Christopher T. Russell

Institute of Geophysics and Planetary Physics
University of California
Los Angeles, California 90024

Originally Published in :
Physics of Solar Planetary Environments (edited by D.J. Williams) pp 526-540, American Geophysical Union, Washington D.C., 1976.

          In this paper we present a brief review of theoretical work on the reconnection of magnetic field lines and the terminology of this area of research. Then, we review the various evidence for reconnection that we find in nature: on the sun, in the solar wind, in planetary magnetospheres and in particular on earth. Temporal variations in the merging rate particularly imbalances between dayside and nightside rates lead to substorms. Reconnection can also occur in the steady state. Periods of prolonged strong reconnection lead to main phase geomagnetic storms even without significant sub storm activity. One of the standing problems of substorm research is what initiates the sudden reconnection on the night side. Sudden impulses are known to trigger sub storms if the magnetosphere is preconditioned by day side reconnection, i.e., if the tail has excess flux. We show evidence for two other interplanetary triggers: northward impulses in a southward field and northward steps in a field that remains southward.



          Reconnection is known by many names: merging, field annihilation, field cutting, and, in the United Kingdom, reconnexion. The latter name, which was chosen to be the title of this paper, is perhaps the most appropriate for it has an X in it, as also has the magnetic field in the reconnection process. We cannot really proceed without first defining what we mean by reconnection. Following Vasyliunas (1975) in his excellent review, we define reconnection to be the process whereby plasma flows across a surface that separates regions containing topologically different field lines. The magnitude of the plasma flow is a measure of the merging rate.

          Figure 1 shows what we mean by topologcally different magnetic field lines. The top panel shows the earth's magnetosphere. The field lines labelled '2' do not touch the earth; those labelled '3' touch the earth on one end; those labelled '1' have both feet on the ground. Reconnection is the flow of plasma from one to another of these topologically different field regions. In three dimensions these different field regions are separated by surfaces called the separatrix surfaces which touch at the X's in this diagram. In three dimensions the separatrix surfaces touch along a line running around the magnetosphere, called the neutral line, or merging line. The line is not neutral in the sense that the magnetic field has zero field strength, for the field may be parallel to the merging line. However, there really are two neutral points, one in the front and one in the rear where the field strength does go to zero. Merging in laboratory plasma machines and in solar flares illustrated in the bottom two panels is entirely analogous.


          Figure 2 shows the two limits of merging usually treated by theoreticians, both two-dimensional: one has a finite north south component of the field and is infinite in the dawn-dusk direction using the magnetospheric tail geometry; the other has a zero north-south component of the field and is infinite in extent along the earth-sun line. The former is usually treated with a hydromagnetic approach and the latter with a single particle approach.


          In the hydromagnetic approach, diffusion of field is restricted to a small region near the X-point and hydromagnetic waves deflect the flow of the remaining incoming plasma so that it flows away from the X-point on the merged field lines. Around the X-point is the diffusion region where the field lines change partners. Field lines change partners all the time. The field lines in the magnetosphere are continually changing partners with the field lines coming out of the core, but it is only when they change partners and form topologically different field lines that reconnection is said to occur.

          It is impossible in a brief overview such as this to pay tribute to all those who have worked on the reconnection problem. Figure 3 is an attempt to credit some of those most frequently referenced in the literature. For proper tribute, the interested reader is referred to the review by Vasyliunas (1975)




The Sun

          Reconnection occurs almost everywhere in the solar system that there is a magnetized plasma. The most obvious phenomenon attributed to reconnection that comes to mind is the solar flare. During solar flares the magnetic field reconfigures rapidly. On the other hand, the solar magnetic field can reconfigure much more slowly as for example when bipolar regions slowly approach one another and exchange field loops as discussed by Hansen and Hansen (1975). This also seems to be associated with reconnection, and thus reconnection appears to be able to occur either rapidly or slowly depending on the various boundary conditions in a particular situation.

