## August 17, 2006

### Dark Matter in the Bullet Cluster

#### Posted by John Baez

Welcome to the n-Category Café!

David Corfield, Urs Schreiber and I have decided to join forces and take over the universe - with Jacques Distler providing invaluable technical support.

But we’re just getting started here…so for now, check out this picture of the "Bullet Cluster" - actually two galaxy clusters colliding at 5000 kilometers per second:

The basic idea: on Wednesday August 15th, a press release came out entitled NASA Announces Dark Matter Discovery. It didn’t say what they found. They’re trying to build up suspense for a mysterious press conference on Monday the 21st. But since one of the folks at the teleconference is Maxim Markevitch, who has found dark matter in the Bullet Cluster, I can guess what this press conference will be about - and I’ll tell you! It’s pretty cool.

Posted at August 17, 2006 11:32 AM UTC

TrackBack URL for this Entry:   http://golem.ph.utexas.edu/cgi-bin/MT-3.0/dxy-tb.fcgi/892

### Re: Dark Matter in the Bullet Cluster

You can read more about this stuff on the Chandra website. It’s just like I guessed, with not much new except some really cool pictures.

Posted by: John Baez on August 22, 2006 2:40 AM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

I’m back from Suzhou - and I’m so lonely I’m commenting on my own blog!

Actually, Toby Bartels raised an interesting question via email, and it deserves comment:

“Are there any indications of dark matter’s interacting nongravitationally?”

On the contrary, they’ve put upper limits on the nongravitational self-interaction of dark matter.

The funny thing is that Markevitch believed so much in dark matter that this is all he wanted to do with the Bullet Cluster. It was Clowe who realized some people would like better proof that dark matter exists - and that eventually led to the NASA press conference.

From the Chandra website:

Question: What did you think when you realized you’d found direct evidence of dark matter?

Markevitch: It was Doug Clowe’s realization. Initially, we were planning to use the lensing maps to answer a more technical question: whether the dark matter particles are perfectly collisionless, which is the current assumption, or they can collide with each other (see our paper 2004 ApJ 606, 819). But Doug looked at the mass/X-ray overlay and realized that it’s also a direct proof of the dark matter existence.

Posted by: John Baez on August 22, 2006 7:33 AM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

Very interesting indeed.

but, but, but …

Why wouldn’t the “blue” just be the ordinary matter that did NOT collide, and the “red” be the ordinary matter that DID collide (and therefore emitted x-ray radiation) ??

Why would ordinary matter that did not collide be dragged down instead of just carrying on undisturbed?

Regards, Hans

Posted by: Hans de Vries on August 22, 2006 2:02 PM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

Why wouldn’t the “blue” just be the ordinary matter that did NOT collide.

As far as I understood from a cursory reading, the blue areas indeed coincide pretty well with the distribution of the galaxies. It is pretty much only the gas (made from ordinary matter) that “collided”. The gas makes up most of the (ordinary) mass.

BUT, the mass of the galaxies is way less than that which must be in the blue areas, according to the measurements of weak lensing of objects behind the cluster. Therefore the rest of the mass in the blue areas is evidently invisible $\simeq$ dark.

Posted by: urs on August 22, 2006 4:01 PM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

Cosmic Variance gave JB credit as a “scientific sleuth” for figuring out the super-secret discovery before the official announcement.

TWF is now one link away from a Slashdotted page, so you may see a traffic spike.

Posted by: Blake Stacey on August 22, 2006 2:49 PM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

Hello!

I’ve been trying to post this over at CV but with no success. My connection seems to be quite poor at this moment… Since you are also discussing some details on this matter, I´ll try here as well. (BTW, John Baez did an outstanding job on figuring out about all this last week!) Here is what I´d like to point out.

