The Physics of CMBR Anisotropies

[meteor]

This paper is quite clear explaining all-you-ever-wanted-to-know about the CMB anisotropies and related phenomena. So, i will be glad that everybody post here their doubts about the CMB power spectrum, E-mode polarization, tensor modes,... The physics behind the CMB anisotropies is quite fascinating

http://arxiv.org/abs/astro-ph/0411049
"The Physics of CMBR Anisotropies"

 

[meteor]

If you ever wanted to know how the CMB power spectrum looks like, I've found it for
you
http://www.jb.man.ac.uk/research/cmb/vsa/CMB_power_spec.html
Observe the flat part of the left known as the Sachs-Wolfe plateau, and the dominant acoustic peak (the Doppler peak). It's important to comprehend how these features arise, there's a lot of information hidden in the power spectrum

 

[hellfire]

This paper is quite clear explaining all-you-ever-wanted-to-know about the CMB anisotropies and related phenomena.

It may be quite clear, but at least for me it is too difficult. I prefer any of the papers on this subject written by Wayne Hu.

So, i will be glad that everybody post here their doubts about the CMB power spectrum, E-mode polarization, tensor modes,...

OK. There are many aspects I do not understand properly, but let's start with one point I have never seen treated in the references: when an electron is trapped by a proton, it has to step down the energy levels until the neutral hydrogen reaches stability. This leads to an electromagnetic emission (as far as I know, usually called inverse photoemission). During the recombination epoch this phenomenon should have taken place massively and the emitted photons should have found free path inmediately, because no scattering followed after recombination (last scattering surface). It seams obvious to me that these emitted photons were not in thermal equilibrium with the radiation of the previous plasma and therefore this emission should be visible as a anisotropy source today. It may be very faint because the photon to baryon ratio is enormous. Is this correct? Is this effect measurable?

 

[Garth]

If you ever wanted to know how the CMB power spectrum looks like, I've found it for
you
http://www.jb.man.ac.uk/research/cmb/vsa/CMB_power_spec.html
Observe the flat part of the left known as the Sachs-Wolfe plateau, and the dominant acoustic peak (the Doppler peak). It's important to comprehend how these features arise, there's a lot of information hidden in the power spectrum

Thank you meteor for these interesting links.

Notice the predicted Sachs-Wolfe plateau does not seem to be there in the actual data;
see http://arxiv.org/abs/astro-ph/0304558 figure 1, right hand diagram.
As they go to the largest angular scales the data points tail off and do not remain on the S-W plateau. The probability of this happening given the best-fit LCDM standard model is two parts in a thousand!

This may be evidence that the universe is not infinite, it can be explained by saying there was just not enough space for these largest anisotropies to form.

But how can this be? The data is also consistent with a flat universe!

This data seems inconsistent until we realize that it is also consistent with conformally flat models.

Therefore the WMAP data may be evidence of a model such as a linearly expanding spherical, "freely coasting" universe, as a cone is conformally flat; or a static cylindrical model as that too is conformally flat.
[You can split them up the time axis and roll them both out flat]

Just a thought.

Garth

 

[meteor]

It may be very faint because the photon to baryon ratio is enormous. Is this correct? Is this effect measurable?

Hi, hellfire. Yes, the photon/baryon ratio was very huge, the cipher that I have is 3*10⁹. So, it's sensible to think that the contribution to the anisotropy was not important. I have searched info about this and have found nothing. So don't take my word for it, i could be wrong

But how can this be? The data is also consistent with a flat universe!

Hi Garth. Don't forget that a flat universe is not necessarilly infinite. It can also be a finite flat hypertorus
In fact there are 18 possible different flat topologies according to this page!
http://luth2.obspm.fr/Compress/jan03_riaz.en.html

 

[hellfire]

Notice the predicted Sachs-Wolfe plateau does not seem to be there in the actual data;
see http://arxiv.org/abs/astro-ph/0304558 figure 1, right hand diagram.
As they go to the largest angular scales the data points tail off and do not remain on the S-W plateau. The probability of this happening given the best-fit LCDM standard model is two parts in a thousand!

As far as I know, the question whether the data from the CMB anisotropies suggests a finite universe is still open. The problem is that on high angular scales the "cosmic variance" problem arises: to define a correlation function and compute the values of the coefficients of the spherical harmonics one needs a statistical sample of points. In the very high angular scales the sample is small, since we have only one universe to get information from. It seams (I do not know exactly how and in which extent) that this leads to this small "dip" in the SW plateau (may be someone could elaborate on this in the light of the formulas (3) - (9) of the paper referenced by meteor).

 

[Garth]

Hi Garth. Don't forget that a flat universe is not necessarilly infinite. It can also be a finite flat hypertorus
In fact there are 18 possible different flat topologies according to this page!
http://luth2.obspm.fr/Compress/jan03_riaz.en.html

That web page is about 2 years old and pre-WMAP data. That data constrained the geometry to flatness to within close limits. The paper "Simulating Cosmic Microwave Background maps in multi-connected spaces" is suggesting possible topologies of nearly flat universes.

There are also conformally flat universes that would explain the data - as I suggested.

Garth

 

[selfAdjoint]

In the perfect case, a torus is locally Riemann flat. I know it doesn't look like it but there is a unique "parallel" to each "line" therough a given point. Where the lines are circles that curve through the hole and around the surface.

 

[Garth]

Thank you, my problem with that though is the WMAP data is of almost isotropic distant cosmological fluctuations. Would not a toroidal topology produce different results in different directions?
Garth