Pages 539-548

THE GEOTAIL MISSION: PRINCIPAL CHARACTERISTICS AND
SCIENTIFIC RESULTS

A. Nishida1, T. Yamamoto1, and T. Mukai1

1 Institute of Space and Astronautical Science, 3-1-1 Yoshinodai, Sagamihara, Kanagawa 229, Japan, E-mail: nishida@fujitubo.isas.ac.jp


ABSTRACT

Strength of the GEOTAIL mission derives from several factors: the optimal choice of the orbit, installation of the advanced instrumentation, use of numerical simulation as the basic intepretational tool, and formation of a satellite network with other IASTP missions. Representative scientific results which reflect these assets are reviewed.

INTRODUCTION

Since its launch on July 24, 1992, GEOTAIL has been operating satisfactorily and has yielded a wealth of new information that has substantially advanced our understanding of the Earth’s magnetotail. Several characteristics of the mission can be counted as the cause of these achievements. These are (a) an orbit that is optimized for a thorough survey of the magnetotail from 10 to 220 Re, (b) state-of-the-art onboard instrumentation, (c) extensive use of numerical simulations for interpreting the observations, and (d) formation of a satellite network with IMP-8, Akebono, WIND and INTERBALL/TAIL. This paper reviews the representative results of the GEOTAIL mission where these characteristics played a major role. References are limited to those which were written after the Special Section "Initial Results of Geotail Mission" appeared in the Geophysical Research Letters (Volume 21, No. 25, December 15, 1994).

OPTIMIZED ORBIT STRATEGY

Figure 1 shows the orbit of GEOTAIL for about 3.5 years since its launch. During the first 2 years GEOTAIL surveyed the distant tail by adjusting its apogee via double lunar swingby maneuvers. The left panel is the GEOTAIL orbit in this distant-tail phase as projected to the GSM (Geocentric Solar Magnetospheric) xy plane. The orbit was altered in October-November 1994 to conduct observations in the near-earth region of the magnetotail. The apogee in this near-tail phase was designed to be 30 Re but it was set temporarily at 50 Re for about 4 months during the decent in order to cover the intermediate distant range well. The right panel is the orbit in the near-tail phase. The perigee was about 10 Re in the distant-tail phase and is more closely so in the near-tail phase, and the inclination of the near-tail orbit is chosen to be -8° so that the apogee will often be at the neutral sheet around the December solstice.

Fig.1. GEOTAIL orbits in the GSM x-y plane. The inclination is always small.

Basic Structure Figure 2 is the plot of basic parameters of the magnetic field and plasma flow against the GSM y coordinate. The left and the right panels are for the nearest (20 to 30 Re) and the furthermost (beyond 150 Re) tail regions covered by GEOTAIL. The top panel shows the orbits in the y-z plane. The second to the last panels show the 0 angle of the magnetic field, proton density, proton temperature, x component of the bulk flow velocity, ion , fast-mode Mach number, and Alfven Mach number.

Fig.2. Basic parameters plotted against y for distance ranges

In the left panel for the near tail, it is clearly seen that all the parameters show distinct changes at y of about 䔸 Re which represents the boundary between the magnetotail and the magnetosheath. The 0 angles are clustered around 0#176 or 180° in the tail with a slight increasing trend due to the flaring of the tail. Density in the tail tends to be below 1 /cm3 and an order of magnitude lower than the sheath density. Ion temperature is higher in the tail. The flow speed is usually much slower than the sheath flow but occasionally fast earthward flows are observed. The ion  takes a wide range of values in the tail while it is around 1 in the sheath. The fast-mode Mach number is distinctly smaller in the tail than in the sheath; so much so that this can be used as a distinguishing characteristics of the magnetotail relative to the magnetosheath. The Alfven-mode Mach number is very often less than 1, so that the communication with the ionosphere via Alfven waves is maintained most of the time. Essentially the same features are seen in the right panel for the distant tail as well, although the plots are much more scattered due to swinging of the tail axis. The flow in the distant tail is mostly tailward and is often as fast as the flow in the sheath.

