Story posted February 08, 2006
Astronomy is an irresistible mix of what can be seen and what can only be imagined. So how does one begin to give definition to something apparently endless and unknowable? One way, according to Visiting Assistant Professor of Physics and Astronomy Joshua Kempner, is to map what you can see and try to infer from that what you can't.
Kempner and student research assistant Eli Sidman '06 are trying to refine and map X-ray images of gasses emitted by galaxy clusters some 10,000 light years away using X-ray images taken with NASA's Chandra satellite. They hope the information will shed new light on how and when clusters form and help give shape to the shapeless - the invisible dark matter and dark energy that scientists believe makes up the majority of the universe.
Kempner recently sat down with writer Selby Frame for a mind-bending conversation about galaxy clusters, cosmic gas and black holes.
SF: First of all, what exactly are galaxy clusters and why are you studying them?
Kempner: Well, clusters are the largest gravitationally bound structures in the universe and so from a cosmological perspective, they can tell us a lot about the parameters of the universe. They form when small clusters of galaxies merge to form larger clusters. The current total number of galaxy clusters discovered is in the neighborhood of between 4,000 and 5,000. These clusters contain hundreds of galaxies, clouds of hot gas, and roughly 80 percent dark matter, the latter being invisible matter that does not emit light.
SF: Tell me more about this dark matter.
Kempner: Sure. Our current best guess for what dark matter is, is something called Weakly Interacting Massive Particles (WIMPs) - essentially, microscopic fundamental particles that interact only gravitationally with other things. These are not made up of protons and electrons or other ordinary matter, but something else entirely. We know it is there because we can see it interacting gravitationally with stars and galaxies ....
SF: ... And with gasses within clusters?
"I'm interested in astronomy for many of the same reasons I'm interested in art. Astronomy can tell us about the universe around us and our place in it. Why do we see what we see in the night sky? How does the universe work? How did it begin? How will it end? These sorts of overall questions help inspire us in our own lives much in the same way art does by helping us learn about ourselves and the human experience."
Sidman received a Burns Student Research Fund grant last summer to work with Kempner on his research, which they co-presented recently at the 2006 Annual Meeting of the American Astronomical Society in Washington, D.C.
Kempner: Exactly. It doesn't emit light, but you can see its effect on X-ray emitting gas - which Chandra now allows us to see in more detail. For example, when two clusters merge together, the particles of gas collide with each other. So, if clusters were just made purely of gas there would be shocks and turbulence throughout the clusters, but they would eventually settle down to form one big, fairly relaxed-looking cluster pretty quickly.
On the other hand, particles of dark matter do not actually hit other particles of dark matter. They pass right through each other. And so it takes a couple of billion years for the dark matter to settle down. So we actually see evidence in some clusters in X-rays of gas being dragged along by something that we do not see and we have to assume that that is dark matter.
SF: So by understanding the gasses, you also learn about dark matter?
Kempner: Hopefully it will add to that body of knowledge, but it's not my primary research question. What we're trying to do is to understand every detail about how galaxy clusters work, and their evolution has mostly to do with the extremely hot gas in them. And by hot, I mean typically about 10 million degrees and extremely low density of about one hydrogen atom per liter.
SF: I'm afraid you've left me at the laboratory door. Do you mind making one of those Earth-bound comparisons for the astrophysically impaired?
Kempner: [laughing] Well, let's see. It is so hot we would have a hard time producing such temperatures on Earth. The surface of the Sun is about 6,000 degrees, so this is a thousand times hotter than that and a trillion trillion times less dense than water. By comparison, the density of what we think of as empty space between the planets in our solar system is about a thousand times higher than the density of this hot gas in a cluster of galaxies.
SF: Wow ... how do you get your mind around all of this? The immensity of it all?
Kempner: I find the easiest thing to do is not to think about it, because it is very easy to get wrapped up in that and to lose sight of the science. I mean, the scales involved are kind of absurd in terms of our daily experience, and if you spend too much time thinking about that it is too easy to get lost.
SF: But it is tantalizing.
Kempner: Yes, it really is.
SF: But to get back to your work, what are you trying to understand about these gasses?
Kempner: The X-ray images tell us where the gas is, but they don't tell us anything about the properties of the gas. The work I've been doing with Eli is looking at the centers of clusters - and clusters typically are about 10 million light years across - so I'm looking at the central 10,000 light years, a pretty tiny region in the center.
Most galaxies - in fact, every galaxy that we know of - has a super massive black hole at its center. A black hole with a billion times the mass of the sun. Those black holes have a tendency to suck in the gas that is around them, and not all of it falls into the black hole. Some of it comes out in very powerful jets. You have a sort of disk of gas swirling in the black hole and then coming out in two jets at the speed of light. These are called active galactic nuclei.
SF: And this is true of all galaxy clusters?
Kempner: Absolutely. Those jets carry a lot of energy with them, and they make it out into the clusters and have a very profound effect on the evolution of the cluster, because carrying all that energy out they have a tendency to heat the gas around them and to push it further out in the cluster. So, what Eli and I have been looking at is in the centers of clusters, which in most cases have active galactic nuclei, we are trying to see how these affect the long-term evolution of the clusters.
SF: Do these jets continue indefinitely?
Kempner: No. These jets turn on for a while and then the black hole uses up all the gas that is around it and they turn off for a while. In some clusters we can actually see the echoes of previous jets, and they never look the same as the current ones. Each rebirth of that active galaxy is different from the previous one because the conditions that created the first one are never quite the same.
Another reason we are interested in jets is because we know that further out in the clusters there are metals like iron. One of the questions we are trying to answer is where do these metals come from? It would seem that these jets should be pushing it out, but in fact we have been finding they are not as efficient as we'd expected at pushing gas around. Somehow the iron has to get from inside the galaxies to out in this gas that exists in-between the galaxies.
SF: Is that where your mapping comes in?
Kempner: Yes, we have been taking data from Chandra and developing maps that combine the X-ray images with measurements of temperature from the spectra of gas. Eli has been taking these raw data through a long process of making them usable, including getting rid of cosmic rays that affect the data. We're measuring spectra at so many different locations within the cluster that we're able to then turn that into an image in temperature, rather than an image in brightness - which is what we normally see.
We are running the data on a program written by John Houck at MIT, which I modified so that it could work on data from other detectors on Chandra. It is running on six Linux stations at Bowdoin, which Larry Hughes in IT helped us to set up. They all work together as if they were a single super computer - it's called distributed computing.
SF: What new data are the maps revealing?
Kempner: Well, as I mentioned, they are revealing that iron concentrations aren't higher, but actually lower, where jets have been. Also, one of the assumptions is that the various properties within the cluster are basically the same all the way around. So if you were to draw a ring around the center you would find the same thing everywhere within that ring.
What we are discovering is that once you look at a high enough level of detail, that is not necessarily true. There are actually more metals on one side than another in many cases. Clusters are not necessarily symmetrical from side to side
SF: So does this firm up a little more of what we know about the universe?
Kempner: Well, it is pretty specialized data, but it's interesting to the handful of us working in galaxy clusters. The main thing, I suppose, is the new questions it spurs. Since we learned that these jets are not very good at pushing metals from the inside of clusters to the outside, then how the heck did that metal get out there? That may not be something we can answer with this generation of X-ray telescopes ...
SF: But it will send you off and running?
Kempner: Yep, I guess it will.