Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.iki.rssi.ru/people/cz99.pdf Дата изменения: Mon Dec 22 14:04:03 2003 Дата индексирования: Mon Oct 1 22:58:19 2012 Кодировка: Поисковые слова: п п п

GLOBAL SUBSTORM EFFECT AND CONVECTION JET UNDER THE CONDITIONS OF CONTINUOUS EXTERNAL DRIVING: MULTI-SPACECRAFT OBSERVATIONS ON DECEMBER 22 23, 1996
Yu. I. Yermolaev, L. M. Zelenyi, N. L. Borodkova, R. A. Kovrazhkin, V. N. Lutsenko, A. A. Petrukovich, S. P. Savin, A. A. Skalsky Space Research Institute, Russian Acad. Sci., Moscow, Russia V. A. Sergeev St. Petersburg University, St. Petersburg, Russia T. Mukai Institute of Space and Astronautical Science, Sagamihara, Japan S.Kokubun Solar-Terrestrial Environment Laboratory, Nagoya University, Japan K. Liou, C.-I. Meng Applied Physics Laboratory, Johns Hopkins University, Laurel, USA G. Parks Geophysics Program, Universityof Washington, Seattle, USA J.-A. Sauvaud Centre d'Etude Spatiale des Rayonnements, CNRS, Universityof Toulouse, France

Multipoint observations by ground-based stations and a eet of ISTP satellites allowed us to study the plasma processes in di erent regions of the near-Earth space during a very interesting interval on December 22 23, 1996 which was characterized by 20 hour southward IMF z and almost constant solar wind pressure 1 2nPa. Five substorm events were observed during this interval of continuous external driving. Several global e ects of these substorms are described. In particular, comparison of measurements in the plasma sheet on both anks showed 1 the similar loading unloading processes in the tail correlated with substorm development, and 2 a strong bursty convection concentrated in a narrow 15 E channel. Observations show also that earthward bursty bulk ows BBF on both satellites likely coincide with appearance of auroral activityat the footpoints of corresponding satellites but there is no correlation between these ows observed on both anks.
B : Y R



Wind-Magnetosphere-Ionosphere Interaction: Interbal l Observations

Presented at the NATO Advanced Research Workshop

Coordinated Studies of the Solar

1998 in Kosice, Slovakia.

, held on September 7 11,

Czechoslovak Journal of Physics, Vol. 49 1999, No. 4a

625


Yermolaev Yu. I. et al.

1 Introduction
One of the most important and challenging problem in the Solar-Terrestrial Physics program is the study of various manifestation of substorm dynamics and mechanisms of their generation. In direct or indirect way the solar wind is a driver of these processes and especially elucidating might be the study of this problem under special conditions of prolonged periods of southward interplanetary magnetic eld IMF. Southward IMF is favourable for reconnection of interplanetary and magnetospheric magnetic elds on the dayside magnetopause and should result in the enhancement of the magnetic ux in the tail, i.e., in the input of the solar wind energy into magnetospheric system 1, 2. When the solar wind energy ux reaches a su cient marginal level, substorms may be triggered directly by the changes in the solar wind and IMF 3 or by the IMF northward turnings 4 . On the other hand, under a su ciently high level of the magnetic energy content in the tail the internal magnetospheric dynamics may result in substorm development 5 . Statistical studies of the substorms supports this point of view 6, 7. Therefore, directly driven and loading unloading processes are important for substorms 1. The substorm is a large-scale phenomenon which essentially disturbs the plasma and magnetic eld both in the magnetotail and in high latitude regions. It is important to make observations not only in these key regions in the tail and the auroral regions but also to study di erentchannels of energy transport steady convection, plasmoids, BBFs and so on in the magnetosphere and to establish a cause-e ect links between physical processes in the di erent parts of the system 1,2, 8,9 . The traditional picture of laminar earthward convection in the plasma sheet under the homogeneous large-scale dawn-dusk electric eld has been questioned by the observations of plasma ows made at high temporal resolution. It appears e.g. 10, 11 that the main transport of mass, energy and magnetic ux in the midtail plasma sheet is realized via the short-lived high-speed plasma ows bursty bulk ows, BBFs. The origin and spatial temporal characteristics of transient features in the plasma sheet are not yet completely understood. An unclear but important aspect is also the cross-tail scales of the convection pattern on di erent time scales. Simultaneous measurements with at least two spacecraft are required to resolve this issue. At the small 1 min time scale, the BBFs appeared to have a small size less than 1 10 of the tail diameter e.g. 12,9 . However, there were also some indications in both event 13 and statistical studies, 14 that the time-averaged 1 h convection features can also be very inhomogeneous in some events. Direct multi-spacecraft measurements of the average convection pattern, including its possible relationship to the BBF characteristics are highly desirable. In this paper we present a unique case of 20 hour southward IMF and almost constant solar wind ram pressure. Simultaneous measurements of solar wind conditions on WIND, of mid-tail plasma sheet on INTERBALL Tail Probe and GEOTAIL, and of auroral region on INTERBALL Auroral Probe and POLAR are used for this analysis. This type of continuous external forcing is rarely observed and few aspects of 626
Czech. J. Phys. 49 1999


