Pages 789-800

INTERBALL OBSERVATIONS OF THE DAYSIDE MAGNETOPAUSE 

O.L.Vaisberg1, L.A.Avanov1, V.N.Smirnov1, J.L.Burch2, J.H.Waite, Jr.2, A.A.Petrukovich1, and A.A.Skalsky1  

1 Space Research Institute, 84/32 Profsoyuznaya str., 117810 Moscow, Russia, E-mail:   olegv@afed.iki.rssi.ru
2 Southwest Research Institute, 6220 Culebra Rd., P.O.Drawer 28510, San Antonio, TX   78228-0510,   USA
 

ABSTRACT 

The high-apogee Interball Tail Probe crosses the magnetopause in two latitude ranges: one is close to the equator, and the other is at middle and high latitudes. A brief description of dayside magnetopause observations from the fast plasma analyzer SCA-1 is given. We give examples of plasma regimes observed at low latitudes that show evidence for plasma cloud penetration into the magnetosphere. Nonstationary reconnection is suggested as a possible mechanism for the observed near-magnetopause structures. 

INTRODUCTION 

The magnetopause is considered to be a primary site for the transfer of energy, mass, and momentum from the shocked solar wind to the geomagnetosphere. Three major processes were considered as candidates for this coupling: reconnection (Dungey, 1963), diffusion (Eastman et al., 1976, Eastman and Hones, 1979), and the Kelvin-Helmholtz instability (Dungey, 1955, Axford and Hines, 1961). Another important entry region is the polar cusp (Haerendel et al., 1978). 

Signatures associated with reconnection include acceleration of the magnetosheath plasma to speeds greater than those in the adjacent magnetosheath upon open field lines (Sonnerup et al., 1981), Paschmann et al., (1979, 1986, 1990), (Gosling et al., (1990a), FTEs (Russell and Elphic, 1978), and the separation of the electron edge earthward of ion edge during accelerated flow events (Vaisberg et al., 1980), and (Gosling et al., 1990b). 

Quasiperiodic pulses of magnetosheath-like plasma on magnetic field lines near the dawn magnetopause observed on ISEE 1/2 were interpreted as evidence for strong diffusion of magnetosheath plasma across the magnetopause and the Kelvin-Helmholtz (K-H) instability at the inner edge of the LLBL (Sckopke et al., 1981), as evidence for magnetic merging and the formation of twisted flux ropes of interconnected magnetosheath and magnetospheric filed lines (Paschmann et al., 1982), and as quasi-periodic magnetopause motion and the observation of draped northward magnetosheath magnetic field lines in the plasma depletion layer (Sibeck, 1990). The K-H instability can accelerate the reconnection rate (LaBelle-Hammer et al., 1988), (Liu and Hu, 1988). Reconnection and the K-H instability may be interconnected. Observations suggest that accelerated flow associated with reconnection can excite the K-H instability (Saunders, 1989). 

Three-dimensional observations of ions provide a useful tool for the analysis of different processes at the magnetopause. We give a description of the three-dimensional fast plasma analyzer SCA-1 installed on Interball-1/Tail Probe satellite. We present the results of observations of two dayside magnetopause crossings, discuss quasi-periodic plasma pulses of magnetosheath-like plasma near dayside magnetopause, ion dispersive beams, and the evolution with time of fluid and kinetic behavior of ions after the first magnetopause crossing. Evidence of reconnection signatures is given. We also discuss multiple magnetopause crossings vs. the plasma penetration alternative, and give evidence of impulsive penetration of magnetosheath-like plasma into the magnetosphere, suggesting a possible source for magnetospheric plasma.  

INSTRUMENTATION. 

