Z. Nemecek1, J. Safrankova1, G. Zastenker2, P. Triska3
1Faculty of Mathematics and Physics, Charles University,
V Holesovic‡kach 2, 18000 Prague 8, Czech Republic, E-mail:
2Space Research Institute, Moscow, Russia
3Institute of Atmospheric Physics, AV CR, Prague, Czech Republic
The INTERBALL mission is predominantly dedicated to the magnetospheric study; nevertheless, simultaneous measurements of two similarly equipped satellites in the solar wind are of principal importance. The solar wind flow carries many inhomogeneities on different scales and the determination of their evolution or relative velocity of propagation requires a real multipoint measurement.
The present paper describes briefly the scientific payload of both INTERBALL satellites from the point of view of the solar wind investigation but the main attention is paid to a comparative study of the disturbances in the solar wind flow on different spatial scales. The small scale is represented by two INTERBALL satellites separated by a few thousands of km. The data set is completed by the WIND and IMP 8 plasma and magnetic field measurements for the study of the large-scale variations.
During more than three decades of intensive experimental investigation of the solar wind our knowledge has progressed to the point that it can be claimed that its basic properties are well understood. The cold solar wind originates at the Sun surface, afterwards it is heated in the solar corona up to ~ 1.5 X 106 K and cooled down by the adiabatic expansion into interplanetary space to ~ 5 X 104 K observed near the Earth orbit. The observations indicate that the solar wind exhibits two distinct states, namely high speed streams characterized by remarkably smooth properties of the plasma and its composition on the one hand, and on the other, low speed streams whose properties are highly variable with considerable fluctuations in density and composition. These two states are connected with different regions on the Sun and they are relatively steady. The interaction of these two streams can lead to the generation of the shock (forward and reverse) pairs but these shocks can be observed at 1 AU only rarely (Burlaga et al., 1985).
In situ measurements of solar wind plasma reported by Schwenn et al. (1978) established the existence of latitudinal gradients in solar wind speed. The observation suggested that the solar wind stream structure itself might vary with the heliographic latitude (Newkirk and Fisk, 1985; Gazis, 1995). These studies have been completed and confirmed by the latest observations of the ULYSSES spacecraft far away from the ecliptic plane (Gosling et al., 1994). Beside the mentioned regular variations there are many irregularities of the solar wind flow which can be detected as fluctuations of the plasma density, velocity or temperature, usually accompanied with the change of the direction or magnitude of the interplanetary magnetic field, which is frozen in the solar wind plasma. Probably the most pronounced disturbances are magnetic clouds (Burlaga et al., 1982), which last often more than one day and which are characterized by an increase of the magnetic field magnitude, magnetic field rotation and low proton density and temperature. The propagation of magnetic clouds or other kinds of perturbations through the supersonic solar wind flow leads often to the creation of interplanetary shocks (Burlaga et al., 1981). The plasma behind the shock is heated and compressed and part of the particles is accelerated to a high energy. Spatial dimensions of the magnetic clouds and other perturbations connected with the coronal mass ejections (CMEs) can vary from hundreds to thousands of RE near the Earth's orbit. On the other end of the spatial scale, one can find various kinds of the MHD fluctuations (Tu, 1991), which generally lead to the heating and acceleration of the solar wind plasma.
The sources of the perturbations mentioned so far are the processes on the Sun and in the solar corona but there are other sources which can affect the solar wind parameters even if the changing solar activity is not considered. It is well known that the planetary bow shocks are sources of the reflected backstreaming particles which generate the MHD waves of large amplitudes. The plasma inside the affected region (foreshock) exhibits substantial fluctuations and the extension of such region is large enough to affect our satellite observations of the solar wind.
In the course of study of the solar wind perturbations the following questions should be answered:
Part of the questions connected with the solar wind perturbations can be answered by the processing of the data measured in one point of space (for example, the frequency of their occurrence, Zastenker et al. 1991) but most problems require simultaneous measurements in two or more points.