The Solar Wind

          There are at least five studies studying reconnection in the solar wind. Four of them have studied the field signature and some of these, the plasma parameters across discontinuities in the solar wind (Uinti et al., 1972; Burlaga and Scudder, 1974; Formisano and Amata, 1975; Bavassano et al., 1976). The results of these studies are less than satisfactory in part because of inherent ambiguities in single-point measurements and in part because of the incompleteness of the available data. The fifth study inferred reconnection of the interplanetary field to form bubbles of field unconnected to the sun in order to explain the occasional occurrence of cool electron periods in the solar wind (Montgomery et al., 1974).

The Terrestrial Planets

          All the terrestrial planets have intrinsic magnetic fields. Mercury has a moment of 5 X 10 Gauss-cm (Ness et al., 1975); Venus, a moment of about 6.5 X 10 Gauss-cm (Russell, 1976a,b, c); the Earth, a moment of 8 X 10 Gauss-cm and Mars a moment of about 2 X 10 Gauss-cm (Dolginov, 1976). The magnetosphere of Mercury has been shown to be quite responsive to the direction of the interplanetary magnetic field (Siscoe et al., 1975). The magnetosphere of Venus also appears to be responsive to changes in the interplanetary field. It has been suggested that variations in the merging rate controls the absorption of solar wind by the Venus ionosphere such that a southward interplanetary field results in little absorption and a northward field allows the solar wind to penetrate the ionosphere (Russell, 1976d). The magnetic moment of Venus is opposite that of earth. Perhaps, the variable nature of planetary interactions controlled by the direction of the interplanetary field explains the sometimes contradictory results of the Soviet Mars investigators who at the same time see evidence for a planetary field and ion pick-up (Vaisberg and Bogdanov, 1974).

          More evidence of merging at Venus is provided by the preliminary data from Venera 9 (Dolginov et al., 1976) shown in Figure 4. When Venera 9 passed through the southern lobe of the Venus magnetotail, it saw a quiescent flaring field directed towards the planet. Just before periapsis marked with a , the solar-directed field component drops in magnitude, and the Z-component becomes increasingly negative. In other words, the field strength drops and the field becomes more dipolar. Furthermore, this event is bounded by By fluctuations which are the signatures of field aligned currents. Thus, this event has all the characteristics of a plasma sheet expansion seen during a substorm on earth (Russell, 1976c).


          Every thirteen months enhancements of Mev electrons are seen in the interplanetary medium at the earth which are not solar related. Rather these enhancements occur when the field line through the earth also intersects the Jovian magnetosphere. The enhancements cease when the field line reaches the end of the Jovian tail approximately 2 AU behind the planet (Mewaldt et al., 1976; Pesses and Goertz, 1976). This observation suggests that the magnetotail of Jupiter is connected to the


interplanetary field. The nature of the energetic proton bursts seen on Pioneers 10 and 11 as Jupiter was approached (Chenette et al., 1974) also suggests the interconnection. The synchronism of the bursts with the Jovian rotation over vast distances suggests direct access from the magnetosphere to the spacecraft with little or no diffusion.

The Earth

          The magnetosphere is the place where most of us have studied merging. The magnetosphere is open. In other words the field lines in the polar cap are connected, not to each other, but to the interplanetary magnetic field. There is overwhelming evidence for the openness of the magnetosphere. There is the response of the magnetosphere to a southward field: the magnetopause erodes, the polar cusp moves equatorwards, the polar cap gets bigger, the tail enlarges, and there is an increase in geomagnetic activity (cf. Russell, 1974; Burch, 1974). There is also the structure of the magnetopause which I trust is covered in the next paper. Finally, there is the evidence from energetic protons entry into the polar cap whose behavior is so neatly explained by the open model of the magnetosphere (Morfill and Scholer, 1973, Fennell, 1973) and from electron shadowing by the moon, which shows the tail has the topological characteristics predicted by the open magnetospheric model (Lin, 1968).



