Icarus 160, 220–223 (2002)
doi:10.1006/icar.2002.6942

NOTE

An Origin for the Linear Magnetic Anomalies on Mars through Accretion
of Terranes: Implications for Dynamo Timing

Alberto G. Fairén

Seminar on Planetary Sciences, Universidad Complutense de Madrid, 28040 Madrid, Spain
E-mail: Fairen.ag@terra.es

Javier Ruiz

Departamento de Geodinámica, Facultad de Ciencias Geológicas, Universidad Complutense de Madrid, 28040 Madrid, Spain

and

Francisco Anguita

Departamento de Petrologı́a y Geoquı́mica, Facultad de Ciencias Geológicas, Universidad Complutense de Madrid, 28040 Madrid, Spain

Received April 8, 2002; revised June 6, 2002

The proposed existence of magnetic lineations in the Terra
Cimmeria and Terra Sirenum regions of Mars was initially ex-
plained by Earth-like sea-floor spreading. Here we argue instead
that these lineations could have been formed at a convergent plate
margin through collision and accretion of terranes. A similar pro-
cess produced banded magnetic anomalies, similar in geometry and
even in size to those in Earth’s North American Cordillera. Be-
cause only sparse and generally weak anomalies have been detected
in the martian northern lowlands, which could constitute an ana-
log to the terrestrial oceanic crust, it is possible that the magnetic
field stopped its activity while crustal recycling was still active in
Mars. c© 2002 Elsevier Science (USA)

Key Words: Mars, surface; tectonics; magnetic fields.

I. Introduction. The discovery of large expanses of the martian southern

et al. 1999), repeated dike intrusions with the locus of intrusion moving with
highlands crust with strong remanent magnetism and some localized (and gen-
erally weak) magnetic anomalies in the northern lowlands (Acuña et al. 1999,
2001) has resulted in major revisions to previous ideas concerning martian dy-
namics. One of the most significant findings of the Mars Global Surveyor (MGS)
mission was the east–west trending linear crustal magnetization anomalies of
surprisingly great intensity centered in the Terra Cimmeria and Terra Sirenum
regions, which can be traced for more than 2000 km (Connerney et al. 1999,
2001, Purucker et al. 2000, Acuña et al. 2001; see Fig. 1). These anomalies have
alternating strong and weak magnetic signatures, a pattern which is compara-
ble to much weaker linear magnetic anomalies observed in the Earth’s oceanic
floor. Thus, this pattern on Mars was proposed to be genetically related to sea-
floor spreading in the presence of a reversing dipolar magnetic field (Connerney
et al. 1999, Nimmo 2000, Sprenke and Baker 2000). Sleep (1994) had already
proposed that plate tectonics operated on Mars in the Noachian, although he
considered the lowlands instead of the highlands as the ancient oceanic crust.

Several alternative hypotheses have since been proposed to help explain the
magnetic lineations, including folding or hydrochemical alteration (Connerney

22
0019-1035/02 $35.00
c© 2002 Elsevier Science (USA)

All rights reserved.
time (Nimmo 2000), and accretion of terranes (Baker et al. 2002, Dohm et al.
2002, Fairén et al. 2002). Even the existence of anomalies may be due to impact
demagnetization of a previously globally magnetized crust (Sprenke and Baker
2000, Arkani-Hamed 2001, Hood et al. 2002).

Although as of this writing, there is no definitive evidence for ancient martian
plate tectonism, here we assume this possibility as a working hypothesis, because
of its relevance in relation to the dynamics and evolution of early Mars. In
this note we show that the characteristics of the magnetic lineations in Terra
Cimmeria and Terra Sirenum can be fairly explained as the result of a process
of collision in a zone of crustal destruction, causing continental accretion, like
the one that took place in the west coast of North America in Jurassic to early
Tertiary time (e.g., Coney et al. 1980), which produced a distinctive pattern of
magnetic anomalies (Saltus et al. 1999). We also discuss the implications of
this hypothesis for the temporal evolution of both martian crustal dynamics and
magnetic field.