Fig.3. Relations between speed and density normalized by solar-winds values.

Figure 3 is the plots of the normalized speed Vx against the normalized density at the four distance ranges. The normalization has been performed by dividing the tail data by the corresponding parameter of the upstream solar wind observed by IMP-8 or WIND satellite. The observed points form two groups. The first delineates a distinct trend that extends from very low speeds and densities to high speeds and densities that are almost the same as the solar wind values. The second group comprises the points which are widely scattered outside this trend; the scatter seems to be more pronounced in speed than in density. The ions of the first group are cold while those of the second group are hot; the dividing temperature is a few hundred eV. The first group represents the plasmas that are directly connected to the solar wind, while the second group has experienced significant energy change after the ions have entered the tail [Yamamoto et al., in preparation].

Convection With GEOTAIL the convective motion of plasma and magnetic field lines can be studied from the near-earth region to the distant magnetotail. The analysis of the distant tail data has shown that there is an apparent difference in the features of the convection between geomagnetically quiet and active times. Here, active and quiet times are operationally defined as follows; the active times are when Kp   3for 24 hours or more, and the quiet times are when Kp  2 for more than 24a day. The distance range covered is roughly from 40 to 200 Re.

Fig.4. Relations between Bz and Vx in geomagnetically active times.

In order to express the character of the convection, we have plotted Bz against V,x, where V is the convection velocity and V,x, is its x component. Two of such plots are reproduced for both active and quiet times. During active times (Figure 4) the convection is tailward in the distant tail at 169 Re (right), but the convection direction is almost equally divided between tailward and earthward in the near tail at 44 Re (left). The average Eyof is positive (namely, directed from dawn to dusk) at both distances. This suggests that the site of magnetic reconnection is often located between these two distances. In quiet times (Figure 5), the convection direction is predominantly tailward both in the distant tail at 200 Re (right) and in the near tail at 30 - 65 Re (left), and the average of is negative (namely, directed from dusk to dawn) at both distances. Thus the convection in quiet times is the tailward motion of the northward-directed field lines.

Fig.5. Relations between Bz and V in geomagnetic quiet times.

From a quantitative analysis of the above observations, it has been derived that for active times that (1) the distant neutral line tends to be located at about 140 Re and mainly the open field lines are reconnected there, while (2) the near-earth neutral line tends to be formed earthward of 50 Re and mainly the closed field lines are reconnected there. Substantially more open field lines are reconnected at the distant neutral line than at the near-earth neutral line [Nishida et al., 1996a]. In quiet times, the plasma and field lines are convected parallel to the neutral sheet across the tail at the same time as they flow tailward. The direction of this cross-tail convection is governed by the polarity of IMF By, and is antisymmetric with respect to the neutral sheet which is twisted. This convection agrees in topology with the expected consequences of the cusp reconnection, but not only the cold ions but also the hot ions take part in this convection [Nishida et al., 1995]. In the nightside magnetosphere at the distance of 10 Re, on the other hand, northward magnetic field lines are injected toward the Earth so that Ey > 0 in quiet times too [Nishida et al., 1996b]. The observations in both quiet and active times, as well as in different distance ranges, can be explained in a consistent manner by a unified model in which reconnection occurs in the neutral sheet which is twisted with respect to the xy plane under the influence of the IMF By, being more so in quiet times [Nishida et al., 1996c].

Energetic Particles:  Dependence of the intensities of energetic ions (9 to 212 keV/e) on x has been studied down to 210 Re for both protons which are mainly of the solar-wind origin and the ions such as O+ which originate from the ionosphere. It was found that 90-100 Re downtail region in the plasma sheet is a strong division between more equal sunward and tailward fluxes closer to the Earth to more strongly tailward flux farther out. The tailward flux beyond that region does not decrease appreciably with x [Christon et al., 1996a]. This observation is consistent with the presence of the neutral line in the above region. The Kp dependence of the ion fluxes presented in Christon et al. [1996b] suggests that the proton flux is about a decade and a half higher at high Kp (up to 5) than at low Kp, while the range of variations of the ions of the ionospheric origin seems to be larger than this (about 2.5 decades).

