counter-part co-rotating aurora breakup region is a strong support to our model. In another substorm-like EEC event on 7 August 2009 (Extended Data Fig. 2), Cassini only detected the Saturnward part of the co-rotating magnetic reconnection region twice (dipolarization event and post-event). We suggest that this is because Cassini was far from the reconnection site. The agreement between our co-rotating reconnection model and the Cassini in-situ measurements is striking, indicating that the reversal of Bθ cannot be fully explained by a 2D magnetic reconnection retreat model, which can either be driven by a solar wind or internal
source. However, our co-rotating reconnection model can only be driven by an internal source, since the tailward region of solar wind driven reconnection is disconnected from Saturn’s magnetosphere and thus cannot be observed after one rotation. The reloading process of magnetic energy after a substorm-like EEC is longer than 11 hours, and hence significantly longer than the 3-4 hours typically measured at Earth11. The discovery of co-rotating reconnection can be widely applied in spinning stars with plasma environments. We also expect a similar process in Jupiter’s magnetosphere, which generates the most powerful aurora in solar system. Moreover, reconnection in rotating pulsars and magnetar magnetosphere is important in accelerating the pulsar wind, or generate soft X-ray28-30. Acknowledgements: ZY, IJR, CJO, AJC and GHJ are supported by a UK Science and Technology Facilities Council (STFC) grant (ST/L005638/1) at UCL/MSSL. ZY is also a Marie-Curie COFUND postdoctoral fellow at the University of Liege. Co-funded by the European Union. ZY warmly thanks the very useful and fresh discussion from Jennifer Y. H. Chan at MSSL/UCL. Cassini operations are supported by NASA (managed by the Jet Propulsion Laboratory) and ESA. The data presented in this paper are available from the NASA Planetary Data System http://pds.jpl.nasa.gov. Author contributions: Z.Y. identified the event from Cassini dataset, and led the analysis in this paper. A.J.C., L.C.R., I.J.R., D.G., C.J.O., G.H.J. and J.H.W. provided detailed assistance with the analysis of magnetic field and particle measurements. M.K.D. provided Cassini/MAG data. R.L.G., W.D., A.R., Z.Y.P. and J.C.G provided detailed discussions on the interpretation and the broaden significance. G.R.L. provided support in Cassini/ELS data analysis. All authors have participated in writing the manuscript. References 1 Li, C. et al. Observation of megagauss-field topology changes due to magnetic reconnection in laserproduced plasmas. Physical Review Letters 99, 055001 (2007). 2 Yokoyama, T. & Shibata, K. Magnetic reconnection as the origin of X-ray jets and Hα surges on the Sun. Nature 375, 42-44, doi:10.1038/375042a0 (1995). 3 Delamere, P., Otto, A., Ma, X., Bagenal, F. & Wilson, R. Magnetic flux circulation in the rotationally driven giant magnetospheres. Journal of Geophysical Research: Space Physics 120, 4229-4245 (2015). 4 Kivelson, M. & Southwood, D. Dynamical consequences of two modes of centrifugal instability in Jupiter's outer magnetosphere. Journal of Geophysical Research: Space Physics 110 (2005). 5 Kronberg, E. et al. A possible intrinsic mechanism for the quasi‐periodic dynamics of the Jovian magnetosphere. Journal of Geophysical Research: Space Physics 112 (2007). 6 Baker, D. N., Pulkkinen, T. I., Angelopoulos, V., Baumjohann, W. & McPherron, R. L. Neutral line model of substorms: Past results and present view. Journal of Geophysical Research-Space Physics 101, 1297513010, doi:10.1029/95ja03753 (1996). 7 Slavin, J. A. et al. MESSENGER observations of magnetic reconnection in Mercury’s magnetosphere. Science 324, 606-610 (2009). 8 Hill, T. et al. Evidence for rotationally driven plasma transport in Saturn's magnetosphere. Geophysical Research Letters 32 (2005). 9 Roussos, E. et al. Quasi-periodic injections of relativistic electrons in Saturn’s outer magnetosphere. Icarus 263, 101-116 (2016). 10 Beck, R. Magnetic field structure from synchrotron polarization. EAS Publications Series 23, 19-36 (2007). 11 Akasofu, S. I. The Development of the Auroral Substorm. Planetary and Space Science 12, 273-282, doi:10.1016/0032-0633(64)90151-5 (1964). 12 McPherron, R., Russell, C., Kivelson, M. & Jr. Coleman, P. Substorms in space: The correlation between ground and satellite observations of the magnetic field. Radio Sci. 8, 1059-1076, doi:10.1029/RS008i011p01059 (1973). 13 Clarke, J. T. et al. Morphological differences between Saturn's ultraviolet aurorae and those of Earth and Jupiter. Nature 433, 717-719 (2005). 14 Vasyliunas, V. Plasma distribution and flow. Physics of the Jovian magnetosphere 1, 395-453 (1983). 15 Dungey, J. W. Interplanetary magnetic field and the auroral zones. Physical Review Letters 6, 47 (1961).
