1. Institute of Geophysics, Chinese Academy of Sciences, Beijing, China
2. Institute of Geophysics and Planetary Physics, University of California, Los Angeles
3. Applied Physics Laboratory, Johns Hopkins University, Laurel, Maryland
Abstract. On April 22, 1979, from 0840 to 1050 UT, conditions varied from quiet to disturbed both on the ground and in the geomagnetotail corresponding to the northward and southward interplanetary magnetic field (IMF) conditions, respectively. In this interval, ISEE 3 monitored the solar wind, 206 RE upstream and 82 RE to the dawnside of the Earth. Very near the center of the tail, ISEE 1 and ISEE 2 were 17 and 16 RE behind the Earth. IMP 8 was also located in the tail at about 35 and 6 RE above the expected position of the current sheet and 16 RE to dawnside of the Earth. The combination of the three spacecraft monitoring the tail in key locations, the strong northward IMF switching to southward, and the excellent coverage of the midnight sector by the North American magnetometer chains make this a unique and ideal case to study. A global perspective of the observations of these four spacecraft, observations of two synchronous spacecraft, GOES 2 and GOES 3, and of ground-based magnetometers is presented herein. For IMF northward, a flapping wave in the tail is seen to propagate away from the Earth at about 95 km/s, while simultaneous plasma convection in the center of the tail was usually in the reverse direction, toward the Earth. A dipolarization of the tail field was observed by the ISEE pair near the current sheet center, and it moved toward the Earth at close to the speed of the convecting plasma. The current sheet was significantly twisted at the position of IMP 8 under the influence of the IMF By conditions, even though Bz was northward. For IMF southward, a reconnection point appeared to form inside the location of ISEE, and then it moved down the tail rapidly. The onset of this activity at the ISEE pair was simultaneous with the onset of ground activity. Plasma convection was mainly away from the Earth. The current sheet became bent at 16 RE, and the bend appeared to move down the tail some time after the plasma flow was earthward. Highly structured magnetic fields were observed at the ISEE pair, the configuration of which is difficult to discern with only two spacecraft. A substorm that developed rapidly after the arrival of southward IMF, accompanied by an enhanced electrojet at L values normally well inside synchronous orbit, had little effect at synchronous orbit at 0100 LT. However, the near-tail region was immediately affected. Nevertheless, despite the strong southward fields observed in the tail and the strong flows, the energetic electron data indicate that the ISEE spacecraft remained on closed field lines for the entire period under study. The extensive documentation of the input and output of this event should make it an excellent candidate for computer simulation.
The magnetotail is a critical region in both the coupling of the magnetosphere to the solar wind and in the storage of energy for later release. The dynamics of the magnetotail play a key role in most substorm theories, and these dynamics are mainly controlled by the solar wind and the direction of the interplanetary magnetic field (IMF). Despite the effort of many individuals, there is still much controversy about how the tail responds to solar wind and interplanetary magnetic field conditions. Of the leading contenders, the oldest paradigm is that of the near-Earth neutral point in which reconnection on the dayside of the magnetosphere transports magnetic flux to the nightside for later release from the magnetotail because of the formation of a neutral point in the tail near the Earth [Russell and McPherron, 1973]. Alternatively, some believe that reconnection of the opposing tail lobe magnetic fields does not trigger the substorm but that the current systems in the near tail become disrupted [Lui et al., 1988]. Still others believe that it is the coupling of the tail to the ionosphere that becomes disrupted at substorm onset [Kan and Akasofu, 1989].
There are many studies using the magnetic signatures measured in the lobe and solar wind aboard the ISEE and IMP 8 spacecraft, in the current sheet and plasma sheet at different substorm stages [e.g., Sergeev et al., 1993, 1995; Sauvaud et al., 1995; Jacquey et al., 1993; Lin et al., 1991; Kaymaz et al., 1994a; Baker et al., 1993; Sanny et al., 1994]. The magnetic variation measured on the ground has also been compared with the observations in space in many studies [e.g., Wiens and Rostoker, 1975; Russell et al., 1994a, b, c; Lopez et al., 1991]. Here we present a study for a series of events under both northward and southward conditions, using observations from the solar wind (ISEE 3), and from the extensive array of International Magnetospheric Survey ground-based magnetometers during a rare circumstance when we have three spacecraft in the tail (ISEE 1, 2 and IMP 8). First, we have chosen this interval to study because, as we will see, ISEE 1, 2 and IMP 8 are all in key positions to observe the dynamical behavior of the tail. Second, the IMF was strongly northward at the beginning of the interval and switched abruptly to a relatively steady and strong southward orientation. Third, at the universal time of this event, the North American chains were providing extensive coverage of the auroral currents in the midnight sector. Thus events in the auroral currents could be timed accurately, and the location of these currents could be determined well. Finally, the two GEOS spacecraft were near midnight, providing additional constraints on the initiation of geomagnetic events.
In this paper we present a global perspective of the conditions in the solar wind, in the near and midtail, and on the ground, and general conclusions about the behavior of the magnetotail based on these observations. We thoroughly document the input to the magnetosphere and its response in the belief that this interval will provide an excellent testbed for future computer simulations. We begin with a discussion of the positions of the satellites with respect to the magnetopause and current sheet positions and the solar wind and magnetospheric conditions. In later sections, we examine the flapping of the current sheet, dynamical changes in the magnetic structure of the tail, flows, and southward turnings. Zhou et al. [this issue] examine in detail the thickness and structure of the current sheet under both northward and southward IMF conditions.
