GEOPHYSICAL RESEARCH LETTERS, VOL. 29, NO. 9, 10.1029/2001GL014513, 2002

Factors controlling the location of the Bow Shock at Mars D. Vignes,1 M. H. Acun˜a,1 J. E. P. Connerney,1 D. H. Crider,1 H. Re`me,2 and C. Mazelle2 Received 5 December 2001; revised 19 February 2002; accepted 28 February 2002; published 11 May 2002.

[1] During the first year of the Mars Global Surveyor (MGS) mission, 553 shock crossings have been identified from a total of 363 orbits. The shape of the shock has been determined by examining the MGS spacecraft Magnetometer/Electron Reflectometer (MAG/ER) data. The location of the shock was found highly variable. The present study shows that the high crustal magnetic sources, found in the southern hemisphere, do not seem responsible for the Bow Shock (BS) variability. The present study shows that contrary to many expectations there is no obvious strong one to one correlation between the location of the highest crustal sources and the variability of the shock position. On the other hand, the shock appears farthest from Mars in the hemisphere of locally upward interplanetary electric field consistent with the idea that mass loading play a role in controlling the BS location, which confirms previous results. INDEX TERMS: 2780 Magnetospheric Physics: Solar wind interactions with unmagnetized bodies; 6225 Planelotogy: Solar System Objects: Mars; 2152 Interplanetary Physics: Pickup ions

1. Introduction [2] The solar wind (SW) is deflected around all the planets of the solar system, creating a bow shock. Depending on the planet, the nature of the obstacle is quite different. Magnetized planets like Earth deflect the solar wind far above their atmosphere while for unmagnetized planets, like Venus, the SW is diverted at low altitudes. In the case of the solar wind interaction with Venus, a lot of results on global scale were found with the Pioneer Venus Orbiter data concerning the variation of the terminator bow shock BS position. The location of the Venus BS is sensitive to (1) the solar cycle and solar EUV flux, (2) the upstream solar wind parameters, and (3) the orientation of the interplanetary magnetic field [Russell et al., 1988]. An equator to pole asymmetry and a North-South asymmetry, by respect to the interplanetary magnetic field direction, were found on the Venus BS location [Alexander et al., 1986]. The North-South asymmetry was due to asymmetric mass loading of the sheath by picked-up oxygen. The additional mass added to the flow in the North slows the flow more and creates a bigger effective obstacle there. [3] The nature of the solar wind interaction with Mars was not fully determined before Mars Global Surveyor (MGS) mission. However, the Phobos 2 magnetic field data were used to examine the location of the bow shock, and it was found that the interplanetary magnetic field controls the shock location and that the Mars shock location varies with solar activity [Zhang et al., 1991b]. The Magnetometer/Electron Reflectometer (MAG/ER) experiment on Mars Global Surveyor (MGS) includes two magnetometers, which acquire continuous vector magnetic field measurements at a rate as high as 32 samples/sec, and an electron

1 2

NASA Goddard Space Flight Center, Greenbelt, USA. Centre d’E´tude Spatiale des Rayonnements, Toulouse, France.

Copyright 2002 by the American Geophysical Union. 0094-8276/02/2001GL014513$05.00

reflectometer, which measures electrons fluxes for energy between 10 eV – 20 keV with a time resolution as high as 2 seconds [Acun˜a et al., 1992]. Recent observations by the MAG/ ER experiment show crustal magnetic sources of high magnitude are mainly confined to the southern highlands [Acun˜a et al., 1999; Connerney et al., 1999]. The most intense magnetic crustal sources detected by MGS-MAG/ER lie in the Terra Sirenum region (150
E to 240
E; 30
S to 85
S). As the planet rotates, a strong Martian remnant field should locally increase the total pressure presented by Mars to the solar wind resulting in an asymmetric obstacle when this region is in the sunlit hemisphere. It is therefore essential to first investigate the bow shock location dependence on crustal magnetic sources and then on interplanetary magnetic field directions.