The ion spectrometer SCA-1 (Vaisberg et al., 1995) has full 3-D capabilities. Its two identical sensor heads EU-1/1 and EU-1/2 cover two hemispheres. Each sensor head consists of a toroidal electrostatic analyzer (ESA) followed by a channel electron multiplier with 8-sectored anode. The energy range of the instrument is 50 eV/Q to ~ 5.0 keV/Q,. An electrostatic scanner in front of each electrostatic analyzer provides measurements over a nearly-2 field of view. In basic fast mode the SCA-1 measures E/Q spectra over 15 energy steps in 64 directions: 8 equally spaced (by 45o) azimuthal directions by 8 polar angles relative to the Sun-directed satellite’s axis: 2o, 17o, 40o, 65o, 115o, 140o, 163o, and 178o. A narrow field of view (2o), and narrow energy passband
(~ 10 %) provide non-averaging velocity space measurements. Measurements of complete energy-angular distribution of ions are performed about every 10 s.  

OBSERVATIONS.

A number of SCA-1 fast mode measurement periods were focused at the magnetopause region. Figure 1 gives the locations of the dayside magnetopause crossings observed in the fast mode of SCA-1. The solar magnetospheric coordinates of the satellite are given when SCA-1 observed the magnetopause and boundary layer. Two magnetopause crossings are discussed in this paper: on September 2, 1995 and on February 15, 1996. They are marked on Figure 1 by dates of measurements. Figure 2 shows September 2, 1995 magnetopause crossing that occurred at about 01:45 UT as seen in SCA-1 plasma data and FM-3 magnetic field data. This magnetopause crossing was already discussed by Vaisberg et al., (1996). Interball-1 was moving approximately along the -Y-axis and crossed the magnetopause at SM coordinates: sm = 10.2o (latitude), and LMT =06.13 hours. The color scale on the right shows "integrated" counts over these cones over an accumulation time of 0.09 sec.

Fig.1.
Fig. 2.

Panels (c), (d), and (e) are calculated flow parameters (moments of the ion distribution function, calculated assuming that all ions are protons): number density (N), total flow velocity, (Vo). and ion temperature, (Ti). Panels (f), (g), and (h) are FM-3 (Nozdrachev et al., 1995) magnetic field components in normal coordinates (Russell and Elphic, 1978): Bn along the normal to the magnetopause, Bl along the magnetospheric magnetic field, and Bm completing the right-hand coordinate system. The magnetic field magnitude is given on panel (i). The normal to the magnetopause was calculated as a minimum variance direction for magnetic field measurements in a time interval of about 40 min centered at the magnetopause crossing.  

Magnetosheath flow with the maximum counts at about 300 eV and a highly anisotropic distribution is seen in panels (a) and (b) before 01:45 UT. At that time the magnetosheath spectrum changes to a magnetospheric spectrum which is much more isotropic and is seen at all angles as an increased counting rate in the upper part of the energy range. Additional detail at the magnetopause is a narrow beam in panel (b),  = 140o. This beam has an energy dispersion such that lower energies are observed first, close to the magnetopause. 

Number density in the magnetosheath remained nearly constant during about a 45 min time interval. Two dropouts (at ~ 01:27 UT, and at ~ 01:39 UT) are associated with velocity jumps (panel (d). The velocity of the magnetosheath plasma slightly increased during time the interval shown, and exhibits several jumps, ranging from about 30 km/s to 70 km/s. The first velocity jump, at ~ 01:13 UT, is accompanied by leakage of magnetospheric particles, and the largest jump, at about 01:26, marks the start of the region of permanent magnetospheric ion leakage to the magnetosheath (the ions in the energy range above 1 keV are seen as a separate part of the distribution in the counts of the antisunward-looking analyzer) at about 01:26 UT and continued to be registered through the magnetopause. A velocity increase of about 30 km/s is also seen just before the magnetopause crossing. Ion temperature was approximately constant before 01:26 UT, and increased after the time, when magnetospheric ions are added to magnetosheath flow. This increase is apparently associated with this admixture. However, the temperature drops somewhat before the magnetopause crossing. Slight temperature increases are associated with velocity bursts, except for a burst at ~ 01:13 UT, that marks the magnetosheath ion leakage region, in front of which the temperature increased by a factor of two. The leakage of magnetospheric ions through the magnetopause is considered as an indication of reconnection (see, for example, Phan et al., 1994). Permanent presence of magnetospheric ions on the magnetosheath magnetic field lines may be an indication of a steady reconnection.  