In this study we will use magnetic field and solar wind data from the ISEE 3 spacecraft at the forward libration point; magnetic field, energetic particle data and plasma data, (only for ISEE 2) from the ISEE 1 and 2 spacecraft in the near-geomagnetic tail; magnetic field and energetic particle data from the IMP 8 spacecraft farther down the tail; and ground-based magnetic records from the stations in North America established for the International Magnetospheric Study. The solar wind instrument on ISEE 3 is an electrostatic analyzer [ Bame et al., 1978a]. We use the values adjusted according to the formulas of Petrinec and Russell . The ISEE 3 magnetometer has been described by Frandsen et al. . The IMP 8 Faraday cup and the IMP 8 orbit have been described by King , and the magnetometer has been described by Lepping et al. . The energetic particle experiment (EPE) measures the directional flux in the ecliptic plane for ions and electrons over the energy range from 50 keV to several MeV [Williams, 1977]. The ISEE 1 and 2 magnetometers have been described by Russell . The energetic particle instruments on ISEE 1 and 2 are solid-state detector telescopes that measure the directional flux of ions and electrons over the energy range from ~25 keV to 3 MeV [Williams et al., 1978]. The ISEE 2 plasma instrument is a 16-channel electrostatic analyzer [ Bame et al., 1978b]. We have adjusted the gain for detector aging according to the method of Song et al. , that assume that the temperature is measured correctly and ion pressure gradients are balanced by magnetic pressure gradients at quiet discontinuities. The ground-based magnetometers have been described by Rostoker , and the database has been described by Russell .
3.1 Spacecraft Location and Expected Neutral Sheet and Magnetopause Configurations
ISEE 3 was in its libration point halo orbit during this interval, 206 RE upstream from the Earth. If we assume a solar wind discontinuity convecting past ISEE 3 with its normal aligned along the solar wind flow (approximately the X axis) with a speed Vsw=450 km/s, then it would arrive at the nose of the magnetosphere after about 46 min. Table 1 gives the positions of the four spacecraft in GSM coordinates at 3 times from 0840 to 1050 UT.
Table 1. Four Spacecraft Positions in GSM From 0840 to 1050 UT
|Spacecraft||Time, UT||X, RE||Y, RE||Z, RE|
|Fig. 1. Magnetotail cross section and current sheet positions at 0840 and 1050 UT, calculating the tail boundary from the formula of Petrinec and Russell  and the current sheet position by Hammond et al.  with the solar wind conditions measured by ISEE 3. The lower panel shows the magnetopause cross section in the GSM X-Y plane at 1050 UT. The solar wind dynamic pressure is 4 nPa, and the interplanetary magnetic field IMF Bz is 5 nT at 0840 and 4 nPa and -9 nT at 1050 UT respectively. The dipole tilt angle of the Earth is 6.2o toward the Sun at 0840 UT and 12.6o toward the Sun at 1050 UT. The top panels show the cross-tail sections viewed from the Sun at ISEE at 17.7 RE and IMP 8 at 35 RE downtail at 0840 UT and at 16.3 and 34.5 RE at 1050 UT. On the left, the tail lobe radius is 19 RE and on the right 22 RE at 0840 UT and 20 RE and 24 RE at 1050 UT respectively.|
Figure 1 gives the expected tail cross section and current sheet configurations corresponding to interplanetary conditions and dipole tilt angle of the Earth at 0840 and 1050 UT and gives the magnetopause cross section at 1050 UT appropriate to the solar wind conditions at that time. The magnetopause location is calculated from the Petrinec and Russell  model and the current sheet position by the Hammond et al.  method. The top four panels show the tail vertical cross section corresponding to the ISEE 1, 2 and IMP 8 positions, respectively. The curves across the center of the tail are the expected current sheet positions. The bottom panel shows the X-Y plane cross section of the magnetopause with the three spacecraft positions at 1050 UT together with the positions of the synchronous spacecraft GOES 2 and GOES 3 in the interval of 0900-1100 UT, shown by two arcs. At 0900 UT, GOES 2 was at 0218 LT, and GOES 3 was at midnight. Over this period, we expect ISEE 1 and 2 to slowly cross the current sheet in the center of the tail from the north to the south and for IMP 8 to stay firmly in the northern lobe. As we will see below, the former expectation is correct but the latter is not, implying a significant twist in the tail current sheet.
3.2. Solar wind and geomagnetic conditions
|Fig. 2. Solar wind conditions measured by ISEE 3 at the forward libration point. From top to bottom are 5 min averages of the interplanetary magnetic field in GSM coordinates, the solar wind proton density, the bulk speed, and the dynamic pressure. The vertical line is at 0906 UT.|
Solar wind conditions are shown in Figure 2, from top to bottom are the IMF Bx, By, Bz components in GSM coordinates and the field strength B, then the solar wind proton density, bulk speed, and dynamic pressure. The IMF Bz remained mostly northward from 0753 to 0906 UT and then remained steady at about -9 nT for over 90 min. If we assume that the scale size of structures in the IMF is large compared to the distance from Earth to ISEE 3 we can calculate that the solar wind convection time from ISEE 3 to the nose of the magnetosphere is almost 46 minutes. Then the period at ISEE 1, 2 and IMP 8 can be roughly divided into two parts: the first hour from 0840 to 0950 UT, corresponding to IMF northward Bz at the dayside magnetopause; and the second hour in the tail from 0950 to 1050 UT, corresponding to IMF Bz southward. For the whole duration shown in Figure 2, the dynamic pressure V2 remained about 3.5 nPa. The solar wind bulk speed Vsw was steady near 460 km/s. The proton density varied from 8 to 9 cm-3, and IMF magnitude increased slowly from 11 to 13 nT. The IMF By remained negative during this period, at about -8 nT during the period of northward Bz and -4 nT during the period of southward Bz. The interval from 0840 to 1050 UT was in the recovery stage of a moderately strong storm and in the middle of a series of auroral electrojet activations with AE index increases from 300 to 600 nT, as shown in Figure 3.