Substorm Development:  Development of the plasmoid (or flux rope) has been studied in the distance range of 30 to 200 Re by a superposed epoch analysis of 140 substorms whose expansion phase onset is determined by Pi2 and mid-latitude magnetic signatures. Time development of plasma density, velocity, temperature, and magnetic field are examined in x- Bx space from -20 to + 60 min relative to the expansion phase onset. Fast plasma flows stream tailward in the plasma sheet boundary layer while the flow in the plasma sheet is slowed as it impacts the pre-existing plasma sheet population. The ion temperature at the center of the plasmoid is about 6 keV when it is formed but decreases as it travels down the tail, and becomes about 2 keV at 140 Re [Machida et al., 1996]. At the distance of 90 Re the reversals of the anisotropy and flow from tailward to earthward is observed with a median delay time of 94 min after the expansion onset, a possible interpretation being the tailward retreat of the neutral line [Angelopoulos et al., 1996].

Substorm features at closer distances of 10 to 50 Re have also been studied with 213 cases of isolated substorms. Change in the flow direction is often observed in association with the onset of the substorm expansion phase, and the sense of this change tends to depend on the distance of the observing site. The flow change is earthward inside 20-30 Re and is tailward outside 20-30 Re, suggesting that the near-earth neutral line is formed in the 20-30 Re range at the substorm onset [Nagai et al., 1996]. This result is most gratifying from the mission-planning point of view, as it justifies our choice of the near-earth orbit which covers the region of the flow reversal, and hence the site of the near-earth neutral line, ideally well.

Sometimes the earthward field-aligned ion beam is observed in the distant tail prior to the fast tailward flow. This suggests formation of a quasi-stagnant plasmoid due to coexistence of the two reconnection sites, one in the near tail and the other in the distant tail beyond the satellite position. The near-Earth X-line is formed inside the separatrix defined by the distant neutral line [Hoshino et al., 1996a; Kawano et al., 1996].

Storms Unusually strong magnetic fields (> 30 nT) are observed beyond x = -60 Re in the tail lobe during a growing stage of the ring current. Obviously this means that the dynamic pressure of the solar wind is very high, but in addition it suggests that additional magnetic fluxes that are transported to the magnetotail by the dayside reconnection prevent the tail radius from decreasing in spite of the strong compression [Kokubun et al., 1996].

Slow Shocks:  Of 303 boundary crossings between the plasma sheet and the lobe in the distance range of 30 to 210 Re, 32 have been identified as slow mode shocks. They satisfy the Rankine-Hugoniot relation, and backstreaming ions are found on the upstream side of the slow shocks. The cold ions of the lobe are accelerated and rotate around the magnetic field as they enter the plasma sheet, and at times a ring-shaped velocity distribution is formed. The slow shocks observed on the front side of plasmoids have a different orientation from that of the ordinary shocks observed at the plasma sheet-lobe boundaries, which suggests the existence of "heart-shaped" plasmoids predicted by a numerical simulation [Saito et al., 1995].

Magnetopause and Low-Latitude Boundary Layer:  The 10 Re perigee often allows GEOTAIL to skim along the dayside magnetopause, so that numerous encounters have been made with the magnetopause and the low-latitude boundary layer. From a preliminary study of the magnetopause crossings, a leakage of ions from the low-latitude boundary layer to the magnetosheath has been identified [Nakamura et al., 1996].

A detailed analysis of the dayside/dawnside LLBL under the condition of weakly southward Bz (where |By| > |Bz|) has shown that the LLBL consists of two parts, the outer LLBL and the inner LLBL. The outer LLBL is where the magnetosheath-like plasma is flowing tailward on reconnected open field lines, and these field lines are being convected tailward in a draped manner along the magnetopause. The inner LLBL is where the mixing of magnetosheath and magnetospheric plasmas has taken place on closed field lines. The flow in the inner LLBL tends to be sunward directed, so that the mixing of the plasmas takes place somewhere down the tail [Fujimoto et al., 1996c].