Magnetic field parameters for this magnetopause crossing were obtained from measurements of the FM-3 magnetometer that has a sampling rate for the September 2 pass of one vector every 32 sec. This low sampling rate complicates attempts to make a detailed comparison with temporal variations of the plasma. The magnetic field in the magnetosphere is about 25 nT and was directed approximately along the Z axis. The magnetic field in the magnetosheath had a negative Bm component that facilitates reconnection. All velocity bursts in the magnetosheath after 01:27 UT are accompanied by changes of B, and the magnetic field within these events approaches the magnetospheric direction. This is another indication of reconnection processes. 

Many plasma features seen during the time interval after the first magnetopause crossing until ~ 02:21 UT resemble successive crossings of the magnetopause. These bursts of ions have nearly the same energy spectrum as the magnetosheath spectrum on the panel (a) (sunward hemisphere). However, trailing parts of the bursts as well as bursts observed after ~ 02:21 UT show decreased counting rate in sunward hemisphere and increased counting rate in antisunward hemisphere. In the lower panel (antisunward hemisphere) these bursts are wider in time scale, and appear as envelopes of bursts seen in the sunward hemisphere. One notice the energy dispersion in the antisunward hemisphere (panel (b), resembling the beam observed at the magnetopause crossing. It is easy to see that where the flux has a maximum in the solar hemisphere, there is minimum flux in the antisunward hemisphere (see the bursts at 01:47 UT, 02:19 UT, and 02:33 UT). The lower energy cutoff and the average energy in the plasma burst are progressively displaced to higher values. Plasma sheet-type particles are slightly depressed within the burst (see one at 02:49 UT). Part of these features are also seen in some earlier bursts (02:14, 02:29).

The trend in the properties of plasma bursts along the spacecraft trajectory is also seen in the flow parameters (panels (c), (d), and (e) on Fig. 2). Number density in the plasma bursts decreases as satellite penetrates deeper into the magnetosphere, although the maximum number density reaches its magnetosheath value in the first half of the magnetospheric region shown. Maximum velocity values observed in the leading parts of plasma bursts in the magnetosphere initially reach higher values than the average velocity in the magnetosheath (time interval 01:48 UT - 02:09 UT), then sometimes reach the magnetosheath value (within the time interval 02:18 UT - 02:32 UT), but the average flow velocity in the bursts continually decreases with increasing distance from the magnetopause, and all plasma bursts observed after about 02:10 UT are non-convected (except for front edge of burst at 01:44 UT and part of burst at 01:59 UT). At this time the plasma ram pressure drops below the magnetic field pressure. Velocity correlates quite well with number density. The ion temperature in plasma bursts in the magnetosphere shows the systematic increase from its magnetosheath value (about 70 eV) to the temperature value of magnetospheric ions (about 2 keV), and is approximately inversely correlated with ion number density. 

The ion distribution function within plasma clouds also changes with time from the first magnetopause crossing (Vaisberg et al., 1996). In the dense parts of the plasma bursts the ion velocity distribution resemble a heated convected beam of magnetosheath-like plasma. The ion velocity distribution in the less dense parts of the bursts is quite different.  

Figure 3 shows the cross-sections of the ion velocity distributions as measured by SCA-1. Each square panel is a plane cross-section of the velocity distribution with horizontal axis Vx directed to the Sun (Sun is on the left) and vertical axis as the other velocity component Vy on the plane. The scales are in km/s. The horizontal row is a complete set of 4 cross-sections along meridional planes of the instrument separated by 45o (meridional direction of the plane is shown on the top of each frame). Velocity components as well as total velocity in the plane were calculated from the plane cross-section and are given under respective frame. Three successive sets of measurements within one of the first plasma bursts after magnetopause crossing (time of measurements is indicated above respective row) are shown.

 Fig. 3.

Fig. 3 indicates that the ion velocity distribution within plasma burst is no longer magnetosheath distribution. It consists of the beams with velocity spread close to those in the magnetosheath flow, suggesting the break-up of magnetosheath velocity distribution into the beams. As the burst density and convection velocity decrease the energy of beams constituting it increases. Least dense bursts contain ion beams that have the energy approaching the energy of magnetospheric plasma. Ion velocity distributions observed in the leading and in the trailing parts of the bursts reflect an asymmetry seen in dynamic spectra and in flow parameters: velocity distributions in the trailing parts have more complicated, beam-like structure than those in the leading parts of the bursts. 