Substorm e ect and convection

:::

it are quite interesting: the occurrence of substorm sequence and modi cation of convection in the plasma sheet due to the occurrence of substorms. A one more lucky coincidence in this interval is that the two spacecraft provided the measurements at comparable distances 25 30 RE downstream but in dawn and dusk parts of the plasma sheet at 20 RE distance between them. This allows us to study a large scale behaviour and cross-tail structure of convection.

2 Observations
2.1 Survey of observations
For monitoring of interplanetary conditions we use the measurements of the WIND spacecraft located at X 80 RE and Y 20 RE during interval under investigation. To compare the WIND data with observations in the plasma sheet at the distance of X = ,30 RE we take into account of time of plasma and IMF propagation between satellites with the solar wind velocity V . After such time shifting the southward IMF commenced at 12 UT on December 22, 1996 and remained southward until 08 UT on December 23. As a result Figure 1, the solar wind-induced cross-tail potential drop computed as kV = 16V=4002 + B sin3 =2 according to Boyle et al. 15 stayed at a high level throughout this time interval with a few drops at 13 14, 17 20 and near 24 UT and gradually declined after 04 UT on December 23 IMF Bz was negative all that time. The nearly constant and low1.2 nPa ram pressure of the solar wind facilitates to view the variations of the lobe magnetic eld and tail pressure as being related to the external driving and or to the internal substorm e ects. Continuous strong external driving caused both the gradual development of the ring current with maximal Dst depression -40 nT reached 17 hours after the commencement of the southward IMF and considerable electro jet activity in the auroral zone Figure 1. Five substorm events could be recognized throughout this disturbed period as enhancements of the westward electro jet in Figure 1. The substorm onsets see Table 1 occurred at 1251, 1625 and 2200 UT on December 22 events 1 3, as well as at 0220, and 0815 UT on December 23 events 4,5. These determinations were based on a large set of observations, including the UVI images of POLAR spacecraft available for events 1,2,4, ground Pi2 pulsations events 1,2,3,. . and both mid-latitude and auroral zone magnetograms some subtle di erences between di erent determinations are not important for present study but will be published separately. 1st and 5th substorm onsets were probably triggered by sharp positive variations of IMF Bz seen as sharp drops of computed potential drop in Figure 1, whereas no large triggers could be noticed for events 2,3,4 referred later as the spontaneous onsets. The plasma sheet parameters have been monitored by two spacecraft, INTERBALL Tail Probe and GEOTAIL. Their GSM coordinates and MLT at substorm onsets are presented in Table 1. Most of the time both spacecraft stayed near the
Czech. J. Phys. 49 1999

627


Yermolaev Yu. I. et al.

