What does the seasonal cycle detected in Curiosity's record of methane concentration mean? Last week, I published a paper in Nature Geoscience with eight of my colleagues that attempts to answer that question. In the spirit of what I put together for my Nature penitentes paper back in January of 2017, I've decided to summarize the work here. This work was three years in the making, but the story of how it came to pass is much less exciting than with the penitentes - simply a combination of hard work and tenacity by my colleagues and I, like all good science. You can't tell the story of my paper without knowing the story of Martian Methane. So I've put together a primer. For those of you who have been following this closely, skip down to Chapter 2.
Chapter 1: The methane story so far
Whither methane, Mars' most mysterious gas? Since it was first discovered telescopically by Michael Mumma's group in 2003, methane has captivated our interest in the red planet. And with good reason: on the Earth, most methane is produced by biology.
Could the same be true on Mars? That's a complicated question to answer. For starters, there's very little of the stuff to go around. On Earth, methane makes up just under 2 parts per million (ppmv*) of the atmosphere, whereas on Mars the values recorded by different groups place the value at between 0.5 and 50 parts per billion (ppbv*) of the much thinner atmosphere. That allows other processes to have a significant effect. For instance, serpentinization of rocks at depth can also produce enough methane to explain these values, as can seeps of methane stored during Mars' early history either as organic carbon compounds or in methane clathrates. Heck, there's even enough accretion of organic matter via Interplanetary Dust Particles each year to produce 11 ppbv if it all gets converted into methane! None of this even considers processes that remove methane from the atmosphere.
Just a few of the many processes which may be important in supplying to and sequestering methane from the martian atmosphere.
Obviously, to understand which processes operate and how much is produced at different times and in different locations, we need to have measurements. On the Earth, even single-station measurements have high value for linking processes to concentrations. For instance, the so-called Keeling curve shows how carbon dioxide concentration changes in response not only to anthropogenic release, a long-term increase, but also due to uptake by the boreal forests, an annual cycle:
By Delorme - Own work. Data from Dr. Pieter Tans, NOAA/ESRL and Dr. Ralph Keeling, Scripps Institution of Oceanography., CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=40636957
Surprisingly, the orbital measurements (from the Planetary Fourier Spectrometer on Mars Express) and the telescopic measurements from various groups that have been acquired do not tell a coherent story. Some have suggested that's because all the measurements are wrong; pointing out that all the detections have been either close to the noise level of the instruments or are small residuals to large corrections. Others have tried hard to find a way for everyone to be right, positing mechanisms that allow for methane to be destroyed much faster than we currently expect. If you ask me, the jury's still out on this one.
Into this controversy waded the Sample Analysis at Mars (SAM) Tunable Laser Spectrometer (TLS), which would make the first and most accurate in-situ methane measurements beginning in the fall of 2012. Many of us expected a quick resolution to all our puzzles. But mother nature is rarely so quick to give up her secrets. Over the next six years, the story slowly unfolded and we told that story in three separate Science papers. First of all, it seems that the total amount of methane is quite a bit lower than we expected, on average about 0.41 ppbv. But, intriguingly, the amount varies over the course of the year from 0.23 ppbv to 0.65 ppbv:
The seasonal cycle in methane observed by Curiosity's SAM-TLS instrument. The double-humped appearance of the curve is because of Curiosity's equatorial latitude which makes the distance between the sun and Mars much more important than the planet's axial tilt in determining the season.
A seasonal cycle is a big deal. Because methane is estimated to have a 300-year lifetime in the Martian atmosphere, the positive detection of any methane means that there was a release some time in the last few centuries. Indeed, it could mean something completely benign - like a cometary impact. But a cycle means that methane is being released and removed from the atmosphere every single year! That screams out for an endogenous martian process or processes and tells us that whatever is happening we can investigate it today.
Unfortunately, however, we were unable to determine a specific process in that Science paper. Certainly, we had our suspicions - you'll notice that the figure above has cracks, fissures and microseepage along with surface release and uptake. But it turns out that the cycle doesn't correspond neatly to a simple model of what's going on in the surface and subsurface. A more in-depth examination of the underlying physics was needed.
Chapter 2: The NatGeo article
Investigating that underlying physics was my role and is the subject of my paper in Nature Geoscience. It turns out that you can indeed reproduce the seasonal cycle that Curiosity sees. Here's the recipe:
First, take a little methane source, one that's nice and constant, and place it below the annual temperature wave (the depth at which the temperature stays the same year round). You want to use a constant methane source because one that changes with time can be dialed up and down during the year to create any pattern at all. There's no science that way, only madness. Besides, if you put the methane source down below the annual temperature wave then conditions are constant over very very long timescales and there's no reason to expect anything other than a nice constant source.
