Full text of DOI:10.1007/BF02900578

based observations affords unprecedented opportunities to
study the magnetospheric response to strong disturbances
in the solar wind, and the interaction of the shocks in the
solar wind, particularly those associated with interplane-
tary pressure shocks and coronal mass ejection. The ob-
servations show that the effects of solar wind pressure
suddenly increase on the magnetosphere is remarkable.
When the solar wind dynamic pressure suddenly increases,
the magnetosphere is compressed, beginning at the day-
side magnetopause and then continuing down the tail. If
the compression occurs, the principal change in the mag-
netosphere will be to shrink in size and increase its mag-
netic field strength accompanied by rapid motion of the
plasma “tied” to those field lines, and leads to the changes
in magnetic field and the plasma in various regions in the
Auroral substorm response to
solar wind pressure shock
1
HONG Minghua , WANG Xianmin , D. Chua
2
3
3
&
G. Parks
1
. Institute of Geology and Geophysics, Chinese Academy of Sciences,
Bejing 100101, China;
2
3
. Institute of Space Medico-Engineering, Bejing 100094, China;
. Geophysics Program, University of Washington, Seattle, WA98195,
USA
Correspondence should be addressed to Hong Minghua (e-mail: mhhong
mail.c-geos.ac.cn)
@
Abstract Two cases of auroral substorms have been stud-
ied with the Polar UVI data, which were associated with
solar wind pressure shock arriving at the Earth. The global
aurora activities started about 1ü2 min after pressure
shocks arrived at dayside magnetopause, then nightside
[8]
[9]
magnetosphere . Sibeck et al. found that the sudden
sharp increase in solar wind pressure initiated both the
ground response and the ringing of the outer magneto-
[
10]
spheric magnetic field. Gonzalez et al. have shown that
most storm sudden commencements (SSCs) are caused by
interplanetary shocks associated with coronal mass ejec-
tions (CMEs). There is a close relationship between solar
wind pressure shock and auroral activities. The pressure
auroras intensified rapidly 3ü4 min later, with auroral sub-
storm onset. The observations in synchronous orbit indicated
that the compressing effects on magnetosphere were ob-
served in their corresponding sites about 2 min after the
pressure shocks impulse magnetopause. We propose that the
auroral intensification and substorm onset possibly result
from hydromagnetic wave produced by the pressure shock.
The fast-mode wave propagates across the magnetotail lobes
with higher local Alfven velocity, magnetotail was com-
pressed rapidly and strong lobe field and cross-tail current
were built in about 1ü2 min, and furthermore the substorm
was triggered due to an instability in current sheet.
[11,12]
shock may cause dayside auroral activity
, and also
[13]
can trigger substorm. Kokubun et al. proposed that the
expansion phase of substorm can be triggered by the sud-
den solar wind dynamic pressure increase when the IMF
must have been initially southward for a period of the
order of 30 min or longer. A quite similar relation between
the IMF direction and the effects of solar wind pressure
increase on substorm activity was pointed out by Pretrinec
Keywords: auroral substorm, solar wind pressure shock, hydro-
magnetic wave.
[14]
[15]
et al. . Shue et al. recently presented another study in
which they demonstrate how increased solar wind plasma
density correlates well with substorm activity, but only as
Substorm is a basic solar wind-magnetosphere-
ionosphere coupling process. It has been the key problem
in magnetospheric physics in the past decades. Substorms
are closely related to interplanetary variations. Solar wind
dynamic pressure variations, IMF Bz component, inter-
planetary shocks all can trigger substorms and produce
geomagnetic effects. In different interplanetary conditions,
the solar-magnetosphere coupling mechanisms may be
different, which are responsible for the transfer of energy,
momentum, and plasma from the solar wind to the mag-
netosphere and the ionosphere. Up to now, several sub-
[16]
long as IMF Bz is negative. Jacquey
investigated the
effect of changes in the IMF and solar wind plasma pa-
rameters on that seen by the ISEE1 satellite. He also noted
that effective substorm triggering by solar wind pressure
pulse only occurred during the periods of already previ-
ously enhanced lobe magnetic field. The compressing
trigger of substorm can be readily understood as the onset
of the reconnection caused by the increased pressure on
the enhancement of reconnection at an existing reconnec-
tion site.
All the above studies emphasized the fact that the
southward IMF plays an important role in the substorm
onset associated with solar wind pressure shocks. In this
note, we try to illustrate the possibility that the sharp in-
crease in solar wind pressure triggered substorm onset
without southward IMF with two new observations.
