the solar wind interaction with mars...

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 87, NO. B12, PAGES 10,285-10,296, NOVEMBER 30, 1982

The Solar Wind Interaction With Mars Revisited

JAMES A. SLAVIN AND ROBERT E. HOLZER

Institute of Geophysics and Planetary Physics, University of California, Los Angeles, California 90024

Owing to a paucity of observational data, no clear consensus has been reached concerning the general nature of the solar wind interaction with Mars. In particular, the previous analyses are still at odds regarding the existence of a small intrinsic field magnetosphere at Mars as opposed to a Venus-type ionospheric interaction (e.g., Russell, 1978a,b; Dolginov, 1978b,c). This study contributes to the resolution of this question in three ways. First, an improved determination of effective obstacle altitude and shape is obtained from the Mars 2, 3, and 5 bow shock encounters through the use of a recently published catalog of gasdynamic flow solutions (Spreiter and Stahara, 1980a,b). Second, building upon the Pioneer Venus findings at a field-free planet (Brace et al., 1980; Elphic et al., 1980), it is shown that the Martian ionosphere cannot support a Venus-type ionopause at the obstacle altitudes inferred through our modeling of the bow wave observations even when maximal induced ionospheric magnetic fields and solar maximum EUV levels are assumed. These results allow an effective Mars magnetic dipole moment of 1.4 (_ 0.6) x 1022 G-cm 3 to be determined that stands off the solar wind over the dayside hemisphere at altitudes ranging from ,• 500 km at the subsolar point to •-1000 km near the terminator with no direct aid from the ionosphere under average solar wind/magnetospheric conditions. Third, a search of published Mars and Mariner radio occultation measurements produced no evidence for the existence of an ionopause at Mars in agreement with the Viking study of Lindal et al. (1979). Rather, the electron density altitude profiles appear qualitatively consistent with the Martian ionosphere terminating in a chemopause associated with the effects of magnetospheric convection as first proposed by Bauer and Hartle (1973). After a review of the various arguments in the literature, as supplemented by the results of this study, we conclude that Mars most probably possesses a small intrinsic field magnetosphere.

INTRODUCTION

The successful operation of the Pioneer Venus orbiter (PVO) with its periapsis in the Cytherean ionosphere has left Mars as the only terrestrial planet for which we lack definitive in situ observations of its interaction with the solar wind. This dearth of low-altitude (i.e., < 103km) particles and fields measurements at Mars has led to controversy (e.g., Russell [1978a,b] versus Dolginov [1978b, c]), over whether or not magnetic fields intrinsic to the planet dominate its interaction with the solar wind (see also reviews by Michel [ 1971 ],Hill and Michel [ 1975],Bauer [ 1976],Ness [ 1979],Russell [1979], and Siscoe and Slavin, [1979], Gringauz [1980]). In this study three specific questions that bear on the size of the Martian magnetic moment are considered:

1. What is the altitude of the obstacle to the solar wind implied by the shape and position of the Mars bow wave as determined by the Mars 2, 3, and 5 (M2, M3, and M5) orbiters?

2. When the possible contributions of magnetic fields induced in the ionosphere are considered, do the Viking descent measurements [Hanson et al., 1977] indicate the need for an intrinsic magnetic field to stand-off the solar wind; and if so, how large must the planetary magnetic moment be?

3. On the basis of both the various radio occulation measure-

ments and the Viking descent observations, at what altitude does the upper boundary of the Martian ionosphere lie and is it formed by the action of magnetospheric convection (i.e., a chemopause) or solar wind ram pressure (i.e., an ionopause)?

We found in this study that all available observations are consistent with the solar wind being diverted about Mars by a modest intrinsic magnetic field without any direct involvement by the ionosphere during average solar wind/magnetospheric conditions. However, during intervals of enhanced solar wind dynamic pressure and/or dayside magnetic merging it is possible that the solar wind

Copyright 1982 by the American Geophysical Union.

Paper number 1A 1202. 0148-0227/82/001 A- 1202505.00

will be able to penetrate to heights <•300 km where the ionosphere, and eventually the neutral atmosphere [e.g., Cloutier et al., 1969; Wallis, 1973; Gombosi et al., 1980], will be able to participate in deflecting/absorbing the flow.

IMPLICATIONS OF THE MARS BOW WAVE

Detection of the Mars bow shock by Mariner 4 [Smith, 1969] and later with the Mars 2, 3, and 5 orbiters [Vaisberg, 1976; Gringauz, 1976; Dolginov, 1976; and references therein] has established that the Red Planet presents a largely nonabsorbing obstacle to the solar wind as has been found to be the case for all planetary bodies examined thus far save the earth's moon with its poor electrically conducting surface and low level of global surface magnetization. Figure 1 plots in Mars-centered solar wind aberrated ecliptic coordinates 14 available crossings of the Mars bow wave from the observations of Mars 2, 3, and 5 (O. L. Vaisberg and V. N. Smimov, private communication, 1979) for x' > -1RMs (1RMs = 3390 km). Solid line segments specify intervals over which the transitions between shocked and free streaming solar wind took place, while in one instance dashes connect multiple encounters by Mars 3 on the same pass. Fitting the mean shock positions per inbound or outbound pass with a 2nd order curve symmetric about the x' axis produces the ellipse of eccentricity 0.94 and semi-latus rectum 1.94Rus with its focus atx' - 0.5Rus shown as a solid curve in Figure 1. In contrast to earlier models of this type [Bogdanov and Vaisberg, 1975;Russell, 1977] we have obtained superior fits to the observations by allowing the conic focus to lie anywhere along the symmetry axis as opposed to being fixed on the origin (see also Slavin and Holzer [1981 ]). In addition, consideration has been restricted to only the shock crossings forward of x' - -1R•s because of both the poorer sampling farther downstream and the desire to investigate the dayside flow that is linked by characteristic lines of the gasdynamic solutions to this forward region of the bow wave [Spreiter et al., 1966].

For comparison, also displayed in the first figure are the mean observed bow shock locations at Venus obtained from the Venera 9

10,285

10,286 SLAVIN AND HOLZER: MARS SOLAR WIND INTERACTION REVISTED

o. 4 , •'[ Pioneer Venus MARS BOW SHOCK ]•N• ½S•ovin eto/., 1980)

/ ,• .... •Mors

t • ,• >2 (Verigin

Fig. 1. Bow shock crossing locations during 14 passes by M•s 2, 3, and 5 (O. L. Vaisberg and V. N. Smi•ov, private communication, 1979) have been plotted in abe•ated ecliptic coordinates with solid line segments spanning the intervals in the obse•ations over which the entire transition took place. A best fit 2nd order curve through the M•s bow wave encounters is shown as a solid line and comp•ed with the mean shock surfaces from Venera 9 and 10 and PVO at solar minimum and maximum, respectively. Note the apparent solar cycle v•ation in the Cytherean bow wave position.

and 10 orbiters [Verigin et al., 1978] and Pioneer Venus [Slavin et al., 1980] in 1975-6 and 1978-9, respectively. In the past, great emphasis has been placed on the relative locations of the Mars and Venus bow waves on the premise that if the Martian bow shock is significantly more distant, then the cause is an intrinsic magnetic field standing-off the solar wind at altitudes higher than are possible at 'field-free" planets such as Venus. Accordingly, prior to the Pioneer Venus mission the large difference in altitude between the Martian bow wave defined by the Mars orbiters and that of Venus from the Venera satellites (the dashed line in Figure 1) was interpreted as strong evidence for the solar wind being deflected about Mars predominantly through an interaction with magnetic fields intrinsic to the planet [Verigin et al., 1978; Breus and Gringauz, 1980]. However, as also shown in the first figure, PVO has found the Venus bow wave to be more distant than was reported by the earlier Venera missions [Slavin et al., 1979b, 1980]. This discovery has been considered by Slavin et al. as evidence for a solar cycle variation in Venus-solar wind interaction with the distance to the

shock observed by PVO being comparable to that of Mars near solar maximum, but much smaller at solar minimum as seen by V9 and V 10 despite the similar solar wind Mach numbers during the two periods.

Accordingly, simple comparisons between the Venus and Mars bow waves no longer appear adequate for determining the type of obstacle Mars presents to the solar wind. For this reason in Figure 2 a gasdynamic model of the solar wind flow about Mars has been constructed by taking advantage of the one-to-one relationship between shock shape and obstacle shape for a given set of upstream conditions [Fan Dyke, 1958]. A best fit shock was selected from among the many published hypersonic gasdynamic solutions for flow about obstacles of various shapes that have recently become available [Spreiter and $tahara, 1980a, b; $tahara et al., 1980]. This solution using 7 = 2 andM = 7.4 has been plotted as plus signs in Figure 2 with the corresponding obstacle displayed as a dashed curve. The choice of 7 = 2, the adiabatic exponent, is consistent with the earth observations [e.g., Fairfield, 1971; Zhuang and

Russell, 1980], while a sonic Mach number of 7.4 is about 20% lower than the expected mean at 1.5 AU. However, hypersonic flow in the near field (i.e., x'• > -1RMs ) is relatively insensitive to changing Mach number beyond M = 5 so that the use ofM = 7.4, because of its availability [$preiter and $tahara, 1980b], is not a significant source of error in the model. Hence, in response to the first question posed by the introduction the model in Figure 2 indicates that the average effective Martian obstacle to the solar wind ranges from 510 km (+20%) at sun-planet-satellite angles of 0 ø to nearly 1000 km at 90 ø. This modeling of the Mars bow wave with a complete gasdynamic flow field represents an important advance over many of the earlier studies [e.g. ,Russell, 1977] which have examined shock crossings to determine the subsolar height of the bow wave and then assumed the Mars magnetosheath to have the same relative thickness as that of the earth in order to infer obstacle

altitude. In fact, the width of the magnetosheath is a function of obstacle shape with thicker magnetosheaths being necessary when redirecting flow about blunter obstacles [Fan Dyke, 1958; $preiter and $tahara, 1980b ]. The model in Figure 2 shows an obstacle that is slightly less blunt than the terrestrial analogue and indicates that studies assuming the same relative magnetosheath thickness as at earth underestimate obstacle altitude by about 20%. This variation in obstacle shape between earth and Mars could be due to many causes including differences in their intrinsic field strength and multipolar composition, the distribution of plasma behind the obstacle, and internal current systems. By comparison with the other works that have used gasdynamic flow calculations [Dryer and Heckman, 1967; $preiter and Rizzi, 1972; Gringauz et al., 1973; Bogdanov and Vaisberg, 1975; Gringauz et al., 1976; Dolginov et al., 1976a,b ] this study obtains a better representation of the observations due largely to the greater number of complete flow models that have become available through the efforts of Spreiter and

GASDYNAMIC MODEL OF SOLAR[ WIND FLOW ABOUT MARS •13

Mean Shock GD Best F•t BS + and Obstacle---

MP Locahons ? ß

CP Locations ? ß -

,

15 0'5 0 -0.5 - t X' (RMs)

Fig. 2. The mean Mars shock surface from the first figure is modeled with a theoretical gasdynamic flow calculation (shown with plus signs marking the shock and a dashed curve indicating the corresponding obstacle) selected from the literature to match the shape of the bow wave (i.e., 7 = 2, M = 7.4, and a shape parameter of H/R o = 0.03; Spreiter and Stahara [1980b]). As shown the fit to the mean empirical shock is quite good. Also displayed are two purported entries and exits from the Martian magnetosphere by Mars 2 and 3 indicated by connected triangles. The three occassions on which the top of the ionosphere may have been observed are then shown with squares. Hence, the high altitude triangles and much lower squares bracket the inferred mean obstacle surface and serve to set upper and lower limits on the observed height of the magnetopause.