2. Dependence Analysis [4] In order to determine the location of the Bow Shock, magnetic and electron data provided by the MAG/ER experiment on MGS were used. During the first year of the mission, the orbit was far enough from Mars to enable the spacecraft to cross the BS. An axisymetric fit of all these BS crossings was made using a conic section. The equation of the shock surface is r = L/(1 + ecosq) where the polar coordinates (r, q) are measured about the focus of the conic, L is the semi-latus rectum and e is the eccentricity. The shock was found highly variable and the solar activity does not seem to play the major role on the BS variation. MGS data from medium solar activity and Phobos 2 observations from solar max give nearly the same mean BS fit suggesting the mean BS position is independent of the phase of the solar cycle [Vignes et al., 2000]. Others factors that could explain this variability include crustal magnetic sources, the interplanetary magnetic field (IMF) orientation, and upstream solar wind parameters. In order to study the variability of the BS position and to compare with previous studies, we extrapolate each crossing to the terminator plane. Conic section fits were used to project them to the terminator plane using the same eccentricity and focus location as for the mean fit, allowing L to change to fit the observed point. 2.1. Influence of the Crustal Sources [5] During the Phobos 2 circular orbit, it was found that the time series magnetic field data contain spectral peaks at 12 and 24 hours. This periodicity was interpreted as a corotating part of the magnetic field, and thus in favor of existence of a planetary magnetic field [Mohlmann et al., 1992]. Recent MGS discovery have found that Mars has no intrinsic magnetic field but has strong localized crustal magnetic fields in the southern hemisphere [Acun˜a et al., 1999]. The most intense magnetic crustal sources detected by MGS-MAG/ER lie in the Terra Sirenum region (150
E to 240
E; 30
S to 85
S). Connerney et al. [1999] presents analytical models of the sources in this region. The total magnetic moment of this region ( 1.3  1017 A.m2) is sufficiently high to increase the total pressure presented by Mars to the solar wind. Thus, corotating features are capable of producing the variability of the shock. Figure 1 shows the calculated terminator BS position versus the longitude of the subsolar point at the time of the BS crossing in order to study the bow shock location dependence on

42 - 1

42 - 2

VIGNES ET AL.: FACTORS CONTROLLING THE SHOCK LOCATION

Figure 1. Location of Mars bow shock extrapolated to the terminator plane versus the subsolar longitude of the spacecraft at the time of the crossing. Medians every 20 degrees are given.

crustal magnetic sources. If the variability of the bow shock location is caused by corotating features, the figure should show a distinct maximum for the Terra Sirenum region. Hence, the shock position does not obviously appear to depend strongly on the planetary longitude. It is worthwhile to mention that Schwingenschuh et al. [1992] and Slavin et al. [1991] also have found that the distance of the BS is independent of subsolar longitude with the data of Phobos 2. 2.2. Influence of the IMF [6] In order to investigate the relation between the terminator shock distance and the interplanetary magnetic field orientation, we have to determine the direction of the IMF for each orbit. There is no other spacecraft upstream to the Martian shock which measures the solar wind parameters. Thus, upstream of each shock crossing, a measure of the magnetic field direction was made. Measurements of IMF were made exclusively for intervals of a few minutes upstream from a specified bow shock crossing where the IMF was steady. Assuming a 1 nT incertitude over the interplanetary components, we restrained the data base to the cases where the 1 nT uncertainty in components translated into an uncertainty of the interplanetary magnetic field direction of less than 20 degrees. 2.2.1. q Bn. [7] To calculate the angle between the upstream magnetic field and the shock fit normal (qBn) at the time of the crossing, a conic section was used, with an eccentricity of 1.03, a conic parameter of 2.04 and a focus location at 0.64 Martian radius (RM) from the center of the planet toward the Sun (a 4 degree angle aberration has been taken into account) [Vignes et al., 2000]. qBn was estimated by the angle between IMF and mean fit shock normal at the point closest to the shock crossing. Figure 2 shows the extrapolated terminator distance with respect to the qBn angle for each orbit. There are more crossings with large qBn angle because the determination of the precise location of bow shock crossing is difficult for low qBn angle, or quasi-parallel type shock. If we use the classical denomination of quasi-parallel (qBn  45
) and quasi-perpendicular shocks, (45
 qBn  90
) the mean value of the terminator distance for the 23 quasi-parallel shock crossings is 2.67 RM while for the 93 quasi-perpendicular shock crossings,