[
1ü7]
storm models have been proposed
to explain the in-
ternal triggering processes due to plasma instability in
current sheet and external triggering processes due to IMF
Bz northward turning for substorm onsets in southward
IMF conditions. All these works were based on the fact
that the magnetic field connection between IMF and
magnetic field at dayside magnetopause play a main role
in the solar wind-magnetosphere coupling.
1
Observations
Recently, the end-to-end monitoring of Sun-Earth
environment by the fleet of International Solar-Terrestrial
Physics (ISTP) spacecraft and complementary ground-
The two events presented here were recorded in a
sequence of UVI images which onboard on Polar space-
craft. During these events Polar was passing through the
Chinese Science Bulletin Vol. 46 No. 18 September 2001
1547

NOTES
Northern Hemispheric polar cap sky, and the UVI was
operating with a single Lyman-Birge-Hopfield Band filter
lower-latitude region. The intensity center was over
70 MLAT in the 18 : 00 21 : 00 MLT, and less than
q
q
ü
: 00 ü24 : 00 MLT. A few minutes
(
LBHlong: 160ü180 nm) at 36.8 s time resolution, and
70 MLAT in the 21
recorded the whole processes of auroral substorm evolu-
tions. Prior to the auroral activities, the Wind and Geotail
located in upstream of solar wind observed the inter-
planetary shocks propagating to the Earth. The pressure
shocks were due to solar wind ion densities and velocities
increase discontinuity associated with CMEs. Prior to
shock arrival, the corresponding IMF with IMF Bz|0 and
southward in a short interval favor to examine the role of
solar wind pressure shock playing in Solar-Terrestrial
coupling processes. The observations in synchronous orbit
and other regions are very helpful to studying the re-
sponses of magnetosphere and ionosphere to solar wind
pressure shocks.
later, the aurora extended to poleward over 80qMLAT,
equatorward lower 60qMLAT, and propagated eastward.
By 07 : 30 UT, the auroral activity gradually quenched,
and recovered to the previous level. In contrast to the iso-
lated substorm, this kind of auroral substorm did not ap-
pear the west-traveling surge and the initial brightening
region was extensive.
(
i ) 1998-08-26 event. The solar wind pressure
shock first was observed at about 06 : 41UT by Wind
which was located at (x, y, z)GSM | (117, ꢀ23, 3)R (see
E
fig. 1). The data show that there are IMF, ion density, and
velocity discontinuities, and produce dynamic pressure
shock. At downshock (unshocked) IMF has southward
component in a short period. At upshock (shocked) IMF Bz
turned northward in about 1 min, the magnitude of mag-
netic field jumped from about 6 nT to 22 nT, ion density
ꢀ
3
ꢀ3
from about 6 cm to 14 cm ; the velocity from 450 km/s
to 600 km/s; and dynamic pressure from about 2 nPa to 8ü
10 nPa. This shock structure kept for about 50 min, then
decreased slightly, but still in a higher value (7ü8 nPa).
The pressure shock front fluctuated slightly due to the
change in density. At about 06 : 49 UT, this shock was also
observed by Geotail located at (x, y, z)GSE|(25, 7, 0)RE, its
structure did not change except that IMF Bz only has about
10 min southward component at down shock. The arrival
times of shock at Wind, Geotail and dayside magnetopause
are different. With the magnetic field data from Beijing
Ming Tomb Station, we determined the SSCs time due to
the shock arrival being about 06:51 UT, and inferred that
E
at
Fig. 1. The IMF and solar wind plasma data observed at Wind for
1
the shock arrived at dayside magnetopause at 10 R
about 06:51 UT.
998-08-26 event.
During this event, the UVI of global auroral activity
was taken by Polar spacecraft which was positioned at
During this event, the geomagnetic field in the polar
region response to pressure shock appeared dramatically.
The AL index decrease began at about 06:53 UT, then
decreased rapidly to – 1200 nT with a ratio of a200 nT/
min. This indicates that the AL index is very sensitive to
the change in solar wind pressure. At the same time, the
change of local electric ejection index CU, CU from
CANOPUS, and the H component of magnetic field from
SESAME, Antarctica, appear similar to that of AL. It is
evident that the auroral electric ejection increased rapidly
in polar ionosphere, and was consistent with the interval
of the auroral activity.