Some overall consistency checks on this phenomenon should be performed:

1) The two clusters would show up as outliers from the “fundamental plane” relation of clusters in the space of 2-component (baryonic+DM) virial theorem parameters [Dantas et al. ApJ Letters, 528, L5, 2000, astro-ph/9910541];

2) N-body gravitational simulations of merging of 3-component cluster models (galaxies+DM+ gas) with appropriate masses and orbital configurations should be performed in order to check the overall consistency of the observed dynamical timescales.

3) It would be still valuable to perform similar N-body simulations with gravity law modified to include MOND terms, that is: mixed simulations with DM and MOND in order to verify whether the phenomenon can still be reproduced under (some level of) MOND.

Concerning the DM cross section, I have some old calculations on it. This is a pdf file of a workshop presentation (around 2001, unpublished), where you can find some estimates based on the 2-component virial theorem. The values seem to be higher than those of Markevitch et al., though.

Best wishes,
Christine

Posted by: Christine Dantas on August 22, 2006 4:45 PM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

Dear Christine -

It was fun being a “scientific sleuth” and trying to figure out what that NASA press conference would be about before it happened.

Your work on galaxy simulations could become quite hot now! Analyzing the Bullet Cluster in more detail, and perhaps similar clusters yet to be discovered, could be the best road to understanding the properties of dark matter in more detail. Clowe’s work is clearly just a first step in this direction - he admits as much.

It’s great: what once was just a theoretical entity, is now something whose viscosity we can start to measure!

Models combining MOND and dark matter seem unbearably inelegant to me, but then I’m just a mathematician - if it were up to me to design the universe, I’d have made it a lot simpler, and not nearly as interesting. Maybe not even interesting enough to allow the existence of mathematicians! So, you’re right: such models should also be simulated, to see how well they can fit the Bullet Cluster.

Posted by: John Baez on August 23, 2006 2:25 PM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

Dear John Baez,

Yes, combining MOND and dark matter is awkward, but we all agree that it is part of standard scientific procedure to scrutinize the available models/theories against observation until they can be clearly falsified. And perhaps it is not yet the end of MOND-like theories, although I am not particularly convinced on it.

Concerning the topic of simulations being hot, well, I believe they have been quite hot for a while, and will probably tend to be more and more necessary in the future, in mostly all areas in science.

In any case, I presently lack the necessary resources to run the high-performance simulations that are necessary for this kind of study.

BTW, I hope numerical quantum gravity will be a growing discipline as well. That’s something I’m particularly looking forward.

Best wishes,
Christine

Posted by: Christine on August 23, 2006 6:54 PM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

I hope numerical quantum gravity will be a growing discipline as well.

If just we had any observations to compare such computations to.

Currently the only plausible measurement to eventually maybe compare quantum gravity computations to (that I am aware of) are QCD measurements, as described for instance in hep-th/0003119. This would involve at least quantum gravity in an “auxiliary” AdS space (not supposed to be our spacetime). As far as I can tell from what I read in the news, there seems to be growing activity on this by QCD and lattice people. But I haven’t really followed the details.

Posted by: urs on August 23, 2006 8:18 PM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

Well, I had mostly on my mind those simulations aimed at clarifying major “internal” problems of the theories, as for instance those CDT simulations of hep-th/0404156 .

However, what is desirable of course is to get to the point where one can have quantum gravity simulations realistic enough to allow for comparisons with observations. Given that we do not have a consistent quantum gravity theory (well, perhaps such a theory already exists, although some wouldn’t claim that), I do not know how far simulations based on nebulous grounds would be of any help.

Best wishes
Christine

Posted by: Christine on August 23, 2006 9:02 PM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

BTW, concerning observations (quantum gravity phenomenology), I believe this is a nice review:gr-qc/0412136.

Christine

Posted by: Christine on August 23, 2006 9:07 PM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

Christine wrote:

I hope numerical quantum gravity will be a growing discipline as well.

Urs wrote:

If just we had any observations to compare such computations to.