The aim of our paper is to show the possibility of the multipoint studies using the data from different spacecraft rather than to bring a comprehensive study of the motion and evolution of the solar wind perturbations. The present situation offers a great opportunity for studying the motion and dimensions of the perturbation because the IMP-8 is still working on its circular orbit, the WIND was launched on November 1, 1994, and the INTERBALL is scanning the near-Earth regions of the interplanetary space from August 3, 1995. We have concentrated our attention on the correlation of the plasma and magnetic field data measured by the mentioned spacecraft in the temporal scale which ranges from minutes to days. Our study supports the widely accepted concept of the solar wind as being homogeneous on the scale of tens Earth's radii if the changes of the solar wind velocity are small and smooth. On the other hand, we show examples of the propagating small structures.
The INTERBALL project was devoted to a detail study of the energy, momentum and mass transfer in the important regions of the solar wind - magnetosphere system. It is conceived as a four-satellite mission of two closely spaced pairs moved on orbits of different altitudes. The first pair of the INTERBALL satellites, the TAIL PROBE, which consists of two satellites (INTERBALL 1 and MAGION-4), was launched into an elongated elliptical orbit with the inclination of 63 degrees, apogee of ~ 195,000 km, and perigee of ~ 800 km, on August 3, 1995. Being launched into the pre-noon sector, the INTERBALL has scanned a broad range of local times throughout the magnetospheric tail toward the subsolar region, which it reached in May 1996. The distance between both satellites is controlled in accordance with the scientific tasks - it is higher in the Earth's magnetospheric tail and smaller for the study of the fine structure of the bow shock and the magnetopause. The temporal satellite separation is given in Figure 1 but the spatial separation varies along the orbit from a hundred of km to about 1 - 3 RE due to an orbital profile of the satellite velocity. The MAGION-4 subsatellite moved ahead of the INTERBALL 1 main satellite in all cases reported in the present study. The main satellite rotates with a period of ~ 120 s and the period of rotation of the subsatellite is ~ 60 s. The rotational axes are parallel to the Sun-Earth line.
Fig1: The separation of the INTERBALL 1 and MAGION-4 satellites along their common orbit as a function of time. The separation is shown for a distance of 1 X 105 km from the Earth.
Both satellites are equipped with a plasma diagnostic payload including instruments for measuring characteristics of magnetic and electric-fields, waves, thermal plasma, and energetic particle distributions (INTERBALL Mission and Payload, 1995). Despite the INTERBALL project being dedicated mainly to the study of the solar wind - magnetopause interaction and to the active plasma processes within the tail where powerful currents, substorms and other large-scale non-linear magnetospheric phenomena occur, many instruments are designed to make comprehensive and quantitative measurements of the dynamics of the interplanetary medium, the Earth's bow shock, foreshock regions, and the magnetosheath.
The observations presented in this paper were made with the omnidirectional plasma detector (VDP and VDP-S) placed onboard both satellites, and with the ion and electron spectrometer (MPS/SPS) placed onboard the MAGION-4. The plasma experiment VDP is intended to determine an integral flux vector and an integral energetic spectrum of ions and electrons in the 0.2 ~ 2.4 keV energy range. For simultaneous measurements in all directions the VDP device is equipped with six independent wide-angle Faraday's cups with the axes forming a three-dimensional orthogonal system. A similar system of four Faraday's cups (VDP-S) is used for two-point measurements onboard the MAGION-4.
The MPS/SPS spectrometer (MAGION-4) is assigned primarily to the measurement of the ion and electron distribution function. The MPS part is designed for the study of the fine structure of the ion flow in the sunward and antisunward directions, the SPS part is intended for the study of the energy and angular ion and electron distribution. The temperature and the velocity modulus can be computed from both (MPS and SPS) electrostatic analyzer data. The determination of ion density requires a joint processing of the analyzer and the Faraday cup data. A complete description of the VDP, VDP-S and MPS/SPS instruments can be found in Nemecek et al. (1996); Safrankova et al. (1996).
For the present study of the spatial correlation of the solar wind parameters three different time intervals have been chosen - the first event lasts for two days, the second for four hours and the third for ~ 20 minutes.
