First, what do we know?
There appears to be methane in the atmosphere of Mars, though it's not a lot of methane. Methane doesn't last long in the atmosphere of Mars and it doesn't get made in the atmosphere of Mars. So the methane must come from somewhere down below.
Let's go over that a bit slower, with some caveats and discussion
There appears to be methane in the atmosphere of Mars
Three different groups have now reported this, one using the PFS instrument on Mars Express the other two using ground based instruments (see the first post for more). It's conceivable they are all mistaken. The data they have taken have not yet been formally peer-reviewed, nor, more importantly, have they been published in a way that the community of experts in this field can get their teeth into them. Spectroscopy is a very tricky kind of science, and the data it provides need a lot of work done to them. The Mars Express team has not yet, as I understand it, done all the work that it might in order to be absolutely sure that it's not being misled by other features in the spectrum. The team led by Victor Krasnopolsky, which had the advantage of using the earthly spectrometer best suited to this work (its designer, Jean Pierre Maillard, is one of the team members), has finished its analysis and submitted the data for publication. The team led by Michael Mumma is still working on its data, because Mumma wants to be able to make a really cast iron case for not one but two spectral features due to methane (the other teams are basically working on just one). I think Mumma hopes to have a finished article to send off to a journal -- a prestigious one, I imagine -- with a month or so (he's not being definite, but that's the impression I've got from him).
As I say, it's possible that they're all mistaken. But the fact that they all three found pretty much the same thing independently and that they found it using the best equipment for the job means, I think, that it's reasonable to be relatively optimistic about them being right.
It's not a lot of methane
The Mars Express group and Krasnopolsky's group both put it at about 10 parts per billion. My understanding (which no one has yet corrected, but which should absolutely not be taken to the bank) is that this means about 90,000 tonnes of the stuff.
Methane doesn't last long in the atmosphere of Mars
Models of martian atmospheric chemistry show methane breaking down in a few hundred years -- 300 or so seems to be a consensus. The idea that it would last a lot longer than that is not credible. (Some planetary atmospheres have lots of methane and have had lots of methane since the year dot. But they are chemically quite different from the martian atmosphere.)
It is possible, in fact, that the lifetime is a bit less than this. It's widely believed that the soil on Mars has the power to oxidise organic molecules (a subject I looked at a bit in the fourth post). Michael Mumma wonders whether it is possible that the soil eats up methane, too -- maybe more on some parts of the surface than on others.
If the three centuries lifetime is right, then the source needed to keep the methane levels constant (we don't know if they are constant, as yet) would be about 300 tonnes a year (previous caveat on my numbers re-emphasised). If the lifetime is less, then the source needs to be bigger.
Methane doesn't get made in the atmosphere of Mars
That, at least, is the general belief. It could conceivably be wrong. I've heard rumours that there is a proposed mechanism for making methane from water and carbon monoxide in the martian atmosphere, but this certainly isn't part of the mainstream consensus.
If the methane is made in the atmosphere, it might be more concentrated at altitude than near the surface, and observations might be able to say whether this is true or not. Mumma's measurements allow the possibility of estimating the temperature of the methane, and since we have some sense of the way temperature varies with altitude on Mars, that might be a way of getting a sense of how high up most of the methane is. Worth checking, because an atmospheric source would make this discovery a lot duller (except to atmospheric chemists). But it seems an outside possibility right now.
It's also possible that the methane comes from outer space. One possibility is a comet (see the first post) but it seems something of a stretch. Another possibility is organic carbon that arrives as interplanetary dust and is then turned to methane by sunlight, or by those chemicals in the martian soil. That possibility is also gone into in the fourth post. Again, I think it's unlikely.
So the methane comes from somewhere down below
Having reached that first conclusion, what can we guess about the source?