Figure 4 shows magnetopause crossing observed with SCA-1 at ~ 22:50 UT on February 15, 1996 at SM coordinates sm = 24.8o, and LMT = 18.48. It has the same format as Figure 2. There is striking similarity between this magnetopause crossing and that of September 2, 1995, except for smaller number of plasma bursts in February 15, 1996 case. The reconnection signatures in the magnetosheath include velocity jumps and number density dropouts accompanied by the leakage of magnetospheric particles. Magnetic field measurements in the magnetosphere show that the satellite was on the closed field lines. This is supported by the observations of the hot nearly isotropic ions. The duration of observations of disturbances in the magnetosheath and in the magnetosphere is close to what was seen in the case discussed above. The plasma bursts have very similar appearance and properties to those observed in September 2, 1995 case. The high-density bursts have high-density front edge, depleted magnetospheric population and energy-dispersive envelope of ion flux in antisolar hemisphere. The weaker the burst, the higher average ion energy and the larger relative ion flux coming from antisolar hemisphere.

Fig. 4.

Advantage of February 15, 1996 observations is a high sampling rate of the magnetic field measurements (16 Hz). This gives better perspective of magnetic structure of bursts compared to September 2, 1995 case. It is clearly seen from Figure 4 that about 2-min duration plasma burst at ~ 23:35 UT and about 5-min plasma burst at ~ 23:54 UT have distinct FTE signatures with bipolar Bn profile.  

So many features of February 15, 1996 magnetopause crossing are similar to those of September 2, 1995 crossing. The ion velocity distributions in the plasma bursts on February 15, 1996 case also show the transition to beam-like and quasi-isotropic distribution both becoming more pronounced with the decrease of ion density. However, the ion beams in the February 15, 1996 case do not increase their energy so fast as they do in September 1995 case. This may be associated with different magnetospheric locations of two magnetopause crossings.  

DISCUSSION 

Two cases at the low-latitude dayside magnetopause illustrate the transient plasma phenomena at the magnetopause, including velocity bursts in the magnetosheath, the mixture of magnetosheath and magnetospheric plasma and plasma bursts in the magnetosphere. The two cases discussed are close to the equator, but significantly separated in local time. The September 2, 1995 case is on the dawn flank of the geomagnetosphere, while the February 15, 1996 case is on the dusk flank.  

Transients in the September 2 case are seen as velocity bursts in the magnetosheath and as plasma bursts in the magnetosphere (Fig. 2). The magnitudes of the velocity increases are in the range of 10s km/s. Only in 2 cases out of 8 velocity bursts does the magnitude reach about 0.5 VA, with an average of about 0.3 VA. Some of the velocity bursts are accompanied by density decreases. Only a small fraction of magnetosheath bursts are accompanied by temperature increases. Plasma bursts on the magnetospheric side of this pass also show velocity increases above the mean magnetosheath value by about 50 km/s, comparable in magnitude to velocity jumps in magnetosheath. However, also in these cases noticeable temperature jumps were not observed.  

The angle between magnetic field directions in the magnetosheath and in the magnetosphere was about 103o for September 2, 1995 case and ~ 159o for February 15, 1996 case, that suggests reconnection as a reason for velocity jumps in the magnetosheath and for plasma bursts in the magnetosphere. This is supported by the leakage of hot magnetospheric ions in the magnetosheath. Another possible reason for the velocity bursts in the magnetosheath and plasma bursts in the magnetosphere observed on September 2 could be the Kelvin-Helmholtz instability.  