the mean value is 2.76 RM. Therefore, we found a difference of about 3%. This difference creates a dawn-dusk asymmetry in the bow shock location because according to the IMF orientation, quasi-parallel shocks dominate the dawn side while quasiperpendicular shocks dominate the dusk side. This asymmetry is consistent with a previous study on Venus and Mars [Zhang et al., 1991a], which found the quasi-parallel shock is closer to the planet than the quasi-perpendicular shock. However, in Figure 2, no restriction was made concerning factors like cone angle, clock angle or susbolar longitude. 2.2.2. Clock Angle and Cone Angle. [8] The Venus bow shock was found to be strongly sensitive to the phase of the solar cycle [Russell et al., 1988]: the terminator bow shock distance is greater for solar maximum than for solar minimum. Alexander and Russell [1985] have invoked mass loading to explain the dependence of the terminator BS position with the solar cycle. Photo-ionization, charge exchange with solar wind protons, and impact ionization convert atmospheric neutral particles into ionized atoms. Then, these ions are picked-up by the convective electric field of the solar wind. The increasing solar EUV flux added additional mass to the planetary sheath and caused the BS to move outward. In the case of Mars, by comparison of MGS and Phobos 2 mean fit results, the mean Mars BS does not appear sensitive to the phase of the solar cycle. The sunspot number during the operational period of Phobos 2 mission, from 29 January to 27 March 1989, was between 140 and 180 while during the first year of MGS mission, it was between 30 and 90 [Vignes et al., 2000]. Let us note that Trotignon et al. [1991], and Russell et al. [1992], found that the Martian BS location respond to the variation of the solar EUV flux. [9] Further, in the case of the solar wind interaction with Venus, the magnetic field direction was found to control the shock location. The pickup effect is at its maximum value when the solar wind flow and the IMF are perpendicular. This particular angle, the angle between the interplanetary magnetic field and the direction of the solar wind velocity, is called the cone angle. The Venus BS was found to be sensitive to the cone angle [Alexander et al., 1986]. A three-dimensional global hybrid simulation of the solar wind

Figure 2. Location of Mars BS extrapolated to the terminator plane versus qBn. Medians of the shock position are given for quasiparallel and quasi-perpendicular shock.

VIGNES ET AL.: FACTORS CONTROLLING THE SHOCK LOCATION

Figure 3. Location of Mars BS extrapolated to the terminator plane versus Clock Angle. Large and small cone angle are distinguished. Medians of the shock position are given depending on the hemisphere and depending on large or small cone angle. interaction with Venus dayside shows that the planetary pickup ions cause a number density asymmetry in the direction of the convective electric field [Moore et al., 1990]. Thus, to show the effect of the pick-up ions, we also need to look to the hemisphere of locally upward convective electric field. The relative clock angle corresponds to the angle between the projection of the radial vector at the shock crossing location into the terminator plane and the direction of the IMF also projected in this plane. Then, a bow shock crossing in the northern hemisphere with respect to the direction of the interplanetary magnetic field, occurs for clock angles between 0
to 180
. A recent study by Dubinin et al. [1998] using the data of Phobos 2 observes the bow shock at a larger distance in the northern hemisphere. [10] Figure 3 shows the extrapolated terminator distance as function of the clock angle. We distinguish cases of large and small cone angle in order to show the cumulative effect of large cone angle in the hemisphere of locally upward electric field. In the northern hemisphere, the mean value of the terminator distance for large cone angle (2.9 RM) is 13% greater than for small cone angle in the same hemisphere. While, in the southern hemisphere we found a small differences (3%) between cases of large and small cone angle. Thus, the shock appears significantly farther from the planet in the hemisphere of locally upward electric field when the angle between the solar wind direction and the interplanetary magnetic field is large. This North-South asymmetry is consistent with the idea of asymmetric mass loading of the sheath by picked-up oxygen. The additional mass added in the northern hemisphere slows the flow more and creates a bigger obstacle. Thus mass loading appears to be playing some role in controlling the BS location. This role is expected to be more important during solar maximum since EUV fluxes increase and, thus photo-ionization rates.