II. Banded magnetic anomalies in north american cordillera and Mars. A
pattern of linear and arcuate magnetic anomalies of alternating polarity is found
on Earth in the North American Cordillera; it can clearly be seen in the aero-
magnetic maps of Alaska (Fig. 2a) and British Columbia (Fig. 2b). There was a
great difference in the height at which data were obtained for Earth and Mars:
∼300 m for data of North America, 85–170 km for data acquired in MGS
aerobraking orbit, and ∼400 km for data acquired during the MGS mapping
orbit. Nevertheless, the magnetic features on Mars are about one order of mag-
nitude stronger than those of the Earth’s continents (Connerney et al. 1999),
and these high magnetic intensities on Mars suggest materials with remanent
magnetization comparable in intensity to that of fresh oceanic basalt on Earth.
Even so, the geometry and even the sizes (∼100- to 200-km broad) of North
American roughly linear anomalies are similar to those of martian lineations
(Connerney et al. 1999, 2001, Purucker et al. 2000) and are interpreted to be
the result of sheared terranes, which are accreted to the continent through a pro-
cess of oblique collision [for an exposition of the accretion process in the North
American Cordillera, see Moores and Twiss (1995)]. These accreted terranes
have several origins: oceanic crust, oceanic arcs, distal parts of continental edges,
rift remainders, or oceanic plateaus (Coney et al. 1980).

0

ecob
Lápiz

ecob
Lápiz

ecob
Lápiz

ecob
Lápiz

ecob
Lápiz



NOTE 221
FIG. 1. Radial magnetic field of Mars, overlain on topographic gradient map, from Purucker et al. (2000).
FIG. 2. (a) Composite aeromagnetic map of Alaska that depicts total field magnetic data values (after Saltus et al. 1999).



222 FAIRÉN, RUIZ, AND ANGUITA
FIG. 2. (b) Aeromagnetic map of the North American Cordillera, over Britis

and Wyoming, United States (from http://crustal.usgs.gov/namad/images.html).

The sources of the Alaskan anomalies have a significant thickness on a crustal
scale (Saltus et al. 1999), the same as that proposed for Terra Cimmeria and
Terra Sirenum based on magnetization intensity (Connerney et al. 1999, Nimmo
2000). Moreover, starting from magnetic features common to several global
magnetic maps of Mars, Purucker (2002) suggests that the southern boundary
between magnetic and nonmagnetic terranes is sharp for longitudes between
120◦ and 240◦E (corresponding to strongly magnetized crust) and gradational
for other longitudes. This could indicate deep-seated discontinuities in the crust
(e.g., fault blocks), which is consistent with magnetic linear features of the North
American Cordillera that are fault-bounded. Dohm et al. (2002) has mapped

numerous macrostructures that have similar trends and appear to correlate with
some of the anomalies.
h Columbia, Canada, and Washington, Idaho, Montana, and portions of Oregon

Importantly, the magnetic character shown by the southern continental margin
of Alaska has been considered (Saltus et al. 1999) as a typical example for
what is to be expected in any region on Earth that undergoes convergent margin
processes. If plate tectonics operated on early Mars, then terrane accretion would
be anticipated. Likewise, if this process occurred in the presence of a reversing
magnetic field, then a pattern of magnetic anomalies similar to the one observed
in North American Cordillera would be expected.

Therefore, the strongly magnetized crust of Mars could be analogous to that in-
terpreted for the North American Cordillera: a jigsaw of continental and oceanic
pieces related to a past period of plate tectonics, but to a process of crustal con-

vergence instead of generation. If the martian boundary in Terra Cimmeria–Terra
Sirenum is a former convergent plate border, the area southward of the dichotomy



T
NO

would be a zone where fragments of crust would accrete, and their geometry
would be defined, as on Earth, by the angle of collision between the two plates.
The magnetic linear bands proposed for these regions would thus be sheared
terranes, and their northward transition to more round-shaped magnetic features
would indicate face-on collision conditions. So, our model of early martian
plate tectonics is not entirely equivalent to the one proposed by Sleep (1994), in
which the highland–lowland boundary northward of Terra Cimmeria and Terra
Sirenum is interpreted as a passive margin instead of a plate-convergence zone.