The duration of disturbances observed in the magnetosheath is nearly equal to the duration of plasma bursts in the magnetosphere. For September 2, 1995 magnetopause crossing the time interval where velocity bursts and the leakage of magnetospheric ions are observed continuously (01:26 UT - 01:45 UT) is about 19 min, comparable to about 20 min duration of disturbances seen on the magnetospheric side, where the plasma bursts are strongest (01:45 UT -02:05 UT). The total time interval of transients in the magnetosheath from the first velocity burst (01:13 UT - 01:45 UT) is about 32 min, and can be compared to the time interval of observation of moderate bursts still bearing some signatures of magnetosheath plasma (01:45 UT - 02:25 UT), about 40 min. This suggests that the same process is responsible for the disturbances in the two regions. If we assume that the magnetopause was, on average, in nearly the same location during the satellite transition through disturbance regions, we may estimate the width of the strong interaction regions on both sides of the magnetopause to be about 4-5 thousand km.  

The duration of strong disturbances seen in the magnetosheath on February 15, 1996 is about 37 min, comparable to September 2, 1995 case. The duration of observation of moderate bursts in the magnetosphere still bearing some signatures of magnetosheath plasma in February 15, 1996 case is slightly above 1 hour, again of comparable magnitude with September 2, 1995 case.  

Plasma bursts observed on the magnetospheric side of the September 2 case show evolution with time or rather distance from the magnetopause. Initially they very much resemble successive magnetopause crossings. The trend in the properties of plasma bursts along the spacecraft trajectory is seen in the flow parameters. The average number density and transport velocity decrease as the satellite moves deeper into the magnetosphere, and the temperature increases. As one can see in the counting rate spectrograms (panels (a) and (b) on Fig. 2), the average energy of ions and the low-energy cut-off increases with distance from the magnetopause, and the relative flux from the antisunward hemisphere increases. Ion thermal pressure remains approximately constant along the satellite’s pass. It seems reasonable to assume that we are observing the evolution of plasma clouds as the satellite moves deeper into the magnetosphere. This trend is also seen in the bursts observed in February 15, 1996 magnetopause crossing. 

The evolution of dynamic spectra and velocity distribution moments is confirmed by the evolution of the ion distribution function that was discussed for the September 2 case (Vaisberg et al., 1996) and is seen in Figure 3 for September 2, 1995 case. Initially the ions have velocity distribution similar to that of the magnetosheath. Kinetic effects are seen at that time only on the edges of the burst. Later the ion distribution function changes drastically. The central part of the plasma burst has a quite different velocity distribution, and ion beams constitute a significant portion of it. This trend is seen through the succession of plasma bursts and supports the supposition that we observe evolution of magnetosheath plasma clouds that entered the magnetosphere.  

The mean energy of ions in plasma cloud systematically increases with time on September 2, 1995 as the satellite moves deeper into the magnetosphere. The energy per ion increases well above that available from convective and thermal ion motion in the adjacent magnetosheath and in the plasma clouds observed close to the magnetopause crossing. It means that the energization mechanism operates during the same time as the plasma cloud dissipates. This mechanism may probably be identified with a detailed analysis of plasma and magnetic field data.  

Energization of ions in plasma clouds observed further from the magnetopause crossing and the observed evolution of the distribution function suggest, that this plasma may subsequently integrate to a hot magnetospheric population on the closed field lines. This is clearly seen in the very dilute plasma clouds in which average ion energy in the beams are higher and they are barely distinguishable from the hot magnetospheric population. Ions in plasma clouds moving up in energy tend to join the hot magnetospheric plasma population. The evolution of plasma bursts observed in two magnetopause crossings suggests that it may be a relatively common phenomenon. Penetration of clouds from the magnetosheath into the magnetosphere may be one possible source of the plasma observed in the outer magnetosphere on closed magnetic field lines. Its importance of this mechanism is unclear at this time and requires analysis of a statistically significant number of cases.  

Of 21 low-latitude dayside magnetopause crossings (Fig.1) observed in the fast mode of SCA-1 for which WIND magnetic filed data were available all but two show plasma bursts. For the majority of crossings Z-component of IMF changed sign during period of observations of the bursts. During negative Bz component of IMF the number of observed bursts varied from 1 to 11 with the median value of 5. Two magnetopause crossings without the bursts occurred during positive Bz and one more magnetopause crossing occurred during positive Bz showed two bursts. This may be another evidence of the influence of reconnection on the bursts’ appearance.  