SLAVIN AND HOLZER: MARS SOLAR WIND INTERACTION REVISTED 10,287

co-workers and the use of shock crossings from only the adequately sampled portion of the bow wave sunward ofx' = - 1RMs (seeSlavin and Holzer [1981 ] for a discussion of spatial biases in the downstream observations associated with the Mars 2, 3, and 5 trajecto- ries).

BALANCING SOLAR WIND DYNAMIC PRESSURE

The Mariner 4 and 5 fly-by missions to Mars and Venus, respectively, provided the first detailed examination of their ionospheres by means of radio occultation experiments [Kliore et al., 1965;

Mars to support the solar wind pressure were' also raised by Gringauz et al. [1974], Whitten and Colin [1974], Wallis [1975], Gringauz [1976], Breus and Gringauz [1980], and Gringauz [1980]. However, only Gringauz, Wallis, and Breus have used the lack of ionopause signatures and weaknesses of the ionosphere as evidence for the existence of a significant intrinsic magnetic field.

As a result of the retarding potential analyzer measurements during Viking 1 and 2's descent, the ability of the Martian ionosphere to stand off the solar wind without aid from intrinsic magnetic fields has again been called into question because the maximum ion pressure observed in the ionosphere was found to be only

1967; Fjeldbo and Eshlernan, 1968, 1969]. In the case of Venus a •2 x 10 -9 dynes/cm 2 [Hanson et al., 1977]. For the theoretically dense atmosphere was confirmed (see reviews by Whitten and Colin [ 1974] and Schunk and Nagy [ 1980]) with the ionosphere standing off the solar wind through the formation of an ionopause at an altitude, on that occasion, near 500 km at a solar zenith angle of 33 ø [Fjeldbo and Eshleman, 1969]. More recently the Pioneer Venus orbiter has unequivocally established the ionospheric nature of the Cytherean solar wind interaction by means of in situ observations such as the electron density profile in Figure 3 showing a typical ionopause crossing [Theis et al., 1980]. However, at Mars the atmosphere is far thinner than at Venus and earth with a corre- spondingly weak ionosphere [Kliore et al., 1965]. No ionopause was detected by Mariner 4 [Fjeldbo and Eshleman, 1969] even though Spreiter et al. [1970] calculated that for average solar wind dynamic pressure conditions the Martian ionosphere might be able

predicted ratio of electron to ion temperature near these altitudes (i.e.,•-,250-300km) of 1.5 [Rohrbaugh et al., 1979] a maximum total plasma pressure of only about 5 x 10 -9 dynes/cm 2 is implied which is a factor of 2 below the expected long term average solar wind dynamic pressure at 1.5 AU of•-,1 x 10 -8 dynes/cm 2 [e.g., Slavin and Holzer, 1981 ]. .•

This lack of sufficient ionospheric pressure was demonstrated by Intriligator and Smith [ 1979] and Breus and Gringauz [ 1980] who then attributed the ability of Mars to deflect the solar wind to the existence of an intrinsic magnetic field. However, they did not examine the possible role of inductive ionospheric magnetic fields. In fact, the Pioneer Venus mission has observed ionospheric magnetic fields of interplanetary origin (i.e., Venus has been found to have no measureable intrinsic field, Dolginov et al. [1969] and

to stand-off the solar wind at heights of 155-175 km over the Russell et al., 1980a] with intensities of order 10-102 nT on about subsolar point which is just above the peak ionization layer near 30% of the orbits with solar zenith angles less than 50 ø [Elphic et 120-150 km [e.g., Kliore et al., 1973]. Similar doubts about the al., 1980b; Luhrnann et al., 1980]. However, while the Cytherean presence of sufficient plasma pressure within the ionosphere of ionosphere can sometimes become magnetized through the 'cap-

Mars Venus h (kin) h (kin)

•Pioneer Venus OETP

it 177 SVS =58 ø _37o 300 -

- •4

•50 - f57

psis)

200

f50

1oo I I I I lO '1 1o 2 lO 3 10 4 fo 5 10 6

Ne (c m -$) Fig. 3. Electron density is plotted as a function of spacecraft height across an ionopause at Venus from the PVO observations [Theis et al., 1980] and compared with the Viking 2 measurements [Hanson et al., 1977] showing a similar appearing boundary at Mars. However, if it were displayed, the earth's plasmapause would also look much like these boundaries in in terms of showing a sudden density decrease. The more important consideration is the weakness of the Martian ionosphere with nearly a two order of magnitude of difference between the plasma densities observed just beneath the Cythe- rean and Martian boundaries.

ture' of interplanetary field lines, statistical studies of pressure balance across the ionopause which neglect these induced fields have obtained good results [Elphic et al., 1980a;Brace et al., 1980; Slavin et al., 1980]. These findings suggest that, on average, these magnetic fields make only a minor contribution to the total ionospheric pressure, Ptotal, standing off the solar wind. Still, for the purpose of estimating their greatest possible contribution to support- ing the solar wind at Mars we will assume that inductive magnetic fields are always present and can be so large as to produce equality between particle and field energy densities in the upper ionosphere so that

13 -- 8,rnk(T i + Te)/B 2 • 1 (1)

Such a limit is reasonable in that the primary sources of field compression are the solar wind dynamic and ionospheric thermal pressures which are in equilibrium on average. Expressed in another form

B % (8rrnk(T i + Te)) • (2)

we are suggesting that an upper limit on the magnetic flux density that may be acquired by an ionosphere in such interactions at a planet, or even a comet, is determined by the ionopause plasma pressure. The condition in (2), which is generally satisfied in the upper ionosphere of Venus [Elphic et al., 1980a,b], then sets a range on the total ionospheric pressure available to balance solar wind dynamic pressure with

nk(T i + T•) <• mtotal ?• 2nk(ri + T•) (3)

Thus it appears that during the epoch of the Viking observations, solar minimum, the greatest solar wind pressure that the ionosphere


can support assuming maximal induced magnetic fields, is'--2 x 5 x 10 -9 = 1 x 10 -8 dynes/cm2; the meanPsw near 1.5 AU. Hence after including induced magnetic fields we find that the ionosphere alone might be able to divert the solar wind, at most, the half of the time when dynamic pressure is below average. However, the higher solar EUV fluxes at other phases of the solar cycle are expected to make the Viking epoch (i.e., solar minimum) a 'worst case' analysis for standing off the solar wind by a factor of"-'4 with respect to the available ionospheric pressure near solar maximum due to the larger ionospheric plasma temperatures and densities present during the rest of the cycle [Whitten and Colin, 1974; Lindal et al., 1979; Wolff et al., 1979]. We show below that under these conditions the largest mean subsolar height at which an ionopause could form in the absence of a significant intrinsic magnetic field is about 350 km.

While the availability of enough ionospheric back pressure is a necessary requirement for a Venus-type solar wind interaction, it is not a sufficient condition. That there may be enough plasma pressure somewhere in the ionosphere does not mean that this is the case at the actual mean obstacle height. To this end we note that the Viking 1 descent observations showed a relatively constant plasma density scale height between---150 km and the first measurements just above 360km with no indication of the top of the ionosphere having been seen [Hanson et al., 1977]. If plasma temperature is assumed to vary slowly [Chen et al., 1978; Johnson, 1978; Rohrbaugh et al., 1979] at altitudes above•350 km while retaining the observed 29 km density scale height, then the plasma pressure at the obstacle altitudes in figure 2 may be given approximately by

He ++ assumed in solar wind). The effective dipole moment obtained is 1.4 x 1022 G-cm 3, which corresponds to a surface flux density of 36 nT at the magnetic equator and 72 nT at the poles. It must be remembered that this result is subject to the many assumptions implicit in (5) that cannot be tested with the available observations. If the dipole moment is tilted away from the spin axis as has been suggested in other studies [Dolginov et al., 1973; Smirnov et al., 1978 ], then the moment computed above is an overestimate by up to •10% depending upon the size of the tilt. Equation (4) also assumes that all of the magnetospheric current systems are similar in relative magnitude to those of the earth. Other studies have already pointed out that this is probably not the situation due to the highly conducting Martian ionosphere as a magnetospheric boundary condition [Rassbach et al., 1974]. In particular, if the Mars magnetotail current system is relatively stronger than at the earth as may be the case [Intriligator and Smith, 1979], then the moment calculated from (5) will be an underestimate by an uncertain amount probably also up to--,10%. Finally, there are also the uncertainties in the gasdynamic flow model and the shock observations themselves, but these are expected to be smaller than those associated with the assumptions discussed above. For all of these reasons we estimate that the Martian magnetic moment may only be determined at this time with a probable error of approximately 40%. Thus we arrive at M = 1.4(_+0.6) x 1022G-cm2 for the effective dipole moment of this planet.