|Fig. 3. Geomagnetic conditions presented by hourly AE and Dst indices. The time interval studied is between the vertical lines. It is in the middle of a series of electrojet activations and during the recovery phase of a moderate storm.|
|Fig. 4. A map of the North America ground stations we used for timing the arrival of the IMF southward turning and studying the electrojet. These stations are mainly on the nightside during the event period. The grey lines show the location of magnetic midnight at 0840, 0950, and 1050 UT, respectively. The solid line with an arrow presents the rough position of the westward electrojet.|
Figure 4 is a map of all ground stations used in our study, which were near midnight during the event period. Also shown are the expected footprints (using the Tsyganenko 1995 model) N. Tsyganenko, (personal communication, 1995) of the field lines passing through the geosynchronous satellites GOES 2 and 3 stationed at 100oW and 135.3oW longitude, respectively. The vertical bars represent the range of locations associated with varying geomagnetic activity. Several polar cap magnetograms (Cambridge Bay, Resolute, Pelly Bay, and Alert) are used for timing the arrival at Earth of the IMF Bz southward turning. As shown in Figure 5, the H component at Alert began to decline at 0947 UT, and the others began to rise. We believe this time marks the arrival time of IMF southward turning of 0906 UT at ISEE 3 based on our estimate of the convection time. A strong westward electrojet with an onset of about 0953 UT was detected by the ground-based magnetometers, as shown by Figure 6a. If this electrojet intensification is a standard substorm expansion phase, the preceding growth phase is rather short, about 6 min, as suggested by the gradual change in H at Sitka beginning at about 0947 UT. From College to Whiteshell the electrojet was very strong, but at Ottawa and Saint John's, it was weak, perhaps because of those stations being far from the center of the electrojet. The vertical line drawn at 0953 UT in Figure 6a is the onset time of southward turning of the magnetic field seen by ISEE 1 and 2 at 17 RE in the tail. The position of the inferred center of the electrojet is shown by the heavy east-west line in Figure 4. We interpret the delay in onset at Meanook as the time required for the northward expansion to reach that station.
|Fig. 5. One min averages of H components measured at several polar cap stations on April 22, 1979. The time at the vertical line is 0947 UT, at which time H components begin to increase or decrease.|
One min averages of H and Z components measured along a chain at constant latitude from College to Saint John's. The time at the vertical line is 0953 UT, which is the onset of the southward magnetic field in the tail.
|Fig. 6b. Three components of the magnetic field and the total field measured at synchronous orbit by GOES 2 and 3 at 3 s resolution. The Hp component is along the north rotational pole, the He component is eastward, and Hn is radially outward.|
At GOES 2 and GOES 3 synchronous orbit, the changes in the magnetic field showed surprisingly little correlation with the changes in the tail or on the ground. Figure 6b shows the magnetic field records obtained by GOES 2 and 3 from 0800 to 1200 UT (H. J. Singer, personal communication, 1995). The magnetic field had no evident variation until 1030 UT at GOES 2. At 1030 UT, the magnetic field became more "dipole-like" as the component parallel to the rotation axis, Hp, suddenly increased by about 20 nT and the eastward component, He, decreased about by 40 nT. These variations occurred significantly later than the sudden decrease in the H component at Meanook (1015 UT) which we attribute to be due to the poleward expansion of the electrojet. In the magnetotail at about 1015 UT, a strong disturbance was detected at the current sheet center that may have corresponded to this poleward expansion but there seems to have been little effect of this on GOES 2 and 3 except perhaps an increase in the level of turbulence at GOES 3 at about 0100 LT. GOES 3, whose conjugate point is a few degrees to the west of Meanook, sees a slight change in the slope of each of the components at 0942 UT, while GOES 2 whose conjugate point is close to Whiteshell sees nothing. At the time of the sudden onset of a southward field at ISEE 2 in the tail (0953 UT) and the sudden onset at College, GOES 3 shows a slight change in the Hn component, and GOES 2 sees nothing. The Hn component on GOES 3 turns positive at about 1007 UT and then becomes noisy on all components at 1010 UT.
3.3. Magnetic Conditions
|Fig. 7a. The magnetic variation in the tail from 0840 to 0950 UT measured by ISEE 1 (grey line), ISEE 2 (thick line) and IMP 8 (thin line) under the IMF northward conditions. Magnetic data shown here and on subsequent plots are 12 s averages every 4 s.|
|Fig. 7b. The plasma variation in the tail from 0840 to 0950 UT measured by ISEE 2 under the IMF northward conditions. Plasma data shown here and on subsequent plots were obtained every 12 s. The top panel includes the plasma total velocity and Vx component. Here Vx>0 means flow toward the Earth, and Vx<0 means flow downtail. The ratio of Tmax to Tmin in the third panel gives a measure of temperature anisotropy. In the bottom panel is the total pressure, calculated by (B2/2µ~˜+ nkT). PB is the magnetic pressure and PT is the thermal pressure.|
|Fig. 7c. Energetic particle measurements from ISEE 1 for the period 0840-0940. The top panel shows the E>30 keV flux of electrons. The next four panels show the speed, direction of ion flow, and gradient scale length deduced from >30 keV ions. Toward designates the range of flow longitudes (90o-270o) in which there is a component of flow toward the Earth. Away designates flow in the opposite direction. Up designates flow with a positive Z component.|
The magnetic and plasma variations at ISEE 1 and 2 in the near tail for northward IMF conditions are shown in Figures 7a, b, and c, respectively. This period provides an ideal opportunity for studying the current sheet structure under moderately quiet conditions using the spaced observations of ISEE 1 and 2 as they crossed the current sheet. The most visual manifestation of the current sheet crossing was the variation in the Bx components in Figure 7a as the current sheet moved up and down. We attribute this variation to a wave propagating down the tail, and because the ISEE 2 variation (as determined by the minima and maxima in the Bx component) led that of ISEE 1 by about 1 min on average, the velocity was almost 95 km/s directed downtail. We note that since ISEE 2 is further south than ISEE 1, it did not enter as far into the northern half of the tail as did ISEE 1. During this time the By and Bz components were highly variable but remained principally negative and positive, respectively. The By components of ISEE 1 and 2 were quite similar, and in general the variations in By arrived at ISEE 1 first and then ISEE 2 second, indicating the structure was propagating or convecting toward the Earth. The same situation occurred in the Bz components. This is most readily seen in the dipolarization observed at ISEE 1, 2 and IMP 8 at about 0847 UT. There was a gradual northward turning seen first at IMP 8 at 0845 UT. Then these followed a very sharp northward turning first at ISEE 1 and then at ISEE 2. The ISEE 2 plasma data, shown in Figure 7b, reveals that the velocity is earthward from 0840 to 0900 UT, suggesting that the By and Bz structure is convected toward the Earth from the more distant tail. The top panel gives the total plasma velocity (top trace) and Vx component (bottom trace). Between 0847 and 0900 UT, Vx was almost the same as the total velocity, reaching a value as high as 240 km/s. By integrating the flow velocity during this flow burst, whose average velocity was 107 km/s, we estimate that the plasma moved about 11 RE toward the Earth. After 0920 UT, since Vx was mainly positive and close to the total velocity, the plasma moved mainly toward the Earth at a very low speed. During the IMF northward period, the plasma sheet remained roughly isotropic as shown in the third panel. The ratio Tmax/Tmin was less than 1.2. The total pressure (B2/2µ~˜+nkT) was nearly constant, as the magnetic pressure and thermal pressure varied out of phase (we corrected the plasma density by a factor of 2.15 in order to match the total pressure to that found in the tail lobes away from the plasma sheet. This also led to constancy of the total pressure during the quiet tail period). In summary, during the entire IMF northward period, a surface wave moved down the tail while the plasma in contrast was stationary or moved toward the Earth.