Kinetic effects at the magnetopause associated with steady reconnection were observed by Fuselier at al., 1991, Fuselier, 1995 and by Smith and Rogers, 1991. Their results showed reflected and transmitted populations both for magnetospheric and for magnetosheath plasma. The observations discussed here also show kinetic effects at the magnetopause and within the boundary layer. What we think is new is a demonstration of evolution of plasma clouds within the boundary layer as an indication of penetration of solar with plasma clouds in to the magnetosphere (initially proposed by Lemaire and Roth, 1978), and possible role of this mechanism in population of the dayside magnetosphere.  

At least in some cases magnetic signatures of observed plasma bursts or plasma clouds on magnetospheric side bear the signature of FTEs. By analogy with FTEs this may be called PTEs for Plasma Transfer Events. 

CONCLUSIONS 

1. The dayside magnetopause crossings show transient phenomena: number density, velocity, and temperature jumps and a leakage of hot magnetospheric ions into the magnetosheath that suggest non-stationary reconnection as the most probable reason for observed transients. 

2. Plasma bursts observed after the first magnetopause crossing show nearly-continuous evolution in time as the satellite moves deeper into the magnetosphere: a decrease of transport velocity and number density, and a strong increase of temperature. The distribution function of ions in plasma bursts changes from fluid-like to beam-like systematically as time passes from the first magnetopause crossing. These changes suggest that the satellite observed the evolution of plasma clouds penetrating into the magnetosphere as a result of an instability at the magnetopause.  

3. Energy per ion increases as plasma clouds penetrate deeper into the magnetosphere suggesting the action of heating or an acceleration mechanism. 

4. Dispersive ion beams were observed at the magnetopause and at the edges of plasma clouds in the magnetosphere. This suggests the gyromotion of ions on the edge of the cloud, and allows one to estimate the spatial scale of observed events. These observations also suggest the erosion of plasma from the cloud’s boundary.

5. Injected plasma clouds may provide a source of magnetospheric plasma.  

ACKNOWLEDGMENTS 

Work at IKI was supported by RSF 94-02-04232 grant, ISF MQ8300 grant, and INTAS-93-2031 grant. Authors are grateful to E.B.Ivanova for the help in data visualization.  

REFERENCES 

Axford W.I. and C.O.Hines, A unifying theory of high-latitude geophysical phenomena and geomagnetic storms, Can.J.Phys., 39, 1422, 1961.

Dungey, J.W., Electrodynamics of the outer atmosphere, in: Proceedings of the Ionosphere Conference, Physical society of London, p.225, 1955.

Dungey, J.W., 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. 505-550, Gordon and Breach, New York, 1963.

Eastman T.E., E.W.Hones, Jr., S.Bame, and J.R.Asbridge, The magnetospheric boundary layer: Site of plasma, momentum and energy transfer from the magnetosheath into the magnetosphere, J.Geophys. Res., 3, 695, 1976.

Eastman T.E., and E.W.Hones, Jr., Characteristics of the magnetospheric boundary layer and magnetopause layer as observed by IMP 6, J.Geophys. Res., 84, 2019, 1979.

Fuselier S.A., Kinetic aspects of reconnection at the magnetopause, in: Physics of the magnetopause, ed. by P.Song, B.U.O.Sonnerup, and M.F.Thomsen, Geophysical monograph 60, AGU, 1995, pp. 181-187.

Fuselier S.A., D.M.Klumpar and E.G.Shelley, Ion reflection and transmission during reconnection at the Earth’s subsolar magnetopause, J.Geophys. Res., 98, 3935, 1991.

Gosling J.T., M.F.Thomsen, S.J.Bame, R.C.Elphic, and C.T.Russell, Plasma flow reversal at dayside magnetopause and the origin of asymmetric polar cup convection, J. Geophys. Res., 95, 8073, 1990a.

Gosling J.T., M.F.Thomsen, S.J.Bame, T.G.Onsager, and C.T.Russell, The electron edge of the low latitude boundary layer during accelerated flow events, Geophys. Res. Lett., 17, 1833-1836, 1990b.

