Saturday, January 20, 2007

On Galaxies and Globulars

Last night I attended a talk by Dr. Keith Ashman from UMKC. His topic was galaxy formation. The majority of it is things that are covered in an intro astronomy class, but there were a few points that tied everything together that I wasn't aware of before. So I figured I'd share with everyone.

In looking around the universe, it seems that there are two main types of galaxies. The first is the familiar spiral galaxy, like M 83. These galaxies tend to be thin, fast rotating disks, with a bright center region, and beautiful spiral arms defined by the dark dust lanes and glowing regions where star formation occurs.

The second major type is the elliptical galaxy. The nearest major one to us, is M 87, featured on the right. These galaxies are characterized by typically being far more massive than the typical spiral galaxy, rotating slower, having spheroidial shapes, and being devoid of gas and dust that allows for new star formation.

The challange for astronomers is to determine how such things form.

Fortunately, it's actually relatively simple to get the basics of a spiral galaxy. During the first few billion years of the universe, things were expanding. Due to slight density variations, some regions began to collapse gravitationally. As one part would slow down, it would "rub against" the adjacent material which wasn't being slowed as much. This would cause a sort of torque and cause a slow rotation in that contracting cloud.

As something rotating contracts, conservation of angular momentum tends to make things spin faster and flatten out. Thus, from the initial contraction, we can get the flat, fast spinning disks.

What's been harder to understand is how to form the more massive ellipticals. These galaxies present a far greater problem. One way this has been approached was to try to establish a relationship between the two major categoires.

One of the earliest attempts with which I'm familiar for this, is work done by Edwin Hubble. His idea was that elliptical galaxaies evolved into spirals. This was the idea behind what is known as the tuning fork diagram in which he proposed that elliptical galaxies would contract, and in doing so, flatten out to become spirals, travelling from the left side of the diagram to the right, branching off into either the upper branch (spirals without bars in their core) or the lower (barred spirals).

However, this concept had many flaws. The first is that there is no mechanism to cause the collapse. Without something to slow the orbits of stars down so they would move towards the core, ellipticals should not do this. Additionally, this would require spontaneous generation of the gas and dust which is not present in ellipticals, but is in spirals. Lastly, it did not account for the large amount of mass that would disappear.

Thus, Hubble's theory was abandoned. In the mean time, astronomers still use the tuning fork diagram as a convenient way to label galaxies. Our own galaxy is believed to be an Sb or Sc, although I've occasionally heard arguments for the presence of a bar.

The emerging consensus now is that Hubble got it backwards. Instead of ellipticals turning into spirals, astronomers now believe that the collision of the easily formed spiral galaxies results in the formation of an elliptical galaxy. This would seem to make sense since the largest elliptical galaxies are typically found in the center of clusters of galaxies where the chance for collision would be the greatest. This idea isn't exactly new (in fact I think even Hubble entertained this notion), but until recently, there have been fatal flaws in it.

The first is that spiral galaxies have a lot of gas and dust. If you stick two together, you should still have a lot of gas and dust.

Another problem is that of globular clusters. If you go back to the image of M87 I stuck in at the beginning of this post and blow it up, you'll see several speckles in the general glow of the galaxy. Until last night, I hadn't ever paid much attention to them and just assumed they were stars in our own galaxy that were along the line of sight.

But what Dr. Ashman pointed out is that these are in fact globular clusters. Elliptical galaxies tend to have a lot of them. Far more than can be accounted for by the simple addition of two spiral galaxies.

So at this point, any theory seeking to explain ellipticals as the combination of spirals would have to explain where these new clusters appeared from, and where all the gas went.

The explanation finally came about when it was realized that one of these problems could provide the solution to the other. Galaxies colliding tends to be a very violent process. The interaction will cause compression in the gas and dust, which should trigger new star formation, in particular, the formation of new globular clusters.

Fortunately, there are many ways to test this hypothesis. One of the ways is to look at some interacting galaxies and see if we see new star formation. In NGC 4038/39, we can see precisely this effect. There are gigantic regions of star formation, highlighted by the bright blue knots scattered all over.

So clearly, the idea that interaction can cause massive star formation is sound.

But what about all that annoying gas and dust? It turns out that one of the places it goes, is into the formation of new stars. However, with the formation of new massive stars, comes intense stellar winds as well as radiation pressure from the light itself. This should help to blow remaining gas and dust out of the galaxy. And fortunately, we can observe just this in galaxies like M82, shown here on the left. M82 is currently undergoing the long process of being canabalized by it's nearby neighbor M81.

So observations of other galaxies confirm that the idea is at the very least, plausable. When spirals collide, it can both create new globular clusters (in the amounts needed to get up to ellipticals), and blow out massive amounts of gas and dust.

However, Dr. Ashman has been going even further to confirm this theory.

If there are indeed globular clusters being formed from the interaction, then we should see at least two distinct groups: Ones from the original spirals, and then a younger generation formed at the time of the merger. Fortunately, it turns out that this is precisely what we see.

So nearly a century after Hubble began trying to work out the relationship between spirals and ellipticals, it looks like a well supported theory is finally coming around. However, there's still more work to be done.

As images like the Hubble Ultra Deep Field show, even shortly after the big bang, there were already elliptical galaxies. The challange now, is to attempt to discover how they could have formed so quickly. This will most likely come from a better understanding of how spiral galaxies form.


mollishka said...

The Milky Way, by the way, definitely has a bar, according to General Consensus these days. It's just difficult to see since, well, both it and we are in the plane of the galaxy.

Jon Voisey said...

I'd suspected that having a bar was becoming the consensus, but I haven't heard much about it in a few years.

I also seem to remember that it was also difficult to see because we were looking very close to straight down it.

Anonymous said...

Great posting Jon. Very easy for a noob like me to understand and to learn from.

mollishka said...

The Milky Way, by the way, definitely has a bar, according to General Consensus these days. It's just difficult to see since, well, both it and we are in the plane of the galaxy.