Pages 655-658


D.E. Larson1, R.P. Lin1, R.E. Ergun1, J.M. McTiernan1, J.P. McFadden1, C.W. Carlson1, K.A. Anderson1, M. McCarthy2, G.K. Parks2 H. Reme3, T.R. Sanderson4, M. Kaiser5 ,and R. P. Lepping5

1Space Sciences Laboratory, University of California, Berkeley, E-mail:
2Geophysics Program, University of Washington, Seattle
3Centre d'Etude Spatiale des Rayonnements, Toulouse, France
4Space Science Department of ESA, ESTEC, Noordwijk, Netherlands
5NASA Goddard Space Flight Center, Greenbelt, Md.


We present ~0.1-102 keV electron observations from the 3-D Plasma and Energetic Particles experiment of a magnetic cloud which passed by the WIND spacecraft on October 18 - 20, 1995. Although the magnetic field exhibits the smooth, continuous rotation signature of a helical flux rope, the electron observations show numerous abrupt discontinuities in flux level and anisotropy within the cloud. In addition, five solar impulsive electron events, accompanied by solar type III radio bursts, were observed while the cloud was passing by the spacecraft. The electron/type III events showed that some of the cloud field lines were magnetically connected to solar active region 7912. Analysis of the velocity dispersion in the arrival times of the solar impulsive electrons provides the length of magnetic field lines. The field line length is found to be longest (~4 AU) near the exterior with a minimum (~1.5 AU) near the center of the cloud, consistent with a helical flux rope model. We interpret the discontinuities in electron fluxes as evidence of disconnection of magnetic field lines from the Sun.


Magnetic clouds are characterized by relatively strong magnetic fields and a smooth rotation of the magnetic field direction over a ~1 day period, consistent with the passage of an approximately force-free helical magnetic flux rope of diameter ~0.25 AU (Burlaga et al. 1988). The ends of the flux rope presumably are connected to the solar surface. Magnetic clouds are a type of Coronal Mass Ejection (CME) and often exhibit CME characteristics such as enriched alpha content, suppressed proton temperature and bidirectional electron streaming. The ~0.1 to ~102 keV electron observations from the WIND 3-D Plasma and Energetic particles experiment can provide substantial new information on the magnetic structure and topology of clouds. Electrons are excellent tracers of magnetic field lines since they are fast and have very small gyroradii. At low energies, ~0.1 - 1 keV, the interplanetary electron fluxes are normally dominated by the continuous outflow of hot coronal electrons--the solar wind halo and strahl. McComas et al. (1989) pointed out that, on occasion, these heat flux electrons are observed to drop out (HFDs), suggesting the disconnection of the interplanetary magnetic field from the hot solar corona. Lin and Kahler (1992) examined higher energy, >2 keV, solar electrons at the times of McComas et al.'s HFDs and found that in most of them the IMF was still connected to the Sun. Disconnections are required to prevent the indefinite buildup of magnetic flux in the interplanetary medium. The problem, as first discussed by Gosling (1975) and then by MacQueen (1980), is that we observe the expulsion of new magnetic flux from the Sun into the interplanetary medium in coronal mass ejections (CMEs) but we do not observe enough compensating disconnection of magnetic flux, necessary to keep the observed values at roughly constant levels in time (King, 1979).

At higher energies (~>1 to ~102 keV), the Sun often impulsively accelerates electrons in flare or flare-like events (see Lin 1985 for review). As these electrons escape into the interplanetary medium they produce solar type III radio bursts, which can be tracked by radio observations. By analyzing the velocity dispersion of the arrival of these impulsively accelerated electrons at 1 AU, information about the field line length can be obtained. Here we use the WIND electron observations to trace the October 18-20, 1995 cloud magnetic field back to a specific solar active region, and to show that the cloud field line lengths vary in a way that is consistent with a twisted flux tube. In addition we show evidence that much of the cloud may already be magnetically disconnected from the Sun in a patchy way.