E
about 8.8 R altitude in the polar region (shown in Plate
ĉ(a)). At 06:50:52UT, prior to arrival of pressure shock,
the remnants of previous auroral activity in the 18:00ü
2
4 : 00 MLT were still visible, at about 06:53 UT, post-
noon aurora brightened suddenly, at the same time, night-
side aurora also started intensification, but faint. Then
about 1ü2 min later, the nightside aurora intensified rap-
idly. At about 06:57 UT, the intensity approached the
maximum. The auroral luminosity initiated almost at all
local times in nightside, it was most intense in 18: 00ü
24 : 00 MLT, and east-west asymmetrical. The dawnside
auroral zone is a little narrow and located in the
At the synchronous orbit, LANL1, LANL4 and
E
GOES8 were located at about (6.6R , 2.3MLAT,
1548
Chinese Science Bulletin Vol. 46 No. 18 September 2001

0
7:30MLT), (6.6R
E
, ꢀ10.2 MLAT, 13:40MLT), and (x, y,
, respectively. All of them had ob-
650 km/s, producing dynamic pressure from 2 to 15 nPa.
The shock arrived at magnetopause at 10 R at about 23:44
z)GSM Ĭ (ꢀ6, ꢀ3, 1)R
E
E
[17]
UT
.
served the response of inner magnetosphere to pressure
shock in their observation sites. LANL4 observed that the
energy fluxes of protons and electrons were enhanced
rapidly. This is consistent with the time of dayside auroral
activity; then about 1 min later, LANL1 observed that the
density and energy flux of high energetic protons in-
creased, in addition, the D flux at both LANL1 and
LANL4 sites increased apparently. At about 06:53 UT,
GOES8 at nightside observed the magnetic field enhanced
pronouncedly. This suggest that the compressing effec-
tiveness by pressure shock propagated to that region and
made the magnetic field become shock-like structure due
to the wave speed being slower in inner magnetosphere
than that in outer magnetosphere.
Meanwhile, Polar spacecraft was at the altitude
about 8.9 R distance around the center of polar cap, and
recorded the progression of the auroral substorm (see
E
Plate ĉ-7ü12). At 23:42:53 UT, prior to arrival of the
pressure shock, the nightside oval was more intense due to
remnants of previous activity. The auroral luminosity was
first observed to increase near noon at about 23:45 UT, the
remnant of previous aurora brightened again in the night-
side oval, and about 2 min later it increased rapidly, the
aurora was more intense at 18:00ü24:00 MLT. At first
the intensity center was around 65qMLAT, and then
drifted to higher latitudes. The poleward edge of oval
moved quickly from
a70q to
a q
75 MLAT. The auroral
The magnetic field was measured in northern tail
luminosity reached the peak at about 23:52 UT, and then
gradually decayed, by about 00:30 UT, Sept. 25, 1998, the
auroral activity recovered to lower level.
lobe by IMP8 which was positioned at (x, y, z) GSM
Ĭ
(
ꢀ24, 0, 19). The characteristics of magnetic field are as
follows: before 06:53 UT, the direction of magnetic field
was sunward with a small southward component, this is
the feature of the northern tail lobe field. It shows that
IMP8 was in the lobe at that time. At about 06:55 UT, the
magnetic field changed suddenly, IMF Bx changed from
sunward (|20nT) to antisunward (| ꢀ20nT), IMF By in-
creased from zero to about 20 nT, and IMF Bz turned
northward, got to about 10 nT. The changes caused by the
shock arrived at the site of IMP8. The magnetic field with
Bx |ꢀ20 nT, By|20 nT, and Bz|8 nT was just the mag-
ꢁ
netic structure in shock. The change in magnetic field in-
dicated that the magnetosphere was compressed when the
shock arrived at that region, the magnetopause boundary
moved inward, and led IMP8 to get out of tail lobe, and
into magnetosheath. The magnetic field observed by IMP8
suggests that the interplanetary shock got into magne-
tosheath and propagated along the magnetosheath to the
E
down tail, by about 06:55 UT reached tail about 24 R .
Assuming that the shock moved with a uniform velocity,
then at about 06:53 UT, it may get near-magnetotail x |
ꢀ
7R
E
, and acted on the near-Earth magnetotail.
(
ii ) 1998-09-24 event. Another event associated
with pressure shock occurred on September 24, 1998. The
first evidence of pressure shock arrival was a rapid increase
in dynamic pressure and magnetic field with the rotating
direction observed by Wind about 23:21 UT, which was
located upstream of solar wind (x, y, z)GSM Ĭ (184, 15,
Fig. 2. IMF and solar wind plasma parameters observed at Wind for
998-09-24 event.