In Table 1 we have compiled a list of the previously inferred Martian magnetic moments and the methods used to determine them. The initial reports of the Mariner 4 experimenters [Van Allen

P (dynes/cm 2) = 1 x 10 -8 e -(h - 200 km)/29km (for h > 350km) (4) et al., 1965; O'Gallagher and Simpson, 1965; Smith et al., 1965'

where an upper ionospheric T i = 3000øK and Te/T i - 1.3 (i.e., for h • 350 km; P is overestimated by (4) below this level) is taken from the ionospheric model ofRohrbaugh et al. [ 1979]. Hence, from this expression we see that even assuming thatPtota I is twice this amount due to the presence of inductive magnetic fields, the available pressure at the 500 km obstacle height in Figure 2 is only about 6 x 10 -13 dynes/cm 2 which is 4-1/2 orders of magnitude below the

Lazarus et al., 1967] indicated that with a closest approach distance of 3.9RMs the bow shock was not crossed. Upper limits of 2.4 - 8 x 10 22 G-cm 3 were then inferred from this negative result. However, a later interpretation of the magnetic field data showed a pair of bow wave crossings which were then used by Smith [1967, 1969] to estimate a moment of 0.8 x 1022 G-cm 3 by scaling the terrestrial analogue to the Mars observations. In addition,Dryer and Heckman [1967] utilized an earth magnetopause shaped obstacle and

mean Psw. In the case of solar maximum condition a 1-1/2 orders of gasdynamic model to obtain an intrinsic moment of 1.7 x 10 22 G magnitude pressure deficit still results when ionospheric density, cm 3 from the Mariner 4 observations. However, Spreiter andRizzi temperature, and scale height are all doubled in (4). Thus even at solar maximum (4) when set equal to the average dynamic pressure, --- 1 x 10 -8 dynes/cm 2, shows that the average subsolar height of a Martian ionopause near the subsolar point would be no more than 350 km as contrasted with the 510 km required by the bow shock

[ 1972] reduced that value to an upper limit by showing that theoretical ionospheric obstacles could also be used to account for the position of the Mariner 4 shock crossings. The main difficulty in the interpretation of the Mariner 4 shock observations is that they occurred far downstream (i.e., - 12RMs < x' < - 2RMs) so that their

observations. Accordingly, we find an answer to the first part of position is more dependent upon the ambient solar wind flow question 2 from the introduction that the Viking observations of the direction and Mach number than upon obstacle height. The Soviet weakness of the Martian ionosphere do indicate the need for an intrinsic field to create the obstacle to the solar wind inferred in from

bow shock location.

With this knowledge and the subsolar obstacle altitude from the preceding bow wave analysis an effective intrinsic magnetic moment may be calculated from the classical pressure balance condition for a magnetic dipole in the solar wind giving rise to a terrestrial type magnetosphere [e.g., Spreiter et al., 1966]

M = (2rr Psw R• 6 k/Je) '/' (5)

where k is half the obstacle drag coefficient taken to be the earth value of 0.88,f is half the enhancement factor of the subsolar point magnetic flux density also assumed to have the terrestrial magnitude of 1.22, R s is the subsolar stand-off distance of 510 + 3390 = 3900 km, and the mean solar wind dynamic pressure for the shock crossings in Figure 1 was•l.5 x 10 -8 dynes/cm 2 (Dolginov [1976]; 4%

analyses based upon their lowest altitude and near wake orbiter measurements are mostly supportive of a small intrinsic field magnetosphere, 1.2 - 2.6 x 1022 G-cm 3, and are discussed in a later section as are the studies of Russell. Intriligator and Smith [ 1979] arrived at a moment of 0.8 x 1022 G-cm 2 from their estimate of the

magnetic field magnitude necessary to make up the deficit between the solar wind and Viking epoch ionospheric pressures. Thus our moment of 1.4(_+0.6) x 1022 G-cm 3 is an intermediate value with respect to the previous studies. Given our estimated uncertainties, only the most extreme moments of Russell and Dolginov appear inconsistent with this analysis.

VERTICAL EXTENT OF THE MARTIAN IONOSPHERE

Another aspect of the Mars environment with a great bearing on the solar wind interaction problem is the vertical extent of the ionosphere and the nature of its high altitude termination. In the


TABLE 1. Summary of Inferred Martian Magnetic Moments

Spacecraft Mms ( 10 22 G-cm 3) Study Basis

Mariner 4

Mars 2 and 3

Mars 2, 3, and 5

•<8

•<8

Van Allen et al. [1965]

O' Gallagher and Simpson [ 1965]

2.4 Smith et al. [1965]

•< 8 Lazarus et al. [1967] 0.8 Smith [1967, 1969]

1.7 Dryer and Heckman [1967]

1.7 Spreiter and Rizzi [1972]

2.4 Dolginov et al. [1972, 1973]

2.4 Gringauz et al. [1974]

2.55 +_ 0.36 Dolginov et al. [1975, 1976a,b] and Dolginov [ 1976]

2.0 Gringauz et al. [1977]

•< 0.9 Russell [1977]

•< 0.2 Russell [1978a, b,c]

1.7 - 2.5 Dolginov [1978a, b,c]

1.2 Stairnov et al. [ 1981 ]

Lack of interaction signature (particles)

Lack of interaction signature nature (particles)

Lack of interaction signature (magnetic field)

Lack of interaction signature (plasma) Modeling bow shock position.

Gasdynamic modeling of bow shock position


Modeling in situ magnetic field (dayside)


Modeling in situ magnetic field (dayside and nightside) and bow shock/magnetopause position


Modeling bow shock position

Lack of magnetospheric signatures

Modeling in situ magnetic field (dayside and nightside) and bow shock/magnetopause position

Modeling in situ magnetic field (dayside and nightside)

O. 8 Intriligator and Smith [ 1979] Pressure balance across solar wind-obstacle interface

1.4 _ 0.6 This study Gasdynamic modeling of bow shock position

preceding sections strong evidence for Mars possessing a small ductivity be such as to permit the flow of shielding currents to intrinsic field magnetosphere has been presented. This being the prevent any large influx of plasma. The marginal ability of the case, the ionospheric measurements should not show the presence Martian ionosphere to balance solar wind dynamic pressure under of a Venus-type ionopause at Mars. However, as displayed in typical conditions has already been discussed in the preceding Figure 3, one of the Viking landers did pass through a relatively section. It is emphasized again in Figure 3 by comparison with the sharp boundary region marking the top of the ionosphere [Hanson et Pioneer Venus electron density altitude profile across the Venus al., 1977]. The Viking 2 observations of the ionosphere terminating iønopause •heis et al., 1980]. As shown, the Cytherean ionosphere abruptly near 300 km do not appear to correspond to the average is significantly stronger than that of Mars even remembering the conditions portrayed by the numerous radio occultation experi- solar maximum (minimum) nature of the PVO (Viking) observa- ments. Instead, it was the Viking 1 measurements made over an tions. The plasma density behind the Venus ionopatise is nearly two altitude range of 140 to 360 km •r-lanson et al., 1977], which orders of magnitude larger than in the case of the purported Martian showed a great similarity to the radio science results [e.g., Whitten ionopause. Elphic et al. [1980b ] have already demonstrated that the and Colin, 1974]. This lander found a nearly constant topside scale pressure balance condition is satisfied across the PVO orbit 177 height of 29 km with no ionopause signatures even though densities ionopause displayed. For the moment, let us assume that a Mars as low as 102 cm -3 were measured at the highest altitudes [Hanson ionopause was in fact observed by Viking 2 as suggested by Cloutier et al., 1977]. In constrast the Viking 2 measurements in Figure 3 and Daniell. The magnetic flux density required to balance external have been modeled by Cloutier and Daniell [1979] in terms of a pressure at the time and place of these measurements may be purely ionospheric interaction with an ionopause at 300 km below calculated directly which the shocked solar wind does not penetrate. However, their B2/8rr = 0.88PswcOS2(SMS) - nk(T i + Te) (6) model does not require the ionospheric plasma pressure to equal the external pressure of the solar wind, but only that ionospheric con- AlthOugh Psw was not measured by Viking at 1.5 AU, we have


inferred a value of 1 x 10 -8 dynes/cm 2 from the 1 AU observations (King [ 1977]; 4% H ++ assumed) after introducing appropriate time lags as has been done in the past by other studies [e.g., Dolginov, 1978a ]. From the Viking 2 results [Hanson et al., 1977] we have n '•1 x 103 cm -3 andT/•2.3 x 103øK near the 'ionopause.' Given an assumed Te/T i - 1.5 [Rohrbaugh et al., 1979], the flux density obtained from (5) is 25 nT with an estimated uncertainty of ,---40% due to predominantly to the need to infer Psw from earth orbit measurements and the lack of in situ ionospheric electron temperatures. This value is similar to those oflntrilligator and Smith [ 1980] and Breus and Gringauz [1980] who assumed average solar wind conditions without examining the 1 AU observations. We conclude from this result that Viking 2 did not observe a Venus-type ionopause at Mars on two grounds. First, in the absence of intrinsic magnetic fields an inductive field of 25 nT would produce/g •'0.3 in the upper ionosphere of Mars. As noted with respect to (1) and (2), a simple unipolar generator cannot self-consistently produce a steady-state magnetic field with Maxwell stresses exceeding the pressures it can support as would be the case for a highly ionized ionosphere with/g < 1. This requirement appears to be generally fulfilled by the Venus ionosphere [Elphic et al., 1980b ]. Secondly, if the 25 nT magnetic field were intrinsic to the planet, then the effective magnetic dipole moment, assuming a factor of 2 enhancement by Chapman-Ferraro boundary currents, would only be 6(_+ 2) x 1021 G-cm 3, which is less than half the amount necessary to account for the average height of the obstacle in Figure 2.