The cold and dense plasmas are observed also in the tail LLBL at x = -15 to -30 Re and their features are consistent with the LLBL observations on the dayside; Sometimes these ions are on the tailward flowing open field lines, while at other times they are on the earthward flowing closed field lines. Since the Alfven speed in the sheath/boundary region was only about 150 km/s for the case studied, reconnection alone cannot explain the sunward turning of the magnetosheath ions. Reconnection at the flanks would have taken place at locations further downtail of the observing site at -27 Re and the tension of the closed field lines would have further accelerated the earthward flow. Alternatively, the magnetosheath plasma may have entered the closed field lines at the tail flanks [Fujimoto et al., 1996a].

Cold Dense Ion Flows in the Lobe:  Cold ions that flow tailward in the lobe are often quite dense and have densities of about 1 /cm3. The ions species deduced from the energy ratio are H+ and O+, and sometimes He+ ions are also involved. In some cases H+ continuously stream tailward both in the lobe and the magnetosheath, and they are smoothly coupled through the magnetopause. Faint O+ fluxes sometimes exist in the magnetosheath near the magnetopause, suggesting that the ions of ionospheric origin leak from the tail [Hirahara et al., 1996].

Polar Rain: The polar rain is usually observed as bidirectional beams of electrons in the tail lobe. The tailward beam is produced by the mirroring of the electrons at lower altitudes, while the earthward beam represents the electrons entering from the solar wind. However, only one of these beams tends to be seen at the boundary of the plasma sheet. This feature can be explained by the blocking of an electron beam due to the field-line geometry involving the X-type magnetic neutral line, and the position of the neutral line that is deduced from this interpretation is in the range of 50 to 150 Re [Shirai et al., 1996]. The entry region of the polar rain electrons from the magnetosheath to the tail has been identified; sometimes the electrons follow the connected IMF and magnetotail field lines smoothly, but in some other times a layer of the isotropic pitch angle distribution exists at the boundary [Shirai et al., in preparation].

Plasma Waves:  From a thorough survey of the plasma waves by GEOTAIL, characteristic wave modes have been identified for each of the regions in the magnetotail and the magnetosheath. There are: Magnetic noise bursts (MNB) in the central plasma sheet, electrostatic solitary waves (ESW) in the plasma sheet boundary layer, and electrostatic quasi-monochromatic waves (EQMW) in the lobe. All these modes are also seen in the magnetosheath [Kojima et al., 1996]. Since GEOTAIL represents an ideal platform for monitoring the Auroral Kilometric Radiation (AKR) continually, an AKR index has been defined by exploiting this property. It has been found that AKR tends to intensify at the expansion phase and show rapid dropoff when the recovery phase starts [Murata et al., 1996].

ADVANCED INSTRUMENTATION

Electron Beam Boomerang Experiment:  The electron beam boomerang method has been used to measure the electric field. Unlike the GEOS electron experiment, the boomerang method measures the time of flight of the artificially emitted electron beam returning to the spacecraft. Since the test electrons with energies of 500 to 800 eV travel a few tens of kilometers before returning, most of their orbit is outside the Debye sphere. Hence the measured electric field is not affected much by local electric disturbances around the spacecraft. The flux of the emitted electrons is modulated in order to distinguish them from the background.

  Fig.6. An example of the electric an magnetic field observation by electron boomerang method. GEOTAIL was on the dusk side at GSM coordinates of (5,7,2) Re when this observation was made.