Large Scale Variations
An example of the long term correlation of the measurements of the WIND and the INTERBALL 1 spacecraft during the period from March 30, 1996 to April 1, 1996 is plotted in Figure 2. The velocity profile shown in the top panel of Figure 2 is nearly the same in the measurements of both spacecraft. The small difference in the velocity magnitude, less than ~ 5%, is probably due to the different methodology of the velocity calculation rather than to the calibration of the devices because the velocity determined by the INTERBALL 1, which is systematically higher in Figure 2, can be lower than that of the WIND in other time intervals (see e.g., Figure 4). The nonlinearity of the calibration can be ruled out in this case because the mean values of the velocity are nearly the same in both Figures.
Fig. 2: A comparison of the solar wind velocity (VSW), number flux (FSW) and IMF components and magnitude measured by INTERBALL 1 (thick line) and WIND (thin line) during the period March 30 - April 1. Separation vector is given in Fig. 3.
The second panel of Figure 2 demonstrates good correlation of the ion fluxes. We note that the profile of the ion flux corresponds to the changes of the plasma density because the bulk velocity is nearly constant during the entive interval. Difference between the ion flux magnitudes is higher than that of the velocities but it is still only about 10 ~ 15%. The differences of this order are a common feature if one compares measurements of two spacecraft (Petrinec and Russell, 1993) and we believe that such accuracy is fully acceptable for most tasks connected with the solar wind and its interaction with planets or comets. We do not think that the observed difference is caused by the separation of the spacecraft which can be deduced from the drawing of the spacecraft positions in Figure 3. This drawing is a projection of the position of both spacecraft into the ecliptic plane. The signs show spacecraft positions in the middle of the interval under study, the heavy lines mark the part of the trajectory which corresponds to our interval, and the arrows indicate the direction of the spacecraft motion.
Fig. 3: Portion of the INTERBALL 1 and WIND orbits in the GSE X-Y plane during March 30 - April 1.
Since the solar wind velocity and density are well correlated in both points of the space, one would expect that the same correlation would be found in the interplanetary magnetic field (IMF) because the hypothesis of the IMF being frozen into the solar wind plasma is generally accepted. The BTOT panel of Figure 2 illustrates that the correlation of the IMF measured in two points separated by ~ 50 RE is rather poor and the strength of the IMF at one point can be twice or more times larger for a long period of time. Examination of the behaviour of the magnetic field components (the last three panels of Figure 2) shows that the maximum difference is in the BX component. This difference can be caused partly by the small shift of the zero level of one of the magnetometers. It should be noted that the measurement of the DC magnetic field of the order of ~ 1 nT is a very difficult task under satellite conditions (temperature changes, high level of electromagnetic noise) and the permanent calibration of the measured data is of principal importance. If we suppose that the satellite is spinning around one of the magnetometer axes, the zero level of two spinning components can be determined with an accuracy better than 0.1nT but the accuracy can be smaller for the non-spinning component (Lepping, 1996). The WIND is spinning around the Z-axis while the INTERBALL 1 around the X-axis and, according to Skalsky (1996), the error of the determination of zero level can reach ~ 1 nT for the INTERBALL BX component from the pre-processed data used in Figure 2. Nevertheless, even if the possible uncertainty of the magnetic field determination is taken into account, the magnetic field seems to be more variable than the plasma parameters. The different profile can be caused by the separation of the spacecraft along the YGSE axis (~40 RE) but, in any case, one should be careful when using the IMF data.
Middle Scale Variations
For the study of the motion of the structures in the solar wind we have chosen the simultaneous measurements of three spacecraft (INTERBALL 1, WIND and IMP-8) from September 5, 1995. This day is characterized by a very low solar wind velocity and thus one can expect a higher variability of other solar wind parameters. As can be seen from Figure 4, the velocities observed by the three spacecraft are nearly constant within our 4-hour interval and all spacecraft register the same value. The common feature of the velocity behaviour is a small increase which occurred at ~ 07:15 UT on the WIND, at 08:10 UT on the IMP-8 and at 08:20 UT on the INTERBALL 1. This particular event will be discussed later.