Having given reasons why the V2 measurements do not appear consistent with the crossing of an ionopause, there remain two classes of interpretations for the observations. The first type still retains the view that Viking 2 passed through the solar wind- obstacle interface near an altitude of 300 km, while the second

proposes that the boundary encountered was within the magnetosphere as is, for example, the case with the terrestrial plasmapause. From the pressure balance discussion and (6) it can be seen that the V2 event cannot be reconciled with a simple compression of the Martian magnetopause down to a height of 300 km unless the magnetic moment determined earlier is an overestimate by a factor of 2 or the value of P•w for this event inferred from 1 AU an underestimate by a factor of 4, neither of which is probable. Further, the errors would actually have to be somewhat larger than these amounts because ionospheric induction currents opposing the compression would be driven in much the same way as planetary induction currents at Mercury are thought to aid in standing-off the solar wind during sudden dynamic pressure enhancements [Hood and Schubert, 1979; Suess and Goldstein, 1979]. However, a more plausible method of lowering the solar wind-obstacle interface down to the 300 km height of the V2 descent observations is that a 'contraction', as opposed to a compression, of the dayside magnetosphere took place such as are observed at the earth [Holzer and Slavin, 1978]. In these events magnetic reconnection transfer magnetic flux from the dayside magnetosphere to the tail (i.e., the magnetotail currents become stronger) thereby decreasing the vol- ume and vertical extent of the dayside magnetosphere without requiring any increase in dynamic pressure. In fact, Rassbach et al. [1974] have already suggested that the entire forward magnetosphere might be 'eroded' away by reconnection so as to expose portions of the ionosphere to the shocked solar wind of the magnetosheath. Such events are expected to be facilitated by the high

should be noted that the rate and total amount of dayside reconnection is expected to grow as the intensity of the anti-parallel magnetic fields increase (e.g., Slavin and Holzer [1979] and references therein). Thus large tilts of the intrinsic magnetic field away from the Mars spin axis (i.e., toward the average plane of the interplanetary magnetic field) as proposed by Dolginov et al. [ 1973] andSmirnov et al. [978] will enhance the effects of erosion over half of the daily rotation period. Accordingly, we propose that the Viking 2 descent observations may have been taken during a large erosion event that reduced the magnetopause height down to that of the sharp upper bound on the ionosphere shown in Figure 3. Such an occurrence would also be consistent with the slightly compressed appearance of the V2 measurements between 250 and 300 km relative to those of V1. The Viking 2 values of ion density and temperature exceed those observed by the earlier lander in this altitude range as would be the case if the Martian ionosphere were bearing a portion of the solar wind pressure during the V2 descent as opposed to none during the V1 observations. Unfortunately, the lack of suitable particles and fields experiments on the Viking spacecraft makes it impossible to prove or disprove this hypothesis.

Alternatively, the boundary detected by V2 could lie within the Martian magnetosphere and correspond to the greatest height to which the ionospheric plasma could reach on that occasion without being lost to magnetospheric convection. Owing to the large size of Mars in comparison to its magnetosphere, the observed boundary could not have been a terrestrial type plasma-pause as the level at which the corotational and cross-magnetospheric electric fields become equal would lie beneath the surface of the planet. Noting this fact, Bauer and Hartle [1973] suggested that the Martian ionosphere could terminate in a 'chemopause' at whose altitude the rate of ion loss due to magnetospheric convection would equal ion production from the ambient neutral population. Assuming a Mar- tian dipole moment of 2.4 x 10 22 G-cm 3 and a cross-magnetosphere electric field of a kilovolt per planetary radius, they estimated that the chemopause would form at the height where ion density falls below about 5000 cm -3. However, if we reduce the assumed magnetic moment down to the 1.4 x 1022 G-cm3 found by this study and use a more realistic electric field of 440 V/Rp derived from scaling a moderately disturbed terrestrial value of 800 V/Rp (i.e. ,Kp = 3 from Kivelson [ 1976]) by the ratio of the IMF magnitudes at 1 and 1.5 AU (e.g. ,Intriligator andSmith [1979] and references therein), then the Bauer-Hartle critical density for the formation of a chemopause falls to---1300 cm -3 which is near the •,1000 cm -3 seen by Viking 2 in Figure 3. Hence, another interpretation consistent with the existence of a small Martian magnetosphere is that the V2 descent took place during an interval of magnetospheric disturbance (i.e., an areomagnetic storm or substorm) accompanied by cross magnetospheric electric fields of--,500 V/Rp. The Viking 1 descent, which failed to see the top of the ionosphere even though densities down below 100 cm -3 near an altitude of 360 km were observed, could then have occurred during a quiescent period with the magnetopause altitude above 500 km and convection electric field

strengths of order 10 V/R e. Such fields are close in strength to those estimated by Rassbach et al. [1974] andHill et al. [1976] while the 500 V/R e value is significantly larger. The order of magnitude range in cross-magnetospheric electric field strength needed to span the V1 and V2 conditions is large in comparison with the factor of'--,4 change in the field at the earth asKp varies from 0 to 5 (see review by

electrical conductivity of the Martian ionosphere relative to that of Kivelson [1976]). However, our knowledge of conditions within the earth due to the lower magnetic field strength at Mars. This will in Martian magnetosphere is poor. It may be that the proposed large tilt turn limit the rate of convection within the Martian magnetosphere of the magnetic field, a situation with which we have no past through 'line tying' [e.g., Hill et al., 1976] and slow and rate of experience, could produce great changes in the electric fields return of magnetic flux to the dayside magnetosphere. In addition, it through the large role of magnetic reconnection in determining

SLAVIN AND HOLZER: MARS SOLAR WIND INTERACTION REVISTED 10,29 1

magnetospheric conditions. Again the Viking instrumentation was not such that any definitive resolution can be reached concerning the very different upper ionospheres observed by Viking 1 and 2 during their descents to the surface.

For this reason we have also conducted an examination of the

published radio occultation measurements at Mars by the American Mariner and Soviet Mars missions. These experiments have sup- plied reliable electron density altitude profiles between the ionization peak of 1 - 2 x 105 cm -3 near 120-150 km in altitude and the level in the upper ionosphere where density falls below,,. 1 - 3 x 10 3 cm -3 [e.g., Eshleman, 1970]. In the absence of significant intrinsic magnetic fields it can be shown that the radio science observations should be able to detect the formation of an ionopause at least on some occasions. For example, from (6) the typical density just below such a boundary near an intermediate solar zenith angle of 45 ø should be more than--.3500 cm -3 for T i •,• 3000øK, Te/T i • 2 and Psw '•' 1 x 10 -8 dynes/cm 2. Consistent with the existence of a small intrinsic field Martian magnetosphere no ionopause signatures are apparent in the radio occultation profiles derived from Mariner 4 (MN4; Fjeldbo and Eshleman [1968]), Mariners 6 & 7 (MN6, MN7; Hogan et al., [1972]), Mariner 9 (MN9; Kliore et al. [1972]), and Mars 4 (M4; Vasil'ev et al. [ 1975]). In each case a relatively constant topside scale height was observed up to between 250 and 400 km where the density gradually moved below the experimental threshold. These lower limits on the vertical extent of the Martian ionosphere along with that determined by Viking l's descent observations have been plotted against solar zenith angle in Figure 4. For purposes of comparison, the Venus ionopause surface at solar minimum and maximum from the Venera 9 & 10 radio occultation studies [Ivanov-Kholodny et al., 1979] and the Pioneer Venus in situ measurements [Brace et al., 1980], respectively, are also displayed along with the Mars obstacle surface from Figure 2. The lower limits on vertical extent, excepting V1, all correspond to similar density levels (i.e., thresholds of ,-• 700- 3000 cm -3 in a•--40 km scale height ionosphere) so that any strong variation in ionospheric density with solar zenith angle such as are present in the Cytherean ionopause traces [Theis et al., 1980] would be observable in Figure 4. However, all that is evident is a dependence on solar cycle phase with MN4 lower limit at solar minimum in 1965 near 250 km and the solar maximum

• '1000- VERTICAL EXTENT OF THE IONOSPHERES OF..'•

-,-. - MARS AND VENUS .,.,.,.,.,• • 800- .." - So,ar M urn Venus lonop.... ....

o .

•> 200 - Solar M•ntmum Venus Zonopause 0(•0 • I • I , I • I I 20'

Solar Zenith Angle

Fig. 4. Lower limits on the vertical extent of the Mars ionosphere along with three instances in which the top of the ionosphere may have been detected are displayed as a function of solar zenith angle. For purposes of comparison both the solar maximum and minimum Venus ionopause positions [Ivanov-Kholodnyi et al., 1979; Brace et al., 1980] as well as the solar wind obstacle at Mars have also been displayed. Relative to the ionopause surfaces, no large solar zenith angle dependence is seen in the Martian upper

MN6 and MN7 points at almost 400 km in altitude. In only two cases, Mars 2 [Vyshlov et al., 1975] and Mars 6 [Vasil'ev et al., 1975] were any signatures at all similar to that of V. iking 2 in Figure 3 found. As with V2, the ionospheric density beneath each of the boundaries appeared too low (i.e., < 3500 cm -3) to be consistent with pressure balance across in ionopause although in neither of these two cases were solar wind dynamic pressure measurements available. The M2, M6, and V2 topside ionospheric boundaries have been plotted in Figures 2 and 4 as squares. We have suggested that these points may correspond to the lowest altitude excursions of the top of the ionosphere and/or the magnetopause. In either event, Figures 2 and 4 show in response to the final question of the introduction that the Martian ionosphere appears to extend up to altitudes between 300 km and the height of the dayside obstacle to the solar wind determined by this study without the formation of an ionopause. Thus we have found that the ionospheric observations constitute an independent, and often under utilized, source of information which supports the existence of a small Martian magnetosphere generally shielding the upper ionosphere from the solar wind.

DISCUSSION

As was reported in the preceding sections, our modeling of flow about Mars and examination of the Martian ionospheric measurements have produced results that are inconsistent with a Venus-or cometary-type solar wind interaction for this planet. The effective magnetic moment inferred is sufficiently large to produce the small magnetospheric cavity needed to account for both the location of the bow wave and the general lack of ionopause signatures in the radio occultation observations. Accordingly, we have compiled a list of the principal arguments cited in the literature for and against the existence of an intrinsic field magnetosphere at Mars and incorpo- rated into it our own findings and comments. However, it must be remembered that the shielding afforded the Martian ionosphere and neutral atmosphere by the modest intrinsic fields proposed here, and elsewhere, will certainly be greatly inferior to that found at the earth. Under certain circumstances, such as high solar wind dynamic pressure or significant dayside reconnection, it appears inevi- table that the ionosphere and atmosphere will be exposed to the solar wind. It is, rather, the conclusion of this study that the Martian ionosphere typically does not play a direct role i.n standing-off the solar wind due to the magnitude of the planetary magnetic field.

PRO-Small Intrinsic Field Magnetosphere

A. Bow shock position. 1. Verigin et al: [1978] compared the relative locations of the Mars and Venus bow shocks as observed by the Mars and Venera orbiters. They. attributed the-more distant location of the Martian bow wave to the.. existence of an intrinsic field standing off the solar_wind at heights,•greater than those associated with an ionospheric obstacle such as at:•Venus. The Pioneer Venus shock observations [Slavin-et al., 1979a, 1980] displayed in Figure 1 invalidate this argument by demonstrating that the Cytherean bow wave can indeed reside at the•same relative heights as does the shock at Mars under:normal solar wind conditions.