Next, bury your source beneath some regolith. For our model, we buried the methane deep, 30 m down. The methane will want to diffuse away from the source and in order to get through to the surface, it will need to travel through that regolith. As it travels, the methane will want to stick to regolith grains in a cycle of adsorption and desorption which actually looks a lot like the process of scattering. Like a pile of gravel, the regolith itself has very thin passageways in between grains, likely so small that the gasses collide more frequently with the walls than each other. Making the problem worse, martian regolith has an awful lot of surface area. This has previously been measured as 18,000 to 30,000 square meters per kg (yes, you read that right!) and martian analogue materials examined on Earth can have specific surface areas of as much as 100,000 square meters per kg or more!
This tells us that adsorption and therefore how much time each methane molecule spends on each surface can be awfully important. I had seen this in action before when I looked at water diffusing through regolith in the lab during my PhD at the University of Arizona. Just adding in a cm or two of crushed JSC-1 would increase the time required for water molecules to diffuse to my detector from a few tens of minutes to tens of hours! Luckily, from that work, I had a subsurface model which could be easily adapted from water to methane. This model is able to take into account both adsorption and diffusion at the same time and therefore could properly plot the progress of each methane molecule as it proceeded from the source to the surface.
However, to do this properly, you need to know just how "sticky" your regolith is, something we call the "adsorption enthalpy." We have data from lab experiments on the Earth that say this should be somewhere between 16 and 20 kJ/mol (and certainly no higher than 25 kJ/mol). But we don't know enough about Mars' surface to really say what this value should be - all we know is that the stickiness of martian regolith is not what we expect, even for abundant molecules like water vapor. So instead of specifying the adsorption enthalpy, we let Mars decide by keeping it as a free parameter and solving for the appropriate value. Turns out that's about 32 kJ/mol, more like the calculations done by others in considering plumes which found 37 kJ/mol.
Third ingredient: an atmosphere. In our model we coupled the methane diffusing outwards from the subsurface source directly to a simplified model of the atmosphere. The most important variable here is the eddy diffusivity of the atmosphere which will mix methane away from the source. We can get that number from models, but there's a problem: since we don't know the concentration of methane in the air away from the rover, we can't say how effective that mixing should be! To solve that problem, we came up with an interesting way of looking at the atmosphere: we defined an "effective atmospheric dissipation lifetime" or "EADT" for the emitted gas. If the methane being emitted from the surface is mixing with loads of methane, that material will persist for a very long period, perhaps all the way up to the photochemical lifetime of ~300 years. However, if the external air is methane-poor, the EADT will be low, perhaps as low as the mixing timescale of ~1 day. For our model we found that EADTs of a few 10s to 100s of sols worked best, but we couldn't statistically rule out much longer values.
When your model is well mixed, place into the computer and bake according to the REMS record of temperature at the Gale landing site! Allow everything to come to a steady state and your methane seasonal cycle is ready.
Chapter 3: What does it all mean?!?
I've been asked about this work: so what? Well, there are a few interesting implications. First, we now have a plausible mechanism of producing the SAM-TLS seasonal cycle, which adds strength to our understanding of this effect. We reproduce the curve so well that a phase lag in the data is now explained.
Secondly, our high values of the adsorption enthalpy suggest that the surface is able to interact strongly with the atmosphere. Not only with methane but perhaps also with other volatile materials. For methane it further means that if you had large plumes of material, you could actually have those materials reabsorbed into the soil fairly quickly, limiting how long material spends in the atmosphere. One of the most surprising effects with Mumma's 2003 plumes was that they only persisted for ~120 sols. Perhaps this is a way to put all that methane back into the ground.
Thirdly, we found that our model worked best when operating with a subsurface source of methane. This therefore strengthens the case that the source of martian methane is in the deep subsurface. While we're agnostic about what is producing that methane - be it ancient burial of organics, methane clathrates, water-rock reactions or life - the location of release allows for any of these mechanisms to be the ultimate source.
Chapter 4: Where do we go from here?
Rumors abound of what TGO is seeing, but it sounds as if there may not be much methane in the upper atmosphere, at least during the times they acquired observations. I'm looking forward to see what they will publish and am keen to know if there is any seasonal cycle in methane at altitude. Once we have those measurements, we can begin to piece together the whole story of martian methane. Clearly, adsorption in the subsurface will be an important part of that discussion.
___
*Strictly speaking, these values are in molecules per total molecules of air, a measure sometimes known as a concentration by volume (hence the "v" in the abbreviation) because in a mixture of gasses at low pressure, each species has the same molar volume. This distinguishes our measure from mass-based approaches used in other fields where a kg/kg formalism is simply denoted as ppm and ppb.
No comments:
Post a Comment