Figure 6 shows an example of the boomerang observation. In the top panel the return flux is shown in the spin-phase vs time plane. The electrons are observed twice per spin when the gyration plus ExB drift brings them back to the spacecraft. The second panel shows the time of flight of the returning electrons as normalized by the period of the modulation. The separation between the times of flights of two returning beams gives the electric field. The bottom three panels show three components of the electric field where the spin axis component Ez is derived by assuming E•B = 0. Good agreements are seen with the measurements by the double probe (solid curves). The third panel shows the magnetic field which is obtained by summing the times of flights of the two return beams. An advantage of the boomerang method is that it is free from an offset, so that comparison with the fluxgate measurement (solid curve) reveals an offset of the latter [Tsuruda et al., 1996].

Waveform Capture:  The plasma wave experiment on board GEOTAIL includes the waveform capture in which two components of the electric field and three components of the magnetic field are recorded at a sampling frequency of 12 kHz and stored in the onboard memory for a period of 8.7 s. The stored data are read out and telemetered to the ground in 275 s. This waveform capture has made it possible to measure the wave form and determine the wave vector, polarization, and Poynting flux with a very high time resolution.

One of the notable accomplishments of the waveform capture experiment is the clarification of the nature of what used to be called the Broad Band Electrostatic Noise (BEN). The waves of this kind has been so designated as they appear as broadband noise in the spectral data. The waveform capture has revealed, however, that BEN in fact consists of isolated bipolar pulses (or trains thereof). From the comparison of the observations by a pair of spinning antennas, it has been shown that the pulses represent isolated potential wells propagating in the direction parallel to the ambient magnetic field [Kojima et al., 1996]. On the other hand, the Poynting directions of the chorus emissions have been derived with the high time resolution data and they suggest generation around the dayside geomagnetic equator [Nagano et al., 1996].

Three-dimensional Velocity Distribution Functions of Ions and Electrons:  GEOTAIL carries onboard two sets of plasma detectors. One of them (LEP) has large geometric factors (on the order of 10-4 and 10-3 cm2 str eV/eV for electrons and ions, respectively) to measure the tenuous magnetotail plasma with high time resolution. The capability of the instrument is expressed, for example, in the studies of the multiple ion species (H+, He+, O+) that stream tailward in the tail lobe. While density and streaming velocity of H+ are well correlated, such correlation are not seen for O+ reflecting the difference in their supply mechanisms [Seki et al., 1996].

The other plasma instrument (CPI) has demonstrated that, even when the bulk flow speed is very low, the three-dimensional velocity distribution of ions in the near-Earth plasma sheet shows pitch angle anisotropies that appear to be the memory of their acceleration and transport in the more distant tail [Frank et al., 1996].

USE OF SIMULATION AS A BASIC INTERPRETATIONAL TOOL

Generation Mechanism of ESW:  Computer simulations have demonstrated that ESW is generated as a result of the nonlinear coalescence of strong electrostatic waves excited by electron beam instabilities. The instability is driven by an electron beam drifting relative to the other electrons and ions. Four different cases have been studied; cold bi-stream instability, weak-beam instability, bump-on-tail instability, and warm bi-stream instability. The ESW is generated by the cold bi-stream instability when ions are hot, and by the bump-on-tail instability when the phase space density of the majority electrons is high enough to make the distribution function flat at the resonance frequency [Omura et al., 1996].

Counterstreaming Ion Beams in Plasmoids:  Mukai et al. [1996a] have noted that the ion distribution function in plasmoids is often characterized by two counterstreaming beams. The relative velocity of the two components is often much higher than the local Alfven speed. Numerical simulations are being performed to interpret this feature. Fujimoto et al. [1996b] have shown that the counterstreaming beams are formed as the ions from the lobes of both hemispheres meet on the reconnected field lines. Hoshino et al. [1996b] have shown that four classes of ion distributions, namely, (1) anisotropic, high speed ion beams in the plasma sheet boundary layer, (2) counterstreaming ions inside the plasmoid, (3) non-gyrotropic, dumbbell-like distributions near the X-type neutral line, and (4) the thermal distribution downstream of the slow shock, can be obtained by the simulation.

