WMAP 5 Year Data Release: Cosmological Interpretation

[cristo]

The WMAP 5 year data was released this afternoon. In total, there are seven papers available from the following website: http://lambda.gsfc.nasa.gov/product/map/current/map_bibliography.cfm

One of the papers is entitled "cosmological interpretation," and lists the new values of the cosmological parameters taking into account this new data release from WMAP along with Supernovae and Baryon Acoustic Oscillation data. I've not really read through much of this yet, but from first glance nothing much seems to have changed!

Anyway, I imagine we'll all have more comments to make after digesting some of the information!

 

[Wallace]

Thanks for the heads up. I agree that on first glance there isn't anything qualitatively different compared to previous data releases. LCDM is still a great model with best fit parameters that haven't really changed and extensions such as time varying dark energy are still poorly constrained.

 

[marcus]

Thanks for the heads up. I agree that on first glance there isn't anything qualitatively different compared to previous data releases. LCDM is still a great model with best fit parameters that haven't really changed and extensions such as time varying dark energy are still poorly constrained.

Christo, thanks for the alert! Wallace getting more data that just confirms the WMAP3 results is reassuring and indicates that you astronomers are doing your job right. I'll drink to that! Though at my advanced (but vigorous) age it cannot be ethanolic.

 

[marcus]

Look at the caption to Figure 6 on page 15 of the "cosmology implications" article.
They have almost ruled out that the universe is spatially infinite. they have a 95% confidence interval for Omega which is
[0.9929, 1.0181]

that is where they use all the relevant data WMAP + BAO + SN

WMAP is microwave observation
BAO depends on galaxy surveys. It is baryon acoustic oscillation data.
SN is supernovae.
As they indicate in the abstract, with WMAP alone they get a 95% confidence interval that is almost as good
[0.9915, 1.0175]They also show how the wind is shifting because they publish a lowerbound estimate for the radius of curvature of the universe. I would say that determining the radius of curvature of the universe is likely to become an important activity in cosmology in the next few years.

They also show how the wind is shifting because they publish a lowerbound estimate for the radius of curvature of the universe.
See Figure 2 on page 4

That comes to 23 *(1/0.72)*3.26 billion lightyears
a lowerbound of 104 billion lightyears. In the middle of last year I was saying a 'best fit' estimate derived from Ned Wright's January paper was 130 billion lightyears------not far off the lowerbound they give with 95 percent confidence.

The figure of 130 billion lightyears corresponds to a 3-sphere circumference of 820 billion LY. Length of the longest geodesic---longest straight line, like the equator is on the Earth surface.

Here are some arxiv links
Komatsu et al (cosmology implications)
http://arxiv.org/abs/0803.0547

Ned Wright et al (just because it is Wright)
http://arxiv.org/abs/0803.0577

Hinshaw et al (the basic results, the overview, maps)
http://arxiv.org/abs/0803.0732

Dunkley et al (estimates of parameters)
http://arxiv.org/abs/0803.0586

 

[xantox]

they have a 95% confidence interval for Omega which is
[0.9929, 1.0181]that is where they use all the relevant data WMAP + BAO + SN

..
As they indicate in the abstract, with WMAP alone they get a 95% confidence interval that is almost as good [0.9915, 1.0175]

They are both WMAP + BAO + SN, the second one is a constraint depending on the equation of state while the first is for w=-1. Concerning the finiteness, I would stick with the paper conclusion :

We conclude that, if dark energy is vacuum energy (cosmological constant) with w = −1, we do not find any deviation from a spatially flat universe.

 

[Wallace]

I agree with xantox here, these results don't really give evidence for non-flatness. The 95% confidence interval contains \Omega_k=0. You can't then suggest that flatness is 'almost' ruled out because the 95% interval is not symmetric around 0. If this statement was supported by the data it would sit outside the interval. You can't mentally fit a Gaussian to a confidence interval to try and mine extra data from it (i.e by assuming the values in the middle of an interval are more likely to be the 'real' value).

 

[marcus]

well it's two against one and you guys are more qualified to judge this, so I will go with what you say.

I wasnt imagining any Gaussian shenanigans Wallace. I just thought that 0.9929 is real close to 1.00
and what I have been seeing is a trend----a rightwards or upwards trend in the lower limit of the confidence interval. But I won't argue. If you two see this as confirming the 1.00 idea I won't try to maintain otherwise. At least for now

 

[Wallace]

Sure 0.9929 is close to unity, but so is 1.0181!