          Perhaps because of the intense effort by the magnetospheric community to understand sub storms, the nature of reconnection as a steady-state process is generally not appreciated except by theoreticians. When the interplanetary field turns southward and maintains a constant or increasingly southward component the magnetosphere appears to enter a steady-state. At such times Kokubun et al. (1976) have shown the existence of an S - like current system. Pytte et al. (1976) has shown that during such occasions the auroral oval is continuously disturbed, not spasmodically. They have called this the convection bay. At such times, Hones et al. (1976) have observed continual earthward plasma sheet flow. Caan et al. (1973) have shown that under these conditions the auroral zone and the magnetic field at synchronous orbit near midnight continuously disturbed while at mid latitudes and in the geomagnetic tail the signature of discrete substorms is absent. Finally, Russell et al. (1974) have shown that ring current injection depends only on the southward component of the interplanetary field, and not on the strength of auroral zone activity as measured by the AE index.



          The dependence of the merging rate on interplanetary conditions is perhaps the most important unsolved problem in magnetospheric physics. While the work on the two-dimension merging models provides a guide, this work does not aid us in solving for the effective length of the merging or neutral line which also governs the merging rate in the three dimensional magnetosphere. The apparent dependence of the magnetospheric day side merging rate on the north-south component of the interplanetary field is somewhat surprising as shown in Figure 5 (Burton et al. 1975a). The dependence has the shape of a half-wave rectifier. When the interplanetary magnetic field is northward, i.e., the electric field is from dusk-to-dawn, there is no injection of energy into the ring current. When the interplanetary field is southward, the dependence of injection rate on southward field (dawn-to-dusk electric field) is linear. This simple injection rate has been used together with a constant decay rate for the ring current to predict the Dst index quite successfully using only measured interplanetary parameters (Burton et al., 1975b). On the other hand, the half-wave rectifier has been replaced in this model with the merging rate law of Sonnerup (1974) for large ratios of the magnetospheric to magnetosheath field and found to work essentially equally as well (Burton and Russell, 1976).




The success of the simple prescription of Burton et al. (1975b) leads to an equally simple concept for the geomagnetic storm. It is simply the result of deep prolonged convection in the magnetosphere. The stronger the southward interplanetary field the greater the merging rate and the deeper convection penetrates into the magnetospheric cavity. The deeper the penetration of convection and hence plasma sheet penetration the more energy can be stored in the magnetospheric ring current. Thus, shock waves and other pressure waves in the interplanetary medium are not, per se, responsible for geomagnetic storms. This was realized by Piddington (1963) who pointed out that Sugiura and Chapman's (1960) study of geomagnetic storms of different main phase sizes all had similar sizes for their sudden commencements. The role of the pressure pulse is to compress the interplanetary field. Thus after the pulse has passed, the interplanetary field is large. If it is both large and southward a storm ensues. If it is large and northward there is no storm (Russell et al., 1974).



          Although the magnetosphere can enter a steady-state when day side and night side merging are in quasi-static balance, most of the time this state is not achieved because of the constantly changing orientation of the interplanetary magnetic field. When the interplanetary magnetic field changes from northward to southward, newly merged flux is added to the tail for some period of time before reconnection suddenly begins, i.e., a substorm is triggered. For a recent review of all the various phenomena associated with this buildup see the review by Russell and McPherron (1973). The triggering of sub storms is a big mystery. What determines the instant of time after the southward turning that night-time reconnection is initiated is not completely understood. Sudden impulses and shock waves do trigger substorms, if and only if, the magnetosphere has been preconditioned by a period of southward interplanetary magnetic field (Burch, 1972; Kokubun et al., 1976). Figure 6 shows two


other apparent interplanetary signatures which trigger substorms. In a study of 18 periods in which a clear interplanetary southward turning was observed after a prolonged period of northward field, Caan et al. (1976) observed a sub storm after about 1-2 hours in every case. In 9 of these events the onset of the sub storm in the auroral zone coincided within about 5 minutes with either a northward transient of the field which then returned to its previous level or a sudden northward step in the field which did not result in a net northward component. In every case, however, in which the field turned northward and stayed northward, geomagnetic activity soon ceased. In six of the remaining 9 events, such a transient or step was within about 30 minutes and was consistent with being the trigger if the timing of the arrival of the event at the earth was in error. In three cases, no clear event occurred in the interplanetary field at or near the onset of the sub storm. As can be seen in Figure 6, these apparent triggers are the dominant transients in the records and the association does not appear to be by chance.