2. Both empirical [e.g.,Russell, 1977] and gasdynamic [e.g., ionosphere. In addition, it should be noted that heights where the Dryer and Heckman, 1967] modeling of shock position have been ionospheric density falls below the order 10 3 cm -3 radio occultation experi- employed to infer the height of the solar wind obstacle at Mars. The ment threshold are still typically 3 scale heights below the altitude of the previously determined stagnation point altitudes range from near solar wind-obstacle interface suggesting that the ionosphere usually plays no significant direct role in the deflection of incident solar wind plasma about 400 km [e.g., Bogdanov and Vaisberg, 1975] to well over 1500 km the planet. [e.g. ,Dolginov, 1978a ]. In this study two major objections to these


previous treatments have been removed: the lack of data selection to -•2100 km in April, 1972; Dolginov [ 1976]. In agreement with the avoid biasing the shock surface by the poor downstream spatial sampling of Mariner 4 and Mars 2, 3, and 5 [see Slavin and Holzer, 1981 ], and the a priori assumption of obstacle/shock shape in order to know the width of the magnetosheath. The mean subsolar obstacle height obtained is 510 (_+20%) km. The corresponding effective dipole magnetic moment is 1.4 (_+0.6) G-cm, 3, which places it nearly half-way in between the extremes [Dolginov et al., 1974; Russell 1978b] of Table 1.

3. Examinations of the variability in the location of the Mars bow shock (Breus and Gringauz [1980], Dolginov [1978a], Gringauz [1976], and Bogdanov and Vaisberg [1975] and references therein) have shown the absolute magnitude of the changes to be similar to those observed at the earth and anticorrelated with

upstream dynamic pressure. This apparent implied similarity in obstacle compressibility between earth and Mars then argues that an intrinsic magnetic field is also responsible for deflecting the solar wind in the Martian case. Vaisberg et al. [1976] have objected to this reasoning on the grounds that the change in shock position with pressure seen by Mars 5 is much smaller than would be expected for a simple dipole magnetosphere. However, we note that such an effect might in fact be expected for a small Martian magnetosphere for two reasons. First, the highly conducting ionosphere at the feet of the magnetic field lines may be able to effectively 'stiffen' them through induction currents just as has been suggested for the modest magnetosphere of Mercury [Suess and Goldstein, 1979; Hood and Schubert, 1979]. Second the Martian magnetosphere may contain more plasma at magnetopause altitudes than does the terrestrial analogue due to the large relative height of the ionosphere which is expected to be continuously losing particles to magnetospheric convection [see Bauer and Hartle, 1973]. Finally Bogdanov [ 1978, 1981 ] has suggested on the basis of Mars 5 observations that the occasional shadowing of the Martian magnetosphere by an outgas- sing moon, Deimos, results in a larger dayside obstacle and more distant bow wave.

B. Mars 2, 3, and 5 interaction region observations. 1. The only magnetic moment determinations for Mars arrived at through

small magnetospheric cavity from this study displayed in Figure 2, no enhanced magnetic fields were observed near periapsis by Mars 3 during periods of average (i.e.,---1 x 10 -8 dynes/cm2; see Slavin and Holzer [ 1981 ]) or greater than average dynamic pressure [Dol- ginov et al., 1973; Dolginov, 1978b].

2. Mars 2 also measured magnetic fields approaching 30 nT near dayside periapsis on a number of occasions such as 1/8/72, 2/22/72, 2/24/72, 4/6/72, 4/18/72, and 5/12/72 [Dolginov, 1978b ] when the magnetometer was turned on during the pericenter passage (note: the particles and fields experiments onboard Mars 2 and 3 were off during more than half of the low altitude passes; Gringauz [ 1976]). However, no direct determinations of magnetic moment have made from the measurements due to a lack of com-

plete attitude information for Mars 2 with which to construct the vector field and also the resulting uncertainties regarding the magnetometer offsets [Dolginov et al., 1976a]. On the average, solar wind dynamic pressure appears to have been below average during these events [Dolginov, 1978a ] and consistent with the small magnetospheric cavity in Figure 2 which corresponds to a somewhat enhanced pressure of 1.5 x 10 -8 dynes/cm 2. In the case of at least the 1/8/72 periapsis, there was also a large drop in plasma ion and electron flux [Dolginov, 1978b; Bogdanov and Vaisberg, 1975] as would be expected upon penetrating a magnetospheric obstacle to the solar wind.

3. The Mars 2, 3, and 5 particles and fields observations at high altitudes on the nightside have been interpreted by Dolginov [ 1976, 1978b], Gringauz et al. [ 1975, 1976], Vaisberg [ 1976], andSmirov et al. [1978] as indicating the presence of a twin lobed magnetotail making up the nightside portion of an intrinsic magnetosphere. However, conclusions reached regarding the type of dayside interaction from tail observations must be tempered by the experience derived from laboratory experiments [Dubinin et al., 1978], as well as at earth and Venus. These investigations have shown that the magnetotails accompanying intrinsic and induced magnetospheres can appear quite similar [e.g.,Russell et al., 1980a,b]. In addition, as with the M2 and M3 dayside 'magnetospheric observations,'

the inversion of magnetometer measurements are those of Dolginov Russell [ 1978b ] has argued that the observational evidence for entry [1978a, 1976 and references therein] using the January 21, 1972, observations of Mars 3. On that occasion, Dolginov et al. [1972, 1973] were able to identify both the inbound and outbound shock crossings along with a relatively sharply bounded region of enhanced magnetic flux density (i.e., • 7-10 times the upstream interplanetary field strength near periapsis; see Figure 2) which they interpreted as being due to an entry into the Martian magnetosphere. The magnetic dipole moments inferred from various subsets of the data yield values from 1.7 to 2.5 x 1022 G-cm 3 with a 'preferred' result of 1.8 x 10 22 [Dolginov, 1978a ]. The dipole tilts obtained are near 20 ø with a polarity opposite to that of the geomagnetic field. These moments are similar to, but slightly greater than the value obtained in this study from gasdynamic modeling of the interaction. However, the agreement would probably be still better if Dolginov had included the effects of the exterior Chapman-Ferraro magnetopause currents that should have enhanced the field in the region Mars 3 probed near the magnetopause by a factor of about 2. We also note that the • 10 nT magnetic flux densities recorded just inside the 'magnetopause' boundaries [Dolginov et al., 1973] are indeed of the correct magnitude to stand off the solar wind pressure atthe time of the 1/21/72 observations [Dolginov, 1978b]. It should also be considered that 1/21/72 appears to have been the only period of low dynamic pressure during which Mars 3 was in a favora- ble observing position over the dayside hemisphere while it still had the low '--1100 km periapsis (Mars 3 periapsis was increased to

into a magnetotail still allows alternate explanations. C. Ionospheric observations. 1. Following the Mariner 4

radio occultation measurements which revealed the tenuous nature

of the Martian atmosphere and ionosphere [Fjeldbo and Eshleman, 1968], a number of studies have expressed doubts concerning the availability of sufficient ionospheric plasma pressure to stand off the solar wind [Spreiter et al. , 1970; Michel, 1971; Whitten and Colin, 1974; Gringauz et al., 1974; Wallis, 1975; Gringauz, 1976, 1980]. Most recently Intriligator and Smith [1979] and Breus and Gringauz [ 1980] have reexamined this question by using the Viking descent observations of ionospheric plasma ions [Hanson et al., 1977] and found insufficient temperatures and densities to balance external solar wind pressure. We have added to these analyses by estimating the effects of enhanced solar EUV radiati•;n at solar maximum and the induced magnetic fields which PVO has observed within the ionosphere of Venus [Elphic et al., 1980b]. While we found that during solar maximum EUV conditions the Martian ionosphere might be strong enough to sometimes form a Venus-type ionopause with a subsolar altitude as high as c•350 km, the ionosphere could not have stood off the solar wind more than 50% of the time during the Viking epoch even when maximal contributions from inductive ionospheric magnetic fields are included. In addition we have also estimated the ionospheric plasma pressure near the subsolar obstacle surface inferred through gasdynamic modeling of bow wave position (pro-A.2). The resulting pressures are 10 -5 and


10-• the mean solar wind dynamic pressure for solar minimum and maximum EUV conditions, respectively. The implication is that/3 must be quite low just inside the obstacle surface with the magnetic field standing off the solar wind.

2. Our examination of published dayside electron density profiles from the radio occulation experiments conducted by the Mars 2, 4, and 6 and Mariner 4, 6, 7, and 9, as well as the Viking 1 and 2 descent measurements, shows no instances where a Venus-type ionopause was observed even though reliable density values were reported down to 1000 cm -3 and at altitudes extending beyond 350 km as shown in Figure 4. In two cases, Mars 2 and 6, sudden decreases in density were observed at high altitudes similar to that found by Viking 2 in Figure 3. However, just as with Viking, the boundary could not correspond to a simple ionopause due to a lack of sufficient plasma pressure (for any reasonable set of assumed ion and electron temperatures) to support the dynamic pressure of the solar wind. This result is consistent with the absence of ionopause signatures in the Viking orbiter radio science measurements reported by Lindal et al. [ 1979].

3. Breus and Gringauz [1980] have considered the variation with solar zenith angle in the radio occultation determined altitude profiles of the ionosphere at both Venus and Mars. In agreement with their conclusions, Figure 4 shows no tendency at Mars for the large increases in the vertical extent of the ionosphere as the terminator is approached that are observed at Venus [Theis et al., 1980]. It is expected for an ionospheric interaction that the distance to the top of the ionosphere increases as the terminator is approached due to the accompanying drop in external plasma pressure as the angle of incidence of the solar wind increases [e.g., Spreiter and Stahara, 1980b; Vaisberg et al., 1980]. The apparent absence of such growths in scale height as the Martian terminator is approached then imply, as pointed out by Breus and Gringauz, that the situation at Mars is different with the ionosphere not bearing the pressure of the solar wind most probably due to the presence of an intrinsic field magnetosphere as the earth.

