Dried-up river channels on Mars are some of the best evidence that water once flowed on the surface of the red planet, but new analysis of a channel once thought to have been carved by water shows that it was in fact formed from lava. This discovery puts into question the origin of many other channels on Mars.

Water leaves distinguishable traces in the channels that it forms. Braided channels, terracing, islands, tributaries with dead ends and meandering curves are all signatures of the flow of water downstream. However, lava can also create surprisingly similar features, such as in a channel near Mauna Loa in Hawaii, which erupted in 1859. This channel contains a large one-kilometre long island in its middle and terracing on the channel walls, proving that we shouldn’t assume that all channels are formed by water.

Putting this hypothesis to the test, NASA scientists led by Dr Jacob Bleacher of the Goddard Space Flight Center studied a 270-kilometre long section of an apparently water-carved channel on the south-western flank of Ascraeus Mons, one off the three towering Tharsis volcanoes, climbing 18 kilometres high. With images taken by an arsenal of cameras onboard a multitude of spacecraft (Mars Odyssey’s THEMIS thermal camera, Mars Reconnaissance Orbiter’s Context Imager, Mars Express’ High/Super Resolution Stereo Colour imager, and the Mars Orbiter Laser Altimeter that was on the now defunct Mars Global Surveyor), Bleacher announced at the 41st Lunar and Planetary Science Conference (LPSC) in Houston on 4 March that this particular channel turns out to have been formed from lava instead.

At the source of the channel, it does indeed have all the signatures of water – terraces, islands and braids – but at the far end the signs are definitely of lava, even to the point that the channel morphs into a lava tube, with vents where lava has been forced out. If water had been integral to the formation of the channel, there would be clays and sulphate minerals borne out of running water, but according to Bleacher the Tharsis volcanoes act as potent dust traps, literally covering up any evidence for hydrated minerals.

Nor are there any signs that water and lava may have mixed; for instance, on Earth maar craters are steep sided holes in the ground formed from steam explosions as water has come into contact with magma, but no maar craters have been found at Ascraeus Mons. In fact, when one considers the volcanic explanation, it seems there is little evidence for water at all.

“If ice were melted to produce run-off prior to the eruption of lava we would expect to see evidence of the pre-existing fluvial channel being modified or filled in by the subsequent lava flow, which we do not see anywhere along the extent of the feature,” Bleacher tells <I>Astronomy Now<I>. “Because half displays morphologies only seen in volcanic channels, and the other displays features that can form in either volcanic of fluvial channels, it seems most likely to us that the feature is volcanic in origin.”

Clearly this could have important implications for our understanding of Mars’ past climate if channels we thought were carved by water now turn out to have a volcanic origin. Bleacher certainly doesn’t rule out that some, if not many, channels on Mars were created by water; subsidiary evidence such as the presence of hydrated minerals bear this out, along with features such as sharp changes in channel slope called knickpoints and step waterfalls called cataracts, which are seen in fluvial channels on Mars, but not at Ascraeus Mons. However, it does mean that we need to take a closer look before determining the origin of any channels in and around the larger Martian volcanoes in future.

Elsewhere on Mars, high-resolution images taken by the HiRISE camera on NASA’s Mars Reconnaissance Orbiter show a puzzling inconsistency with the movement of sandy ripples on large fields of dunes. At LPSC, Simone Silvestro of the International Research School of Planetary Sciences at G d’Annunzio University in Italy, revealed that pictures of ripples on dark dunes in the Nili Patera region of Mars, taken a few months apart in 2007, indicated that the ripples had migrated seven metres, and that the shapes of the edges of the dunes had changed in that time.

However, in a second study also announced at LPSC, Matthew Golombek of NASA’s Jet Propulsion Laboratory found that ripples around craters in Meridiani Planum had remained stationary for at least 100,000 years. “HiRISE images are so good, you can tell if a crater is younger than the ripple migration,” says Golombek. "There's enough of a range of crater ages that we can bracket the age of the most recent migration of the ripples in this area to more than 100,000 years and probably less than 300,000 years ago."

Why should dune ripples in Meridiani Planum, in Mars’ southern hemisphere, be so different to ripples at Nili Patera in the north? We know that wind still blows in Meridiani Planum, for it has been seen blowing churned up dust in the Mars Exploration Rover Opportunity’s wheel tracks (Opportunity is currently exploring Meridiani Planum). However, Opportunity has also discovered ‘blueberries’ – small pebbles of the mineral hematite just one to three millimetres across – in Meridiani Planum, that formed in liquid water in the past. Despite their diminutive size, Golombek speculates that they may still be too heavy for the faint winds in the thin atmosphere to dislodge. “The blueberries appear to form a layer of armour that shields the smaller sand grains beneath them from the wind,” he says.