Fig. 1. Survey of solar wind, ground, and plasma sheet activity on December 22 23, 1996. From top to the bottom: a superimposed magnetograms of 11 ground stations used to derive the standard AE index and SYM-H index of low-latitude magnetic variations whichisaproxy for st index denoted as st ; b IMF z and solar wind ram pressure measured on WIND shifted in time to = ,30 E taking into account the solar wind speed; c time variations of the expected cross-tail potential di erence computed according to 15 ; d variations of the total plasma plus magnetic pressure and of the magnetic ux transport rate V By in the midtail according to Geotail measurements; the tail lobe magnetic pressure at = ,30 E 30 computed for given solar wind conditions according to 21 is shown as a reference. Substorm onsets are shown by the dashed vertical lines.
D D B X R X R P

628

Czech. J. Phys. 49 1999


Substorm e ect and convection

:::

Table 1. Substorm onsets, interplanetary conditions, and Geotail GT and Interball Tail Probe IT locations

No Onset IMF Bz Time Change UT UT 1 2 3 4 5
Dec.22

INTERBALL GEOTAIL Location Comments SC GSM GSM GSM MLT h E E E
X Y Z R R R

1251 1625 2200

1215 No No No 0745

IT GT IT GT IT GT IT GT IT GT

23.4 28.5 22.4 27.9 20.2 26.3 17.7 24.6 13.0 22.0

12.2 10.8 11.8 12.1 10.9 13.8 9.4 15.9 4.7 18.7

2.4 1.5 2.0 3.0 2.0 6.0 4.4 5.0 8.3 0.2

22.1 Triggered 01.2 22.1 Spontaneous 01.3 22.1 Spontaneous 01.6 22.2 Spontaneous 02.2 22.2 Triggered 02.6

Dec.23

0220 0815

neutral sheet at distances of 15 30 RE from the Earth but GEOTAIL was in the dawn plasma sheet during fth event it exited to the magnetosheath and INTERBALL was in the dusk plasma sheet. During the most interesting time period between substorms 1 and 3, INTERBALL and GEOTAIL were at comparable distances from the Earth 22 and 27 RE but a half way between the tail axis and the dusk and dawn anks, respectively. We used GEOTAIL magnetic eld MFI 16 and plasma LEP 17 measurements at 12 s resolution. On INTERBALL Tail Probe we used the magnetic eld observations 18 and plasma observations made with two instruments. The moments of the distribution function from CORALL ion instrument covering the 0.05 25 keV energy range 19 were available at the spacecraft spin 2 min resolution; we use only the Vx component of the ow along the spacecraft spin axes which, on the one hand, represents the most important component of the magnetospheric convection and, on the other hand, is less a ected by the time variations of parameters during the 2 min spin period. The ELECTRON instrument provided the density and temperature of the plasma sheet electrons 20 . Variations of total plasma plus magnetic pressure in the tail Fig. 1, bottom panel show a systematic substorm response. In each substorm event, with either triggered or spontaneous onset, there was an increase of total pressure loading preceeding the onset. More than a half the time between the substorm onsets ranging from 3.6 to 6.1 hours there were no rapid changes of total pressure. At these times
Czech. J. Phys. 49 1999

629


Yermolaev Yu. I. et al.

the plasma convection was quite strong and variable, dominated by the bursty bulk ows see the bottom panel of Figure 1. The most intense ows are seen during the 2 3 hour periods following the substorm onsets.

