Back in October I summarised the discoveries of large numbers of extremely faint galaxies in the Coma and Virgo clusters. Since then there's been a steady stream of papers on these ultra-diffuse galaxies, at the rate of about one a week. They now seem to be a common feature of many galaxy clusters (e.g. this one).
Although low surface brightness galaxies have been known for many years, these galaxies are much larger - as extended as the Milky Way but a thousand times fainter. The important question is how massive they are. They could potentially be either huge dwarfs (very extended but not very massive) or so-called "crouching giants") (as massive as giant galaxies but with almost all of their mass in dark matter rather than stars). If they were crouching giants, they could be a huge challenge* to cosmological models, which don't predict anything like these objects in such numbers.
* As in, "holy crap what the hell happened ?"
A few very recent developments indicate that these objects are likely to be huge dwarfs. The difficulty has been that to measure the total mass you need to know how fast the stars are moving (to calculate how much mass you need to hold them together). For objects this faint, this is extremely difficult, especially since the little blighters don't seem to have any gas (which is much easier to use to measure motions).
A paper in February of this year used a neat trick to make things easier, measuring the speeds of the globular star clusters of one of these galaxies rather than individual stars. Still difficult, but easier. They found that although the galaxy is extremely dominated by dark matter (around 3,000x as much mass in dark matter than stars - for normal galaxies it's more like 10x), it's just a dwarf galaxy after all.
They also found that they could get a pretty accurate estimate of the total mass from the globular cluster mass - confirming that a relation known to work for ordinary galaxies works even on these very faint objects. Which means we can completely avoid measuring the motions of the stars altogether and essentially just measure the brightness of the globular clusters. Huzzah !
A paper published today uses the globular cluster trick to confirm that another similar object is also a huge dwarf, while a paper from last month shows that these objects are broadly compatible with mainstream theoretical models of galaxy formation. So although we don't have too much data to work with, it seems likely that most of these objects are faint, low mass, but very extended.
Not that this means everything is hunky-dory. Far from it. Although the vast majority of these things don't have any measurable gas content, a very few are extremely gas rich. Why some should have gas (yet apparently are not currently forming any stars) while others have managed to lose their gas completely is a mystery. And still, as a hugely under-cited paper makes clear, the number of these objects is far lower than cosmological models predict. They may not be quite as dramatic as blowing all our ideas out of the water, but they're still bloody interesting little buggers even so.
Subscribe to:
Post Comments (Atom)
-
In the last batch of simulations, we dropped a long gas stream into the gravitational potential of a cluster to see if it would get torn... -
Another day, another paper on how exciting Ultra Diffuse Galaxies are. At first, these large, faint galaxies were just wholly unexpected, a...
-
Of course you can prove a negative. In one sense this can be the easiest thing in the world : your theory predicts something which doesn...
Rhys Taylor One element, I assume being considered, but that I'd like to know more about, is consideration regarding the possibility that those galaxies are in a non "main sequence" stage of their evolution, such as in the process of formation, merging, or tidal disruption caused by other galaxies or dark matter influences.
ReplyDeleteMore generally, as we know with stellar classification evolution is obviously strongly taken into account with the inclusion of the "main sequence", is there a similar scale for Galaxies? Put it more plainly; is there are a "main sequence" in galactic classification for those galaxies that are in the more "steady" phase of their evolution OR is galactic evolution not really much comparable to stellar at all?
Ciro Villa It's complicated. There are the well-known red sequence and blue cloud, which is the equivalent of a galaxy HR diagram :
ReplyDeletehttp://www.txhughes.com/images/sequences.jpg
But if and how galaxies move from one to the other is not known. Stellar evolution appears to be dominated entirely by mass. For galaxies it also depends on mergers (which are at least common - many people would say even the dominate evolutionary mechanism), tidal encounters (which depends on the density of nearby galaxies) as well as ram-pressure gas stripping and other gas-only effects (which depends on the surrounding gas outside the galaxy), and possibly the initial mass function (how many massive stars are formed that can explode/have strong stellar winds which can ionize/blow out the gas completely, which is in turn dependent on the chemical composition of the gas (more metals = more cooling).
And that's the simplified version. Some people think galaxies can move more-or-less freely from the blue cloud to red sequence (by running out of gas for star formation) and even back again (by merging with a gas-rich galaxy or re-accreting material from the intergalactic medium). Others think that movement between the two sequences is a rare and not significant process. The bottom line is that there's no widely-accepted "typical" process of galaxy evolution.
For these UDGs, one idea is that they formed all of there stars early on in a relatively short burst, blowing out all of their gas. That would explain why they're so faint - they just never had the chance to form many stars. The difficulty with this is that they're so dark matter dominated it's not yet clear if the stars/supernovae would have had enough energy to remove their gas. It also wouldn't explain why the number of detected galaxies is so much lower than cosmological models.
But it seems even less likely that these are unstable, transient objects changing from one type of galaxy to another. They have very simple, smooth, spheroidal structures and their globular cluster distribution appears to be quite symmetrical. If they were experiencing significant disruption, they should be messier. An exotic possibility suggested for similar objects in the Local Group was that these aren't stable galaxies at all, but are just "tidal debris" that's evaporating - we happen to see it at the stage where it's still barely detectable. Given the new discoveries of large numbers of similar objects in cluster, that now looks to be very unlikely.
The short answer is that we don't know.