ꢀ
ꢀ
6)R
E
, then observed by Geotail at (x, y, z)GSE Ĭ (20, ꢀ22,
1
2)R . Measurement of the IMF and solar wind plasma is
E
shown in fig. 2. It it shown that IMF and plasma increased
At about 23:46 UT, the AL index varied prominently,
discontinuously, the IMF become stronger mainly in the
decreased from –400 to –1600 nT in a few minutes. The
geomagnetic field observed at Greenland magnetometer
stations shows that the negative bay of H component first
appeared at the southern station NAQ at about 23:46 UT,
then the negative bay propagated to northern, which was
yGSM direction, IMF Bz increased from about zero to 7ü10
nT; B jumped from 14 to 40 nT; the solar wind density in-
ꢀ
3
creased from 8 to 20 cm , velocity increased from 420 to
Chinese Science Bulletin Vol. 46 No. 18 September 2001
1549

NOTES
observed by their northern stations.
The evidence of magnetosphere compressed was
observed by LANL1 at the synchronous orbit. During the
plasma sheet, therefore, nightside magnetosphere would
respond to the solar wind pressure shock in 1 2 min
ü
timescale. This is consistent with GOES observation,
which showed that the magnetic field in night synchro-
nous orbit changed about 06:53:30UT, just tens of seconds
after the arrival of shock. Typical fast-mode speed in
event LANL1 was at (6.6R
E
, ꢀ0.7MLAT, 23MLT). The
plasma data show that the density and energy fluxes of
electrons and protons increased prominently and D flux
increased apparently. The particle densities and energy
fluxes changed with the features of discontinuity.
[
18]
near-tail lobe is of the order of 6000 km/s , implying
traversal of a 25R tail radius in only a few tens of sec-
E
onds. Consequently, it is expected that the effect of com-
pression of shock reached inner or plasma sheet in a few
tens of seconds after it arrived at corresponding tail
magnetopause. The auroral substorm in this study shows
that after initial intensification in the whole auroral oval, it
intensifies rapidly in about 2 min later at nightside, corre-
sponding to AL decreasing dramatically, implying that a
strong field-aligned current had been built. These indi-
cated that the disturbance started in an extensive region,
then a strong auroral precipitation occurred, which may be
due to a burst process in near-tail around the equator. Prior
to the nightside auroral activity pronouncedly intensified,
the observation by GOES8 and LANL1 at the synchronous
orbit also confirmed that the energy of protons and elec-
trons near the magnetic equator both increased suddenly.
This shows that the disturbance in near tail preceded the
nightside strong auroral activities, and near the time of
shock passed over the corresponding magnetopause. Collier
2
Discussion
The two events presented here were corresponding to
the following interplanetary conditions: one with an IMF
Bz|0 at least for about 1 h before the shock, another
with about 10 min southward component. This time inter-
nal is less than that the magnetospheric convective re-
sponse needed. Therefore, we can infer that these two
events were controlled by pressure shock, and they have
no close relation to southward IMF.
In contrast to isolated substorm, these two events
have different spatial-temporal evolution features: The
first important feature is that the time between the arrival
of the pressure shock and the ionospheric response is very
short. The auroral substorm occurred 3 ü 4 min after
pressure shock arrived at Earth’s magnetopause. The re-
sponse is too rapid for magnetospheric convection proc-
esses. The timescale for convection response is at least
[18]
et al. showed with their study of theory and observations
[
4]
about 30 min , while for response to pressure shock it is
only its 1/10, same as Alfven characteristic time. This
suggests that the inner magnetosphere and ionosphere
respond to pressure shock quite quickly. Such a quick
transport of the energy possibly is related to hydromag-
netic wave fast- mode. It was found that during the Earth
on the interplanetary shock passage, the shock compresses
that when pressure shock impulses the magnetotail, the lobe
field increased rapidly and the tail establishes a new equi-
librium in about 2 min, i.e. a strong crossing-tail current
might have been built in the magnetotail corresponding to
the lobe field enhancement. This kind of configuration in
tail is instable, therefore, the rapid intensification in night-
side auroras and dramatic changes of AL indexes must have
been caused by an instability in near-tail plasma, such as
[
18]
the magnetosphere and propagates to down tail . In the
998-08-26 event, IMP8 observed that it took about 4 min
for the shock propagating from the dayside magnetopause
1
[6]
the ballooning instability studied by Pu et al. .