A dipolarization of the magnetic field occurred at 0847 UT, most sharply at ISEE-1 and 2. This change was clearer at ISEE 1 near the current sheet center where the Bz component of ISEE 1 reached 20 nT compared to a total field of 22 nT. This dipolarization area crossed ISEE 2 48 s later at which time Bz peaked at 16 nT in a total field of 26 nT. After the dipolarization event, on average throughout the rest of the period of northward IMF, the gradient of Bz down the tail was very small as indicated by the near equality of Bz at ISEE 1, 2, and IMP 8.
Figure 7c shows energetic electron and ion observations from ISEE 1. From top to bottom are shown the greater than 30 keV electron fluxes, the ion bulk velocity magnitude and direction, and the e-folding scale length derived from ion anisotropies at the same pitch angle but with gyrocenters above and below the satellite. An examination of both the ISEE 1 and 2 energetic particle data shows a slight earthward anisotropy consistent with the Earthward plasma flow seen on ISEE 2 before 0900 UT. The energetic electrons show a local peak in flux at the time of the northward turning and have a weak bidirectional field-aligned anisotropy. There is no evidence for open field topology either before, during, or after this northward turning.
Another very significant phenomena occurred during the northward IMF conditions. IMP 8 entered the southern half of the tail at 0846 UT and remained there about half an hour. According to Figure 1 and Table 1, IMP 8 should have stayed in the northern tail at this time. However, as shown in Figure 7a, the Bx component of IMP 8 became negative and reached a minimum of -21 nT. This implies that the tail current sheet was twisted significantly under these IMF Bz and By conditions. IMP 8 stayed in the southern tail from 0846 to 0916 UT except for two brief excursions back and forth across the current sheet. After 0916 UT, IMP 8 remained in the northern tail for the remaining period under study. A similar twist has been reported by Kaymaz et al. [1994a, b] in a statistical study of IMP 8 magnetometer data. Reference to the IMF shown in Figure 2 suggests that the entry into the southern tail was associated with the northward turning reaching the Earth (convection time about 48 min) and the entry into the northern tail again at 0916 UT associated with the return to a near- horizontal magnetic field at 0832 UT (convection time about 44 min).
|Plate 1. Energetic particle measurements and magnetometer data from IMP 8 for the period 0755 to 1200 UT on April 22, 1979.|
The IMP 8 energetic particle data are shown in Plate 1. The IMP 8 energetic particle flux, along with the magnetic field is consistent with a flapping tail and closed field lines. The maximum flux is found in the middle of the current sheet, and energetic electrons (>30 keV) are present along with the energetic ions (>50 keV) throughout the period 0840-0950 UT.