Haerendel G. Paschmann, N.Sckopke, H.Rosenbauer, and P.C.Hedgecock, The frontside boundary layer of the magnetosphere and the problem of reconnection, J.Geophys. Res., 83, 3195-3216, 1978.

LaBelle-Hammer A.L., Z.F.Fu, and L.C.Lee, A mechanism for patchy reconnection at the dayside magnetopause, Geophys. Res. Lett., 15, 152, 1988.

Lemaire, J., and M.Roth, Penetration of solar wind plasma elements into the magnetosphere, J. Atm. Terr. Phys., 40, 331, 1978.

Liu Z.X. and Hu Y.D., Local magnetic reconnection caused by vortices in the flow field, Geophys. Res. Lett., 15, 752, 1988.

Nozdrachev M.N., V.A.Styazhkin, A.A.Zarutsky, S.I.Klimov, S.P.Savin, et al., Magnetic field measurements onboard the Interball Tail spacecraft: the FM-3I instrument, in: INTERBALL Mission and Payload, IKI-RSA-CNES, 1995, pp. 228-229.

Paschmann G., B.U.O.Sonnerup, I.Papamastorakis, N.Sckopke, G.Haerendel, et al., Plasma acceleration at the Earth’s magnetopause: Evidence for reconnection, Nature, 282, 243, 1979.

Paschmann, G., G.Haerendel, J.Papamastarakis, N.Sckopke, S.J.Bame, and C.T.Russell, Plasma and magnetic field characteristics of magnetic flux transfer events, J.Geophys. Res., 87, 2159, 1982.

Paschmann, G., I.Papamastorakis, W.Baumjohann, N.Sckopke, C.W.Carlson, et al., J. Geophys. Res., 91, 11.099-11,115, 1986.

Paschmann, G., B.Sonnerup, I.Papamastorakis, W.Baumjohann, N.Sckopke, et al, The magnetopause and boundary layer for small magnetic shear: Convection electric fields and reconnection, Geophys. Res. Lett., 17, 1829, 1990.

Phan, T.-D., G.Paschmann, W.Baumjohann, N.Sckopke, and H.Luehr, The magnetosheath region adjacent to the dayside magnetopause: AMPTE/IRM observations, J. Geophys. Res., 99, 121-141, 1994.

Russell, C.T. and R.C.Elphic, Initial ISEE magnetometer results: Magnetopause observations, Space Sci. Rev., 22, 691, 1978.

Sonnerup, B.U.O., G.Paschmann, I.Papamastorakis, N.Sckopke, G.Haerendel, et al., Evidence for magnetic field reconnection at the Earth’s magnetopause,. J.Geophys. Res., 86, 10,049-10,067, 1981.

Sckopke N., G.Paschmann, G.Haerendel, B.U.O.Sonnerup, S.J.Bame, et al., Structure of the low latitude boundary layer, J. Geophys. Res., 86, 2099, 1981.

Sibeck, D., R.P.Lepping, and A.J.Lazarus, Magnetic field line draping in the plasma depletion layer, J.Geophys. Res., 95, 5489, 1990.

Smith, M.F., and D.J.Rogers, Ion distributions at the dayside magnetopause, J.Geophys. Res., 96, 11617, 1991.

Saunders, M.A., Possible Kelvin-Helmholtz waves driven by reconnection accelerated flows, Geophys. Res. Lett., 16, 1031, 1989.

Vaisberg, O.L., A.N.Omel’chenko, and V.N.Smirnov, Observation of the Plasma Structures Injected into the High-Latitude Boundary Layer of the Earth’s Magnetosphere, Cosmic Research, Vol. 18, No. 2, 195-201, 1980.

Vaisberg O.L., A.W.Leybov, L.A.Avanov, V.N.Smirnov, E.I.Ivanovs, et al., Complex plasma spectrometer SKA-1, in: INTERBALL Mission and Payload, IKI-RSA-CNES, 1995, pp. 170-175.

Vaisberg, O.L., L.A.Avanov, V.N.Smirnov, J.L.Burch, A.W.Leibov, et al., Initial observations of fine plasma structures at the flank magnetopause with complex plasma analyzer SCA-1 onboard Interball tail probe, Ann. Geophys., submitted, 1996.