Some of our baryons aren't missing

The missing baryon problem is that Big Bang theory predicts there should be several times more normal (baryonic) matter in the Universe than is actually observed. Where could they be hiding ? The two leading contenders have been in hot, very thin gas in galactic halos, or in huge filaments between galaxies. Hot gas can be detected by X-ray telescopes if it's dense enough, as in the central regions of many galaxy clusters, but it it's too thin then our telescopes won't detect it.

Numerical simulations of the large-scale structure of the Universe have thus far been limited to purely dark matter, because the physics of baryonic matter is too computationally expensive to simulate on a large scale. Fortunately since dark matter dominates the mass, adding the baryons shouldn't be able to change the results regarding the overall structure - and indeed they're very successful at reproducing the complex network of filaments and voids of galaxies (the "cosmic web") that is actually observed. But of course it means we can't really determine from the simulations where that missing gas should be found : filaments or halos, it's anyone's guess.

There have been many, many claims over the years for the direct detection of this "cosmic web" at various wavelengths. Obviously we know the web itself exits, because it's very clearly visible in the distribution of galaxies. But does it contain any of the missing baryons ? Such claims include :
https://arxiv.org/abs/astro-ph/0501335
... from as far back as 2004, who detected hydrogen linking two galaxies, but two galaxies doth not a web make, and :
https://phys.org/news/2008-05-xmm-newton-universe.html
... who found a bridge of X-ray gas between two galaxy clusters, and in particular :
https://phys.org/news/2012-11-planck-filament-hot-gas-linking.html
... who found another, fairly spectacular bridge of hot gas linking two well-separated galaxy clusters. Again though, two clusters doth not a web make.

Then there are claims for the gas being in galaxy halos, such as this one :
https://phys.org/news/2012-09-chandra-milky-halo-hot-gas.html
... which claims there's a large halo of gas around our own Milky way
and this recent claim :
https://phys.org/news/2016-02-fast-radio-discovery-universe.html
... which used measurements of a fast radio burst in a distant galaxy to determine the properties of the intervening matter.

All of these claims are very interesting, but they all have a common problem : small number statistics. Most of them but huge extrapolations from single data points (especially the fast radio burst, which only probes a tiny line of sight) and make wholly unwarranted claims based on limited data ("we have found the missing baryons" - no, you've found some baryons, not all of them, and you don't even really know if they were missing or not).

The problem with these small number statistics is that any individual bridge of gas, either between galaxies or galaxy clusters, could simply be due to tidal interactions pulling out the gas in the galaxies/clusters. And we knew about that gas anyway - it wasn't really "missing" to begin with. What would be much more compelling would be to have evidence of these gas filaments in much greater numbers - if the gas really is primordial from the cosmic web, then it should be found in filaments which are (more or less) everywhere, not just in a few prominent examples.

That seems to be what these independent teams have found. As in the case of that particularly spectacular example, they used data from the Planck telescope to measure the Sunyaev-Zel'dovich effect : the dimming of the cosmic microwave background by this hot, thin gas. But instead of looking for one or two extreme cases where it could be detected directly, they stacked a huge number (>100,000) of galaxy pairs believed to be connected by this cosmic web. Essentially they took the images of multiple galaxy pairs and re-oriented them so that the pairs of galaxies were at the same position, then they added them up. In this way any faint emission between the galaxies that's present but undetectable might reveal itself.

And it did. Convincingly. Stacking has a lot of caveats, but when it works it greatly increases sensitivity. What I particularly like was that they did tests to make sure that their claimed detection, which at 5.3sigma isn't particularly strong, isn't the result of some artifact introduced by this stacking procedure. As a comparison they tried choosing random pairs of galaxies that aren't in the same parts of the web and performing this same stacking procedure, which gave no evidence of the same detection. They also corrected for the fact that individual galaxies will have gaseous halos, but it seems that this signal really is attributable only (or largely) to the cosmic web. It's not likely to be tidal features either, because tidal features usually have tails extending in two opposite directions, which is not observed here.

There are some caveats. The most serious is this :
"the contributions from other types of galaxies, less massive galaxies and galaxies in higher redshift should be present at some level.
So I'd be a little worried that some of the gas in the filaments might be from other galaxies in the web which haven't been accounted for. Additionally, some of this gas is almost certainly from tidal and other interaction features : the problem with stacking is that it washes out all of the details. It will be interesting to see what happens using more limited samples, to see which particular types of galaxies/environments show this signal most clearly.

Of course this doesn't rule out that some of the missing baryons reside in galaxy halos too - again, stacking washes out the details. But it's a very cool result. Does this mean we can legitimately, just this once use that dastardly phrase, "mystery solved" ? I'd say : not quite, but very nearly. I'd like to see more explanations about subtracting the other galaxies present along the web, though these are likely to be too small to make a significant contribution. I think it's safe to stop worrying about the baryons being missing and start worrying instead about exactly where they are.

Rather nice press release here.