The second feature of the auroral substorm events in
this study is that the auroral substorm initiated in an ex-
tensive local time, very different from that of isolated sub-
storm. The isolated substorm described by Akasofu is as
follows: aurora brightens initially in a local region near
the nightside meridian plane of oval, then the region of
bright expansion phase aurora expands poleward, and
forms west traveling surge. This process was thought
caused by the cross-tail current disruption due to plasma
instability in the tail current sheet, and the disruption re-
gion extends to down tail and travels to west. However,
for the two auroral substorms associated with pressure
shock in this study, the most distinguishing feature is the
latitudinal expansion of the auroral luminosity and en-
hancement of incident energy flux at all local time rather
than within a limited longitudinal region as in the isolated
substorm described above. This may be related to the re-
gion of interaction between the pressure shock and mag-
netosphere. Usually, the plasma sheet is thinner in the
to tail x Ĭ ꢀ24RE, On average, at about 06:53UT, the
E
shock would reach x Ĭ ꢀ7R in the magnetotail. The
observations at the synchronous orbit also confirmed
magnetospheric response with that LANL1 at dayside first
observed the changes of particle energy (at about
0
(
6:53UT), when LANL4 located at dawnside observed
06:54UT) the similar phenomena. Hence, we proposed
that the pressure shock, the step function like, impulse to
the magnetosphere when it passed over the Earth. At day-
side the magnetosphere was compressed strongly, and
results in the loss cone instability, and causes the en-
hancement of auroral precipitation, and leads to dayside
auroral activities. While in nightside magnetosphere, solar
wind compression produces hydromagnetic fast-mode, the
fast-mode has a higher Alfven velocity due to stronger
magnetic field and smaller plasma density in magnetotail
lobes. The compressive effects by pressure shock can be
transported crossing the lobe sooner, and get into the
1550
Chinese Science Bulletin Vol. 46 No. 18 September 2001

This work was supported by the National Natural Science Foundation of
China (Grant Nos. 49974033 and 49634160).
center than at two flanks, in particular during southward
IMF. The magnetic field reconnection between IMF and
dayside magnetopause field transports the magnetic field
from dayside magnetosphere to magnetotail by magneto-
spheric convection, makes tail lobe field increase, and
current sheet becomes thin. Since the occurrence rate of
reconnection near the meridian is higher than that in two
flanks, the magnetic field near meridian in the tail is
stronger than at two flanks, and the tail current sheet is
thinner in center region than in two flanks. But the inter-
action between pressure shock and magnetosphere is dif-
ferent from the interaction between IMF and magneto-
sphere. The pressure shock forces on the magnetosphere
in all boundaries simultaneously, the plasma sheet is com-
pressed on both center and flanks, and the current sheet
might become thin in the extending longitudinal region,
the energy density increases in a larger longitudinal region.
This may be the reason why the auroral substorm associ-
ated with pressure shock has no very close relation to lo-
cal times.
References
1. Dungey, J. W., Interplanetary magnetic field and auroral zones,
Geophys. Res. Lett., 1961, 6: 47.
2
.
Axford, W. I., Hines, C. O., A unifying theory of high-latitude
geophysical phenomena and geomagnetic storm., Can J. Phys,
1
961, 39: 1433.
3. Baker, D. N., Pulkkinen, T. I., Angelopoulos, V. et al., Neutral line
model of substorms: Past results and present view, J. Geophys
Res., 1996, 101(A6): 12975.
4
.
Lui, A. T. Y., Current disruption in Earth’s magnetosphere: Ob-
servations and models, J. Geophys. Res., 1996, 101(A6): 13067.
5. Lyons, L. R, Substorms: Fundamental observational features: dis-
tinction from other disturbances, and external triggering, J. Geo-
phys. Res., 1996, 101(A6): 13089.
6
.
Pu, Z. Y., Korth, A., Chen, Z. X. et al., MHD drift ballooning in-
stability near the inner edge of the near-Earth plasma sheet and its
application to substorm onset, J. Geophys. Res., 1997, 102(A7):
14397.
7
8
.
.
Shen Chao, Liu Zhenxing. Substorm onset caused by IMF north-
ward turning, Science in China, Series A, 2000, 30(Suppl.): 69.
Russell, C. T., Zhou, X. W., Chi, P. J. et al., Sudden compression
of the outer magnetosphere associated with an ionospheric mass
ejection, Geophys. Res. Lett., 1999, 26: 2343.
[
20]
Karlsson et al.
found in their study of substorm
9
.