I just spoke to a stats guru about this and it wasn't exactly as a had thought. Here is what the conclusion was from the discussion:

If you assume the Universe is not flat then it is more likely to be finite than infinite (i.e. \Omega_k is more likely to be negative than positive), but the difference is very marginal and not worthy of much of a mention. I had thought you couldn't say anything at all but apparently it is reasonable to make some statement about the difference between the two cases.

The above however started from the assumption that the Universe is not flat. If you ask the question 'is the Universe flat' then these results support that statement and the asymmetry of the confidence interval (which is not very asymmetric in any case, only a factor of 2 difference, i.e. flatness sits one third of the way along the interval) does not tell you anything.

 

[jal]

Could someone clarify:
1. "... drag epoch is slightly later than the photon decoupling epoch .."
2. "... the sound horizon size at the drag epoch happens to be slightly larger than that at the photon decoupling epoch .."
3. "In Table 3 we give the CMB decoupling epoch, BAO drag epoch, as well as the corresponding sound horizon radii that are determined from the WMAP 5-year data."

The wording is confusing me since I'm assuming that the radii of the particles would be smaller than the decoupling radii since particles do not move at the speed of light.
Is there a typo?
jal

 

[wolram]

May be cosmologists will use understandable terms when they have some thing important to say.

 

[Chronos]

I'm still firmly in the quantumly fluctuating dead flat camp. Slightly positive today, slightly negative tomorrow - think of it as a bounce.

 

[jal]

As an amateur, I think of the “sound horizon” in an expanding universe the same way that sound would propagate if I had a bell ring inside a container and I would remove the air, (particles). The particles would get further apart and sound would stop propagating.

I was told by Garth
" If you want to find out how crowded the protons were at the Last Scattering Surface of the CMB there is an easier way.
The CMB has been red shifted by ~ 1100 since it was emitted at the LSS. This means that linear distances between representative galaxies were 1100 times smaller then.
The volume of the universe was therefore ~109 smaller than now.
The present baryon density is ~ 10-30 gm/cc which means it was ~ 10-21 gm/cc at the LSS.
This is far more rarefied than in the best laboratory vacuum, not what you would call crowded!"
--------
Therefore, the “bell ringing” would happen when the universe was more dense, (at a greater z), when the universe was filled with hydrogen gas that would be at least as dense as air, which of course would be after “inflation” and before decoupling.
jal

 

[jal]

Hi Chronos!

I'm still firmly in the quantumly fluctuating dead flat camp. Slightly positive today, slightly negative tomorrow - think of it as a bounce.

I think that they left 2 doors open.

p.6
On the other hand, an inflationary expansion may not be the only way to solve cosmological puzzles and create primordial fluctuations. Contraction of the primordial universe followed by a bounce to expansion can, in principle, make a set of predictions that are qualitatively similar to those of inflation models (Khoury et al. 2001, 2002a,b; Khoury et al. 2003; Buchbinder et al. 2007; Buchbinder et al. 2007; Koyama & Wands 2007; Koyama et al. 2007; Creminelli & Senatore 2007).

http://arxiv.org/abs/hep-th/0702165v2
A smooth bouncing cosmology with scale invariant spectrum
Authors: Paolo Creminelli (ICTP, Trieste), Leonardo Senatore (Harvard U.)
(Submitted on 20 Feb 2007 (v1), last revised 1 Nov 2007 (this version, v2))
The model represents an explicit and predictive alternative to inflation, although, at present, it is clearly less compelling.
---------


p.16
There is also a possibility that non-Gaussianity can be used to test alternatives to inflation. In a collapsing
universe followed by a bounce, there exists a duality relation, with regard to the slow-roll parameter, between this scenario and the conventional inflationary scenario.
Inflation models with featureless scalar-field potentials usually predict that PR(k) is nearly a power-law
(Kosowsky & Turner 1995).

