SAO Guest Contribution

Water on Mars
Malcolm Walter

Malcolm Walter is Adjunct Professor of geology at Macquarie University in Sydney and Director of M. R. Walter Pty Ltd. He has worked for 35 years on the geological evidence of early life on Earth, including the earliest convincing evidence of life. Since 1989 he has been funded by NASA in their "exobiology" and "astrobiology" programs, focussing on microbial life in high temperature ecosystems, and the search for life on Mars. During 1999 his book "The Search for Life on Mars" was published by Allen & Unwin. He has published more than 100 articles and several other books. He also works as an oil exploration consultant and a consultant to museums, and is currently curator of a special Centenary of Federation exhibition on space exploration (for the new Museum of Australia in Canberra, Museum Victoria, and elsewhere).

If you have any questions about this Guest Contribution, don't contact Malcolm directly please.  Instead, post them to the Astronomy News forum and we'll put them together & forward them to Malcolm for comment.

There is water ice at the north pole of Mars, and the dry valleys on the planet have long been considered as strong evidence for the former presence of liquid water at some time in the past. From time to time the evidence for former liquid water is questioned. For example, the apparent lack of incised channels within the valleys has led some scientists to suggest that they formed through "mass wastage", flow of liquefied sediment, rather than water. But even mass wastage involves some water, and last year the Mars Global Surveyor imaged a channel within a valley. The laser altimeter aboard the Global Surveyor has shown that the northern plains of Mars are extraordinarily flat, comparable to the oceanic abyssal plains on Earth. This has given additional credence to the interpretation that there was once an ocean in this area. And recently a truly startling discovery has suggested the recent presence of liquid water at the planet's surface, as I discuss later in this article.

In 1976 Vikings 1 & 2 found powerful evidence that early in its history, Mars was a warmer and wetter place, and therefore potentially habitable.  How can we know the age of features on Mars?  All the inner, rocky, planets in the Solar System formed by the accretion, infall, of rocky debris, from the size of dust grains up to that of small planets.  We can see one result of this on the Moon: lots of impact craters.  The number of impacts was very high as the planets were forming, and then decreased sharply once most of the rocky debris in the Solar System had been swept up. A result of this is that old parts of a planet or moon are heavily cratered, whereas younger parts have fewer and fewer craters.  Most of this record has been obliterated on Earth by later geological processes, but on a dead place like the Moon we can see the record preserved; nothing much has happened there for the last few billion years. Different parts of the Moon have different abundances of craters. Rocks from these areas were collected by the Apollo astronauts and dated back here on Earth. So now we can relate crater abundance to age. This provides us with a crude clock that we can use on other planets, if we make the assumption that cratering history was uniform at least across the inner parts of the Solar System. By counting the numbers of craters on a terrain, the age of its surface can be roughly estimated.

Martian landscapes can be dated to three periods of its history, called Noachian, Hesperian and Amazonian. The absolute ages of these periods are very poorly known, and estimates depend on what assumptions are made about cratering rates. The Noachian is the oldest, and is older than 3 500 to 4 300 million years. This is followed by the Hesperian, older than 1 800 to 3 550, and the Amazonian is the youngest. Noachian landscapes form the highlands of the southern hemisphere, whereas the lowlands of the northern hemisphere are younger. The oldest surfaces have abundant evidence of the former presence of liquid water. First, the craters on these surfaces are deeply eroded, in contrast to those on younger surfaces that are very well preserved. The climate must have been different from now to cause the erosion, probably by rain and wind. Second, there are numerous networks of dry valleys. These are comparable to river valleys on Earth, and though their origin is debated it is generally considered that they were eroded by rivers (of water).  An alternative considered is that they might have formed by ‘mass wasting’, flow of water-logged sediment. However, the recent discovery of a channel within at least one of these valleys strongly supports the river-erosion interpretation.  For a time when the channels were first discovered erosion by liquid carbon dioxide was considered a possibility, but for a number of reasons this interpretation is considered to be untenable.

There are also giant river courses considered to have resulted from brief, catastrophic floods, perhaps when a meteorite impact melted large amounts of permafrost, or underground ice, or sent a shock wave through an aquifer.  These are called outflow channels, and are up to tens of kilometres wide.  They start wide and branch downstream, and where they flow over level plains spread out to be hundreds of kilometres across.  Many start in ‘chaotic terrain’, that looks as if the ground has collapsed.  They are comparable to some giant channels on Earth that formed when natural dams collapsed and lakes drained catastrophically. But some of the floods were 100 times larger than the largest known terrestrial example.