MGS topography and gravity data indicate that the crust in these locations of
strongly magnetized highlands is clearly thicker than that in the lowlands (Zuber
et al. 2000), which favors the hypothesis of an accretional origin versus that of a
sea-floor spreading. Finally, if the accretion model is correct, it has a number of
petrologic implications: accreted terrane packages would presumably contain
many rock types, such as arc volcanics, silicic plutonics, and metamorphosed
sediments. These rock types are not presently known on Mars, and this could be
an interesting target for future investigations.

III. Discussion: martian dynamo timing. The absence of magnetization in
the Hellas and Argyre basins is an apparent indication that the dynamo was
already nonoperative when the forming impacts were produced, and this would
imply a minimum age of ∼4 Gyr for the death of the martian magnetic field
(Acuña et al. 1999, 2001, Arkani-Hamed 2001). However, Schubert et al. (2000)
proposed an alternative post-Hellas and post-Argyre timing for the martian dy-
namo, though the characteristics of the anomalies around both these impact
basins do not favor this possibility (Arkani-Hamed 2001). If magnetization in
Terra Cimmeria and Terra Sirenum was produced in relation to the accretion of
terranes, the absence of significant magnetic anomalies in the northern lowlands
could have deep implications on the temporal evolution of both the martian
dynamo and lithosphere.

In this sense, if the northern lowlands constitute former oceanic floor materials,
which originated from sea-floor spreading centers during the time of accretion to
the south, then they should have had a similar alternating magnetization pattern
as that described for the highlands, because the magnetization process would
have affected them in a similar way. However, only minor anomalies have been
detected on the martian lowlands and no linear bands have been described (Acuña
et al. 1999, 2001).

The absence of magnetic lineations in the northern plains could result from
several possibilities, including data resolution limits, alteration of magnetized
northern plains materials through erosion and hydrothermal processes, or a
marked decrease in the time scale for dynamo reversals. The absence of lin-
ear anomalies in the northern lowlands could be explained by a termination of
magnetic field before the cessation of Mars plate tectonism (Fairén et al. 2002,
Baker et al. 2002). This delay could have been long enough for plate tectonics
to generate new, unmagnetized crust in the lowlands until the definitive stand-
still of crustal recycling. As this process would have been gradual, the localized
anomalies detected in some areas of the lowlands could represent weak signa-
tures of the final stages of magnetic field activity. In this sense, it is interesting
to stress that the surface of the northern lowlands is clearly stratigraphically
younger than that of the highlands (e.g., Tanaka et al. 1992, Strom et al. 1992),
though MGS gravity data seem to indicate that the lowlands basement is roughly
contemporary to the highlands (Frey et al. 2001, Zuber 2001); therefore, and at
least in certain locations, the magnetization of the ancient crust of the lowlands
may have survived underlying younger crust (Stevenson 2001, Zuber 2001).

Finally, the absence of magnetization in extended areas of the southern high-
lands, centered in Noachis Terra, would be due to a demagnetization caused by
the impacts that created Hellas and Argyre (Hood et al. 2002), although perhaps
additional impacts could also be required for this process (Purucker 2002). Alter-
natively, the nonmagnetized southern regions could represent a primitive martian
crust that evolved before the planetary magnetic dynamo began to operate.

ACKNOWLEDGMENTS
We thank Dr. James Dohm, University of Arizona, and two anonymous ref-
erees for valuable suggestions which improved the final version of this work.
E 223

REFERENCES

Acuña, M. H., and 12 colleagues 1999. Global distribution of crustal magnetiza-
tion discovered by the Mars Global Surveyor MAG/ER experiment. Science
284, 790–793.

Acuña, M. H., and 13 colleagues 2001. Magnetic field of Mars: Summary of re-
sults from the aerobraking and mapping orbits. J. Geophys. Res. 106, 23,403–
23,417.

Arkani-Hamed, J. 2001. A 50-degree spherical harmonic model of the magnetic
field of Mars. J. Geophys. Res. 106, 23,197–23,208.

Baker, V. R., S. Maruyama, and J. M. Dohm 2002. A theory for the geological
evolution of Mars and related synthesis (Geomars). Lunar. Planet. Sci. 33,
1586 (abstract) [CD-ROM].