2.2 Plasma sheet during intense bursty plasma ows
Observed variability of the plasma ows can be due to both temporal variations and spatial e ects when spacecraft crosses di erent parts of the apping plasma sheet or even exits to the lobes. A favourable situation for probing the inner part of the plasma sheet IPS, de ned by the restriction that plasma parameter was 0.5 occurred after 1330 UT after the plasma sheet expansion associated with the rst substorm when INTERBALL stayed in the IPS for more that 8 hours, as indicated bylowvalues of the Bx magnetic component in Figure 2 left. GEOTAIL also spent a considerable time near the neutral sheet, but crossed it many times, so the Bx behaviour was much more variable. Therefore, to avoid the spatial e ects we plotted in Figure 2 right only those 2 min averaged parameter values measured on GEOTAIL when it was in the IPS. It should be noted that due to design features, the CORALL ion spectrometer on INTERBALL can underestimate the ion density. Consequently, we used plasma density from ELECTRON instrument together with ion temperature from CORALL to compute the total pressure on INTERBALL. The plasmas measured by GEOTAIL and INTERBALL show some similarities as well as some di erences. First of all, attention is drawn to the fact that a such large-scale phenomena as plasmoids and TCRs they are shown by arrows on the second and third panels in Figure 3 are observed only on one from two satellites: plasmoids in the beginning of 1st and 4th substorms on Tail Probe and TCR in the beginning of 3rd substorm on GEOTAIL. This discrepancy allows us to estimate the sizes of these phenomena and location of magnetic eld reconnection point in the tail during the substorm events. Third right and left panels in Figure 2 and second and third panels in Figure 3 present Vx component of ion velocity obtained on the INTERBALL Tail Probe and GEOTAIL, respectively. Vx bursts are basically earthward and last about several minutes but there is no correlation between Vx velocity components observed on both satellites. The plasma density value after the rst substorm was about 0.1 cm,3 on both spacecraft. The temperatures were somewhat low for the disturbed conditions, on GEOTAIL the proton temperature changed from 2keV to 4keV during the time interval studied where the strongest ows were observed. On INTERBALL the proton temperature was somewhat higher, changing between 3 and 8keV. This di erence could be prescribed either to the azimuthal gradient due to the ion acceleration in the electric eld from dawn to dusk or to the radial gradient due to the smaller radial distance of INTERBALL than GEOTAIL. The electron temperature on INTERBALL changed in parallel to the proton temperature favouring the latter explanation. A clear temperature increase at 17 UT correlates with the Bz increase indicating once more the importance of heating in the compressed magnetic ux tube. Although similar sporadic earthward ows were observed between 630
Czech. J. Phys. 49 1999


Substorm e ect and convection

:::

Fig. 2. Survey of plasma BALL, left panels and in magnetic pressure 30 is storms No. 1 3 are shown
P

sheet parameters measured in the dusk plasma sheet INTERthe dawn plasma sheet GEOTAIL, right panels. The tail lobe shown for reference on the bottom panels. The onsets of subfor reference. Note that only data in the inner plasma sheet are included for GEOTAIL.

Czech. J. Phys. 49 1999

631


Yermolaev Yu. I. et al.

Fig. 3. Comparison of 30 min average values of the Earthward magnetic ux transport x z in the inner plasma sheet under PL B 0.5 as measured on INTERBALL and GEOTAIL spacecraft. For reference we also show the amplitude of the expected crosstail electric eld taken to be homogeneous with the cross-tail potential 80 kV and tail diameter 50 E . The time variations of the cross-tail potential drop computed from solar wind parameters according to 15 are shown on the upper panel.
V B P =P R

632

Czech. J. Phys. 49 1999


Substorm e ect and convection

:::

Fig. 4. Energy-time spectrograms for ions, O+ and electrons during the INTERBALL Auroral Probe pass through polar regions on 22 December, 1996.
Czech. J. Phys. 49 1999

633


Yermolaev Yu. I. et al.

1330 and 1500 UT on INTERBALL, they did not produce comparable acceleration. The Bz values in that case did not change much and stayed small, about 2 nT. The fast ows in the IPS region were exclusively in the earthward direction on both spacecraft, this is also true for high speed observed during the loading and unloading phases see GEOTAIL data for substorm 2, as well as during the episodes of steady total pressure. A remarkable feature of the time interval considered here is the low total pressure between 14 and 19 UT, which is smaller than the average lobe magnetic eld pressure P30 computed according to Fair eld and Jones 21 and shown as the reference level in the bottom plots in Figure 2. During this time interval the strongest ows were observed at both spacecraft, but with higher speed ows on the GEOTAIL. These features are of great interest and they will be discussed in the section 3.2.