It might come from living organisms
The surface of Mars is widely seen as both uninhabited and uninhabitable (some people disagree). There has, however, been speculation about niches for life in the subsurface. (I think the first paper on this was "On the possibility of chemosynthetic ecosystems in subsurface habitats on Mars" by Boston, Ivanov and McKay, Icarus, 95, 300-308, 1992). Such bacterial biosystems, or biomes, or whatever you want to call them, might look something like this, and would in all likelihood be based on microbes that made use of hydrogen and produced methane. There's a bit more on such microbes in the second post
It might come from geological processes
Processes in which molten rock or liquid water move through the crust might also give off methane. This was the main subject of the sixth post. Either molten rock or liquid water would be a fascinating discovery in and of itself. As yet I have not come across the much duller alternative that I've been thinking of as a "cold, dry crust" theory -- an account that gets methane from the depths of Mars to the surface without the help of water or magma, just by slowly seeping up. Doesn't mean that such a theory isn't out there, or won't be soon; but magma and water are the current favourites in this race.
It might not be quite as current as all that
The joker in all this comes from the fact that methane molecules can be trapped in little cages of ice; this mixture of ice and gas is known as a clathrate. There are lots of methane clathrates on the earth, and there's no reason to think they couldn't form on Mars, if there was some methane and the right conditions. Unlike methane loose in the atmosphere, methane locked up in clathrates could be stable for a very long time -- tens, maybe hundreds of millions of years. If the methane in the martian atmosphere is coming from the breakdown of clathrates it could be very old methane, and then there would be no need to invoke any of the above methods for making methane on Mars today (though one of them, or something similar, would have had to be going on when the clathrates were originally formed).
If this is the case, then the next question is why are the clathrates breaking down now. There is some evidence that the climate on Mars is changing in very minor ways at the moment (the number of slope streaks in some areas is growing faster than old ones are fading away, and "swiss cheese" patterns in the south polar cap appear to be growing year on year.) Could this change be associated with clathrates giving up their captive methane?
It is becoming quite clear that Mars, like the earth, undergoes cyclical climate change driven by various sorts of wobble in its orbit: ice ages, more or less. If this cyclical change is breaking down clathrates now, it's a fair bet that at other times in the cycle the clathrates build up. (The ice that melts at the end of an ice age on the earth is ice that built up at the beginning). The exact timescale of Martian cyclical climate change is open to argument (there's a little about the possibilities in this post), but it's not long in geological terms -- from 25,000 years or so to maybe a couple of million. So if methane has been produced in the past few cycles, it is still very recent in geological terms.
It's conceivable that the clathrates involved are much older than the current climate cycles. But there's a probabilistic argument which makes that look unlikely, though not impossible. (Or it may just make me look a fool -- always a risk when you meddle with probability, in my experience). The longer a clathrate has been stable, it seems to me, the less likely it is to be becoming unstable right now, all other things being equal. If this methane has been stored for hundreds of millions of years, why should it choose now to come out?
There appear to me to be two answers to that. One is that methane leaks out all the time and that it has been doing so for millions of years, which implies that there is a truly colossal reservoir in the clathrates. The other is that something is warming the clathrates now that wasn't before. The possibilities there would include geothermal heating of some sort (which takes us back to volcanism) or a long-term climate change superimposed on the cycles we're now seeing evidence of. Any of these things would be a pretty exciting discovery in its own right.
(On the earth, by the way, you can tell freshly-made methane from clathrate-stored methane by carbon-14 dating. Carbon-14 is a radioactive carbon isotope that is continuously produced in the earth's atmosphere by cosmic rays hitting nitrogen atoms; because carbon-14 decays, "fresh" carbon-bearing materials will always have more of it than carbon-bearing materials that have been locked away for some time. Unfortunately, the martian atmosphere contains very little nitrogen and thus production of carbon-14 there will be far too low to help.)
So that seems to be the state of play. The methane is either produced by subsurface ecosystems or by one of a few possible geological processes. It's simplest to assume that these processes are going on today, but possibly they went on sometime in the past.
The next post looks at what happens now.