3. Summary and Conclusions [11] During the first year of the MGS mission, the apoapsis of the orbit was far enough to enable MGS to cross the shock, which was found to be highly variable. The BS dependence on factors such as upstream solar wind parameters and orientation of

42 - 3

the interplanetary magnetic field was examined. Cloutier et al. [1999] have found some similarities between Mars and Venus on small scale for the ionosphere (e.g., magnetic field rotation associated with ion clouds and flux ropes formations in the low field ionospheric regions), and in this study some similarities on the global scale are found. (1) The shock position depends on the qBn angle; the quasi-parallel shock is closer to the planet than the quasi-perpendicular shock. (2) The Martian BS terminator distance is greater for large cone angle than for low cone angle in the northern hemisphere, with respect to the interplanetary magnetic field direction. It is worth noting that this study of the IMF control of the Martian bow shock is made exclusively with cases where the IMF amplitude is strong enough to have an uncertainty in the direction of the IMF that is less than 20 degrees. This data restriction enables this study to look at cases of favorable pick-up effects: strong IMF, large cone angle and north clock angle. [12] Mass loading may be important in the solar wind interaction with Mars. However, the mean BS position was found to be independent of the solar cycle and solar EUV flux by comparison of MGS and Phobos 2 results. Furthermore, Slavin et al. [1991] found a bow shock at Mars more distant than for Venus in the terminator plane, 2.66 RM versus 2.39 RV with a standard deviation of 0.49 RM versus 0.21 RV during solar maximum. With MGS results, during period of medium solar activity, a terminator distance of 2.62 RM is found with a standard deviation of 0.33 RM. The Mars BS position in the terminator plane appears more variable than the case of Venus. Thus the solar wind interaction with Mars is slightly different than the case of Venus even if there are some similarities. Further, that the mean Martian BS position is independent of the solar cycle and EUV flux while the mass loading plays a role in the solar wind interaction with Mars suggests that many different factors control the position of the Martian BS. The present study supports the idea that the presence of crustal sources does not play a significant role in the position of the bow shock. Trotignon et al. [1993] and Schwingenschuh et al. [1992] found that the subsolar shock distance appears to vary slightly with the solar wind dynamic pressure during period of maximum solar activity from the Phobos 2 data analysis. The MAG/ER experiment does not provide the solar wind dynamic pressure. Hence, it would be very interesting to study this correlation during minimum or medium solar activity. The variations of the dynamic pressure could be related to the ion formation process and then to the mass loading process [Breus et al., 1989]. A more comprehensive analysis would require a more complete study of parameters such as solar wind density and velocity. [13] Acknowledgments. This work was performed while D. Vignes held a visitor researcher position sponsored by The Centre National d’Etudes Spatiales (CNES) at NASA Goddard Space Flight Center.