Sibeck, D. G., Baumjohann, W., Lopez, R. E., Solar wind dynamic
pressure variations and transient magnetospheric signatures,
Geophys. Res. Lett., 1989, 16(1): 13.
quenching and multiple onsets that the substorm onset has
been triggered by a solar pressure pulse after a period of
southward IMF. An incomplete substorm occurred when
the solar wind pressure decreased, in spite of a continuous
southward IMF, due to the magnetosphere expansion.
However, the substorm can develop fully when there is a
continuously high solar wind pressure. For the events
presented here, the shock kept high pressure for a long
time, and the magnetosphere was compressed enough
time to let the substorm develop completely. The substorm
developed and recovered both quickly, lasting about 40
min. After that, the auroral activity quenched gradually,
although the solar wind pressure still was high. This indi-
cates that the magnetosphere needs to be refilled after
energy release. This feature is not different from the sub-
storm associated with southward IMF.
1
0. Gonzalez, W. D., Tsurutani, B. T., Criteria of interplanetary pa-
rameters causing intense magnetic storms (Dst˘100nT), Planet
Space Sci., 1987, 35: 1101.
11. Craven, J. D., Frank, L. A., Russell, C. T. et al., Global auroral
responses to magnetospheric compressions by shocks in the solar
wind: Two cases studies (eds. Kamide. Y., Slavin, J. A.), Solar
Wind-magnetosphere Coupling, Tokyo: Terra Scientific, 1986,
3
67ü380.
1
2. Zhou, X. Y., Tsurutani, B. T., Rapid intensification and propaga-
tion of the dayside aurora: Large scale interplanetary pressure
pulses (fast shocks), Geophys. Res. Lett., 1999, 26(8): 1097.
13. Kokubun, S., McPherron, R. L., Russell, C. T., Triggering of sub-
storms by solar wind discontinuities, J. Geophys. Res., 1977, 82: 74.
1
4. Petrinec, S. M., Russell, C. T., Near-Earth magnetotail shape and
size as determined from the magnetospause flaring angle, J. Geo-
phys. Res., 1996, 101: 137.
Our observations point out that the solar wind pres-
sure shock is independent of southward IMF in triggering
substorm. We propose that substorm onset associated with
solar pressure shock may be explained based on hydro-
magnetic wave. When the pressure shock impulses the
nightside magnetosphere, and launched MHD fast mode
wave. The fast-mode wave propagates across the mag-
netotail lobes with higher local Alfven velocity, magneto-
tail was compressed rapidly and strong lobe field and
cross-tail current were built in about 1ü2 min, and fur-
thermore the substorm were triggered due to an instability
in current sheet. We believe that solar wind pressure shock
plays an important role in the substorm process, which is
worthy to be studied farther.
15. Shue, J. H., Kamide. Y., Effects of solar wind density on the
westward electrojet, Proceeding of the Fourth International Con-
ference on Substorms, Terra Sci., 1998, 667ü678.
1
6. Jacquaey, C., Time-variation of the large scale tail magnetic field
prior substorm related to solar wind changes, Proceeding of the
Third International Conference on Substorms, Versailles, France,
ü
12 17 May 1996, Europe Space Agency Spec. Publ. ESA
SP-389, 1996, 295ü300.
17. Russell, C. T., Wang. J. A., Lepping, R. P. et al., The interplane-
tary shock of September 24, 1998: Arreval at Earth, J. Geophys.
Res., 2000, 105: 25143.
1
8. Collier, M. R., Slavin. J. A., Lepping, R. P., et al., Multispacecraft
observations of sudden impulses in the magnetotail caused by so-
lar wind pressure discontinuties: Wind and IMP8, J. Geophys,
Res., 1998, 103: 17293.
1
2
9. Lui, A. T. Y., Road map to magnetotail domains, Magnetotail
Physics (ed. Lui, A. T. Y.), Baltimore: Johns Hopkins Univ. Press,
Acknowledgements We would like to express sincere thanks to Dr.
Lepping, R and the staff of CDAWeb for supplying us with Wind IMF
data and other ISTD data for this study. We also wish to thank the staff
of WDC-C2 and associated observation stations for their help in the
observations and in making AE index data available, to the staff of Bei-
jing Ming Tomb Magnetometer Station for providing magnetic field data.
1
987, 3ü9.
0. Kalsson, S. B. P., Opgenoorth, H. J., Eglitis, P. et al., Solar wind
control of magnetospheric energy content: Substorm quenching
and multiple onsets, J. Geophys. Res., 2000, 105: 5335.
(Received March 26, 2001)
Chinese Science Bulletin Vol. 46 No. 18 September 2001
1551