Fig.4: Solar wind velocity as measured in three different points of the interplanetary space on September 5, 1995. The position of the spacecraft used for this comparison is shown in Figure 5.
The projection of the spacecraft positions into the ecliptic plane is shown in Figure 5; the ZGSE coordinates of all spacecraft were less than 10 RE. A good correlation of the ion fluxes measured by the mentioned spacecraft in the time interval from 07:00 to 11:30 UT can be seen in Figure 6. The time scale of the Figure corresponds to the IMP-8 measurements, the data from other spacecraft are shifted to obtain the maximum of the cross-correlation function for a lag equal to zero. The time shifts (11.7 min for the INTERBALL 1 and -24.0 min for the WIND) are nearly equal to the shifts which were obtained by the integration of the solar wind velocity along the distance of the spacecraft.
Fig. 5: A sketch of the INTERBALL 1, WIND and IMP-8 positions on September 5, 1995. The shadowed area depicts the estimated shape of the solar wind perturbation (see text for details).
Fig. 6: A comparison of the ion fluxes measured by IMP-8 and INTERBALL 1, IMP-8 and WIND, INTERBALL 1 and WIND (three upper panels, respectively) on September 5, 1995. The bottom panel shows the cross-correlation coefficient corresponding to the upper panels.
The top panel in Figure 6 shows that not only the overall shapes and magnitudes of the ion flux profiles are nearly the same but even the structures of 10 minute duration correspond well in the measurements of the IMP-8 and INTERBALL 1 spacecraft (see e.g., the decrease of the ion flux at ~ 10:24 UT or 11:12 UT). Higher level of fluctuations in the INTERBALL 1 data is caused by better temporal resolution of the data used in this particular case and thus the fluctuation level can be considered as an indicator of the presence of the small scale perturbations in different time intervals. The flux enhancement, which is observed by the IMP-8 between 08:05 and 08:15 UT, is shifted by about 6 minutes and its amplitude is higher in the INTERBALL 1 measurements. The velocity of this structure in the solar wind frame computed from the time delay between the observations of two spacecraft and from their distance along the XGSE axis (see Figure 5) is ~ 150 km/s. As this velocity highly exceeds the Alfven velocity, we suggest that the observed delay is caused by the oblique edge of the disturbance rather than by the velocity different from that of the solar wind. Moreover, the velocity measured inside this structure is only ~ 20 km.s-1 in the solar wind rest frame (see Figure 5).
A comparison of the IMP-8 and WIND ion flux data in the second panel of Figure 6 gives again a very good overall correlation but there are two distinct non-correlated structures. The first one at 08:05 UT in the IMP-8 data has been discussed above. The same event can be identified as a small increase of the ion flux between 07:36 and 07:48 UT in the WIND data. The time delay between the IMP-8 and WIND observation of this structure can be again explained by the oblique edge of the structure with respect to the solar wind propagation. Different magnitudes of the ion flux enhancement as measured by the three spacecraft can be caused either by the temporal evolution (steepening) of the structure or by the spatial structure. As the bulk velocity inside the structure is higher than the velocity of the surrounding plasma we prefer the steepening as a more probable cause. A comparison of the enhancements of the ion flux observed by the INTERBALL 1 and the WIND suggests that the discontinuity probably had not been formed on the Sun but that is evolved gradually somewhere in the solar wind.
The second non-correlated structure (at ~ 10:24 UT) is again registered first by the WIND spacecraft but the IMP-8 and the INTERBALL 1 registered it at the same time (top panel in Figure 6). The time delay between the WIND and the IMP-8 observations corresponds to the apparent velocity of the motion of the structure in the range from 60 to 150 km/s in the solar wind frame. A more precise determination is impossible because the shape of the observed flux depletion is different in the measurements of both spacecraft. Since this velocity exceeds the Alfven one and no indication of the shock formation neither in the plasma nor magnetic field data (Fig. 7) is seen, we suppose that the observed time delay is caused by the orientation of the edge of the structure.
