CMB, Sky Surveys, & Computational Cosmology

[saltydog]

Recently, several groups have endeavored to experimentally map the large-scale structure of the Universe with the Sloan Survey being the most ambitious.
At the same time, we've been able to map the cosmic microwave background to 10^-6 degrees. I believe these two maps agree but I'm confused about it. For example, select a portion of the sky, if the CMB indicates a perturbation, will the exact same portion of the sky in the Sloan map indicate some structure such as a super-cluster? And in an area of CMB uniformity, will the same location in the Sloan map be relatively sparse?

Also, the new field of Computational Cosmology has likewise "created" a Universe from theoretical computations alone.

Is the topology of these three maps in agreement? If so, can anyone add more to the interpretation of the results? The info on the net I've been able to read is a bit confusing.

SD

 

[hellfire]

In my oppinion you cannot expect any matching betwen the primary anisotropies (the ones which were generated during recombination at z = 1100) and the current (z = 0) distribution of matter in the universe. However, there are possible matchings between the secondary anisotropies (generated between 0 < z < 1100) and the large scale structures today.

One of the effects which leads to secondary anisotropies is called "Sunyaev-Zel'dovich effect": CMB photons are inverse Compton scattered when passing through hot ionized gas such as the intracluster gas -- these photons are displaced from their original blackbody curve (2.73 °K) and appear as a deviation to that curve. The intracluster gas contains six times more mass than the visible mass in galaxies and occupies a great volume.

Since the Sunyaev-Zel'dovich anisotropies can be identified in the CMB map (I believe mainly due to its angular scale) it is possible, at least in principle, to correlate these anisotropies with the locations of cluster and superclusters. I think this has already been done for some cases. You can find more information in the following very interesting page I highly recommend you to read: http://astro.uchicago.edu/~laroque/ (and go to "Sunyaev-Zel'dovich effect")

 

[saltydog]

Well hellfire . . . thanks. I'm not afraid of the math but the terminology is tough for me to follow. I'll give it and the reference you gave me some time though as I'm very interested in this. Can you clear up the two points below?

1. The COBE images we're all familiar with: Does the topography of this map match the distribution of matter we now see in the Universe?

2. The models of the Universe emerging for studies in Computational Cosmology: Do the topography of these maps reflect the distribution of matter in the Universe?

Thanks a bunch, you can just answer yes or no for each if you wish and I'll follow up with internet searches.

SD

 

[hellfire]

1. The COBE images we're all familiar with: Does the topography of this map match the distribution of matter we now see in the Universe?

The COBE images depict every deviation of the microwave background from a blackbody curve at 2.73°K. Some of these deviations were produced at z = 1100 (360.000 years after the big-bang). At this time electrons and protons formed neutral hydrogen and the background photons did not find anymore electrons to scatter with. A source of deviation from the blackbody curve is, for example, the redshift of photons when escaping from the potential wells of higher density concentrations. These are called "primary anisotropies". Other deviations were produced afterwards. The background photons have to travel from z = 1100 to our detectors and do interact with electrons, gravitational fields, etc. These are called "secondary anisotropies". The Sunyaev-Zel'dovich effect is a kind of secondary anisotropy, which is produced when the background photons interact with the electrons in the intracluster gas. I think the primary anisotropies are too far away in time to be correlated with the matter distribution. If you had a map depicting only the Sunyaev-Zel'dovich effect, then you could correlate.

2. The models of the Universe emerging for studies in Computational Cosmology: Do the topography of these maps reflect the distribution of matter in the Universe?

Only statistically (statistical methods are used to measure the degree of clustering of galaxies), but not for concrete matter structures.

 

[saltydog]

Thanks for taking the time to explain that to me Hellfire. That was helpful. I once inquired about running the simulations on a desktop. They told me it was too difficult. Need a supercomputer. That'll change. Starting to see multi-processor chips on the market now. Do you know if the calculations are ammendable to parallel processing? I'm thinking of "How to Create a Universe on Your Desktop" appearing in Scientific American soon.

SD

 

[hellfire]

Thanks for taking the time to explain that to me Hellfire. That was helpful. I once inquired about running the simulations on a desktop. They told me it was too difficult. Need a supercomputer. That'll change. Starting to see multi-processor chips on the market now. Do you know if the calculations are ammendable to parallel processing? I'm thinking of "How to Create a Universe on Your Desktop" appearing in Scientific American soon.

I'm sorry but I cannot help you. May be you can find some information here: http://www.mpa-garching.mpg.de/Virgo/

 

[saltydog]

Thanks hellfire, those pictures and movies were amazing! I could easily see myself just staring at them in awe for long periods of time if they were longer. We are so much living in a golden age of Astronomy now. How nice to be experiencing it as it happens!

Salty

 

[Chronos]

Indeed, there are others who agree it is the golden age of astronomy:
Observational Implications of Cosmological Event Horizons
http://xxx.arxiv.cornell.edu/abs/astro-ph/0406099
N-body simulations are used to computationally test models of things like large scale structure formation [e.g., galaxies and galactic clusters]. They require mind boggling computational power and time to do all the iterations without using tons of approximations [which pretty much defeats the whole purpose of doing the study]. You can, however, do simple N-body calculations on a desktop. Just don't expect much. It is fun, you can generate some very nice fractal patterns.