3.4. Magnetic and Plasma Variations for Southward IMF, 0950-1050 UT
|Fig. 8a. The magnetic variation in the tail from 0950 to 1010 UT measured by ISEE 1 (grey line), ISEE 2 (thick line) and IMP 8 (thin line) under the IMF southward conditions. The left vertical line at the time of 0953 UT shows the time of the onset of reconnection in the tail. The right vertical line is at about 1000 UT when a cross tail flow is measured by ISEE 2.|
|Fig. 8b. The plasma variation in the tail from 0950 to 1010 UT measured by ISEE-2 under the IMF southward conditions. From the top to the bottom every panel shows the same parameters as Figure 8a with the same Y axis scales.|
|Fig. 8c. Energetic particle measurements from ISEE 1 for the period 0940 to 1040 UT. The top panel shows the E>30 keV flux of electrons. The following four panels show the speed, the direction of ion flow, and the gradient scale length deduced from >30 keV ions.|
The magnetic and plasma variations in the tail from 0950 to 1010 UT are given in Figures 8a and 8b . Because of the increased dynamical behavior of the tail once the IMF had turned southward, this and succeeding panels show only 20 min of data per figure. At 0953 UT, ISEE 1 and 2 both detected a strong southward turning of the magnetic field in the center of the tail. This is only 6 min after we believe the southward IMF reached the nose of the magnetosphere. This time is nearly coincident with the onset of activity on the ground. The southward turning of the magnetic field in the tail was 18 s earlier at ISEE 2 (the satellite closer to the Earth) than at ISEE 1 and reached -20 nT at both spacecraft. The southward turning was similar at the two spacecraft but lasted longer at ISEE 1. However, the By variation was quite different. In fact, the By component of the ISEE-1 magnetic field reversed sign from positive to negative during its negative Bz excursion. This behavior could be due to several quite different field configurations, and we cannot uniquely define such a complex field configuration with just two spacecraft. One geometry that is consistent with the data is that of a very intense flux rope with its axial current flowing tailward (along X) and its center lying between ISEE 1 and 2 in the Z direction and slightly to the dawnside of both (in the Y direction). Both ISEE 1 and 2 were in the edge of the plasma sheet current region to judge by the variation in Bx. A problem with this explanation is that the rope has to be so small that it would be highly coincidental that ISEE 1 and 2 both entered it. Alternatively, the southward turning could represent reconnection over a significant region (several RE?), and the By differences could represent shear in the field in principally the vertical direction. From Figure 8b, we see that before the supposed reconnection onset the plasma motion was directly toward the Earth, but after that time it moved directly away from the Earth with a speed of about 250 km/s. This is the expected behavior of reconnection. While the magnetic field was southward, the plasma was heated slightly anisotropically, and the total pressure rose slightly. As soon as the field turned northward at ISEE 2 (for which we have plasma data), both the accelerated flow and temperature anisotropy stop, as shown in the first and fourth panel of Figure 8b. We interpret this not as a cessation of the acceleration but a motion of the acceleration region. However, after the field turned "northward", the plasma behavior was not a mirror image of the southward behavior. The flow velocity was only weakly earthward, and the Bz component was at most weakly southward. The energetic particles indicate a dawn-to-dusk anisotropy, which if interpreted as a flow shows that the predominant flow at this time is cross tail. At 0950 UT, earthward flow commenced (see Figure 8c; flow velocities and gradients derived from ISEE 1 energetic ions). At that time, electron fluxes began dropping. Over the following hour, they drop faster during tailward flow episodes. Occasionally, measurable energetic electron anisotropies indicating tailward beams appear; however, these are seen as enhancements over the ambient isotropic electron flux, suggesting the existence of an intermittent electron acceleration occurring on these field lines, earthward of the spacecraft. If this signature is due to reconnection, it implies that the reconnection point also moved rapidly, departing tailward at close to the velocity of the earthward accelerated plasma (as measured in its reference frame). We note that the plasma motion exhibited from 0953:30 to 0955 UT would have removed 3.5 RE of the plasma sheet inside of 16.5 RE. Since it appears that the neutral point moved at the speed of the flow, the onset of reconnection may have occurred at 13 RE here. We also note that the flow at ISEE 1 closer to the center of the plasma sheet persists much longer than at ISEE 2.
After 1000 UT, ISEE 1 gradually crossed back and forth through the current sheet, while ISEE 2 stayed in the current sheet but near the current sheet boundary. This verifies that the current sheet became thin, with a half thickness of not more than 4000 km, which is consistent with the supposed necessity of a very thin current sheet to enable reconnection to begin [McPherron et al., 1973]. This current sheet thinning continued at least until 1021 UT when the ISEE pair went into southern tail. The significant difference between plasma total velocity V and Vx tells us that there was a strong cross-tail flow near the current sheet boundary while the current sheet became thinner. Perhaps the current sheet thinning is associated with this cross-tail flow. The total pressure remained constant, and the plasma was nearly isotropic while this flow was in progress.
The energetic particle data from ISEE 1 and 2 are consistent with this current sheet thinning. As shown in Figure 8c, the gradient anisotropy indicates an e-folding scale of about 0.1 RE during the period of fast flow, about 0.3 to 0.4 RE between 1000 and 1010 UT. The fast tailward flow, which begins during the 1010 to 1020 UT period at ISEE 2 (see Figure 9b), starts earlier at ISEE 1. The energetic particles show flows of about 600 km/s beginning at 1011 UT and continuing until 1021 UT in contrast to the 100 km/s flow at ISEE 2. There are several tailward electron bursts above the ambient flux over this interval. The ISEE 2 energetic ions (not shown) also show tailward flow at about 1016 and 1018 UT, consistent with the plasma data (in Figure 9b), along with accompanying tailward energetic electron bursts.
During the entire period as shown in Figure 7d, IMP 8 appeared to be close to the plasma sheet. The magnetic field remained northward until the data gap at 1029 UT. At 1000 UT, IMP 8 temporarily approached the center of the current sheet. This implies that the current sheet was twisted at this time again. The magnetic pressure and the thermal pressure varied out of phase so the total pressure was nearly constant. The only major change in the energetic particles at IMP 8 from the flux levels present before 0950 UT is in the flux of E>250 keV electrons. These particles begin to increase in intensity simultaneously with the onset of what appears to be an expansion phase of a substorm at 0953 UT, coincident with the arrival of southward magnetic field at ISEE 1 and 2.
|Fig. 9a. The magnetic variation in the tail from 1010 to 1030 UT measured by ISEE 1 (grey line), ISEE 2 (thick line) and IMP-8 (thin line) under IMF southward conditions.|
|Fig. 9b. The plasma variation in the tail from 1010 to 1030 UT measured by ISEE 2 under the IMF southward conditions. From the top to the bottom, every panel shows the same parameters as Figure 8a with the same Y axis scales.|
As shown in Figure 9a, the period from 1010 to 1030 UT was extremely disturbed in the magnetic field. At ISEE 2, the Bz component turned southward gradually from 1005 UT, 7 min before the rapidly negative Bz turning at ISEE 1. The tail field remained southward for about 15 min at ISEE 2 and 6 min at ISEE 1. From 1009 to 1016 UT, ISEE 1 stayed on the southern side and very near the current sheet center. Here the By component remained at a high value, reaching 16 nT. Meanwhile, Bz was southward -16 nT, and ISEE 2 stayed near the current sheet boundary with a slightly increased Bx, small By, and small southward Bz. At about 1015 UT, the flow at ISEE 2 changes from being slow and toward the Earth to being a little faster (about 100 km/s) and away from the Earth. We note that flow toward the Earth with a southward IMF is not expected in a simple reconnecting tail geometry.