Among other things we have the inverse square force law, or better yet the Schwarzschild metric, or better yet the quantum corrections to the Schwarzschild and Kerr metrics. We also have the quantum gravity corrections to scattering. And, we have all the phenomena of general relativity. Some of these are actually observed; the rest are uncontroversial things that any decent theory of quantum gravity should get right.

From a perturbative approach all these calculations are in principle straightforward - though in practice getting the exact numbers right is darn tricky, and people got them wrong many times before getting them right! Would you have guessed that the order-hbar quantum correction to the Schwarzschild metric involve the number 62/15?

For nonperturbative approaches like spin foams, which seek to build spacetime “from the ground up”, getting these calculations right is a serious challenge. Dan Christensen and I worked on these things numerically for quite a while, until I got frustrated and gave up. Right now Christensen and Rovelli, Speziale, et al are making the most progress - Christensen is numerically checking their analytical calculations of graviton propagators.

Posted by: John Baez on August 24, 2006 3:03 AM | Permalink | Reply to this

### an analog computers for quantum gravity

If just we had any observations to compare such computations to.

Among other things we have the inverse square force law, or better yet the Schwarzschild metric, or better yet the quantum corrections to the Schwarzschild and Kerr metrics.

We have predictions for effects. But it is hard to imagine a feasible experiment - now or in the foreseeable future - that could make measurements with which to compare these predictions.

That’s what I meant.

If just we had any observations of the quantum corrections to the Schwarzschild metric. Then we could compare them to the calculations which you cite.

That’s why I mentioned AdS/QCD. There they really have measurements that they imagine comparing to some form of “dualized QG”.

Of course, whether or not one is willing to regard a computation in “dualized QG” as a computation in QG is up to taste, I guess.

Given the general lack of feasible experiments in QG, it is at least interesting that one can apparently map much of QCD to a dual QG and pull back along this map all kinds of QG effects, like black hole radiation, for instance, to ordinary laboratory QCD.

As Nastase writes in hep-th/0501068:

Thus RHIC is in a certain sense a [quantum gravity] testing machine, analyzing the formation and decay of dual black holes, and giving information about the black hole interior.

It’s sort of having an analog computer for quantum gravity. Pretty much the next best thing to having QG itself, maybe.

Posted by: urs on August 24, 2006 9:24 AM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

Christine wrote:

Concerning the topic of simulations being hot, well, I believe they have been quite hot for a while, and will probably tend to be more and more necessary in the future, in mostly all areas in science.

Yes, simulations in general have been “hot” ever since computers have gotten good enough. What I meant was something a bit more specific, namely: we’ve entered a phase where understanding a mysterious form of matter that makes up about 25% the universe requires accurate simulations of galaxy cluster collisions. Dark matter is a gaping hole in our understanding of fundamental physics. So, accurate simulations of galaxy cluster collisions are now a top priority for making progress on fundamental physics!

Of course for people like you who were already interested in galactic dynamics, it may be annoying for someone else to walk in the door and say “Hey! Your work is important now!” Despite its pompous name, “fundamental physics” (the quest for the basic laws of physics) is not more important than all the other kinds of physics. But there are a bunch of people who like fundamental physics and feel very frustrated at its slow progress, so to them, this new avenue for progress is very exciting.

And, it’s yet another example of how these days most progress in fundamental physics is coming through astrophysics.

It’s too bad you don’t have the computational resources to tackle these problems as well as you could. Maybe the situation will improve now!

Posted by: John Baez on August 24, 2006 2:41 AM | Permalink | Reply to this

### Re: Dark Matter in the Bullet Cluster

[…] understanding a mysterious form of matter that makes up about 25% the universe requires accurate simulations […]

Indeed, that’s exciting.

By the way, what are the prospects for actually determining special properties of that dark matter. As a theorist, one is dying to know which of the proposed candidates it is, if any. The LSP, or the axion - or something else?

How could one tell? Or at least constrain some possibilities?

Posted by: urs on August 24, 2006 9:34 AM | Permalink | Reply to this

Post a New Comment