There are some young valleys, on steep slopes of crater and canyon walls, and on the flanks of volcanoes. There are enormous volcanoes, including Olympus Mons, at 27 kilometres high the biggest in the Solar System, and though they are currently dormant or extinct they have been active through much of Martian history. The young valleys have been used as evidence for intermittent warm and wet periods, but they might only indicate local flow from groundwater springs in areas of higher heat flow, such as near volcanoes. The water cycle on Mars is not understood, and there are strongly contrasting interpretations. The presence of ancient river channels suggests that Mars was warm and wet early in its history, but this is hard to explain. The problem is that like all such stars the Sun would have been less luminous at the time the channels formed, more than 3 000 million years ago. Even a very powerful greenhouse effect caused by abundant carbon dioxide in the atmosphere might not have been enough to warm the planet above the freezing point of water. While many scientists consider that the evidence indicates that Mars was warm and wet for an extended period, others suggest that might have happened only briefly and intermittently, throughout Martian history. It can be argued that from time to time giant floods caused by volcanic heating of permafrost flooded the lowlands of the northern hemisphere and produced a temporary ocean.

So there is abundant evidence for the former presence of liquid water, and water ice has been directly observed at the north pole. But estimates of the quantity of water on Mars vary widely.  The leading expert on this subject is Michael Carr of the US Geological Survey. He quotes estimates that range from enough water to cover the whole of Mars to depths of more than 10 kilometres, all the way down to less than 10 metres. The erosional features that we observe seem to require the equivalent of ‘a few hundred metres’ of water. This is an estimate of the total quantity of water. It is not suggested that it was evenly distributed across the planet. It could not have been, because the channels and other erosional features formed on land.  Where is this water now ?  Some was lost by reacting with rocks to form clays and other water-bearing minerals. Some would have been broken down and literally blown away from the upper atmosphere by the solar wind. Some is present as ice at the poles. But most is probably underground, as ice. At high latitudes this permafrost could be hundreds of metres thick. Impacts on Mars during the late stage of planetary accretion would have fractured much of the crust to depths of kilometres, creating a highly porous and permeable ‘regolith’ into which the water would have soaked.

It has been calculated that over the last 10 million years the angle of the spin axis of Mars (the ‘obliquity’) to the ecliptic, the plane of rotation around the Sun, has ranged from 13 degrees to 47 degrees. The obliquity varies chaotically, on a time scale of hundreds of thousands to millions of years. In contrast, the obliquity of the Earth is stabilised by the presence of the Moon. When the obliquity of Mars is at a minimum the poles would have permanent caps of frozen carbon dioxide, because as on Earth little of the Sun’s warmth would reach the poles; when it is at a maximum, the polar caps would melt in summer.  At times of high obliquity the water and carbon dioxide stored at the poles would vapourise and be released into the atmosphere, possibly raising the pressure high enough to make liquid water stable for short times. At such times any subsurface microbiota that might exist could migrate to the surface.  So even in recent times there could have been habitable places on the surface, such as lakes and springs.  Even now, landslides on the sides of volcanoes and in canyons could, for a brief time, allow the exposure of subsurface aquifers and any microbes they might contain.

Earlier this year high resolution images from a camera on the Global Surveyor were released. These show pristine channels only a few tens of metres wide by hundreds of metres long, high on the flanks of craters and valleys. The channels are sharply defined, with no sign of erosive smoothing or of superimposed impact craters, both features suggesting a young age (less than one million years, and maybe much less). They look like gullies eroded by water ( It is difficult to make sense of this. At the current very low atmospheric pressure and low temperatures on Mars (5.6 mb average, compared to 1,000 mb on Earth, and -600C), liquid water is not stable; it will sublime, going straight from ice to vapour. While some observers contend that the channels must have another origin, the similarity to water channels on Earth compels us consider a similar origin.

One possibility suggested by the discoverers is that there have been many momentary releases of liquid water that survived just long enough as a sediment-water slurry to make the channels. We can only speculate about how this might have happened, but earthquakes or distant meteorite impacts might have melted or fractured some of the permafrost, sending pulses of liquid water to the surface for a few seconds.

If these interpretations are correct, new strategies for searching for life are suggested. The channels could be searched for evidence of life brought to the surface from underground aquifers.

We clearly have a great deal yet to learn about the hydrosphere of Mars.

Further Reading:
Michael Carr (1996) Water on Mars Oxford University Press, 229 p.
Malcolm Walter (1999) The Search for Life on Mars Allen & Unwin, 170 p.

© Swinburne Copyright and disclaimer information
Maintained by: Glen Mackie (
Authorised by: Prof. Karl Glazebrook (
Monday, 19-Nov-2007 11:17:07 EST

Back to the Guest Contributions Index