The position of the spacecraft allows us to determine some dimension limits of the discussed disturbances. Their thickness along the XGSE axis is about ~30 RE, the extent along the YGSE axis should be greater than the satellite separation (~ 30 RE). The estimated dimensions and the shape of the disturbances are shown as the shadowed area in Figure 5.
A comparison of the magnetic field measurements for the same time interval is shown in Figure 7, where the time scales are shifted in the same way as in Figure 6. One can note that the correlation between the magnetic field is worse than the correlation of the ion fluxes but the hypothesis of the magnetic field frozen in the solar wind plasma is almost valid for the time interval under study. The enhancements/depressions of the ion flux mentioned above are accompanied by the decrease/increase of the magnetic field strengths and they move with the same velocity as the corresponding plasma structures as measured by all three spacecraft. The examination of the changes of the magnetic field direction inside the structures and on their boundaries suggests that they can be classified as tangential discontinuities.
Fig. 7: A comparison of the IMF measurements on September 5, 1995 shown in the same way as fluxes in Figure 6.
Small Scale Variations
The position of the INTERBALL spacecraft during the interval which has been chosen for a demonstration of the small scale solar wind structures is depicted in Figure 8 together with the position of the bow shock, calculated for the actual solar wind condition (Nemecek and Safrankova, 1991). The arrow shows the direction of the magnetic field at 18:50 UT. The INTERBALL 1 is located at the GSE (X,Y, Z) = (13.6,20.2, 12.5) RE and moves outward, mainly along the YGSE axis. The mutual position of the INTERBALL 1 and MAGION-4 spacecraft is plotted in the right part of Figure 8.
Fig. 8: Position of the INTERBALL satellites with respect to the bow shock (dotted line). The detail in the right part of the Figure shows the mutual position of the INTERBALL 1 and the MAGION-4. The dashed line shows a probable orientation of the edge of the foreshock wave.
The magnetic field and plasma data plotted in Figure 9 exhibit substantial fluctuations from the beginning of our interval to about 19:05 UT and then between 20:15 and 20:30 UT. The magnetic field direction is oriented nearly along the bow shock normal till 19:05 UT (inward till 18:40 UT, outward between 18:40 and 19:05 UT) and it has a similar orientation during the second interval of higher fluctuations. It means that the high frequency fluctuations can be attributed to the foreshock of the quasiparallel bow shock, despite the fact that the INTERBALL satellites are located near the apogee, far ahead of the bow shock. The foreshock is usually studied separately but, from the point of view of the solar wind - magnetosphere interaction, it would be taken as part of the solar wind because a great portion of the magnetosphere is affected by the flow formed by the foreshock processes. The time delay between the observations of the INTERBALL satellites as well as the degree of the correlation cannot be seen from the Figure due to the small distance between the satellites and the short duration of the fluctuations. Nevertheless, during the periods when the bow shock behind the satellites is quasiperpendicular (e.g., 19:15 ~ 20:15 UT), the fluxes measured by the two INTERBALL satellites are well correlated.
Fig. 9: The magnetic field and ion flux data obtained by the INTERBALL and the ion flux measured by the MAGION-4 (bottom panel) on March 22, 1996 in the foreshock region.
Figure 10 shows a detail comparison of the ion flux measurements of the INTERBALL satellites for a shorter time interval located on the boundary of the foreshock region. The data from both satellites are depicted in one panel but the flux scales are shifted for a better comparison. The ion flux is modulated by irregular oscillations with a typical period of about 1 minute. The oscillations are well correlated in the measurements of both satellites; the correlation coefficient is ~ 0.81 and peaks for a lag of 3.5 s. It means that the typical time of the motion of the corresponding structures from the MAGION-4 satellite to the INTERBALL 1 satellite is 3.5 s as can be seen in Figure 11. The time is a mean propagation time of all changes of the ion flow which occurred during the time interval under study, regardless of their nature (i.e., waves, discontinuities, structures, etc.). We have chosen the time interval without any significant change in the background value of the ion flow and thus the time delay determined from Figure 11 can be attributed to the observed oscillations. The velocity of the propagation can be a function of the frequency of oscillations but the shape of the correlation function suggests that all modes, except that with a period of ~ 47s, are damped over a distance between the satellites. The separation of the satellites along the XGSE axis is only 300 km and thus the observed delay is rather big. However, it is consistent with an assumption that the observed oscillations are the waves that propagate along the magnetic field with the Alfven velocity and are blown down by the solar wind. In such case the leading edge of the perturbation would be nearly parallel to the shock (see dashed line in Figure 8) and at the time, when the leading edge reaches the MAGION-4 position, the distance of this edge to the INTERBALL 1 along the XGSE axis is ~ 1500 km. This distance corresponds roughly to the determined time delay between the observations of both INTERBALL satellites. The secondary maxima of the correlation coefficient profile indicate a quasiperiodic character of the observed fluctuations which is connected with their wavy nature as mentioned above.