4. Bauer and Hartle [1973] have previously considered the vertical extent of the ionosphere in a small Martian magnetosphere. They proposed that an upper boundary would form at the level where losses due to magnetospheric convection sweeping away the ionospheric plasma were equaled by the production rate of new ions from the ambient neutral population and termed it the 'chemopause.' The Viking descent measurements present this model with two extreme situations in that V2 found a boundary near the 300 km height with a number density of about 1000 cm -3 while V 1 determined only a lower limit on the vertical extent at 360km with density below 100 cm -3' For the Mars magnetic moment determined by this study (pro-A.2) we find that the range in cross- magnetospheric electric fields necessary to span the Viking observations in terms of the chemopause model is,'-, 10-500 V/R e. This variation is large in comparison with that observed in the terrestrial magnetosphere, but not so much as to be ruled out given the possible large effects of dayside reconnection on a small magnetosphere and our general ignorance of the particles and fields environment of Mars. However, it may be that the boundary in the Viking 2 measurements corresponds to the Martian magnetopause during a large erosion event. In this case, the V 1 and radio occultation results must be considered alone and do not require convection electric

fields in excess of 10-102 V/R e estimates ofRassbach et al. [ 1976].

ANTI-Small Intrinsic Field Magnetosphere

A. Bow shock position. 1. Russell [1977] compared the relative heights of the Martian and Venusian bow waves using the Mars

orbiter and mostly Venus fly-by measurements with results similar to those of Verigin et al. [1978]. However, in contrast to their interpretation (pro-A.1) Russell maintained that both of these planets have predominantly ionospheric solar wind interactions, but with the larger shock stand-off distance at Mars due to its ionosphere having greater electrical conductivity. Under these circumstances it might be expected that Mars will absorb less of the incident flow than Venus. Breus and Gringauz [ 1980] have criticized this hypothesis on the grounds that their calculations show the Venus ionosphere being the better conductor of the two. In addition, we have found that the Mars bow shock is too distant from the planet to be consistent with a predominantly ionospheric solar-planetary interaction (pro-A.2,C. 1).

2. The studies of bow shock position by Vaisberg et al. [1975], Bogdanov and Vaisberg [ 1975], and Russell [ 1977] have placed the subsolar altitude of the solar wind-obstacle interface in the 350 to

400-km range. In each case the altitude of the ob. stacle was assumed indicative of an ionospheric interaction. However, while the obstacle altitude determined in these instances may be criticized as being 20-30% too low (pro-A.2' Slavin and Holzer [1981]), it is more important to note that the Martian ionopause, should it exist, can be expected to form at an altitude that is lower than that of Venus due to the weakness of the Martian ionosphere (pro-C.l' e.g., Spreiter et al. [1970]). The height of the subsolar Venus ionopause is shown in Figure 4 and lies far below the obstacle surface found by this study.

B. Mars2, 3, and5 interaction region observations. 1. Both Wallis [ 1975] and Russell [ 1978a ] have challenged the claim that Mars 3 ever entered the Martian magnetosphere. In both cases the essential contention was that the purported magnetospheric field observations were probably just the enhanced magnetic flux densities that are always present behind a fast mode MHD shock wave. The 'bow shock' identified by Dolginov et al. [1972, 1973] could then correspond to entry into the Mars foreshock region [e.g., Greenstadt and Fredericks, 1979] where upstream waves might mimic magnetosheath turbulence when observed with a low sampling rate as on Mars 3. However, the plasma electron observations of the Faraday cup experiment [Gringauz et al., 1974] agree with Dolginov et al.'s identification of bow shock position and show a 'drop out' approximately coincident with the magnetopause crossing identified in the magnetic field measurements as would be expected upon entering a low/3 dayside intrinsic magnetosphere. Finally, it must also be remembered that given the large periapsis altitudes of the Mars orbiters (i.e., 1100 to 2100 km) relative to the small magnetospheric cavity in Figure 2, a lack of intrinsic field signatures in the orbiter data is not an argument against the small magnetosphere model.

2. Plasma ion measurements near Mars 2 and 3 periapses on the dayside were made by the RIEP Spectrometer (e.g., Vaisberg et al. [ 1972]; only electrons were successfully examined with the Mars 2 and 3 Faraday cup; Gringauz [1976]. Within the regions of enhanced magnetic field strength observed around periapsis by Mars 2 and Mars 3 suggested by Dolginov [e.g., 1978b,c] to be within the Martian magnetosphere, significant fluxes ofE < 200 eV ions were detected [Vaisberg et al., 1972; Vaisberg and Bogdanov, 1974]. Their bulk flow speed was less than the thermal speed, ,-,- 50 km/s, with temperatures of ,--10-20 eV and densities near ,--,10 cm -3 [Bogdanov and Vaisberg, 1975]. These findings have been interpreted by the RIEP investigators as evidence for the presence of a 'hot ion cushion' in front of the Martian ionosphere (i.e.,/3 '--'0.2 for B ,• 25 nT and T e • Ti) with the observed magnetic field being of interplanetary, as opposed to planetary, origin. It is their suggestion that the solar wind interaction with Mars observed by the plasma


spectrometer is in fact very similar to that of Venus where the intrinsic magnetic field is negligible [Russell et al., 1980a, b; Elphic et al., 1980a;Brace et al., 1980;Spenner et al., 1980; Vaisberg et al., 1980]. Our principal objections to the Vaisberg ionospheric interaction model are twofold. First, the January 21, 1972, Mars 3 vector magnetic field observations show a large shift in the field direction as it encounters the 'magnetosphere' with the field magnitude usually jumping rapidly in both the Mars 2 and 3 measurements as the enhanced field region is entered [Dolginov et al., 1973]. This change in the orientation of magnetic field observed coincident with an increase in field magnitude is not consistent with the 'draping' of fieldlines about an ionospheric obstacle that is observed at Venus [Elphic et al., 1980b]. In addition, at both Venus [Elphic et al., 1980b; Spenner et al., 1980] and earth [Crooker et al; 1979] the magnetic field increase in the depletion layer is gradual in contrast to the relatively sudden enhancement seen at Mars. Second, it does not appear that the Martian ionosphere can stand off the solar wind altitudes above 300-400 km even at solar maximum (see pro-C. 1), whereas the orbiter observations are all at altitudes above 1100 km.

The requisite depletion layer (or 'hot ion cushion') thickness to produce the periapsis observations is therefore about '--, 103 km which is an order of magnitude greater than expected theoretically [Zwan and Wolf, 1976].

3. Mars 5 RIEP ion observations adjacent to the magnetotail 1 - 3 RMs downstream from the terminator have detected a boundary layer region of reduced flow speed and density similar to that of the earth [Vaisberg et al., 1974]. As such, boundary layers are a general MHD aspect of solar wind flow past large magnetic field dominated volumes upon which they exert tangential surface stresses. How- ever, Vaisberg et al. [ 1976 and references therein] have found their measurements to contain evidence for the presence of heavier ions, such as 0 +, being contained in these flows. These observations have then been used as arguments on behalf of a strong interaction between the solar wind and the Martian atmosphere from whence the heavier ions would be accreted [Vaisberg, 1976]. While contro- versial [Bezrukikh et al., 1978; Vaisberg and Smirnov, 1978], the observation of ions of planetary origin being swept away does give Mars and Venus shared properties as O + has also been detected downstream at Venus by the PVO solar wind plasma experiment [Mihalov et al., 1980]. However, ionoized atmosphere neutrals are expected, in the context of the small mangetosphere model, to be lost to the solar wind not only when high dynamic pressure or dayside magnetic reconnection expose them to the magnetosheath flow, but also during more average conditions through the action of magnetospheric convection [ Bauer and Hartle, 1973]. In addition, as pointed out by Stairnov et al. [ 1978], any large tilt of the magnetic dipole relative to the Mars spin axis could further weaken the shielding ability of the intrinsic field.

CONCLUDING REMARKS

We find that the results of our analyses and those of the previous studies argue quite convincingly for the existence of a modest intrinsic magnetic field at Mars dominating the solar-planetary interaction. Further, the lack of ionopause signatures in the radio occultation observations and the overall weakness of the ionosphere appear to pose major obstacles to the postulation of a Venus-type interaction for this planet. The opposing side to this view proceeds largely by criticizing individual points in the magnetospheric inter-

tion. It is our conclusion that the ionospheric interaction model for Mars is not supported by the available experimental measurements.

Acknowledgments. The authors gratefully acknowledge the contribution of data by O. L. Vaisberg, V. N. Smirnov, and L. H. Brace. In addition, discussions of the Venus solar wind interaction with J. G. Luhmann and R. C. Elphic as well as remaks on preliminary versions of this manuscript by C. T. Russell have proved most helpful. Both referees also made useful suggestions concerning the presentation of the results. This work was supported by NASA grant NAGW-74 and NSF grant ATM 79-16245.

The Editor thanks L. H. Brace for his assistance in evaluating this paper.

REFERENCES

Bauer, S. J., Solar-wind control of the extent of planetary ionospheres, in Solar Wind Interaction with the Planets Mercury, Venus, and Mars, edited by N. F. Ness, pp. 47-62, NASA SP-397, 1976.

Bauer, S. J., and R. E. Hartle, On the extent of the Martian ionosphere, J. Geophys. Res., 78, 3169, 1973.

Bezrukikh, V. V., M. I. Verigin, and N. M. Shyutte, Detection of heavy ions in the interaction region between the solar wind and the planet Mars, Cosmic Res., 16, 476, 1978.

Bogdanov, A. V., A mechanism for Deimos to influence the characteristics of the Martian magnetosphere, Cosmic Res., 16, 590, 1978.

Bogdanov, A. V. Mars satellite Deimos interaction with the solar wind and its influence on flow around Mars, J. Geophys. Res., 86, 6926, 1981.

Bogdanov, A. V., and O. L. Vaisberg, Structure and variations of the solar wind--Mars interaction region, J. Geophys. Res., 80, 487, 1975.

Brace, L. H., R. F. Theis, W. R. Hoegy, J. H. Wolfe, J. D. Mihalov, C. T. Russell, R. C. Elphic, and A. F. Nagy, The dynamic behavior of Venus ionosphere in response to solar wind interactions, J. Geophys. Res., 85, 7663, 1980.

Breus, T. K., and K. I. Gringauz, Nature of the obstacles which slow down the solar wind near Venus and Mars, and the properties of the interaction between the solar wind and the atmospheres of these planets, Cosmic Res., 18, 426, 1980.

Chen, R. H., T. E. Cravens, and A. F. Nagy, The Martian ionosphere in light of Viking observations, J. Geophys. Res., 83, 3871, 1978.

Cloutier, P. A., and R. E. Daniell, Jr., An electrodynamic model of the solar wind interaction with the ionospheres of Mars and Venus, Planet. Space $ci., 27, 1111, 1979.