The October 18-20, 1995 magnetic cloud is an excellent example of a magnetic cloud. The magnetic field strength (|B|) and direction (,) are shown in the top three panels (a, b, c) of Figure 1. The field rotates smoothly from south to north during the 30 hour passage of the cloud. Burlaga et al. (1996) have fit this signature to a force-free flux-rope geometry and estimated the axis of the cloud to be nearly in the ecliptic plane ( = -10 °), near to the Parker spiral angle ( = 291 °) with a right handed chirality. In addition they determined that WIND passed very close to the central axis, (y0/R0=0.087) and the diameter was about 0.27 AU. The leading edge of the cloud was identified by Burlaga et al. (1996) to be coincident with the passage of a tangential discontinuity in the magnetic field at 19:08 on October 18, 1996. The trailing edge of the cloud is not as clearly identifiable. We have chosen the trailing edge to be coincident with a minimum in |B| at 01:36 on October 20.

Figure 1. Summary plot of cloud event showing the magnetic field magnitude (|B|) and direction (, ) in the top three panels (a, b, c). Panel (d) is a spectrogram of the radio observations from 14 MHz to 10 kHz from the WIND Waves experiment (Bougeret et al 1995). Panel (e) shows the electron flux for electrons traveling anti-parallel (135°<<180°) to the magnetic field direction for 17 logarithmically-spaced energy channels ranging in energy from 100 eV to 60 keV. The following spectrogram (panel f), present the ratio,F/F0, electron flux (F) divided by an average flux (F0) measured during a stable period, to show the impulsive solar electron events. From the flare time (marked by red arrows) and the arrival time of electrons as a function of energy, the field line length at each point was determined and is shown in panel (h). Magnetic connection to or disconnection from the Sun from the heat flux dropouts is indicated in panel g.

Figure 1d is a color spectrogram of the radio observations covering frequencies between 14 MHz and 10 kHz provided by the WAVES instrument on WIND (Bougeret et al. 1995). The numerous fast drift bursts, from high to low frequency, are solar type III radio bursts.

Panel e of Figure 1 shows the flux of 135 °- 180 ° electrons in 17 energy channels logarithmically spaced between ~0.1 and 100 keV, respectively. Since the interplanetary field azimuth,  = 291 ° these electrons are streaming away from the Sun. In contrast to the smooth rotation of the magnetic field, the electrons exhibit many abrupt, discontinuous drops in flux level, occurring simultaneously at all energies. When the full 3-D angular distributions measured by the experiment are examined, it is found that these dropouts are heat flux dropouts (HFDs). Assuming that these HFDs indicate magnetic disconnections (see McComas et al. 1989), Figure 1g plots the connectivity back to the Sun. Note that the disconnected regions range from a few minutes to hours long and they are intertwined with connected regions.

Panel f of Figure 1 shows a color spectrogram of the ratio F/F0 of the electron flux, F, for pitch angles 135°<<180°, divided by an average background flux (F0) in the absence of impulsive events, to illustrate the velocity dispersion in the arrival of energetic electrons impulsively accelerated at the Sun. Each of the enhancements in electron flux can be attributed to a flare observed by the Yohkoh Soft X-ray Telescope within solar active region 7912. The accelerated electrons escape outward along open field lines and can be detected by the 3D Plasma experiment if the WIND spacecraft is on a field line connected to the flare site. The crosses in Figure 1f mark the initial arrival time, tarr (at different energies) of the injected electrons. Using the start of the solar type III burst, as identified by the sharp increase in >5 MHz wave power measured by the WAVES experiment, as the electron injection time, tinj (marked by red arrows in Figure 1d), the field line length (L = vel(tarr-tinj)) was determined and plotted in Figure 1h. The field line length varies from ~4 AU near the leading edge of the cloud to ~1.5 AU near the center and rises again on the trailing part.