Comparison with Global MHD Models:  GEOTAIL observations near the dawnside magnetopause at a downstream distance of about 81 Re are compared with the global MHD model where IMP 8 observations of the solar wind ions and IMF are used as the driving input. During this interval the IMF exhibited a series of rotations from northward to duskward. Of principal interest is the dense, cold ion stream which is similar to that expected for the magnetosheath plasmas but different from the expected in the y and z components of the ion bulk flow and in the y component of the magnetic field [Frank et al., 1995].

COMBINATION WITH THE OBSERVATIONS BY OTHER SPACECRAFT

IMP 8 and WIND It has become a common practice to compare the GEOTAIL observations with the solar wind and IMF monitored by IMP 8 or WIND in studying the causative processes of the phenomena observed in the magnetotail.

In addition, a systematic study has been made on the modification of the solar wind flow and IMF due to interaction with the magnetosphere. For this purpose, GEOTAIL observations in the magnetosheath are compared with the WIND observations in the upstream solar wind, and the effect of draping and tension of the magnetic field lines has been identified. It has been found that the enhancement of magnetosheath flow speed due to magnetic tension are not faster than the solar wind. Flows significantly faster than the solar wind have been observed, but are attributed to dynamic processes at the magnetopause [Petrinec et al., 1996].

Akebono:  The observations by the polar-orbiting satellite Akebono have provided complementary information on the low-altitude (at the heights of several to ten thousand km) counterpart of the magnetotail process. It has been demonstrated that IMF By produces asymmetries in the ionospheric convection and in the plasma entry into the tail in a consistent manner [Fujimoto et al., 1996d]. The relation between the dawn-to-dusk electric fields observed in the distant tail and at low altitudes is being studied [Matsuoka et al., in preparation].

Interball/Tail:  GEOTAIL and INTERBALL/TAIL represent a complementary set of spacecraft for the magnetotail study since INTERBALL/TAIL is often in the tail lobe while GEOTAIL is designed to be in the plasma sheet most of the time. The collaboration between these missions has been organized by IACG, and some initial results are presented in this Symposium [Mukai et al, 1996b].

It is anticipated that the comparison with the POLAR observations will be actively conducted in the coming years.

FUTURE OPERATION PLAN

The design limit of the eclipse period of GEOTAIL is 2 hours. Until now this limit has been kept by a series of the orbit correction maneuvers. Since the onboard fuel will be almost completely consumed by the end of 1997, some of the eclipses after spring 1998 will be longer than the design limit. However, ongoing studies show that the spacecraft can stand the eclipses that are expected to exceed 2 hours only by 30 min at most. It is anticipated therefore that an extensive data base of the magnetotail as well as the dayside magnetopause region will be obtained for both quiet and active solar conditions by a continuing operation of GEOTAIL into the next century.

REFERENCES

Angelopoulos, et al., Anisotropy reversals in the distant magnetotail and their association with magnetospheric substorms, J. Geomagn. Geoelectr., 48, 629 (1996)

Christon, S.P., et al., The downtail distance variation of energetic ions in the Earth’s magnetotail region: Geotail measurements at X > -208 Re, J. Goemagn. Geoelectr., 48, 615 (1996a)

Christon, S.P., et al., Average energetic ion flux variations associated with geomagnetic activity from EPIC/STICS on Geotail, in Proc. 3rd International Conference on Substorms, in press (1996b)

Frank, L.A., et al., Observations of plasmas and magnetic fields in Earth’s distant magnetotail: Comparison with a global MHD model, J. Geophys. Res., 100, 19177 (1995)

Frank, L.A., et al., Plasma velocity distributions in the near-Earth plasma sheet: A first look with the Geotail spacecraft, J. Geophys. Res., 101, 10627 (1996)

Fujimoto, M., et al., Plasma entry from the flanks of the near-Earth magnetotail: GEOTAIL observations in the dawnside LLBL and the plasma sheet, J. Geomagn. Geoelectr., 48, 711 (1996a)

Fujimoto, M., et al., A new kinetic evidence for the near-Earth reconnection, Geophys. Res. Lett., in press (1996b)