Cloutier, P. A., M. B. McElroy, and F. C. Michel, Modification of the Martian ionosphere by the solar wind, J. Geophys. Res., 74, 6215, 1969.

Cooker, N. U., T. E. Eastman, and G. S. Stiles, Observations of plasma depletion in the magnetosheath at the dayside magnetopause, J. Geophys. Res., 84, 869, 1979.

Dolginov, Sh. Sh., The magnetosphere of Mars, in Physics of Solar Plane- tary Environments, vol. 1, edited by D. J. Williams, pp. 872-888, AGU, Washington, D.C., 1976.

Dolginov, Sh. Sh., The magnetic field of Mars, Cosmic Res., 16, 204, 1978a.

Dolginov, Sh. Sh., On the magnetic field of Mars: Mars 2 and 3 evidence, Geophys. Res. Lett., 5, 89, 1978b.

Dolginov, Sh. Sh., On the magnetic field of Mars: Mars 5 evidence, Geophys. Res. Lett., 5, 93, 1978c.

Dolginov, Sh. Sh., E. G. Eroshenko, and L. N. Zhuzgov, Magnetic field investigation with interplanetary station Venera 4, Cosmic Res., 6, 469, 1969.

Dolginov, Sh. Sh., E.G. Eroshenko, and L. N. Zhuzgov, Dokl. Akad. Nauk. SSSR, 207(6), pages, 1972.

Dolginov, Sh. Sh., E. G. Eroshenko, and D. N. Zhuzgov, Magnetic field in the very close neighborhood of Mars according to data from the Mars 2 and 3 spacecraft, J. Geophys. Res., 78, 4779, 1973.

Dolginov, Sh. Sh., E. G. Eroshenko, and L. N. Zhuzgov, Magnetic field of Mars from data of Mars 3 and Mars 5, Cosmic Res., 13, 108, 1975.

Dolginov, Sh. Sh., E. G. Eroshenko, and L. N. Zhuzgov, The magnetic field of Mars according to the data from Mars 3 and Mars 5, J. Geophys Res., 81, 3353, 1976a.

pretation of the January 21, 1972 Mars 3 measuements by Dolginov Dolginov, Sh. Sh., E.G. Eroshenko, D. N. Zhuzgov, V. A. Sharova, K. I. Gringauz, V. V. Bezrukikh, T. K. Breus, M. I. Verigin, and A. P.

and Gringauz. While such comments have their place, they are no Remizov, Magnetic field and plasma inside and outside of the Martian substitute for demonstrations of how the existing Mars observations magnetosphere, in Solar Wind Interaction with the Planets Mercury, might be explained in terms of an ionospheric solar wind interac- Venus, andMars, edited by N. F. Ness, pp. 1 - 20, NASA SP 397, 1976b.


Dryer, M., and G. R. Heckman, Application of the hypersonic analogue to the standing shock of Mars, Solar Phys., 2, 112, 1967.

Dubinin, E. M., I. M. Podgornyi, Yu. N. Potanin, and S. I. Shkol'nikova, Determining the magnetic moment of Venus by magnetic measurements in the tail, Cosmic Res., 16, 693, 1978.

Elphic, R. C., C. T. Russell, J. A. Slavin, L. H. Brace, and A. F. Nagy, The location of the dayside ionopause of Venus: Pioneer Venus magnetometer observations, Geophys. Res. Lett., 7, 561, 1980a.

Elphic, R. C., C. T. Russell, J. A. Slavin, and L. H. Brace, Observations of the dayside ionopause and ionosphere of Venus, J. Geophys. Res., 85, 7679, 1980b.

Eshleman, V. R., Atmospheres of Mars and Venus: A review of Mariner 4 and 5 and Venera 4 experiments, Radio Sci., 5, 325, 1970.

Fairfield, D. H., Average and unusual locations of the earth's magnetopause and bow shock, J. Geophys. Res., 76, 6700, 1971.

Fjeldbo, G., and V. R. Eshleman, The atmosphere of Mars analyzed by integral inversion of the Mariner 4 occultation data, Planet. Space Sci., 16, 1035, 1968.

Fjeldbo, G., and V. R. Eshleman, The atmosphere of Venus as studied with the Mariner 5 dual radio-frequency occultation experiment, Radio Sci., 4, 879, 1969.

Gombosi, T. I., T. E. Cravens, A. F. Nagy, R. C. Elphic, and C. T. Russell, Solar wind absorption by Venus, J. Geophys. Res., 85, 7747, 1980.

Greenstadt, E. W., and R. W. Fredticks, Shock systems in collisionless space, in Space Plasma Physics, pp. 807-878, National Academy of Sciences, Washington, D.C., 1979.

Gringauz, K. I., Interaction of solar wind with Mars as seen by charged particle traps on Mars 2, 3, and 5 satellites, Rev. Geophys. Space Phys., 14, 391, 1976.

Gringauz, K. I., A comparison of the magnetospheres of Mars, Venus, and the Earth, Physics of Planetary Magnetospheres, pp. 5-23, ed. K. Knott, Proceedings of the 23rd Plenary Meeting of COSPAR, Budapest, Hungary, 1980.

Gringauz, K. I., V. V. Berzrukikh, T. K. Breus, G.I. Volkov, L. S. Musatov, L. P. Havkin, and G. F. Sloutchenkov, Results of solar plasma electron observations on Mars 2 and 3 spacecraft, J. Geophys. Res., 78, 5808, 1973.

Gringauz, K. I., V. V. Bezrukikh, T. K. Breus, M. I. Verigin, G. I. Volkov, and A. V. Dyachkov, study of plasma near Mars and along the path earth-Mars, 2, Characteristics of the electron component along the orbits of artificial satellites Mars 2 and 3, Cosmic Res., 12, 535, 1974.

Gringauz, K. I., V. V. Bezrukikh, M. I. Verigin, and A. P. Remizov, Studies of solar plasma near Mars and along the earth-Mars path, 3, Characteristics of ion and electron components measured in satellite Mars 5, Cosmic Res., 13, 107, 1975.

Gringauz, K. I., V. V. Bezrukikh, M. I. Verigin, and A. P. Remizov, On the electron and ion components of plasma in the antisolar part of near- Martian space, J. Geophys. Res., 81, 3349, 1976.

Gringauz, K. I., V. V. Bezrukikh, T. K. Breus, M. I. Verigin, and A. P. Remizov, The magnetic field of Mars estimated from the data of plasma measurements by Soviet artificial satellites of Mars, The Soviet-American Conference on Cosmochemistry of the Moon and Planets, pp. 859-863, ed. J. H. Pomeroy and N.J. Hubbard, NASA SP-370, 1977.

Hanson, W. B., S. Sanatani, and D. R. Zuccaro, The Martian ionosphere as observed by the Viking retarding potential analyzers, J. Geophys. Res., 82, 4351, 1977.

Hill, T. W., and F. C. Michel, Planetary magnetospheres, Rev. Geophys. Space Phys., 13, 967, 1975.

Hill, T. W., A. J. Dessler, and R. A. Wolf, Mercury and Mars: The role of ionospheric conductivity in the acceleration of magnetospheric particles, Geophys. Res. Lett., 3, 429, 1976.

Holzer, R. E., and J. A. Slavin, Magnetic flux transfer associated with expansions and contractions of the dayside magnetosphere, J. Geophys. Res., 83, 3831, 1978.

Hogan, J. S., R. W. Stewart, and S. I. Rasool, Radio occultation measurements of the Mars atmosphere with Mariner 6 and 7, Radio Sci., 7, 525, 1972.

Hood, L. L., and G. Schubert, Inhibition of solar wind impingement on Mercury by planetary induction currents, J. Geophys. Res., 84, 2641, 1979.

Intriligator, D. S., and E. J. Smith, Mars in the solar wind, J. Geophys. Res., 84, 8427, 1979.

Ivanov-Kholodnyi, G. S., M. A. Kolosov, N. A. Savich, Yu. N. Alexandrov, M. B. Vasilyev, A. S. Vyshlov, V. M. Dubrovin, A. L. Zaitsev, A. V. Michailov, G. M. Petrov, V. A. Samovol, L. N. Samoz- naev, A. I. Sidorenko, and A. F. Hasyanov, Daytime ionosphere of

Venus as studied by Veneras 9 and 10 dual-frequency radio occultation experiments, Icarus, 39, 209, 1979.

Johnson, R. E., Comment on ion and electron temperatures in the Martian upper atmosphere, Geophys. Res. Lett., 5, 989, 1978.

King, J. H., Interplanetary Medium Data Book, Rep. NSSDC/WDC-A-R 77-04, NASA Goddard Space Flight Center, Greenbelt, Md., 1977.

Kivelson, M. G., Magnetospheric electric fields and their variation with geomagnetic activity, Rev. Geophys. Space Phys., 14, 189, 1976.,

Kliore, A. J., D. L. Cain, G. S. Levy, V. R. Eshleman, G. Fjeldbo, and F. D. Drake, Occultation experiment: Results of the first direct measurements of Mars' atmosphere and ionosphere, Science, 149, 1243, 1965.

Kliore, A. J., G. S. Levy, D. L. Cain, G. Fjeldbo, and S. I. Rasool, Atmosphere and ionosphere of Venus from the Mariner 5 S-band radio occultation measurement, Science, 158, 1683, 1967.

Kliore, A. J., D. L. Cain,G. Fjeldbo, B. L. Seidel, M. J. Sykes, and S. I. Rasool, The atmosphere of Mars from Mariner 9 radio occultation experiments, Icarus, 17, 484, 1972.

Kliore, A. J., G. Fjeldbo, B. L. Seidel, M. J. Sykes, and P.M. Woiceshyn, S-band radio occultation measurements of the atmosphere and topography of Mars with Mariner 9, J. Geophys. Res., 78, 4331, 1973.

Lazarus, A. J., H. S. Bridge, J. M. Davis, and C. W. Snyder, Initial results from the Mariner 4 solar plasma experiment, Space Res., 7, 1296, 1967.

Lindal, G. F., H. B. Hotz, D. N. Sweetnam, Z. Shippony, J. P. Brenkle, V. Hartsell, R. T. Spear, and W. H. Michael, Jr., Viking radio occultation measurements of the atmosphere and topography of Mars: Data acquired during 1 Martian year of tracking, J. Geophys. Res., 84, 8443, 1979.

