Colours indicate the orientation of Mars' paloemagnetic field. Stripes in the field (lower left) do not have a ready explanation. Credit: NASA

After the Mars Global Surveyor (MGS) settled into orbit around the red planet in 1997, a magnetometer on board began sending back measurements that have puzzled planetary scientists ever since. A section of the martian crust appeared to consist of long 'stripes' of iron-bearing minerals permanently magnetized with alternating orientations. Clearly, an ancient dynamo imprinted its field in the rock during the planet's early history. But why the stripes?

Now, Ken Sprenke of the University of Idaho in Moscow, Idaho, has provided a tantalizing theory to explain these stripes: they were created by ancient hotspots beneath the crust. The hotspots would exhume material from the interior that would then 'freeze' with the magnetic field holding sway at the time. Publishing on 20 June in Icarus1, Sprenke points out how the stripes recall a famous, elongated feature made by a hotspot on Earth: the Hawaiian islands. Like teeth on a comb, the hotspots, moving slowly underneath a hard crustal shell, could have left the long, parallel tracks seen in the magnetization maps of MGS. "There could easily have been dozens of hotspots back then," Sprenke says of the first half-billion years of the planet's existence, when the magnetic dynamo — molten iron circulating in the planet's core — was probably active.

Sprenke found that the stripes, tracing out parallel arcs, fall into two families, each defining a distinctive pair of poles. He argues that these ancient magnetic poles — with an episode of polar wandering in between — represent the spin axis for the planet at the time. Finally, Sprenke, needing a mechanism to drag the crust over the hotspots, claims that Mars, in this early epoch of the Solar System, may have captured some satellites. These satellites, he says, could have exerted a gravitational tide that slowed down the crust relative to the hotspots in the interior. Intriguingly, 7 of the 15 giant impact basins on Mars — presumably created as the putative satellites later collided with Mars — fall along equators of the two poles fixed by the magnetic stripes. "We weren't expecting to see that at all, and it just worked out like that," he says.

Scars and stripes

Some scientists say that, like other explanations for the mysterious stripes, Sprenke's will fall short. "I'm certainly very sceptical about this," says Shijie Zhong, a planetary scientist at the University of Colorado at Boulder. Very little about Mars's outer shell is known, he says, and so to assume that tidal forces from satellites could do this — if it were even possible for Mars to capture additional satellites at all — is too big a leap. Zhong has shown that a more likely mechanism for a dragging outer shell would be forces caused by convection in the interior2.

Other scientists find problems with the proposed hotspots themselves. Not only would there have had to be many of them at once, tracing out parallel arcs, but, even more troubling, there is no corresponding topography associated with them, says John Connerney, a planetary scientist at NASA's Goddard Space Flight Center in Greenbelt, Maryland. On Earth, hotspots create island or mountain chains. Sprenke says that later volcanic activity, commencing after Mars's dynamo had stopped, may have erased the surface expression of magnetized stripes that still lurk beneath.

Connerney, who was on the MGS instrument team that made the magnetic map, says that his own team's pet theory for the stripes has also been roundly criticized. "The community is not of one mind on this issue," he says. "There are lot of things about Mars's magnetism that are still unexplained." The MGS team proposed that the stripes were evidence of ancient sea-floor spreading, and a sign that Mars, for a brief period of time, had plate tectonics3. The analogies to Earth were obvious: ships in the 1960s detected alternating magnetic bands of rock at the bottom of the Atlantic Ocean, and this led to the theories of sea-floor spreading that eventually evolved into plate tectonics.

Rob Lillis, a space physicist at the University of California, Berkeley, says the mystery will only be resolved when the MGS data, obtained from an orbit nearly 400 kilometres above the surface, are improved upon. The magnetic stripes at the bottom of the Atlantic were detected by ships, for example, but are not noticed from satellites. Perhaps, Lillis says, the stripes will reveal their true shape with further resolution.

In spite of the lingering mystery, magnetometers have been left off of recent Mars orbiters, often because of the extra hassles in limiting magnetic noise from other onboard instruments. But Lillis will get another chance to find out if the stripes are real. He's a scientist on the Mars Atmosphere and Volatile Evolution (MAVEN) orbiter, due for launch in 2013. It will carry two magnetometers, placed at the end of booms to insulate the instruments from the rest of the spacecraft.

Lillis wants MAVEN to fly in orbits as low as 150 km for several years. NASA also has a need to maintain MAVEN as a communications satellite for other missions and may therefore opt to boost MAVEN to a higher, fuel-conserving orbit. But as long as the team gets some time at a low orbit, Lillis says, "MAVEN will certainly improve the quality of this [MGS] map."