2.3 Auroral observations
The top and bottom color panels in Figure 3 are keograms made from POLAR UVI images at LBHL band 22 showing the auroral dynamics in the AACGM "Latitude - UT" coordinates 23 at two meridians, 22.40 MLT top and 02.30 MLT bottom, i.e., at the footpoints of Tail Probe and GEOTAIL, respectively. Note that the LBHL auroral emission is approximately proportional to the total energy input of precipitating electrons 24, 25 and, therefore, can be used as a proxy of the hemispheric energy input. Both panels show strong auroral activity during the interval studied. Nevertheless, there are signi cant di erences between the two pictures. Increases in brightness and poleward expansions of the aurora in the top panel coincide with substorm activity and have characteristic recurrence period about 3 6 hours. The bottom panel does not show this dependence: increases in brightness occur with periodicity about 1 hour and there is no poleward expansions of the aurora. Data presented in Figure 3 allows to suggest that observations of BBFs on each satellite coincide with appearances of auroral activity in footpoint of corresponding satellite. During the interval studied POLAR and INTERBALL Auroral Probe had several orbits such a way that there were auroral zone and polar cap observations during substorms. Detailed analysis of these data will be the sub ject of further investigations. Here we only present an example of data and note several large-scale phenomena observed by satellites. Figure 4 presents ion, O+ and electron energy spectrograms measured with ION instrument on INTERBALL Tail Probe 26 during the time interval of 1600 1700 UT on December 22. In auroral zone, which began at 1638 UT, the ION instrument observed ion injections with periodicity about 1 2 min and time-of- ight energy dispersion. Simultaneously with the injections the upward O+ uxes with energy up 10 keV were observed. Estimates made on the basis of this dispersion show that these ion uxes probably originated at the distance 15 20 RE from the observation point, i.e., not far but nearer to the Earth from Tail Probe and GEOTAIL locations. At the same time 1640 UT, large deviation of magnetic eld 634
Czech. J. Phys. 49 1999


Substorm e ect and convection

:::

measured on Auroral Probe from the model eld were observed, i.e., there were currents which disturbed the magnetic eld in this region. This time coincides with the poleward aurora expansion observed with POLAR UVI instrument.

3 Discussion
3.1 Substorms during conditions of continuous driving
There have not been many studies of substorms during conditions of continuous driving, especially concerning the characteristics of loading unloading processes. Our study shows that the duration of both loading and unloading episodes 0.5 1 hour is similar to those for the classic substorms. In events 2,4,5 the loading episodes do not correlate with any change of the solar wind parameters which shows that the occurrence of the loading phase during steady energy supply from the solar wind is controlled by the internal dynamics of the magnetotail. It is also of interest that four similar sharp positive IMF Bz excursions sharp decreases of potential drop in the middle panel of Figure 1 have been observed at 1240, 1715, 2340 UT on December 22 and 0800 UT on December 23. However only rst and last of them, which were preceded by the loading episodes, could trigger the 1st and 5th substorms. This is a good illustration to show that the magnetotail should be prepared for the expansion phase to be triggered by the IMF discontinuity. The time intervals between the onsets of ve successive substorms were 3.5, 5.7, 3.7, and 6.1 hours, respectively. Their length seems to be inversely related to the average values of the expected potential drop between substorms 61, 55, 81, and 52 kV, see Figure l. These values of inter-substorm time length are larger than the average substorm recurrence time 2.7 hours reported by Borovsky et al. 7 and much larger than 50 min recurrence time found byFarrugia et al. 27 for similar conditions of strong steady energy supply. In our case the substorms were truly global recon guration events determined from both loading unloading features in the tail. It will be important to revisit this issue with the more extended data set. A new observation is that during a considerable part more than a half of the time interval between substorms the total pressure is nearly constant and plasma sheet convection is quite strong. These episodes resemble the long-duration periods of steady magnetospheric convection SMC, e.g. 13 but in our case these episodes last 2 3 hours and intervene between the expansion and growth phases of successive substorms. POLAR UVI measurements showed that substorms were accompanied by poleward expansions of auroral brightness at 2230 MLT Tail Probe footpoint and such auroral motions were not observed at 0230 MLT GEOTAIL footpoint. BBFs observed in the plasma sheet during substorms and between them are likely correlated with auroral intensi cations in the footpoint of corresponding satellite. During substorms Auroral Probe observed ion injections with energy dispersion correlated with upward O+ ion uxes with energy up 10 keV and disturbances of magnetic eld which are probably connected with local electric currents. Estimates made for 2nd substorm at 1630 UT on December 22 showed that these injecCzech. J. Phys. 49 1999