          In summary, reconnection appears to be a ubiquitous phenomenon in the solar system affecting the sun, the solar wind, and all the planets which spacecraft have visited so far. We also might expect reconnection to be important in cometary tails and in the joining of the interplanetary field to the interstellar field. The filamentary structure of nebula also suggest a complex magnetic structure in which reconnection might be taking place. Siscoe and Heinemann (1974) have suggested that merging of the stellar magnetic fields in colliding solar winds of binary stars might have important consequences. Finally in Figure 7 we show evidence for a galactic magnetosphere which is a radio source (Miley, 1973). There are many other such radio sources (Rudnick and Owen, 1976). Reconnection, thus, appears to be an important process throughout the universe.



          This work was supported by the National Aeronautics and Space Administration under research grant NGR 05-007-004 and contract NAS 2-8808.




Bavassano, B., M. Dobrowolny, and F. Maria Ni(1976): Evidence of magnetic field line merging in the solar wind. J. Geophys. Res ., 81, 1-6.

Bratenahl, A. and P.J. Baum (1975): On flares, sub storms, and the theory of impulsive flux transfer events. Institute of Geophysics, University of California, Riverside Report UCR-75-21.

Burch, J.L. (1972): Preconditions for the triggering of polar magnetic sub storms by storm sudden commencements. J. Geophys. Res ., 77, 5629.

Burch, J.L. (1974): Observations of interaction between interplanetary and geomagnetic fields. Rev. Geophys. Space Phys. 12, 363-378.

Burlaga, L.F. and J.D. Scudder (1974): Sweet's mechanism in the solar wind. Astrophys. J., 191, L149-L152.

Burton, R.K. and C.T. Russell (1976): in preparation.

Burton, R.K., R.L. McPherron and C.T. Russell (1975a): The terrestrial magnetosphere: A half-wave rectifier of the interplanetary electric field. Science, 189, 717.

Burton, R.K., R.L. McPherron and C.T. Russell (1975b): An empirical relationship between interplanetary conditions and Dst. J. Geophys. Res., 80, 4204-4214.

Caan, M.D., R.L. McPherron and C.T. Russell (1973): Solar wind and sub storm related changes in the lobes of the geomagnetic tail. J. Geophys. Res., 78, 8087-8096.

Caan, M.N., R.L. McPherron and C.T. Russell (1976): in preparation.

Chenette, D.L., T.F. Conlon and J.A. Simpson (1974): Bursts of relativistic electrons from Jupiter observed in interplanetary space with time variation of the planetary rotation period. J. Geophys. Res., 79, 3551-3558.

Dolginov, Sh.Sh. (1976): Magnetic field of Mars, in Proc. of Soviet-American Conference on Cosmochemistry of Moon and Planets, in press.

Dolginov, Sh.Sh., E.G. Eroshenko, L.N. Zhuzgov, V.B. Buzin, and V.A. Sharova (1976): Preliminary results of magnetic field measurements at pericenter region of "Venera-9" satellite. Pis'ma Astron. Zh., 2, 88.

Fennell, J.F. (1973): Access of solar protons to the earth's polar cap. J. Geophys. Res., 78(7) 1036.

Formisano, V. and E. Amata (1975): Evidence for magnetic field reconnection in the solar wind, in The Magnetospheres of the Earth and Jupiter, (edited by V. Formisano) 205-217.

Hansen, S.F. and R.T. Hansen (1975): Differential rotation and reconnection as basic causes of some coronal reorientations. Solar Phys., 44, 503-508.

Hones, E.W., Jr., S.J. Bame and J.R. Asbridge (1976): Proton flow measurements in the magnetotail plasma sheet made with Imp 6. J. Geophys. Res., 81, 227-234.

Kokubun, S., R.L. McPherron and C.T. Russell (1976): Triggering of substorms by solar wind discontinuities. J. Geophys. Res., submitted.