Fig. 10: On March 22, 1996 a comparison of the ion fluxes measured in two shortly separated points. The gray and dark lines correspond to the left and right scale (INTERBALL 1 and MAGION-4, respectively).
Fig. 11: Cross-correlation coefficient of the ion fluxes measured by the INTERBALL 1 and MAGION-4 satellites on March 22, 1996.
Figure 12 shows a detail of the observed ion flux fluctuations. Good correlation of the measurements, which is clearly seen in Figure 10, is not so clear on the shorter scale. Nevertheless, the time shift of about 3.5 s can be identified between the corresponding maxima or minima.
Fig. 12: Comparison of the ion fluxes in the foreshock region on a short time scale on March 22, 1996. The gray and dark lines correspond to the scales as in Figure 10.
We have examined the correlation between the solar wind parameters and the IMF values in different points of the interplanetary space with motivation to test the applicability of the data measured at a large distance from the Earth to the study of the solar wind - magnetosphere interaction. For the sake of simplicity we have chosen intervals with constant solar wind velocity and investigated only the variations of the ion density and the IMF magnitude on different temporal scales. The condition of the constant velocity allows us to use the single time shift for a description of the solar wind propagation.
The first example in our study has shown a good correlation of the ion density changes during a period of two days despite the magnitude of the ion flux varying from 2 to 16.108 cm-2.s-1 and the testing points being separated by more than 50 RE along the YGSE axis. On the other hand, a comparison of the magnetic field measurements has shown that the IMF measured far upstream can differ substantially from the value in the vicinity of the Earth's magnetosphere. For the study of the dynamics of the Earth's magnetosphere this would be very important but, unfortunately, at present we are not able to determine which part of observed differences is caused by the calibration of magnetometers and which part is connected with the evolution of the IMF itself. This effect will be a subject of further studies.
The second example in our study has shown that the quiet solar wind can contain a lot of inhomogeneities of different shapes. Some of them come probably from the Sun but there are indications that part of them can undertake a significant evolution when travelling through the interplanetary space. The shape of these structures is probably controlled by the IMF orientation and thus they can reach the Earth's magnetosphere at a time that differs substantially from the time based on the solar wind velocity and the distance between the measuring point and the subsolar point of the bow shock or magnetopause. The dimensions of the structures shown in our study are big enough to affect the whole magnetosphere. The search for the smaller structures should be based on the data with better time resolution and on the measurements of the satellite separated less than those used in the present study.
The third example in our study confirms a well known hypothesis that the foreshock region is highly turbulent and consists of the clouds of enhanced plasma density (Safrankova et al., 1991; Scholer, 1993), moving with the solar wind velocity. Two point measurements allow us to determine the spatial dimensions and the orientation of these clouds in the space. The next study of this phenomenon will be oriented to the cases when the IMF changes its orientation to determine the lifetime of density fluctuations. The systematic study of the foreshock region is very important from two points of view: first, the quasiparallel shock is a significant part of the bow shock and, second, in our contribution it has been shown that for the study of the solar wind-magnetosphere interaction the data taken in the vicinity of the Earth's magnetosphere are more appropriate.
In the introduction of our paper we have outlined a few questions which should be answered in the course of study of the solar wind perturbations. We believe that our preliminary analysis shows that the scientific payload of the INTERBALL satellites is able to contribute to the solution of all mentioned tasks. Moreover, we hope that our multipoint study would invoke further investigation of this exciting topic.
