Thursday, May 24, 2012

A Puzzle of Winds

The composition "Seasonal Winds" by my favourite artist and photographer Greg Martin imagines what 400 km/h would look like ripping down Chasma Borealis on Mars. It's too bad such atmospheric phenomena don't exist since they make for such great art and literature. However, the story of how this misconception has propagated itself is an interesting study in the history of planetary science. Go to http://www.experiencetheplanets.com/ to download a high resolution version of the above image.

This spring I set myself a puzzle to solve. As an atmospheric scientist working on Mars, you get used to hearing from other disciplines about the tremendous winds that this planet has to offer. "They happen twice each year as the winter hemisphere moves from North to South and back again - all that CO2 sublimates and rushes towards the other pole." If not this story, then it is tales of some kind of a self-sustaining instability in the global dust storms which occur periodically. However it happens, this tremendous wind has been invoked to explain all manner of features on the surface of Mars.

By the mid 1990s, the existence of such winds was common knowledge extending even into the Science Fiction world. Two highly awarded novels - Greg Bear's "Moving Mars"and Kim Stanley Robinson's "Red Mars" - both have major plot points where such winds are important. In the former, a large pressure wave kicked up by a dust storm and featuring supersonic winds disturbs Casseia Majumdar and her new husband on their honeymoon and, temporarily, provides the environmental conditions necessary for reviving past martian life. In "Red Mars" colonists Nadia and and Arkady are nearly killed when their dirigible is thrown off course in an intense dust storm. When they go outside to repair damage, they can barely stand up, even though the pressure is only 12 milibars (hence the winds must be enormous!) Both of these occurrences are significant because they postulate Mars as it exists today, not some distant terraformed future.

That would be all well and good except these winds don't actually exist! The highest windspeed recorded by our landed missions, our best determinant of ground level windspeed on Mars, was 16 m/s (~58 km/h or 36 mph) for the Phoenix Lander. You might say: "that's all well and good, but Phoenix only lasted 152 sols during northern summer, a fairly quiescent time." You would be right, however, the much longer 2245-sol (~7 Earth Year/3.5 Martian Year) Viking Lander missions saw the entire seasonal cycle several times over and were operating during the two "great dust storms of 1977." Their record shows no windspeed in excess of 19 m/s (68 km/h or 43 mph).



More to the point, absent the pressure wave of an impact, such high speed winds are impossible. To explain why, we need to consider how winds are generated. In simplistic terms, wind occurs when lateral or vertical pressure differences arise in an atmosphere (say by differential heating). Since air is a fluid and cannot resist pressure, as rock does, it flows from regions of high pressure to low pressure. The greater the pressure difference, the greater the windspeed. On Earth, some of the highest winds occur in Hurricanes with pressure drops of close to 100 milibars. Here you might expect to experience speeds in excess of 250 km/h (156 mph).

It is at this point that many will note that 100 milibars on Earth is about 10% of the total atmospheric pressure whereas on Mars, the seasonal variation due to CO2 condensation is closer to 25%. But it's not the magnitude of the pressure change relative to the atmospheric pressure that drives winds, it is the absolute pressure difference. And on Mars with less than 10 milibars available at most locations, you just cannot drive high winds even if you were to create a local vacuum as your low-pressure region. To even achieve the highest figures noted above for Phoenix and Viking, you need to have very specialized and uncommon circumstances. For instance, typical wind speeds at Phoenix were 5 m/s and below (<20km/h; <12 mph).

This fact is obvious to any atmospheric scientist. The theoretical "geostrophic wind" (i.e. wind produced by a difference in pressure on a smooth rotating sphere) is a first year graduate school topic in this field. Not surprisingly, typical wind speeds on Mars were well known in the late 1970s and discussed at length in a dissertation by Bob Haberle (who would go on to help produce and maintain the famous Ames Mars Global Circulation Model). So where did the apparently common knowledge come from for high windspeeds on Mars?

(above) The huge polar erg (or "sand sea") is shown as the dark material surrounding the martian polar cap in this 1971 Mariner 9 image. A closer look (below) shows that an endless procession of dunes make up this dark circle. Special Thanks to Piotr A. Masek for image processing.

It seems this can be traced to observations of sand dunes made by the first Mars Orbiter, Mariner 9. Why sand dunes? Well let me give you some background: in the second world war a scientist named Bagnold studied sand dunes in the Sahara. In addition to sorting out what kinds of dune surfaces were safe for the British to drive over in jeeps and tanks, he produced a model of dune formation and behavior still used today. This model included relationships between the windspeed and the smallest grain size that could be lifted by the wind and skipped along the surface in a process called "saltation." It is saltation that gives rise to the larger self-organizing arrangements of sand that we call dunes. For the Earth this "threshold velocity" is rather modest: 4.5 m/s (16 km/h or 10 mph) is required at 1 m from the surface to initiate saltation and dune formation.

Since dunes are observed on Mars in the present day it was assumed that, at least periodically, saltation must be occurring on Mars. Since Bagnold's equations involve atmospheric pressure and gravity, they can be easily scaled to the Red Planet and used to derive the necessary threshold velocity. However, the figure is staggering - if saltation worked on Mars as it does on Earth, you would need a windspeed of over 69 m/s (248 km/h or 154 mph) to initiate the process!

The problem was that the planetary scientists drew the wrong conclusion from this. Instead of assuming that such winds must exist and persist long enough for dunes to form, those scientists should have examined the assumptions underlying dune formation more closely. Other parts of the theory did not translate as well as they should have. For instance, the minimum size of the particles required was exceptionally large (~1 mm) and the mean hop distance of such saltating particles was 100 times that of their terrestrial counterparts. If all these parameters were correct the mechanical stresses would not permit exposed rock or even the grains themselves to survive on the planet!
The Viking lander Cameras look a little reminiscent of some of the devices for peering out of tanks or submarines. The tough cover was built to protect against high speed particle impacts from dust storms. 

The Viking landers (cameras especially) were designed to withstand this theoretical punishment which was based on the observations of Mariner 9 of the 1971 global dust storm and of the presence of dunes. Perhaps, the Viking designers worried, the loss of the soviet Mars-3 lander which survived only 14.5 seconds after soft landing in that very dust storm was due to such particles?

But since that time we have accumulated more data that has changed our understanding. Dunes work very differently on Mars then they do here on Earth. In addition to dunes and ripples, a third self-organizing system of "mega-ripples" occurs at intermediate sizes. And while one might assume that perhaps the dune we see on Mars were perhaps produced in the past and lie dormant today, the presence of sand-sized particles saltating onto the solar panels of the MER spacecraft suggests that at least some sand processes remain active today.

____
For more information on dunes across the solar system and the "Martian Problem," check out Chapter 9 of Melosh's "Planetary Surface Processes"

No comments:

Post a Comment