At 1018:30 UT ISEE 1 rapidly crossed back and forth through the current sheet, accompanied by a large By variation, reaching -26 nT, and a Bz northward turning, reaching 11 nT. After the northward going current sheet crossing, the By component reached +20 nT. Thus the change in By associated with the current sheet crossing was more than 45 nT from positive to negative then from negative to positive. Also, the shape of the By variation is roughly symmetric with the northward current sheet crossing at 1018:30 UT denoted by the left vertical line in Figure 9a. When ISEE 1 crossed the current sheet in the southward direction the By component changed direction, from dawn to dusk. The main contribution to BT came from the Bz component, as shown by the right vertical line. During this period, ISEE 2 was in the southern tail, and plasma was moving down the tail. In this region, the thermal pressure was low and magnetic pressure was high, as shown in Figure 9b. Although a measurable temperature anisotropy was detected here, it was not associated with plasma heating. In fact, the average temperature was slowly decreasing. Again, the structure of this strong disturbance is difficult to identify with only two spacecraft.
From 1025 to 1028 UT, a tailward bulk plasma flow with speeds up to 400 km/s was detected by ISEE 2, as shown in Figure 9b, and the plasma temperature anisotropy increased. However, the southward magnetic field is weak at this time. Afterward, the plasma flow suddenly turns toward the Earth at 1028 UT. At the same time, the current sheet possibly became bent. This caused the magnetic field to decrease sharply at ISEE 2 and then sequentially at ISEE 1. ISEE 1 even crossed back and forth through the center of the current sheet. The whole structure moved downtail with a bent current sheet, while the plasma was flowing earthward. A repeat of this behavior on a longer timescale is seen later at 1044 UT. At IMP 8, the midtail field was even quieter than that in the previous period. The magnetic field and three components changed slowly and smoothly.
|Fig. 10a. The magnetic variation in the tail from 1030 to 1050 UT measured by ISEE 1 (grey line) and ISEE 2 (black line) under the IMF southward conditions. There is a data gap in IMP 8 in this period.|
|Fig. 10b. The plasma variation in the tail from 1030 to 1050 UT measured by ISEE 2 under the IMF southward conditions. From the top to the bottom, every panel shows the same parameters as Figure 8a with the same Y axis scales.|
From 1030 to 1050 UT, ISEE 1 and 2 initially stayed in the south as plotted in the top panel of Figure 10a. ISEE 1 gradually entered the current sheet from 1035 UT accompanied by opposite By components at both spacecraft. At the same time, plasma was moving down the tail until 1044 UT, as shown by the velocity panel in Figure 10b. The energetic particles, however, indicate flow toward the Earth. Perhaps there is an acceleration source tailward of the spacecraft sending accelerated energetic ions in the opposite direction from the thermal plasma flow. A gradient anisotropy is also present at this time, indicating a thin structure. After that time, as noted above, a phenomenon very similar to that at 1028 UT occurred with a longer timescale. The plasma density increased at 1044 UT where the plasma motion reversed direction. A large temperature anisotropy also appeared at the two edges of the density peak. However, there was only a slight increase in total pressure because of the perfect complement between magnetic and thermal pressure variations. There was no plasma heating even though some high-temperature anisotropies existed during this period. For this current sheet disturbance, the ISEE pair were a little bit farther from the current sheet center so that ISEE 1 did not cross the current sheet center and the magnetic field decrease was not so large as the previous one.
At about 1100 UT, the field at ISEE 1 dipolarizes significantly. The Z component becomes the dominant component of the field. This indicates a profound change in the current sheet structure to one that is far more distributed in Z, and probably diminished integrated current. At this time, the energetic electrons are abruptly accelerated. Their fluxes rise to levels higher than at any time during the event. We interpret this as betatron acceleration of the population that resided in the weak field region near the center of the current sheet, which no longer exists.
|Fig. 11. The magnetic variation in the tail from 1050 to 1120 UT measured by ISEE 1 (grey line), ISEE 2 (thick line) and IMP 8 (thin line) under the IMF southward conditions. IMP 8 has no magnetic data until 1102 UT. The vector components of the magnetic field are not shown at ISEE 2 after it enters eclipse because of the resulting uncertainties in the despinning process.|
There was a gap in the IMP 8 magnetic data from 1029 to 1102 UT. During this period Bz at IMP 8 eventually turned southward (see Figure 11). This was at least 30 min later than the onset of southward magnetic fields at ISEE. The large negative Bz component (-18 nT) and small Bx, By components imply that the magnetic field at 35 RE had become very untail-like. We do not believe that the IMP 8 spacecraft left the tail at this time. The field is quiet and not magnetosheath-like. Also, the location of the spacecraft is well inside the expected position of the tail boundary.
In this paper, a global perspective of the near and more distant tail during quiet and disturbed conditions corresponding to northward and southward IMF Bz component has been presented for the period on April 22, 1979, from 0840 to 1050 UT. During this period, ISEE 3 monitored the solar wind 206 RE upstream and 82 RE to dawn side of the Earth, and three spacecraft ISEE 1, 2, and IMP 8 were at 17, 16, and 35 RE behind the Earth in the geomagnetic tail close to the expected location of the current sheet. Such coverage of key regions of the tail with three satellites is rare, and one of the objectives of the International Solar Terrestrial Physics program is to obtain much more of such coverage. During this interval, the IMF changed from strongly northward, during which time the tail underwent some relaxation, to strongly southward when the tail became dynamic. Moreover, at this time the North American magnetometer chains were in the midnight sector allowing us to time the onset of events accurately and to determine well both the latitudinal and longitudinal extent of features. These 2 hours are rich in detail about the behavior of the tail under quiet and disturbed conditions and should provide an excellent testbed for future computer simulations. During the first hour when the IMF was northward, we observed the following:
1. A dipolarization of the tail field that was observed by the ISEE spacecraft pair near the current sheet center, moving toward the Earth at close to the speed of the convecting plasma.