References Acun˜a, M. H., et al., Mars Observer Magnetic Fields Investigation, J. Geophys. Res., 97, 7799 – 7814, 1992. Acun˜a, M. H., et al., Global Distribution of Crustal Magnetization Discovered by the Mars Global Surveyor MAG/ER Experiment, Science, 284, 790 – 793, 1999. Alexander, C. J., and C. T. Russell, Solar cycle dependence of the location of the Venus bow shock, Geophys. Res. Lett., 12, 369, 1985. Alexander, C. J., J. G. Luhmann, and C. T. Russell, Interplanetary field control of the location of the Venus bow shock: Evidence for comet-like ion pickup, Geophys. Res. Lett., 13, 917 – 920, 1986. Breus, T. K., S. J. Bauer, A. M. Frymskii, and V. Ya. Mitnitskii, Mass loading in the solar wind interaction with Venus and Mars, J. Geophys. Res., 94, 2375 – 2382, 1989. Cloutier, P. A., et al., Venus-like interaction of the solar wind with Mars, Geophys. Res. Lett., 26, 2685 – 2688, 1999. Connerney, J. E. P., et al., Magnetics lineations in the ancient crust of Mars, Science, 284, 794 – 798, 1999.

42 - 4

VIGNES ET AL.: FACTORS CONTROLLING THE SHOCK LOCATION

Dubinin, E., K. Sauer, M. Delva, and T. Tanaka, The IMF control of the Martian bow shock and plasma flow in the magnetosheath. Predictions of 3-D simulations and observations, Earth Planets Space, 50, 873 – 882, 1998. Mohlmann, D., et al., Magnetic field environment of Mars as studied by Phobos 2, Adv. Space Res., 12, 9221 – 9229, 1992. Moore, K. R., et al., A statistical study of ions and magnetic field in the Venus magnetotail, J. Geophys. Res., 95, 12,005 – 12,018, 1990. Russell, C. T., et al., Solar and interplanetary control of the location of the Venus bow shock, J. Geophys. Res., 93, 5461 – 5469, 1988. Russell, C. T., et al., Bow shocks and magnetotails of Venus and Mars: A comparison, Adv. Space Res., 12, 163 – 167, 1992. Schwingenschuh, K., et al., The Martian magnetic field environment: Induced or dominated by an intrinsic magnetic field, Adv. Space Res., 12, 9213 – 9219, 1992. Slavin, J. A., et al., The solar wind interaction with Mars: Mariner 4, Mars 2, Mars 3, Mars 5 and Phobos 2 observations of bow shock position and shape, J. Geophys. Res., 96, 11,235 – 11,241, 1991. Trotignon, J. G., et al., Plasma wave system measurements of the Martian bow shock from the Phobos 2 spacecraft, J, Geophys. Res., 96, 11,253 – 11,264, 1991. Trotignon, J. G., et al., Position and shape of the martian bow shock: The Phobos 2 plasma wave system observations, Planet. Space Sci., 41, 189 – 198, 1993. Vignes, D., et al., The Solar Wind interaction with Mars: Locations and

shapes of the Bow Shock and the Magnetic Pile-up Boundary from the observations of the MAG/ER experiment onboard Mars Global Surveyor, Geophys. Res. Lett., 27, 49 – 52, 2000. Zhang, T. L., K. Schwingenschuh, C. T. Russell, and J. G. Luhmann, Asymmetries in the location of the Venus and Mars bow shock, Geophys. Res. Lett., 18, 127 – 129, 1991a. Zhang, T. L., et al., Interplanetary magnetic field control of the Mars bow shock: Evidence for Venus like interaction, J. Geophys. Res., 96, 11,265 – 11,269, 1991b.

D. Vignes, NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA. ([email protected]) M. H. Acun˜a, NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA. ([email protected]) J. E. P. Connerney, NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA. ([email protected]) D. H. Crider, NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA. ([email protected]) H. Re`me, Centre d’E´tude Spatiale des Rayonnements, Toulouse, France. ([email protected]) C. Mazelle, Centre d’E´tude Spatiale des Rayonnements, Toulouse, France. ([email protected])