Lin, R.P. (1968): Observations of lunar shadowing of energetic panticles. J. Geophys. Res., 73, 3066.

Mewaldt, R.A., E.C. Stone and R.E. Vogt (1976): Observations of Jovian electrons at 1AU. J. Geophys. Res., in press.

Miley, G.K. (1973): Brightness and polarization distributions of head-tail galaxies at 1415 MHz. Astron. Astrophys., 26 , 413-421.

Montgomery, M.D., J.R. Asbridge, S.J. Bame and W.C. Feldman (1974): Solar wind electron temperature depressions following some interplanetary shock waves: Evidence for magnetic merging? J. Geophys. Res., 79, 3103-3110.

Morfill, G. and M. Scholer (1973): Study of the magnetosphere using energetic particles. Space Sci. Rev., 15(213), 267-354

Ness, N.F., K.W. Behannon, R.P. Lepping and Y.C. Whang (1975): The magnetic field of Mercury: Part One. J. Geophys. Res., 80, 2708-2716.

Pesses, M.E. and C.K. Goertz (1976): Jupiter's magnetotail as the source of interplanetary Jovian MeV electrons observed at Earth. Geophys. Res. Lett., 3, 228-230.

Piddington, J.H. (1963): Theories of the geomagnetic storm main phase. Planet. Space Sci.,11 (11), 1277.

Pytte, T., R.L. McPherron, E.W. Hones, Jr., and H.I. West, Jr. (1976): Multiple-satellite studies of magnetospheric substorms: Distinction between polar substorms and convection-driven negative bays. Planet. Space Sci., submitted.

Rudnick, L. and F.N. Owen (1976): Head-tail sources in clusters of galaxies. Astrophys. J., 203, L107-Llll.

Russell, C.T. (1974): The solar wind and magnetospheric dynamics, in Correlated interplanetary and Maqnetospheric Observations, (edited by D.E. Page), p.3, D. Reidel Publ. Co., Dordrecht, Holland.

Russell, C.T. (1976a): The magnetic moment of Venus: Venera-4 measurements reinterpreted. Geophys. Res. Lett., 3, 125-129.

Russell, C.T. (1976b): The magnetosphere of Venus: Evidence for a boundary layer and a magnetotail. Geophys. Res. Lett., submitted.

Russell, C.T. (1976c): Venera-9 magnetic field measurements in the Venus wake: Evidence for an Earth-like interaction. Geophys. Res. Lett., submitted.

Russell, C.T. (1976d): The Venus bow shock: Detached or attached? J. Geophys. Res., 81, submitted.

Russell, C.T. and R.L. McPherron (1973): The magnetotail and sub storms. Space Sci. Rev., 15, 205.

Russell, C.T., R.L. McPherron and R.K. Burton (1974): On the cause of geomagnetic storms. J. Geophys. Res., 79, 1105.

Siscoe, G.L., N.F. Ness and C.M. Yeates (1975): Substorms on Mercury? J. Geophys. Res., 80, 4359-4363.

Sugiura, M. and S. Chapman (1960): The average morphology of geomagnetic storms with sudden commencement. Ab. Akad. Wiss. Gottingen., Math. Phys. Kl.

Unti, T.W.J., G. Atkinson, C.-S. Wu and M. Neugebauer (1972): Dissipation mechanisms in a pair of solar wind discontinuities. J. Geophys. Res., 77, 2250-2263.

Vaisberg, O.L. and A.V. Bogdanov (1974): Flow of the solar wind around Mars and Venus: General principles. Kosmich. Issled. 12, 279-284.

Vasyliunas, V.M. (1975): Theoretical models of magnetic field line merging, 1. Rev. Geophys. Space Phys., 13, 303-336.

Siscoe, G.L. and M.A. Heinemann (1974): Binary Stellar Winds, Astrophys. and Space Sci., 31, 363-374.

Sonnerup, B.U.O. (1974): Magnetopause Reconnection Rate, J. Geophys. Res., 79 , 1546-1549.

Back to CT Russell's page More On-line Resources
Back to the SSC Home Page