16 17 18 19 20
21 22 23 24 25 26 27 28 29 30
Lui, A. T. Y. Current disruption in the Earth's magnetosphere: Observations and models. Journal of Geophysical Research-Space Physics 101, 13067-13088, doi:10.1029/96ja00079 (1996). Jia, X. et al. Magnetospheric configuration and dynamics of Saturn's magnetosphere: A global MHD simulation. Journal of Geophysical Research: Space Physics 117 (2012). Kronberg, E. et al. Mass release at Jupiter: Substorm‐like processes in the Jovian magnetotail. Journal of Geophysical Research: Space Physics 110 (2005). Dougherty, M. K. et al. The Cassini Magnetic Field Investigation. Space Science Reviews 114, 331-383, doi:10.1007/s11214-004-1432-2 (2004). Jackman, C. et al. Field dipolarization in Saturn's magnetotail with planetward ion flows and energetic particle flow bursts: Evidence of quasi‐steady reconnection. Journal of Geophysical Research: Space Physics 120, 3603-3617 (2015). Gurnett, D. et al. Discovery of a north‐south asymmetry in Saturn's radio rotation period. Geophysical Research Letters 36 (2009). Arridge, C. S. et al. Plasma electrons in Saturn's magnetotail: Structure, distribution and energisation. Planetary and Space Science 57, 2032-2047 (2009). Angelopoulos, V. et al. Tail reconnection triggering substorm onset. Science 321, 931-935, doi:10.1126/science.1160495 (2008). Arridge, C. S. et al. Cassini in situ observations of long-duration magnetic reconnection in Saturn's magnetotail. Nature Physics (2015). Masters, A. et al. Supercorotating return flow from reconnection in Saturn's magnetotail. Geophysical Research Letters 38 (2011). Vogt, M. F. et al. Structure and statistical properties of plasmoids in Jupiter's magnetotail. Journal of Geophysical Research: Space Physics 119, 821-843 (2014). Radioti, A. et al. A multi-scale magnetotail reconnection event at Saturn and associated flows: Cassini/UVIS observations. Icarus 263, 75-82 (2016). Contopoulos, I. The role of reconnection in the pulsar magnetosphere. Astronomy & Astrophysics 466, 301307 (2007). Kirk, J. & Skjaeraasen, O. in International Cosmic Ray Conference. 2418. Rea, N. et al. A low-magnetic-field soft gamma repeater. Science 330, 944-946 (2010).
Figure 1 | Overview of 1-min resolution magnetic field Br, Bθ and BΦ components from the Cassini magnetometer (MAG) (a-c) in Kronographic Radial-Theta-Phi coordinates, and (d) electron differential energy flux from the Cassini electron spectrometer (CAPS-ELS). The two black arrows in (d) represent the electron population in dipolarization region and central plasma sheet region, which are compared in Extended Data Fig. 1.
Figure 2 | A comparison of 1-min resolution magnetic field (a-c) and electron differential energy flux (d-f) during the unperturbed period (dashed blue magnetic field and panel (d) for electron flux); dipolarization period (solid black magnetic field and (e) for electron flux); and postdipolarization (dashed red magnetic field and (f) for electron flux) periods. The unperturbed measurements are the in-situ Cassini measurements shifted forward by 11 hours, and the postdipolarization measurements are the in-situ Cassini measurements shifted backward by 11 hours, (g) Geometry of the magnetic field topology and schematic of Cassini’s location (star) relative to the current sheet. The blue magnetic field line illustrates the stretched magnetic field, which represents a thin current sheet condition. The black and red curves represent the dipolarized and post-dipolarization magnetic field lines, respectively.
Figure 3 | Comparison between the magnetic reconnection model and in-situ measurements. (a) A sketch showing the reconnection line (red curve), and that the spacecraft (green rectangle) was Saturnward of the reconnection line. (b) After a small rotation from (a), the spacecraft was tailward of the reconnection line. (c and d) The expected variation of magnetic components Bθ and BΦ during the rotation (from a to b). Note that the red dotted curves represent the measurements expected from t + 1 rotation period to the black curves.