635


Yermolaev Yu. I. et al.

tions were generated at distances of 15 20 RE , that is, nearer to the Earth than GEOTAIL and INTERBALL Tail Probe, but not far from them. The example of ion injections presented here see Figure 4 is interesting because only during this substorm the plasmoids or TCR as for 3rd substorm were not observed on Tail Probe and GEOTAIL. So, it is possible to suggest that only during this substorm the point of reconnection of tail magnetic eld was located farther from the Earth than Tail Probe and GEOTAIL locations, 22,12, 1 and 24, 13, 3 RE in GSM frame, respectively. In other cases the reconnection region are likely to be nearer Earth than the Tail Probe and GEOTAIL locations, i.e., nearer than 22 RE , and the region of generation of ion injections may coincide with region of eld reconnection. This estimate is in agreement with our recent result 15 16 RE obtained on the basis of INTERBALL Tail Probe and GEOTAIL observations 28 . Figure 5 presents the half-hour averages of the Vx Bz values in the inner part of the plasma sheet the averages were plotted only if the measurements in the IPS occupied at least 10 minutes of entire 30 min averaging time interval. These values give a good proxy of the earthward magnetic ux transport V By as was checked using GEOTAIL data. In steady state con guration the total average earthward ux transport Ey integrated across the tail should correspond to the cross-tail potential drop , and for = 80 kV computed according to Boyle et al. 15 top panel in Figure 5 and tail radius 25 RE computed using the Shue et al. 29 formula the expected average ux transport rate should be E0 0.25 mV m. The INTERBALL-based estimate gives an average Vx Bz value about the E0 value. On GEOTAIL, during three hours 14 to 17 UT the ux transport rate was 3 times larger than on INTERBALL. Note Figure 2 that most of this time the tail pressure was nearly steady and it was considerably reduced as compared to its average value. The observation of stronger convection in the dawn plasma sheet is not inconsistent with the auroral zone magnetometer data. Between 13 and 17 UT the deepest negative magnetic bay which forms a lower envelope in the magnetogram plot in Figure 1 top came from Barrow station 69 of CGLat, magnetic midnightat 12.5 UT. This con rms that the strongest auroral electro jet currents occurred in the high-latitude portion of the postmidnight auroral oval. The observation of long-duration enhanced convection regions is not unique. Nishida et al. 14 compared daily averages of V B measured by GEOTAIL at 95 RE during long active period and found a distinct peak of the magnetic ux transport rate 0.43 mV m in premidnight sector. In the event studies reported by Sergeev and Lennartsson 13 , and Nakamura et al. 30 the average transport rates were about 3 4 times higher when compared to E0 in the premidnight sector of plasma sheet. In both cases the observations were performed during an interval of several hours following the substorm when the reduced total pressure was also observed. For the rst time our results provide direct evidence of strong long-term inhomogeneity in the plasma sheet convection, but here the convection jet occupied the dawn half of plasma sheet. The location of the convection jet, thus, mayvary
a

3.2 Convection jet feature in the plasma sheet

636

Czech. J. Phys. 49 1999


Substorm e ect and convection

:::

Fig. 5. BBF observations on dusk INTERBALL Tail Probe, 3rd plate and dawn GEOTAIL, 2nd plate anks of plasma sheet and POLAR UVI keogram in corresponding footpoints 1st and 4th plates.