Coney, P. J., D. L. Jones, and J. W. H. Monger 1980. Cordilleran suspect terranes.
Nature 288, 329–333.

Connerney, J. E. P., M. H. Acuña, P. J. Wasilewski, N. F. Ness, H. Rème,
C. Mazelle, D. Vignes, R. P. Lin, D. L. Mitchell, and P. A. Cloutier 1999.
Magnetic lineations in the ancient crust of Mars. Science 284, 794–798.

Connerney, J. E. P., M. H. Acuña, P. J. Wasilewski, G. Kletetschka, N. F. Ness,
H. Rème, R. P. Lin, and D. L. Mitchell 2001. The global magnetic field
of Mars and implications for crustal evolution. Geophys. Res. Lett. 28,
4015–4018.

Dohm, J. M., S. Maruyama, V. R. Baker, R. C. Anderson, J. C. Ferris, and T. M.
Hare 2002. Plate tectonism on early Mars: Diverse geological and geophysical
evidence. Lunar. Planet. Sci. 33, 1639 (abstract) [CD-ROM].

Fairén, A. G., M. A. De Pablo, G. Castilla, J. Ruiz, J. Anguita, A. Hácar,
E. Toloba, A. Garcı́a, and S. Rodrı́guez 2002. Three ages for the martian
lithosphere. Lunar. Planet. Sci. 33, 1131 (abstract) [CD-ROM].

Frey, H., K. M. Shockney, E. L. Frey, J. H. Roark, and S. E. H. Sakimoto 2001. A
very large population of likely buried impact basins in the northern lowlands
of Mars revealed by MOLA data. Lunar. Planet. Sci. 32, 1680 (abstract)
[CD-ROM].

Hood, L. L., N. C. Richmond, and J. Halekas 2002. The global distribution of
martian crustal magnetic field: Interpretation and implications. Lunar. Planet.
Sci. 33, 1125 (abstract) [CD-ROM].

Moores, E. M., and R. J. Twiss 1995. Tectonics. Freeman, New York.

Nimmo, F. 2000. Dike intrusion as a possible cause of linear martian magnetic
anomalies. Geology 28, 391–394.

Purucker, M. E. 2002. Why are some martian terranes strongly magnetic and
some non-magnetic? Lunar. Planet. Sci. 33, 1936 (abstract) [CD-ROM].

Purucker, M., D. Ravat, H. Frey, C. Voorhies, T. Sabaka, and M. Acuña 2000. An
altitude-normalized magnetic map of Mars and its interpretation. Geophys.
Res. Lett. 27, 2449–2452.

Saltus, R. W., T. L. Hudson, and G. G. Connard 1999. A new magnetic view of
Alaska. GSA Today 9(3), 1–6.

Schubert, G., C. T. Russell, and W. B. J. Moore 2000. Timing of the martian
dynamo. Nature 408, 666–667.

Sleep, N. H. 1994. Martian plate tectonics. J. Geophys. Res. 99, 5639–5655.

Sprenke, K. F., and L. L. Baker 2000. Magnetization, paleomagnetic poles, and
polar wander on Mars. Icarus 147, 26–34.

Stevenson, D. J. 2001. Mars’s core and magnetism. Nature 412, 214–219.

Strom, R. G., S. K. Croft, and N. G. Barlow 1992. The martian impact crater-
ing record. In Mars (H. H. Kiefer, B. M. Jakosky, C. W. Snyder, and M. S.
Matthews, Eds.), pp. 383–423. Univ. of Arizona Press, Tucson.

Tanaka, K. N., D. H. Scott, and R. Greeley 1992. Global stratigraphy. In Mars
(H. H. Kiefer, B. M. Jakosky, C. W. Snyder, and M. S. Matthews, Eds.),
pp. 345–383. Univ. of Arizona Press, Tucson.

Zuber, M. T. 2001. The crust and mantle of Mars. Nature 412, 220–227.
Zuber, M. T., and 14 colleagues 2000. Internal structure and early thermal evo-
lution of Mars from Mars Global Surveyor. Science 287, 1788–1793.


	FIG. 1.
	FIG. 2. (a)
	FIG. 2. (b)
	ACKNOWLEDGMENTS
	REFERENCES