7 February 2013

The Mysterious Moving Rocks of Mars

Posted by Ryan

Figure from Orloff et al., 2011

How did the boulders in the picture above end up in clumps and arcs instead of randomly distributed across the surface? That’s the focus of the paper “Possible Mechanism of Boulder Clustering on Mars” by Travis Orloff, Mikhail Kreslavsky, and Eric Asphaug that is currently In Press in the journal Icarus.

The picture above is from their previous paper from 2011 about evidence for boulder movement at high-latitudes on Mars. They noticed that in many locations with “patterned ground” – a generic term for polygons and other shapes that form in permafrost areas – large boulders seem to have been moved into clumps and clusters that are decidedly non-random. How do they explain these boulder clumps without invoking little green men with a penchant for rock gardens? The secret is ice. Or rather, ices.
But let’s back up. We have patterned ground here on Earth, and Orloff et al. point out that there are several different mechanisms that explain why clusters of stones form in patterned ground. So why invent a new mechanism for Mars? Well for one thing, many of the mechanisms at work on Earth require at least a little bit of liquid water. But more importantly, the boulders that they have found clustered on Mars are huge – several meters across instead of the 1 cm to 25 cm sized stones seen in patterned ground on Earth. Mars has lower gravity, but that’s not enough to explain the clustering of these very large rocks.

Orloff and his co-authors aren’t the first people to tackle this problem, and they run through several other hypotheses that have been proposed to explain the boulder clumps. The first is called gravitational slumping. But this occurs on steep slopes that are not relevant to the clumps discussed in this paper.

A related process is called gravitational creep, and can occur on shallower slopes. But it requires something in the environment to give the rocks a bit of a nudge. On earth, this is generally living things, wind, or water. Orloff et al. argue that these factors are non-existant or very weak in the areas that they are studying, so gravitational creep is probably not a good explanation. Some authors have pointed out that sublimating ice could provide the nudge necessary for gravitational creep, but Orloff et al. point out that to move boulders, you would need to remove meters worth of ice.

Another idea is “dry cryoturbation”, a process where particles of soil are cycled from the edges of polygons in the patterned ground, down the cracks, and then pushed back up to the surface in the center of the polygon. Orloff et al. rule out this hypothesis because they don’t see evidence of boulders being buried by the circulating soil at the edges of the polygons, and they don’t see boulders being uncovered in the center of the frost polygons.

Orloff et al. do grant that “frost heave” is a possible explanation for the boulder clumping. This is a process where ice forms in porous soil, and since ice takes up more space than liquid water, it causes the soil to expand or “heave”. The process requires freezing and thawing, so it is probably not active on Mars today, but it could have occurred in the last few million years, which would be consistent with the age of the clustered boulder terrain. But, the clustered boulders are seen around craters of varying ages, so if frost heave is the culprit, it would require freeze and thaw over long period of Mars history, so Orloff et al. suspect another process may be at work. Also, the problem of the boulders’ size remains: how do you cluster such big rocks?

Figure 2 from Orloff et al., 2013: "Cartoon of mechanism for boulder clustering. (a) Boulders sitting on the ground at different radial positions on polygons. We describe these boulders as: one very large boulder near the edge, one boulder already in a polygon crack, one boulder in the center of the polygon, one boulder near a crack, and two boulders close to each other near another crack. (b) Seasonal CO2 frost forms a layer. (c) The frost layer solidifies into a slab locking in all small boulders (but not the large boulder) while the near surface ice continues to contract beneath the isothermal frost slab. (d) The frost slab sublimates away and the ground starts to heat up. (e) The boulders are free to move with the expansion of the near surface ice. The boulder already in a crack remains in the crack. The boulder in the center of the polygon stays in the center. The one boulder near a crack falls into the crack. The two boulders close to each other both move towards the crack but separation between them increases."

Enter, Orloff et al.’s new idea, illustrated in the figure above. The idea hinges on the fact that on Mars, during the winter at high latitudes, a layer of CO2 frost forms. This CO2 frost seems to form a slab-like layer rather than fluffy snow, and in doing so it locks any boulders on the surface into place. Once the CO2 slab has formed in early winter, the H2O-ice-saturated soil underneath it will cool down and contract, opening up the polygonal cracks of the patterned ground. Then, as spring arrives, the CO2 slab sublimates away and the underlying ground expands again. As it expands, it moves the boulders toward the cracks in the surface. Once the boulders are sitting on the cracks, they are trapped there for all subsequent seasonal cycles.

Orloff et al. calculate that, with this process, it would take a few hundred thousand years to move boulders from the center of the polygons to the edges, which is like the blink of an eye compared to the estimated age of the clumped boulder terrains.
They wrap up the paper with three ways to test this hypothesis: first, their idea would mean that there is a certain size threshold where boulders can’t be moved, so a careful study of the size of clustered boulders could test the prediction. In particular, as you go toward the pole, thicker CO2 ice slabs would form, allowing bigger boulders to be clustered. Second, big boulders should be the first ones to break free from the CO2 slab in the spring, so vents of CO2 gas from beneath the ice should coincide with big boulders. And third, of course, boulders (and the Phoenix lander) should move toward the edges of polygons. Orloff et al. rightly point out that this would be impossible to measure without extremely high resolution orbital images or a dedicated lander mission. To me, this last one is not really a testable part of the hypothesis since the movements are so small, but the other two tests are certainly do-able.

I chose to summarize this paper because it is a good example of how, despite apparent similarities with Earth, when you are dealing with another planet, you always have to be thinking of processes that could be at work there that have no analog here. Maybe frost heaving can explain everything given enough time to work, but Orloff’s frost slab hypothesis seems plausible too, and they propose two practical tests, so I say it is certainly worth keeping in mind as we continue to try to understand patterned ground on Mars.

ResearchBlogging.orgOrloff, T., Kreslavsky, M., & Asphaug, E. (2013). Possible Mechanism of Boulder Clustering on Mars Icarus DOI: 10.1016/j.icarus.2013.01.002