Fig. 6. Schematic view of the convection jet pattern.
Czech. J. Phys. 49 1999

637


Yermolaev Yu. I. et al.

from one event to another and even change during an event. The upper estimate of the possible cross-tail width of the convection jet in the plasma sheet Y is about 15 RE 1 3 of tail diameter DT with E GT ' 3E0 in y the convection jet region to get the right cross-tail potential drop value, that is, Y ' E0 =E GT DT ' 1=3 50 RE 31 . The large-scale pattern of the convection y jet in geomagnetic equator is illustrated by Figure 6 where the triangle and circle show satellite positions, small arrows | BBFs, and large arrows | average largescale convection at the position of the corresponding satellite. It is of note that the di erences between the 0.5 hour average Vx Bz values on GEOTAIL and INTERBALL decreased with time and after 17 UT following the 2nd substorm, both values became comparable to the expected cross-tail electric eld E0 . Whether it re ects any inherent dynamics of the convection jet or was caused by the spatial variation exit of GEOTAIL from convection stream cannot be stated for sure. This long-term inhomogeneity of convection could be an important factor which in uences the 3D current system and magnetic con guration of the tail e.g. 12 . The origin of the convection jet phenomenon should be related to the high intensity occurrence rate of the bursty bulk ows at distances r 30 RE which, being averaged, forms the observed enhanced convection. One possibility concerns the inhomogeneous reconnection rate in the far tail which depends on the Alfv speed en in the lobe adjacent to the plasma sheet and, therefore, on the distribution of the cold plasma density e.g. 13 . More observations are required to understand the physics of these phenomena.

4 Conclusions
20 hour interval of southward IMF and low steady solar wind pressure on DeWe presented several preliminary results of multi-point observations during a cember 22 23, 1996. Five substorm events with distinct loading unloading features in the midtail were observed during the continuous external forcing. The loading features did not necessarily correlate with the changes of solar wind parameters which implies that switching to the loading can be determined by the internal magnetotail dynamics. The substorm recurrence time, 3.6 to 6.1 hours, appeared to be larger than the values previously reported by Borovsky et al. 7 and Farrugia et al. 27 and inversely dep ended on the average rate of magnetic ux supplied to the magnetotail. During about half the time interval between substorms the con guration was locally steady with no considerable change of total pressure but with intense convection in the plasma sheet resembling the steady convection episodes. A unique possibility of simultaneous observations of convection in the inner plasma sheet by GEOTAIL and INTERBALL Tail Probe during more than 8 hours gave direct evidence that on the long term the magnetic ux transport rate may be very inhomogeneous in the tail for many hours. The short 1 min earthward bursty bulk ows BBF on both satellites likely coincide with appearance of auroral activity at the footpoints of corresponding satellite but there is no correlation between these ows observed on both anks. The large-scale phenomenon of con638
Czech. J. Phys. 49 1999


Substorm e ect and convection

:::

vection jet appears to be due to the grouping of the most intense bursty bulk ows in the localized portion of the plasma sheet. The location of convection jet in the dawn plasma sheet presented here di ered from that in previous observations of enhanced convection 13, 14, 30 which placed the convection jet in the premidnight plasma sheet. More observations are necessary to establish the morphology and origin of the convection jet phenomenon.
We are grateful to INTERBALL Tail and Auroral Probe magnetic eld group for providing data. We thank R. P. Lepping and K. W. Ogilvie the WIND spacecraft for making available to us the data of their experiments. We also thank T. Iyemori WDCC1, Kyoto for the high-resolution data of SYM index. The work at IKI was supported by the INTAS Grant No. 96-2346. The work byVAS was nancially supported by the RFBR grant No. 98-05-04114. The Editor thanks Tuija Pulkkinen for her assistance in evaluating this paper.

