J. G. Luhmann and C. T. Russell

Originally published in:
Encyclopedia of Planetary Sciences, edited by J. H. Shirley and R. W. Fainbridge,
454-456, Chapman and Hall, New York, 1997.


Mars, at it heliocentric distance of approximately 1.5 AU, exemplifies a terrestrial body that is affected by both a slightly greater distance from the Sun than Earth and a lower gravitational field at the surface. The latter results from Mars' smaller size. The radius of Mars is about 3395 km on average (compared to the 6371 km radius of Earth). As a result, much of the inventory of the lighter gases of Mars has escaped to space over time. Some atmospheric constituents, such as the predominant CO2 and a small contribution of H2O periodically freeze out into the polar caps in a seasonal cycle. Ground ice, like permafrost, is most probably present down to latitudes quite close to the equator.

A hypothesized early dense CO2 atmosphere of Mars is considered to be today sequestered in carbonate rocks on the surface. These carbonates are thought to have formed in the presence of liquid water, when Mars had a greenhouse effect (q.v.) contributing to the warmth of its early climate. Mars is too small to have plate tectonics and the resulting volcanic activity that together would have recycled the CO2 back into the atmosphere. It is likely that some of the early Martian atmosphere has been lost to space because Mars, like Venus has no substantial intrinsic magnetic field to protect the atmosphere from solar wind scavenging.



The mean density of Mars deduced front spacecraft encounters is anomalously low at 3.9 g cm-3 (compared to ~ 5.3 g cm-3 for Venus and Earth). However, estimates of the size of its iron-bearing core are more typical, at about half of the planet's radius. Like the other planets, Mars probably had a molten core in the first billion or so years following accretion. The core of Mars may now be solid, if the small size of the planet led to more rapid evolution following the onset of solid core formation. Although the Viking landers carried a seismometer, which could have given some definitive information about the state of the Martian core, the device was not well coupled to the surface and suffered from vibrations caused by the winds. Surface features indicate that seismic activity related to plate tectonics is minimal, although volcanic activity was clearly present in the last billion years, as shown by the giant dormant volcano Olympus Mons and others.


Magnetic Field

The first indication of the weakness of the magnetic field of Mars was obtained during the Mariner 4 spacecraft flyby in 1965. At a closest approach of 3.9 Mars radii, no indication of the Earth-like dipole magnetic field predicted by scaling arguments from theory was detected. Still, a shock-like disturbance in the solar wind signaled the presence of an obstacle approximately the size of Mars. Most subsequent magnetic field measurements in the vicinity of Mars were carried out on a series of five MARS spacecraft launched by the Soviet Union between 1971 and 1974 (see Soviet MARS missions). Several of these successfully operated in orbit for periods long enough to both confirm the Mariner 4 results and to measure the disturbance of the interplanetary magnetic field caused by the obstacle. However, none of these spacecraft approached Mars closer than ~ 1300 km or ~ 1.3 Mars radii from the center of the planet, and none probed the solar wind wake inside of the optical shadow, where the magnetotail of an intrinsic magnetosphere resembling a weak version of Earth's would be found. The Viking landers reached the surface of Mars in 1976, but did not carry magnetic field experiments its part of their scientific payloads. although they made ionospheric measurements of relevance to the magnetic field question. Because the available measurements could be interpreted front the viewpoint of either a small Earth-like magnetosphere, or a Venus-like ionospheric obstacle, different researchers have adopted both of these paradigms for over a decade. Their divergent views depended on the techniques and arguments used in analyzing the still ambiguous data (Luhmann and Brace, 1991).

These differences in opinion have to some extent been altered by the most recent magnetic field measurements on the Soviet Phobos-2 spacecraft in 1989 (e.g. Nature. 341. 19 October 1989, describes the first results). The orbit of Phobos 2 went into the deep wake of Mars, for the first time providing magnetic field data in the optical shadow at distances as close as ~ 2.7 Mars radii and as distant as ~ 20 Mars radii. These data unambiguously showed that the magnetic fields in the wake of Mars are determined by the interplanetary field orientation, and are thus not Earth- like, at least in the near-equatorial spacecraft orbit plane. The current upper limit on the dipole moment remains at ~ 10-4 times that of Earth, a value established on the basis of the previous observations. This moment is derived not from the wake data but from estimates of the subsolar altitude of the Martian obstacle to the solar wind of ~ 400 km. Additional indirect information concerning the magnetic field of Mars derived front ionospheric observations and the understanding of solar wind interactions is described below.

Today, the only other 'direct' information that Martian magnetism is from a special class of meteorites known as the SNC meteorites (q.v.) which are thought to come from Mars. Magnetic field analyses of these possible samples of the Martian crust indicate that magnetic fields of ~ 1000 nT may have been present on the surface of Mars at the time that these meteorites were ejected by a giant impact some 180 million years ago. (For comparison, the present field on Earth near the equator is about 3 X 104 nT. The present upper limit on the dipole moment implies surface fields of only a few tens Of nanotesla.)