Figure 2 is a schematic representation of a flux rope model consistent with the observations. The impulsive solar electron events can be correlated with x-ray flares observed by Yohkoh in active region 7912, thus implying the local field line maps back to this region. From the dispersion in arrival times, the field line length is determined for these periods. The smooth curve drawn in Figure 1h, with the shortest lines near the middle of the cloud and the longest lines nearer to the edges, qualitatively supports the flux rope model with field lines twisted about a central core.

Figure 2. Schematic picture of possible magnetic cloud topology. In a magnetic flux rope with constant alpha, magnetic field lines wind around a central core axis in a helix pattern. Two closely spaced field lines of a magnetic flux rope are shown. These two field lines are nearly parallel but one is connected to the Sun, while the other is not.

As the magnetic cloud passes by, the spacecraft crosses field lines that may have different connectivity. During the heat flux dropout periods, the spacecraft is presumably on field lines that have completely disconnected from the Sun. Assuming the disconnections are the result of reconnection near the Sun, an important question is when these reconnection events occurred. Gosling et al. (1995) have suggested that magnetic clouds are formed with the 3-dimensional reconnection of sheared magnetic arcades overlying an erupting filament in the solar corona. They have argued this process could produce field lines that are connected at one or both ends to the sun and even completely disconnected from the Sun. However, another possibility remains that the cloud emerges from the solar surface as a fully evolved flux rope with little or no reconnection. As the cloud moves through the interplanetary medium, the ends of parts of the flux rope could become disconnected from the solar surface. As mentioned earlier, such a process must occur eventually for all CMEs that expel magnetic flux into the interplanetary medium.


Bougeret, J.-L., M.L. Kaiser, P.J. Kellogg, R. Manning, K. Goetz, S.J. Monson, N. Monge, L. Friel, C.A. Meetre, C. Perche, L. Sitruk, and S. Hoang, Space Sci. Rev., 71, 231 (1995).

Burlaga, L. F., Magnetic clouds and force-free fields with constant alpha, J. Geophys Res., 93, 7217-7224 (1988).

Burlaga, L. F., R. P. Lepping, W. Mish, K. W. Ogilvie, A. Szabo, A. J. Lazarus, J. T. Steinberg, A magnetic cloud observed by WIND on October 18-20, 1995, NASA-GSFC, Laboratory for Extraterrestrial Physics Internal Document, February 1996.

Gosling, J. T. Large-scale inhomogeneities in the solar wind of solar origin, Rev. Geophys, 13, 1053 (1975).

Gosling J. T.; J. Birn; M. Hesse, Three-dimensional Magnetic Reconnection and the Magnetic Topology of Coronal Mass Ejection Events, Geophysical Research Letters, V22 N8:869-872., April 15, 1995.

King, J. H., Solar cycle variations in IMF intensity, J. Geophys. Res., 84, 5938 (1979).

Lin, R. P., Energetic solar electrons in the interplanetary medium, Solar Phys., 100, 537 (1985).

Lin, R. P.; S. W. Kahler, Interplanetary magnetic field connection to the Sun during electron heat flux dropouts in the solar wind, J. Geophys. Res., Vol 97, No A6, 8203-8209, June 1, (1995).

Lin, R. P., K. A. Anderson, S. Ashford, C. Carlson, D. Curtis, R. Ergun, D. Larson, J. McFadden, M. McCarthy, G. K. Parks, H. Reme, J.M. Bosqued, J. Coutelier, F. Cotin, C. d'Uston, K.-P. Wenzel, T.R. Sanderson, J. Henrion, and J.C. Ronnet, A three-dimensional plasma and energetic particle investigation for the Wind spacecraft, Space Sci. Rev. 71, 125 (1995).

MacQueen, R. M., Coronal transients: A summary, Philos. Trans. R. Soc. London, Ser. A., 297, 605 (1980).

McComas, D. J., J. T. Gosling, J. L. Philips, S. J. Bame, J. G. Luhmann, and E. J. Smith, Electron heat flux dropouts in the solar wind: Evidence for interplanetary magnetic field reconnection?, J. Geophys. Res., 94, 6907 (1989).