The present work was supported by the Czech Grant Agency under Contracts No 205/96/1575 and 202/94/0467 and by the Charles University Grant Agency under Contract No 180. The authors express their thanks to N. Ryb'eva and A. Skalsky for supplying the MIF-M magnetic field data and to A. Lazarus, R. Lepping, and K. I. Paularena for the WIND and IMP-8 plasma and IMF data.
Burlaga, L. F., E. Sittler, F. Mariani, R. Schwenn, Magnetic loop behind an interplanetary shock: Voyager, Helios and IMP-8 observations, J. Geophys. Res., 86, 6673 (1981).
Burlaga, L. F., L. W. Klein, N. R. Sheeley, Jr., D. J. Michels, R. A. Howard, H. J. Koomen, R. Schwenn, H. Rosenbauer, Magnetic cloud and coronal mass ejection, Geophys. Res. Lett., 9, 1317 (1982).
Burlaga, L. F., V. Piezo, A. Lazarus, P. Gazis, Stream dynamics between 1 AU and 2 AU: A Comparison of Observations and Theory, J. Geophys. Res., 90, 7377 (1985).
Gazis, P. R., Synoptic maps of solar wind parameters from in situ spacecraft observation, J. Geophys. Res., 100, 3383 (1995).
Gosling, J. T., S. J. Bame, D. J. McComas, J. L. Phillips, B. E. Goldstein, M. Neugebauer, The speeds of coronal mass ejections in the solar wind at mid-heliographic latitudes: Ulysses, Geophys. Res. Lett., 21, 1109 (1994).
INTERBALL Mission and Payload, ed. by Yu. Galperin, T. Muliarchik, and J.-P. Thouvenin, CNES-IKI-RSA (1995).
Lepping R., private communication, (1996).
Nˆemecek, Z., and J.Safrankova, The Earth's bow shock and magnetopause position as a result of the solar wind - magnetosphere interaction, Journal of Atmospheric and Terrestrial Physics, 53, 1049 (1991).
Nemecek, Z., Fedorov A., Safrankova J., Zastenker G., Structure of the low-latitude magnetopause: MAGION 4 observations, Annales Geophys., in press.
Newkirk, G., and L. A. Fisk, Variation of cosmic ray and solar wind properties with respect to the heliospheric current sheet, 1, Five GeV protons and solar wind speed, J. Geophys. Res., 90, 3391 (1985).
Petrinec, S. M., and C. T. Russell, Intercalibration of solar wind instruments during the international magnetospheric study, J. Geophys. Res., 98, 18963 (1993).
Skalsky A., private communication, (1996).
Safrankova, J., Z. Nemcek, O. Santolik, Some comments on the ion distribution function evolution in the quasiparallel shock, Adv. Space Res., 11, (9)223 (1991).
Safrankova, J., G. Zastenker, Z. Nemecek, A. Fedorov, M. Simersky, L. Prech, Small scale observation of the magnetopause motion: Preliminary results of the INTERBALL project, Annales Geophys., in press.
Scholer, M., Upstream waves, shocklets, short large-amplitude magnetic structures, and the cyclic behaviour of oblique quasi-parallel collisionless shocks, J. Geophys. Res., 98, 4 (1993).
Schwenn, R., M. D. Montgomery, H. Rosenbauer, H. Miggenrider, K.-H. Mulhauser, S. J. Bame, W. C. Feldman, and R. T. Hansen, Direct observations of the latitudinal extent of a high-speed system in solar wind, J. Geophys. Res., 83, 1011 (1978).
Tu, C.-Y., The interplanetary MHD fluctuations and its relation with the heating and acceleration of the solar wind, J. Geomag. and Geoelect., 43, 101 (1991).
Zastenker, G., L. Avanov, Yu. Yermolaev, P. Bochsler, Z. Nemecek, J. Safrankova, Variability of coronal structures and ion components in the solar wind, Czech. J. Phys., 41, 1001 (1991).