Astronomy is too big to fail

Alas the title, "A new astrophysical solution to the Too Big To Fail problem" does not refer to astronomers finding a way to support the banking sector, which would have been more unexpected...

The "Too Big To Fail Problem" is a variant of the "missing dwarf galaxy" problem, where there aren't as many dwarf galaxies detected as expected from simulations. Specifically, the most massive dwarf galaxies should be so large that there's no way - according to the simulations - that their dark matter "halos" can avoid attracting enough gas to start forming stars. It's not just that we're not finding many of the really small dwarfs (that's a problem too), it's that this is a problem even for the bigger guys. It only stops when you get to really massive galaxies.

Supernovae explosions and other processes might be able to explain why the littlest galaxies don't form any stars - a few big blasts could send the remaining gas off into the void, never to be seen again. So they form just a few stars, which eventually die off and thus we can't detect the smallest dwarfs any more. But this explanation shouldn't be possible for the largest dwarfs, which are so massive they ought to be able to hold on to their gas more strongly.

This latest paper uses a bunch of more advanced simulations with better resolution and (they think) treatment of the various processes at work. They make "synthetic observations" of their simulations so they can directly compare their results to real observations. They find that the observations would be underestimating the mass of the dark matter present. Although supernovae can't blast the gas out of the largest dwarfs, it can still cause disruption - the nice thin disc that we see in larger galaxies, and we assume is present in dwarves, gets "puffed up" and becomes substantially thicker. The gas is prevented from collapsing into stars both by its rotation and by this pressure support provided by the various feedback processes from the stars.

This means that the gas discs in the largest dwarves don't extend as far as we expect them to, so their rotation underestimates their true mass. So the dwarf galaxies we detect are actually quite a bit more massive than we've been estimating, if this scenario is correct. They haven't failed, they just aren't quite what we expected from a simpler assumption.

It's a rather technical paper, and personally I think they could have explained their results in a simpler way. I don't think they explain their various scaling relations in a very clear way, making it hard to check if their interpretation of their simulations is sensible. Also (although this isn't crucial), their sample is currently rather small. It would be useful to include more galaxies at different masses to see how well the gas traces the true rotation for other objects.
http://adsabs.harvard.edu/abs/2017arXiv170303810V