2. Plasma temperature that was roughly isotropic at 16 RE downtail.
3. A total pressure that remained stationary at 16 RE downtail.
4. A Bz component that was often surprisingly uniform from 16 to 35 RE down the tail.
5. Closed field lines at ISEE 1, 2 and IMP 8 for the entire period.
During the second hour when the IMF was southward, we observed the following:
1. A southward turning of IMF Bz that arrived at the Earth 41 minutes after measured at ISEE 3.
2. An onset of activity at the ISEE pair at 16 RE that was simultaneous with the onset of ground activity.
3. A current sheet half thickness that was not more than 4000 km thick.
4. Plasma convection that was mainly away from the Earth but which often differed significantly between ISEE 1 and 2 only 1 RE apart.
5. An energetic particle flux that continued to decline but never to levels indicative of open field lines.
6. A large negative Bz detected at the ISEE pair on two separate occasions at only 16 RE downtail.
7. An onset of southward Bz at IMP 8 delayed by more than 30 min.
8. Inactivity in the geomagnetic field near midnight at synchronous orbit while in the ionosphere strong electrojet activity was occurring well equatorward of the expected conjugate point of the synchronous orbit spacecraft.
From our overall study of this event period, we conclude that while not static, the magnetotail is much quieter under the northward IMF conditions than southward. A wave formed by the tail flapping and propagated away from the Earth at about 95 km/s, while plasma convection was usually toward the Earth. The current sheet was significantly twisted at the position of IMP 8 under the influence of the By conditions even though Bz was northward. For southward IMF conditions, the magnetotail was extremely disturbed, and a strong westward electrojet on the ground arose when the tail became active. A reconnection point appeared to form inside of ISEE and move down the tail rapidly, but still ISEE remained on closed field lines according to the energetic particle data. In the center of this apparent reconnection point, there was a shear in the field that was probably principally in the vertical direction. On two occasions, a bent current sheet formed at 16 RE moved downtail opposite the direction of the plasma flow. The magnetic field was highly structured, and its configuration was difficult to discern with only two spacecraft. The extensive activity of the equatorial electrojet at the onset of activity at ISEE, despite the inactivity at GOES 2 and 3 that should have been between the auroral zone activity and ISEE in invariant latitude, indicates that our geomagnetic models do not indicate the true extent of the distortion of the magnetic field at disturbed times. Clearly, synchronous orbit was deep in the moderately undisturbed magnetosphere, but the field lines above Whiteshell nominally at a lower L value went far into the geomagnetic tail, roughly to 16 RE. Zhou et al., [this issue] discuss these phenomena and the current sheet structure in more detail.
The authors wish to thank N. F. Ness, R. P. Lepping, S. J. Bame, E. J. Smith, and J. T. Gosling for providing us and the NSSDC with the correlative solar wind and tail data used in this investigation and also H. J. Singer for providing the synchronous orbit data. This work was supported by research grants from the National Science Foundation ATM 94-13081 and from the National Aeronautics and Space Administration, NAGW-3974. This work was also supported by the National Science Foundation of China under the grant of HT 49384007.
The Editor thanks three referees for their assistance in evaluating this paper.
Baker, D. N., T. I. Pulkkinen, R. L. McPherron, J. D. Craven, L. A. Frank, R. D. Elphinstone, J. S. Murphree, J. F. Fennell, R. E. Lopez, and T. Nagai, CDAW 9 Analysis of magnetospheric events on May 3, 1986: Event C, J. Geophys. Res., 98, 3815, 1993.
Bame, S. J., J. R. Asbridge, H. E. Felthauser, J. P. Gore, H. L. Hawk and J. Chavez, ISEE-C solar wind plasma experiment, IEEE Trans. Geosci. Electron., GE-16, 160, 1978a.
Bame, S. J., J. R. Asbridge, H. E. Felthauser, J. P. Gore, G. Paschmann, K. Lehmann, and H. Rosenbauer, ISEE-1 and ISEE-2 fast plasma experiment and the ISEE-1 solar wind experiment, IEEE Trans. Geosci. Electron., GE-16, 216, 1978b.
Frandsen, A. M. A., B. V. Connor, J. Van Amersfoort, and E. J. Smith, The ISEE-C vector helium magnetometer, IEEE Trans. Geosci. Electron., GE-16, 195, 1978.
Hammond, C. M., M. G. Kivelson, and R. J. Walker, Imaging the effect of dipole tilt on magnetotail boundaries, J. Geophys. Res., 99, 6079, 1994.
Jacquey, C., J. A. Sauvaud, J. Kandouras, and A. Korth, Tailward propagating cross-tail current disruption and dynamics of near-Earth tail: A multi-point measurement analysis, Geophys. Res. Lett., 20, 983, 1993.
Kan, J., and S.-I. Akasofu, Electrodynamics of solar wind - Magnetosphere ionosphere interactions, IEEE Trans. Plasma Sci., 17, 83, 1989.
Kaymaz, Z., G. L. Siscoe, N. A. Tsyganenko, and R. P. Lepping, Magnetotail views at 33 Re: IMP 8 magnetometer observations, J. Geophys. Res., 99, 8705, 1994a.
Kaymaz, Z., G. L. Siscoe, J. G. Luhmann, R. P. Lepping, and C. T. Russell, Interplanetary magnetic field control of magnetotail magnetic field geometry: IMP 8 observations, J. Geophys. Res., 99, 11113, 1994b.
King, J. H., Availability of IMP 7 and IMP 8 data for the IMS period, in IMS Source Book: Guide to the International Magnetosphere Study Data Analysis, edited by C. T. Russell and D. J. Southwood, p.10 AGU, Washington D. C., 1982.