References
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 D.N. Baker: in Substorms 1, ESA-SP 335, 1992, p.185. V.A. Sergeev, R.J. Pellinen, and T.I. Pulkkinen: Space Sci. Rev. 75 1996 551. S.-I. Akasofu: Space Sci. Rev. 28 1981 121. G. Rostoker: in High-Latitude Space Plasma Physics, B. Hultqvist, T. Hagfors Plenum Publ. Co., New York, 1983, p.189. D.N. Baker, A.J. Klimas, R.L. McPherron, and J. Buchner: Geophys. Res. Lett. 17 1985 41. Y. Kamide, P.D. Perrault, S.-I. Akasofu, and J.D. Winningham: J. Geophys. Res. 82 1988 5521. J.E. Borovsky, R.J. Nemzek, and R.D. Belian: J. Geophys. Res. 98 1993 3807. V. Angelopoulos, W. Baumjohann, C.F. Kennel, F.V. Coroniti, M.G. Kivelson, R. Pellat, R.J. Walker, H. Luhr, and G. Paschmann: J. Geophys. Res. 97 1992 4027. V. Angelopoulos et al.: Geophys. Res. Lett. 24 1997 2271. W. Baumjohann: Space Sci. Rev. 64 1993 141. V. Angelopoulos et al.: J. Geophys. Res. 101 1996 4967. V.A. Sergeev, V. Angelopoulos, J.T. Gosling, C.A. Cattell, and C.T. Russell: J. Geophys. Res. 101 1996 10817. V.A. Sergeev and W. Lennartsson: Planet. Space Sci. 36 1988 353. A. Nishida, T. Mukai, T.Yamamoto, Y. Saito, and S. Kokubun: Geophys. Res. Lett. 22 1995 2453. C.B. Boyle, P.H. Rei , and M.R. Hairstone: J. Geophys. Res. 102 1997 11. S. Kokubun, T.Yamamoto, M.H. Acuna, K. Hayashi, K. Shiokawa, and H. Kawano: J. Geomag. Geoelectr. 46 1994 7. T. Mukai, S. Machita, Y. Saito, M. Hirahara, T. Terasawa, N. Kaya, T. Obara, M. Ejiri, and A. Nishida: J. Geomag. Geoelectr. 46 1994 669.

Czech. J. Phys. 49 1999

639


Yermolaev Yu. I. et al.: Substorm e ect

:::

18 S. Klimov et al.: Ann. Geophys. 15 1997 514. 19 Yu.I. Yermolaev, A.O. Fedorov, O.L. Vaisberg, V.M. Balebanov, Yu.A. Obod, R. Jimenez, J. Fleites, L. Llera, and A.N. Omelchenko: Ann. Geophys. 15 1997 533. 20 J.-A. Sauvaud, P. Koperski, T. Beutier, H. Barthe, C. Aoustin, J.J. Thocaven, J. Rouzaud, E. Penou, O. Vaisberg, and N. Borodkova: Ann. Geophys. 15 1997 587. 21 D.H. Fair eld and J. Jones: J. Geophys. Res. 101 1996 7785. 22 M.R. Torr et al.: Space Sci. Rev. 71 1995 329. 23 K.B. Baker and S. Wing: J. Geophys. Res. 94 1989 9139. 24 D.J. Strickland, R.E. Daniel, Jr., J.R. Jasperse, and B. Basu: J. Geophys. Res. 98 1993 21533. 25 G.A. Germany, M.R. Torr, D.G. Torr, and P.G. Richards: J. Geophys. Res. 99 1994 383. 26 J.-A. Sauvaud, H. Barthe, C. Aoustin, J.J. Thocaven, J. Rouzaud, E. Penou, D. Popescu, R.A. Kovrazhkin, and K.G. Afanasiev: Ann. Geophys. 16 1998 1056. 27 C.J. Farrugia, M.P. Freeman, L.F. Burlaga, R.P. Lepping, and K. Takahashi: J. Geophys. Res. 98 1993 7657. 28 A.A. Petrukovich et al.: J. Geophys. Res. 103 1998 47. 29 J.-J. Shue, J.K. Chao, H.C. Fu, C.T. Russell, P. Song, K.K. Khurana, and H.J. Singer: J. Geophys. Res. 102 1997 9497. 30 R. Nakamura, S. Kokubun, L. Bargatze, T. Mukai, T. Yamamoto, T. Nagai, K.B. Baker, M.R. Hairston, P.H. Rei , and O.A. Troshichev: in Proc. of 4th Intern. Conf. on Substorms, 1998 in press. 31 Yu.I. Yermolaev, V.A. Sergeev, L.M. Zelenyi, A.A. Petrukovich, J.-A. Sauvaud, T. Mukai, and S. Kokubun: Geophys. Res. Lett. 26 1999 177.

640

Czech. J. Phys. 49 1999