Valley Networks: Connecting the Dots

The Viking 1 orbiter, one of two early providers of Martian data.
Credit: NASA

Was it ever warm enough for rain to fall on Mars? That is one of the most vexing questions for scientists who study the red planet.

The question is of critical importance to astrobiology. An early warm, wet Mars could have provided an environment conducive to the development of life. If, on the other hand, Mars has never gotten much above the freezing point, life may have had a more difficult time taking hold.

For many decades, scientists trying to puzzle out Mars’s early climate had only one planet-wide set of data to work with: images taken by the Viking orbiters that visited there in the late 1970s.

One of the most distinctive features of the Martian landscape visible in the Viking images are networks of what appear to be water-carved valleys. These valley networks are present throughout the Martian southern highlands, on heavily cratered terrain that is judged to be some of the oldest on the planet. But the Viking images are not highly detailed, and scientists found that their shortcomings made it difficult to determine just how extensive the valley networks were.

Valley networks are common on Earth as well. They are the branching systems of tributaries that feed creeks, streams and rivers. The complexity of these terrestrial systems and their drainage density, or close spacing, is a strong indication that they were formed primarily by the surface runoff of rainwater.

"The importance of drainage density is that that gives you inferences about climate. Higher drainage densities generally have more contribution from surface runoff and therefore by inference, precipitation," says Brian Hynek, who recently completed his Ph.D. at Washington University and is now a post-doctoral researcher at the University of Colorado at Boulder.

But rainfall is never solely responsible for creating these landforms. Groundwater processes, which soften and undermine the soil, causing it to collapse, also make a contribution. On Earth, although runoff typically dominates, there are places where groundwater processes play the primary role.

On Earth, it is usually easy to tell what combination of processes was involved in any given location. Groundwater-induced collapse, for example, begins at the lowest point in a valley network and works its way upward, while runoff begins cutting its characteristic channels at high elevations and moves downward.

Data from the Mars Global Surveyor (above) suggests extensive valley networks and a warm, wet early Mars. Credit: NASA

On Mars, however, where the valleys networks were formed billions of years ago, it can sometimes be difficult to read the telltale signs – or even to tell up from down – particularly in the Viking images.

Mike Carr, of the United States Geological Survey in Palo Alto, CA, has studied the Viking images extensively. His conclusion: a combination of groundwater and runoff processes is responsible for forming Mars’s valley networks. Although precipitation was likely required, that precipitation may have been in the form of snow, not rain – so Mars may never have been warm like Earth.

In the Viking images, says Carr, the Martian valley networks don’t appear as fully developed as Earth’s. They are short, not heavily branched and not closely spaced. Their drainage density appears very low. If rainwater were involved, one would expect to see much higher drainage densities.

But according to Hynek, data acquired by a more recent Mars mission, the Mars Global Surveyor (MGS), which began orbiting Mars in September 1997, makes it possible to fill in the gaps and to remove some of the ambiguity of the Viking images. These data, he says, reveal that the valley networks are more extensive than previously thought and bolster the argument for rainy days on a warm, wet early Mars.

MGS contains an instrument known as the Mars Orbiter Laser Altimeter (MOLA). MOLA beamed laser pulses at millions of points on the planet’s surface and then measured how long it took for the light to reflect back. Each time measurement was used to calculate the elevation at a specific location. These data let researchers create a highly accurate topographic map of Mars. MOLA was so precise that scientists now have more detailed global topographic information for Mars than they do for Earth.

MOLA data doesn’t exist as images, but rather as a database of numbers. Hynek used this database to create computerized 3D images of Mars that contain valley networks. Over these synthetic images, he superimposed semitransparent photographic images of the same regions, taken by the Mars Orbiter Camera (MOC), onboard MGS. This enabled him to see some of the lights and darks of different geological formations in combination with the shaded relief of his original image.

Finally, he added false color to his images, using different colors to represent different elevations, much like the color coding used on elevation maps of Earth. Because this work was done in software, Hynek was able to assign colors to highlight certain topographic features, specifically the valley networks.

The Mars Orbiter Laser Altimeter (MOLA). Credit: NASA

"That gives you an idea of the slopes and which direction water would flow if it were even there," says Hynek. "Previously people just had images … and had no way to manipulate them. But now we’re able to bring out very subtle variations that haven’t been seen before."

The result was dramatic. What had appeared in the Viking images as poorly developed systems of valleys suddenly jumped out as fully developed networks.

"With the new technique, using the exact same defining characteristics for valley networks, we find about a 10 or more times increase in the number of valleys in any given place in the highlands. And then measuring their length, you find again about an order of magnitude [10 times] increase in the total lengths of valleys. And then from valley lengths and area, you can calculate the drainage density, which is how closely spaced these are.

"The biggest argument against surface runoff has been, ‘Well, we don’t see high drainage density. These are much lower than what we find on Earth, so how can you have precipitation?’ Now we find much higher drainage densities than previously thought. And in light of the new data, we find that the drainage densities are comparable to what we find on Earth. So now the major argument against surface runoff and precipitation is, at least in this data, no longer valid."

Carr isn’t convinced. "I think Brian’s doing great stuff," he says. But even the increased drainage densities Hynek has calculated, Carr points out "are lower than most places on Earth."

Carr isn’t saying that precipitation was completely absent. "You’ve gotta have precipitation of some kind," he says. Even if the valleys were created primarily by groundwater processes, there had to have been some means of recharging the groundwater reservoirs, or the valley-formation process would have shut off too quickly to create the valley networks.

In this image of Mars, drainage networks are seen to feed into a larger river system. Credit: NASA

But, says Carr, "it could be snow; it doesn’t have to be rain." And if it was snow, Mars may never have been all that warm. It’s not that easy, says Carr, to imagine how Mars ever could have been warm enough for rain to fall and for runoff to carve the valley networks. "I’m just relying on the good judgment of the atmospheric physicists who do this kind of thing, the modeling people, you know, very solid atmospheric physicists who say it’s just incredibly difficult to warm Mars up."

Perhaps, counters Hynek. But "the drainage networks we are looking at are 3.7 billion years old. Something that sits on a surface for billions of years, like on Mars, will undergo extensive modification from erosion and deposition." Furthermore, "frequent bombardment from small impacts … pulverize and mix the upper meters to tens of meters of the [Martian] surface and thus erase small valleys."

The possibility that it snowed, rather than rained, on Mars, Hynek says, "cannot be ruled out…. If valley networks were still forming when Mars ‘turned cold,’ the final imprint would possibly be from melting under a snowpack."

Nevertheless, says Hynek, "There is a lot of evidence for ‘contribution’ by runoff and, by inference, precipitation."

It’s unlikely that the debate will be resolved any time soon. Mars does not yield the secrets of its past easily. "I’ve been puzzling about this for so long," says Carr. But no one has yet developed a self-consistent picture of what took place on early Mars that is embraced by the scientific community as a whole. The differing interpretations of the available data, he says, are "just not converging."

Related Web Pages

Mars Global Surveyor Home
Drainage Basins and Channel Incision on Mars
Groundwater Formation of Martian Valleys