Now you don't...

TOWARDS the end of 2011 a large and hugely expensive robotic rover called Curiosity is due to blast off for Mars from Cape Canaveral. If it makes it safely to the planet's surface in August 2012 (getting down from orbit in one piece has not always proved easy for space probes) one of the first things it will do is sniff the air. Its creators, back on Earth, will be straining to see if that air carries a whiff of methane.

In 2004 three different groups said they had seen signs of methane in the atmosphere of Mars. Since, on Earth, almost all atmospheric methane comes from living things, this provided the biggest news from the planet since ALH 84001, a meteorite purportedly bearing Martian fossils, created headlines in 1996.

The shimmering phantasm that is Martian life helps explain why the exploration of Mars soaks up more effort and expenditure than any other branch of planetary science. The barren, arid, radiation-slaked landscape revealed by early probes put to rest the idea that there might be anything living on the Martian surface. But some people hope there are living things below this inhospitable exterior. Methane would be evidence in favour of that. Thoughts of the gas have therefore shaped Curiosity's scientific programme—and in 2016 the first Mars mission to bring together NASA, America's space agency, and ESA, its European counterpart, is to be devoted to analysing trace gases in the atmosphere, with methane the star attraction.

Natural gas?

There have, though, always been doubts about Martian methane. As with ALH 84001 the eye of faith has, in the view of sceptics, been too willing to interpret observations in the most favourable light—and when observations are as strikingly hard to make as those which seek traces of gas in a thin atmosphere so far away, lighting can be everything.

Those views have now been given a particularly thoroughgoing expression in a paper in Icarus by Kevin Zahnle, a planetary scientist at NASA's Ames Research Centre, in California, and two colleagues. This paper is not the last word on the subject, but the case it assembles is a powerful one, ranging from spectroscopic minutiae to a close look at how hard it is to fit the methane data into what is known about Mars as a whole.

One of the things that everyone agrees about methane on Mars is that it has to be short-lived. Though Mars has relatively little oxygen in its atmosphere (just 0.13% of the total) that atmosphere nonetheless provides what chemists call oxidising power, which gives it the ability to pull methane molecules to pieces. The predicted lifetime of Martian methane, based on observations of the Earth's atmosphere and experiments in laboratories, is just 300 years. This is one of the reasons why methane was seen as an exciting discovery—its constant oxidation in the atmosphere means it would have to be replenished by some occult process. Even if there was no life involved that would, at least, require that some novel chemistry was going on.

The problem is that observations of methane on Mars imply that its lifetime must be far shorter than a few hundred years. They indicate that the gas comes and goes on a seasonal basis, with much more of it around at some times and places than others. If methane waxes and wanes with the seasons, then its lifetime must be on a par with the length of a season—several months, not several centuries.

Various suggestions as to how this might be possible have been made. There could be special catalysts in the soil, for example, perhaps created by the static electricity whipped up by dust devils. Dr Zahnle and his colleagues argue that such ideas merely pass the buck. Whatever oxidises the methane would itself get used up in the process, and so would also need to be replenished. To oxidise the methane at a high rate the planet would need to make new oxidising chemicals at that same high rate. There is no evidence of its doing so.

The known way in which oxidising power gets added to the Martian atmosphere is through the destruction of water molecules by ultraviolet light. This creates hydrogen and oxygen. Left to themselves they would recombine, producing no net change, but some of the hydrogen leaks out of the atmosphere into space, leaving oxygen—the paradigmatical oxidiser—behind. This process works too slowly, though, to account for the sudden drops seen in the methane level. Explaining these requires new ways of producing oxidising power which do nothing to alter the balance of other chemicals in the atmosphere. That seems a tall order.

There are other things which might be happening to the methane. It could be stored on the surfaces of minerals, or locked into exotic ices, or even eaten by yet more bugs. But all these, too, are inconsistent with the big picture. Mars has xenon in its atmosphere, and xenon atoms are similar enough to methane molecules that a physical process which locked up methane would lock up xenon too. Yet the xenon persists, airily unfettered. As to methane-eating bugs, a look at the planet as a whole again seems to rule that possibility out. Carbon monoxide offers a lot more energy per molecule to hungry microbes than methane does—but Mars's carbon monoxide level is stable and much higher than the claimed methane level. Martian life might be different from terrestrial life in many ways, but it is hard to conceive of any form of life that would spurn a rich and abundant energy supply in favour of a scarce and less fulfilling one. Darwin would certainly not have approved of such pickiness.

Or just hot air?

Carl Sagan, a devoted student of Mars (and, as it happens, once an editor of Icarus), used to say that extraordinary claims need extraordinary evidence. Dr Zahnle and his colleagues show that a rapidly fluctuating level of methane is an even more extraordinary claim than has been widely appreciated. That puts a pressure on the evidence which it may not be able to bear.

The three sets of observations that purported to see methane on Mars all relied on the absorption lines which the gas imposes on light at specific wavelengths. Two of these sets of observations were made from the Earth, one from an ESA spacecraft orbiting Mars. The spacecraft's instrument is not particularly precise, and although a model of the Martian atmosphere containing methane gives a better match to the spectra it sees than a methane-free model does, the match is not that great, and does not explain all the spectral anomalies in the data. Were these the only observations of methane on Mars, the subject would be viewed with much more scepticism.

Far more precise observations are possible from Earth. But here there is a different problem. Mars's atmosphere is thin, and even the highest levels of methane it might contain are low. The Earth's atmosphere is thick and contains a lot more methane. A ray of sunlight that passes down through the Martian atmosphere, bounces off the planet's surface, goes back up through the atmosphere, traverses interplanetary space and then comes down through the Earth's atmosphere to a mountaintop observatory encounters about 2,000 times more methane molecules on the Earthly leg of its journey than on the two Martian ones put together. That rather tends to swamp the signal.

The only thing that makes the distinctive spectral features of methane on Mars worth looking for is that spectral lines caused by methane on Mars are offset compared with the same lines caused by methane on Earth, thanks to the relative movement of the two planets. Even so, there are a great many absorption lines of other sorts that can lead to confusion, with different lines potentially masking the signal depending on whether Mars is coming or going.

There is also the problem that the carbon in methane comes in different isotopic forms, and the different forms have subtly different spectra. In some circumstances, a feature caused by a rare isotope in the Earth's atmosphere might mimic the expected feature from normal methane on Mars. Dr Zahnle and his colleagues point out various ways that these factors could have misled observers.

The observers are not taking this criticism lying down. Michael Mumma, of NASA's Goddard Space Flight Centre in Maryland, leads the only Earthbound team still gathering the sort of data in which Martian methane has been seen. Having served as a referee on Dr Zahnle's paper, he withdrew from the process, saying there was too much wrong with the paper for it to be fixed. He will be writing a rebuttal soon, he says. Dr Zahnle's team hope that it includes raw data Dr Mumma has not yet published and which might put some of the arguments to rest. Both will look forward to data from Curiosity.

In the meantime, the debate carries a worthwhile scientific lesson in itself. Observations, which to an outsider might sound like simple things, are often remarkably difficult, and depend on complex models to make any sense at all. Thomas Huxley, Darwin's ally in the fight to get evolution accepted, spoke warmly of the facility with which ugly facts can kill beautiful theories. But that fatal ability should not hide the fact that well-applied theories, beautiful and otherwise, can play a crucial role in deciding which observations get treated as facts in the first place.