Lepping, R. P., A. J. Lazarus, L. J. Moriarty, P. Milligan, R. S. Kennon, R. E. McGuire, and W. H. Mish, IMP 8 solar wind magnetic field and plasma data in support of Ulysses-Jupiter encounter: 13-31 January 1992, internal document, NASA laboratory for extraterrestrial physics Goddard Space Flight Cent., Greenbelt, Md., Dec. 25, 1992.
Lin, N.-G., R. L. McPherron, M. G. Kivelson, and R. J. Walker, Multipoint reconnection in the near-Earth magnetotail: CDAW 6 observations of energetic particles and magnetic field, J. Geophys. Res., 96, 19427, 1991.
Lopez, R. E., H. Spence, and C.-I. Meng, Simultaneous observation of the westward electrojet and the cross-tail current during substorms, in Magnetospheric Substorms, Geophys. Monogr. Ser., Vol. 64, edited by T. Iijima, T. A. Potemra, and J. R. Kan, pp. 123, AGU, Washington, D. C., 1991.
Lui, A. T. Y., R. E. Lopez, S. M. Krimigis, R. W. McEntire, L. J. Zanetti, and T. A. Potemra, A case study of magnetotail current sheet disruption and diversion, Geophys. Res. Lett., 15, 721, 1988.
McPherron, R. L., C. T. Russell, and M. P. Aubry, Satellite studies of magnetospheric substorms on August 15, 1968, 9, Phenomenological model for substorms, J. Geophys. Res., 78, 3131, 1973.
Petrinec, S. M., and C. T. Russell, Intercalibration of solar wind instruments during the International Magnetospheric Study, J. Geophys. Res., 98, 18963, 1993.
Petrinec, S. M., and C. T. Russell, Near-Earth magnetotail shape and size as determined from the magnetopause flaring angle, J. Geophys. Res., 101, 137, 1995
Rostoker, G., High latitude north American networks operative during the IMS, in the IMS Source Book, edited by C. T. Russell and D. J. Southwood, 159-169, American Geophysical Union, Washington DC, 1982.
Russell, C. T., The ISEE 1 and 2 fluxgate magnetometers, IEEE Trans. Geosci. Electron., GE-16, 239, 1978.
Russell, C. T., Increasing the accessibility of IMS magnetograms, EOS, 68, 1601, 1987.
Russell, C. T., and R. L. McPherron, The magnetotail and substorms, Space Sci. Rev., 15, 205, 1973.
Russell, C. T., M. Ginskey, and S. M. Petrinec, Sudden impulses at low- latitude stations: Steady state response for northward interplanetary magnetic field, J. Geophys. Res., 99, 253, 1994a.
Russell, C. T., M. Ginskey, and S. M. Petrinec, Sudden impulses at low- latitude stations: Steady state response for southward interplanetary magnetic field, J. Geophys. Res., 99, 13403, 1994b.
Russell, C. T., M. Ginskey, and V. Angelopoulos, Effect of sudden impulses on currents in the auroral ionosphere under northward interplanetary magnetic field conditions: A case study, J. Geophys. Res., 99, 17617, 1994c.
Sanny, J., G. L. McPherron, C. T. Russell, D. N. Baker, T. I. Pulkkinen, and A. Nishida, Growth-phase thinning of the near-Earth current sheet during the CDAW 6 substorm, J. Geophys. Res., 99, 5805, 1994.
Sauvaud, J. A., C. Jacquey, T. Beutier, R. P. Lepping, C. T. Russell, and R. J. Belian, Dynamics of the magnetospheric mid-tail induced by substorms: A multisatellite study, Adv. Space Res., 18 (8) 35 - 43, 1995.
Sergeev, V. A., D. G. Mitchell, C. T. Russell, and D. J. Williams, Structure of the tail plasma/current sheet at ~11 RE and its changes in the course of a substorm, J. Geophys., Res., 98, 17345, 1993.
Sergeev, V. A., V. Angelopoulos, D. G. Mitchell, and C. T. Russell, In situ observations of magnetotail reconnection prior to onset of a small substorm, J. Geophys. Res., 100, 19121, 1995.
Song, P., C. T. Russell, R. J. Fitzenreiter, J. T. Gosling, M. F. Thomsen, D. G. Mitchell, S. A. Fuselier, G. K. Parks, R. R. Anderson, and D. Hubert, Structure and properties of the sub-solar magnetopause for northward interplanetary magnetic field: Multiple-instrument particle observations, J. Geophys. Res., 98, 11319, 1993.
Wiens, R. G., and G. Rostoker, Characteristics of the development of the westward electrojet during the expansive phase of magnetospheric substorms, J. Geophys. Res., 80, 2109, 1975.
Williams, D. J., The ion-electron magnetic separation and solid state detector detection system flown in IMP 7 and IMP 8: Ep 3 50 keV, Ee 3 39 keV, NOAA Tech. Rep. ERL 393-SEL40, U.S. Dep. of Commer. Boulder, Colo., October 1977.
Williams, D. J., E. Keppler, T. A. Fritz, B. Wilken, and G. Wibberenz, The ISEE-1 and 2 medium energy particle experiment, IEEE Trans. Geosci. Electron., GE-16, 3, 270, 1978.
Zhou, X-Y., C. T. Russell, J. T. Gosling, and D. G. Mitchell, Three spacecraft observations of the geomagnetic tail during moderately disturbed conditions: Structure and evolution of the current sheet, J. Geophys. Res., this issue..
D. G. Mitchell, Applied Physics Laboratory, Johns Hopkins University, Laurel, MD 20723-6099. (E-mail: mitchell@spacemail. jhuapl.edu)
C. T. Russell, Institute of Geophysics and Planetary Physics, 6869 Slichter Hall, University of California, Los Angeles, CA 90095-1567 (E-mail: email@example.com)
X-Y. Zhou, Institute of Geophysics, Chinese Academy of Sciences, Beijing 100101, China (E-mail: firstname.lastname@example.org. cn)
Received April 17, 1996; revised February 24, 1997; accepted February 28, 1997.
Copyright 1997 by the American Geophysical Union.
Paper number 97JA00683.