The dynamo theory of planetary magnetism indicates that Mars may have had a dipole moment of about one-tenth of Earth's when it was first formed (Schubert and Spohn, 1990). The rotation rate Of Mars is approximately that of Earth and is thus sufficient for the operation of this initial dynamo. The other necessary ingredient of a convection driver in the core was supplied by heat left over from the accretion of the planet, which may have been effective for up to a few billion years. If such a field did indeed exist, evidence of it may still be present on the surface in the form of magnetized rocks and crustal regions like those observed on the Moon. No observations indicating the presence of such fields have been reported other than the aforementioned SNC meteorites' magnetization.


Solar Wind Interaction

To date, the observations at Mars suggest that it great deal o similarity exists between near-Venus space and near-Mars space. Mars, like most of the other planetary obstacles, is preceded in the solar wind by a 'bow shock'-like structure that reflects the slightly greater than planet-size scale of the weakly magnetized Martian obstacle. The subsolar distance of the bow shock is ~ 1.5 Rm while its terminator position is ~ 2.7 Rm. Within the bow shock, the solar wind plasma is diverted around the obstacle. Within it, the imbedded interplanetary magnetic field is compressed against the obstacle nose. The field distortion associated with the divergence of the flow gives the 'draped' configuration of field lines illustrated in Figure 1.. This 'magnetosheath' is a common feature of all planetary obstacles.

Fig. 1. Illustration of the disturbance in the solar wind flow (a) and interplanetary magnetic field (b) produced by a planetary obstacle in the solar wind. Early spacecraft to Mars detected this disturbance, the size of which gave an upper limit to the strength of the Martian magnetic field. (J. G. Luhmann and L. H. Brace, Rev. Geophys., 29, 121, 1991, copyright American Geophysical Union).

Within the dayside obstacle boundary implied by the Mars bow shock position, the magnetic field geometry is unknown. However, both in situ measurements on the Viking Landers and radio occultation experiments on the Viking Orbiters and other spacecraft indicated the presence of a substantial dayside ionosphere below about 300 km. (The subsolar obstacle height inferred from the bow shock position is ~ 400 km.) The in situ measurements from the Viking Landers also provided information on the temperatures in the ionospheric plasma which were used to calculate the ionosphere's thermal pressure. This calculation resulted in the conclusion that Mars must have a planetary magnetic field of significance because these pressures were less than the incident solar wind pressure. Nevertheless, we know from the observations at Venus during disturbed solar wind conditions that an ionospheric obstacle can persist in the face of such levels of excess solar wind pressure. One possibility is that the Mars ionosphere, like that of the disturbed Venus counterpart, contains large-scale horizontal magnetic fields that are induced by the solar wind interaction. Models of these induced fields suggest that they should be several tens of nanotesla in strength. The electron temperatures measured in the ionosphere by the Viking Landers are also consistent with fields of this strength and orientation, but it is not clear from these whether the field is planetary or induced by the solar wind interaction.

As mentioned above, the Phobos 2 magnetic measurements in the near- equatorial wake of Mars showed that the fields in that region are controlled by the interplanetary field orientation. The relationship between the interplanetary field draped over the obstacle and the field in such an 'induced' magnetotail is illustrated by Figure 2. In spite of this finding, there are still advocates of an intrinsic field contribution to the Martian magnetotail because the magnetotail appears to be wider than that of Venus relative to the planet radius (~ 2.0 planetary radii in diameter compared to Venus' ~ 1.2). It is argued that this contribution has not been detected because the 'intrinsic' field tail features may be restricted to regions removed from the equator.

Fig. 2. Illustration of the 'induced' magnetotail in the wake of Mars and its connection to the draped interplanetary magnetic field (J. G. Luhmann and L. H. Brace, Rev. Geophys., 29, 121, 1991, copyright American Geophysical Union).

Phobos 2 did detect significant fluxes of planetary ions (mainly O+, as at Venus) that had been scavenged from Mars by the passing solar wind (e.g. see the Nature special issue mentioned above). The details of the acceleration of these ions are not completely understood, but the electric field in the solar wind is expected to remove ions formed in the upper atmosphere that extends above the 'obstacle' boundary into the magnetosheath and undisturbed solar wind. The observed rates of escape for the oxygen suggest that the solar wind scavenging process has the potential to remove all of Mars' present inventory of atmospheric oxygen over the next 108 years. These observations also suggest that the solar wind interaction must have played some role in the Martian atmosphere's evolution over the past 4.5 billion years, or at least after the thermally driven planetary dynamo ceased to operate.



This work was supported by the National Aeronautics and Space Administration under research grants NAGW-1347. and NAG2-2573.



Luhmann, J. G. and Brace, L. H. (1991) Near-Mars space. Rev. Geophys., 29, 121.

Russell, C. T., 1987, Planetary magnetism in Geomagnetism, Vol. 2, (ed. J. A. Jacobs) London: Academic Press, pp. 457-523.

Schubert, G. and Spohn, T. (1990) Thermal history of Mars and the sulfur content of its core, J. Geophys. Res., 95, 14095-104.

Back to CT Russell's page More On-line Resources
Back to the SSC Home Page