Luhmann, J. G., R. C. Elphic, C. T. Russell, J. D. Mihalov, and J. H. Wolf, Observations of large scale scale steady magnetic fields in the dayside Venus ionosphere, Geophys. Res. Lett., 7, 917, 1980.

Michel, F. C., Solar wind interactions with planetary atmospheres, Rev. Geophys. Space Phys., 9, 427, 1971.

Mihalov, J. D., J. H. Wolfe, and D. S. Intriligator, Pioneer Venus plasma observations of the solar wind-Venus interaction, J. Geophys. Res., 85, 7613, 1980.

Ness, N. F., The magnetic fields of Mercury, Mars, and the Moon, Ann. Rev. Earth Planet. Sci., 7, 249, 1979.

O'Gallagher, J. J., and J. A. Simpson, Search for trapped electrons and a magnetic moment at Mars by Mariner 4, Science, 149, 1233, 1965.

Rassbach, M. E., R. A. Wolf, and R. E. Daniell, Jr., Convection in the Martian magnetosphere, J. Geophys. Res., 79, 1125, 1974.

Rohrbaugh, R. P., J. S. Nisbet, E. Bleuler, and J. R. Herman, The effect of energetically produced 02 + on the ion temperatures of the Martian ther- mosphere, J. Geophys. Res., 84, 3327, 1979.

Russell, C. T., On the relative locations of the bow shocks of the terrestrial planets, Geophys. Res. Lett., 4, 387, 1977.

Russell, C. T., The magnetic field of Mars: Mars 3 evidence re-examined, Geophys. Res. Lett., 5, 81, 1978a.

Russell, C. T., The magnetic field of Mars: Mars 5 evidence re-examined, Geophys. Res. Lett., 5, 85, 1978b.

Russell, C. T., Does Mars have an active dynamo?, Geochim. Cosmochim. Acta. Suppl., 120, 3137, 1978c.

Russell, C. T., The interaction of the solar wind with Mars, Venus, and Mercury, inSolar System Plasma Physics, vol. 2, edited by C. F. Kennel, L. J. Lanzerotti, and E. N. Parker, pp. 208-252, North-Holland, Amsterdam, 1979.

Russell, C. T., R. C. Elphic, and J. A. Slavin, Limits on the possible intrinsic magnetic field of Venus, J. Geophys. Res., 85, 8319, 1980a.

Russell, C. T., R. C. Elphic, J. G. Luhmann, and J. A. Slavin, The magnetotail of Venus: Pioneer Venus magnetometer observations (ab- stract), Eos Trans. AGU, 61, 1019, 1980b.

Schunk, R. W., and A. F. Nagy, Ionospheres of the terrestrial planets, Rev. Geophys. Space Phys., 18, 813, 1980.

Siscoe, G. L., and J. A. Slavin, Planetary magnetospheres, Rev. Geophys. Space Phys., 17, 1677, 1979.

Slavin, J. A., and R. E. Holzer, The effect of erosion on the solar wind stand-off distance at Mercury, J. Geophys. Res., 84, 2076, 1979.

Slavin, J. A., and R. E. Holzer, Solar wind flow about the terrestrial planets, 1, Modeling bow shock position and shape, J. Geophys. Res., 86, in press, 1981.

Slavin, J. A., R. C. Elphic, C. T. Russell, J. H. Wolfe, and D. S. Intriligator, Position and shape of the Venus bow shock: Pioneer Venus orbiter observations, Geophys. Res. Lett., 6, 901, 1979a.

Slavin, J. A., R. C. Elphic, and C. T. Russell, A comparison of Pioneer and Venus and Venera bow shock observations: Evidence for a solar cycle variation, Geophys. Res. Lett., 6, 905, 1979b.

Slavin, J. A., R. C. Elphic, C. T. Russell, F. L. Scarf, J. H. Wolfe, J. D.


Mihalov, D. S. Intriligator, L. H. Brace, H. A. Taylor, Jr., and R. E. Daniell, Jr., The solar wind interaction with Venus: Pioneer Venus observations of bow shock location and structure, J. Geophys. Res., 85, 7625, 1980.

Smirnov, V. N., A. N. Omel'chenko, and O. L. Vaisberg, Possible discovery of cusps near Mars, Cosmic Res., 688, 1978.

Smith, E. J., A review of lunar and planetary magnetic field measurements using space probes, in Magnetism and the Cosmos, edited by S. K. Runcorn, p. 271, Oliver and Boyd, Edinburgh, 1967.

Smith, E. J., Planetary magnetic field experiments, in Advanced Space Environments, edited by O. L. Tiffany and E. M. Zaitzeff, American Astronautical Society, Tarzana, Calif., 1969.

Smith, E. J., L. Davis, Jr., P. J. Coleman, Jr., and D. E. Jones, Magnetic field measurements near Mars, Science, 149, 1241, 1965.

Spenner, K., W. C. Knudsen, K. L. Miller, V. Novak, C. T. Russell, and R. C. Elphic, Observation of the Venus mantle, the boundary region between solar wind and ionosphere, J. Geophys. Res., 85, 7655, 1980.

Spreiter, J. R., and A. W. Rizzi, The Martian bow wave--Theory and observation, Planet. Space Sci., 20, 205, 1972.

Spreiter, J. R., and S. S. Stahara, A new predictive model for determining solar wind-terrestrial planet interactions, J. Geophys. Res., 85, 6769, 1980a.

Spreiter, J. R., and S. S. Stahara, Solar wind flow past Venus: Theory and comparison, J. Geophys. Res., 85, 7715, 1980b.

Spreiter, J. R., A. L. Summers, and A. Y. Alksne, Hydromagnetic flow around the magnetosphere, Planet. Space Sci., 14, 223, 1966.

Spreiter, J. R., A. L. Summers, and A. W. Rizzi, Solar wind flow past nonmagnetic planets--Venus and Mars, Planet, Space Sci., 18, 1281, 1970.

Stahara, S. S., D. Klenke, B. C. Trudinger, and J. R. Spreiter, Application of Advanced Computational Procedures for Modeling Solar Wind Inter- actions with Venus - Theory and Computer Code, NASA Contract Rep. 3267, 1980.

Suess, S. T., and B. E. Goldstein, Compression of the Hermaean magnetosphere by the solar wind, J. Geophys. Res., 84, 3306, 1979.

Theis, R. F., L. H. Brace, and H. G. Mayr, Empircal models of the electron temperature and density in the Venus ionosphere, J. Geophys. Res., 85, 7787, 1980.

Vaisberg, O. L., Mars-plasma environment, in Physics of Solar Planetary Environments, edited by D. J. Williams, pp. 854-871, AGU, Washing- ton, D.C., 1976.

Vaisberg, O. L., and A. V. Bogdanov, Flow of solar wind around Mars and Venus--General principles, Cosmic Res., 12, 253, 1974.

Vaisberg, O. L., and V. N. Smirnov, Reliability of heavy ion identification in the plasma flow near Mars, Cosmic Res., 16, 480, 1978.

Vaisberg, O. L., A. V. Bogdanov, N. F. Borodin, E. M. Vasil'ev, A. V. D'yachkov, A. A. Zertsalov, B. V. Polenov, and S. A. Romanov,

Observation of the region of interaction between the solar-wind plasma and Mars, Cosmic Res., 10, 417, 1972.

Vaisberg, O. L., A. V. Bogdanov, V. N. Smirnov, and S. A. Romanov, Initial results of ion flux measurements by the RIED-2801 M instrument on Mars 4 and 5, Cosmic Res., 13, 112, 1975.

Vaisberg, O. L., A. V. Bogdanov, V. N. Smirnov, and S. A. Romanov, On the nature of the solar wind-Mars interaction, in Solar Wind Interaction With the Planets Mercury, Venus, and Mars, edited by N. F. Ness, pp. 21-40, NASA SP 397, 1976.

Vaisberg, O. L., D. S. Intriligator, and V. N. Smirnov, An empirical model of the Venusian outer environment, 1, The shape of the dayside solar wind-atmosphere interface, J. Geophys. Res., 85, 7642, 1980.

Van Allen, J. A., L. A. Frank, S. M. Krimigis, and H. K. Hills, Absence of Martian radiation belts and implications thereof, Science, 149, 1228, 1965.

Van Dyke, M.D., The supersonic blunt-body problem--Review and ex- tension, J. Aerosp. Sci., Vol., 485, 1958.

Vasil'ev, M. B., A. S. Vyshlov, M. A. Kolosov, N. A. Savich, V. A. Samovol, L. N. Samoznaev, A. I. Sidorenko, Yu. N. Aleksandrov, A. I. Danilenko, A. L. Zaitsv, G. M. Petrov, O. N. Rzhiga, D. Ya. Shtern, and L. I. Romanova, Preliminary results of the two-frequency radio occultation of the Martian ionosphere by means of the Mars automatic interplanetary stations in 1974, Cosmic Res., 13, 41, 1975.

Verigin, M. I., K. I. Gringauz, T. Gombosi, T. K. Breus, V. V. Bezrukikh, A. P. Remizov, and G. I. Volkov, Plasma near Venus from the Venera 9 and 10 wideangle plasma analyzer data, J. Geophys. Res., 83, 3721, 1978.

Vyshlov, A. S., G. S. Ivanov-Kholodnyi, A. V. Mikhailov, and N. A. Savich, Interpretation of results from the measurements of the upper Martian ionosphere via a dispersion interferometer on the satellite Mars 2, Cosmic Res., 13, 219, 1975.

Wallis, M. K., Weakly shocked flows of solar wind plasma through the atmospheres of comets and planets, Planet. Space Sci., 21, 1647, 1973.

Wallis, M. K., Does Mars have a magnetosphere?, Geophys. J. R. Astron. Soc., 41, 349, 1975.

Whitten, R. C., and L. Colin, The ionospheres of Mars and Venus, Rev. Geophys. Space Phys., 12, 155, 1974.

Wolff, R. S., B. E. Goldstein, and S. Kumar, A model of the variability of the Venus ionopause altitude, Geophys. Res. Lett., 6, 7641, 1979.

Zhuang, H. (2. and (2. T. Russell, An analytic treatment of structure of the bow shock and magnetosheath, J. Geophys. Res., 86, 2191, 1980.

Zwan, B. J., and R. A. Wolf, Depletion of solar wind plasma near a planetary boundary, J. Geophys. Res., 81, 1636, 1976.

(Received February 9, 1981; revised June 5, 1981;

accepted June 29, 1981.)

the solar wind interaction with mars...

Documents