Fujimoto, M., et. al, Structure of the low-latitude boundary layer: A GEOTAIL observation, J. Geophys. Res.submitted (1996c)

Fujimoto, M., et al., Loading of solar-wind plasma on lobe field lines at the near-tail flank in an IMF By dominated period, Geophys. Res. Lett., in preparation (1996d)

Hirahara, M., et. al., Cold dense ion flows with multiple components observed in the distant tail lobe by Geotail, J. Geophys. Res., 101, 7769 (1996)

Hoshino, M., et al., Evidence of two active reconnection sites in the distant magnetotail, J. Geomagn. Geoelectr., 48, 515 (1996a)

Hoshino, M., et al., Ion dynamics in the magnetic reconnection: Comparison between numerical simulation and GEOTAIL observations, J. Geophys. Res., submitted (1996b)

Kawano, H., et al., A quasi-stagnant plasmoid observed with Geotail on October 15, 1993, J. Geomagn. Geoelectr., 48, 525 (1996)

Kojima, H., et al., GEOTAIL waveform observations of broadband/narrowband electrostatic noise in the distant tail, J. Geophys. Res., submitted (1996)

Kokubun, S., et al., Large field events in the distant magnetotail during magnetic storms, J. Geomagn. Geoelectr., 48,561 (1996)

Machida, S., et al., Generation of plasmoid and their evolution in the Earth magnetotail inferred from GEOTAIL observations, in preparation (1996)

Mukai, T., et al., Structure and kinetic properties of plasmoids and their boundary regions, J. Geomagn. Geoelectr., 48, 541 (1996a)

Mukai, T., et al., IACG campaign 1 summary report: initial results, Adv. Space Res. , submitted (1996b)

Murata, T., et al., Proposal of a new index for substorm studies: AKR index, J. Geophys. Res., submitted (1996)

Nagai, T., et al., Substorms and plasmoids, Adv. Space Res. , submitted (1996)

Nagano, I., et al., Analysis of wave normal and Poynting vectors of the chorus emissions observed by GEOTAIL, J. Geomagn. Geoelectr., 48, 299 (1996)

Nakamura, M., et. al., Leakage ions from the LLBL to MSBL: Confirmation of reconnection events at the dayside magnetopause, J. Geomagn. Geoelectr., 48, 65 (1996)

Nishida, A., et al., GEOTAIL observation of magnetospheric convection in the distant tail at 200 RE in quiet times, J. Geophys. Res., 100, 23663 (1995)

Nishida, A., et al., Magnetotail convection in geomagnetically active times - 1. Distance to the neutral lines -, J. Geomagn. Geoelectr., 48, 489 (1996a)

Nishida, A., et al., Traversal of the nightside magnetosphere at 10 to 15 Re during northward IMF, Geophys. Res. Lett., submitted (1996b)

Nishida, A., et al., A unified model of the magnetotail convection in geomagnetically quiet and active times, J. Geophys. Res., submitted (1996c)

Omura, Y., et al., Electron beam instabilities as generation mechanism of electrostatic solitary waves in the magnetotail, J. Geophys. Res., 101, 2685 (1996)

Petrinec, S.M., et al., GEOTAIL observations of magnetosheath flow near the magnetopause, using WIND as a solar wind monitor, J. Geophys. Res., submitted (1996)

Saito, Y., et. al., Slow-mode shocks in the magnetotail, J. Geophys. Res., 100, 23567 (1995)

Seki, K., et. al., Coexistence of Earth-origin O+ and solar wind-origin H+ / He++ in the distant magnetotail, Geophys. Res. Lett., 23, 985 (1996)

Shirai, H., et. al., Drop-off of the polar rain flux near the plasma sheet boundary, J. Geophys. Res., submitted (1996).

Tsuruda, K., H. Hayakawa, M. Nakamura, Electric field measurements in the magnetosphere by the electron beam boomerang, in Measuring Technique for Space Plasmas, AGU Monograph, in press (1996)