http://arxiv.org/abs/0802.1067
The Phase Transition to Slow-roll Eternal Inflation
Authors: Paolo Creminelli (ICTP, Trieste), Sergei Dubovsky (Harvard U., Physics Dept., and Moscow, INR), Alberto Nicolis (Columbia U.), Leonardo Senatore (Harvard U., Physics Dept.), Matias Zaldarriaga (Harvard U., Physics Dept., and Harvard-Smithsonian Ctr. Astrophys.)
(Submitted on 7 Feb 2008)
-----------
http://arxiv.org/abs/0801.0467v1
Large Non-Gaussianity Implication for Curvaton Scenario
Authors: Qing-Guo Huang
(Submitted on 3 Jan 2008)

p.14
The curvature parameter at the beginning of inflation must be below of order unity, as inflation would not begin otherwise. However, it is plausible that begin k was not too much smaller than 1; otherwise, we have to explain why it was so small before inflation, and probably we would have to explain it by inflation before inflation. In that case Ntot would refer to the sum of the number of e-foldings from two periods of inflation. From this argument we shall take begin k _ 1.The reheating temperature can be anywhere between 1 MeV and 1016 GeV. It is more likely that it is between 1 TeV and 108 GeV for various reasons, but the allowed region is still large.
-------
jal

 

[jal]

Follow up on previous post.

p.21
the detection of non-adiabatic fluctuations between matter and radiation has a profound implication for the physics of inflation and, perhaps more importantly, the origin of matter.

Since there are more than just experts reading this, I’m including references and quotes.
Big Bang Nucleosynthesis (BBN) occurs 13 times (for example, on p. 31 and fig. 15, p. 36 and p. 43)
I prefer a scenario without the inflation phase because “inflation” seems to cause more problems than it cures.

Going from a scalar to a quark phase to a hadron (hydrogen) phase seems the simplest. Holding the Baryogenesis or leptogenesis phase in a balanced state seems to be a much simpler solution to obtain a homogeneous and isotropic state and it does not require inflation.
The quarks can circulate in the whole universe, for as long as you like, and bounce around, but no farther than the confinement of the quarks, (to make a proton) occures. When that occurs, you get the phase change and the extra symmetry.
The time line “fine tunning” problems disappear.
==========
references
http://en.wikipedia.org/wiki/Baryogenesis
The next step after baryogenesis is the much better understood Big Bang nucleosynthesis, during which light atomic nuclei began to form.
Nucleosynthesis is the process of creating new atomic nuclei from preexisting nucleons (protons and neutrons).
-------------
http://en.wikipedia.org/wiki/Big_Bang_nucleosynthesis
Big Bang nucleosynthesis (or primordial nucleosynthesis) refers to the production of nuclei other than those of H-1. There are two important characteristics of Big Bang nucleosynthesis (BBN):
• It lasted for only about seventeen minutes (during the period from 3 to about 20 minutes from the beginning of space expansion); after that, the temperature and density of the universe fell below that which is required for nuclear fusion. The brevity of BBN is important because it prevented elements heavier than beryllium from forming while at the same time allowing unburned light elements, such as deuterium, to exist.
• It was widespread, encompassing the entire universe.
The key parameter which allows one to calculate the effects of BBN is the number of photons per baryon. This parameter corresponds to the temperature and density of the early universe and allows one to determine the conditions under which nuclear fusion occurs.
Sequence of BBN
Big Bang nucleosynthesis begins about one second after the Big Bang, when the universe has cooled down sufficiently to form stable protons and neutrons, after baryogenesis.

-------------
http://en.wikipedia.org/wiki/Timeline_of_the_Big_Bang
The inflationary epoch
Between 10-36 seconds and 10-32 seconds after the Big Bang
Main article: Inflationary epoch
The temperature, and therefore the time, at which cosmic inflation occurs is not known for certain. During inflation, the universe is flattened (its curvature is critical) and the universe enters a homogeneous and isotropic rapidly expanding phase in which the seeds of structure formation are laid down in the form of a primordial spectrum of nearly-scale-invariant fluctuations. Some energy from photons becomes virtual quarks and hyperons, but these particles decay quickly. One scenario suggests that prior to cosmic inflation, the universe was cold and empty, and the immense heat and energy associated with the early stages of the big bang was created through the phase change associated with the end of inflation.
----------
http://en.wikipedia.org/wiki/Inflationary_epoch
http://en.wikipedia.org/wiki/Color_confinement
http://en.wikipedia.org/wiki/Quark-gluon_plasma

==========
jal

 

[BigRedJeffro]

What is the total density? From the three year WMAP data it was thought to be 1.02 +or- 0.02. I see numbers like -0.0179 < Omega[k] < 0.0081 which reduces to -0.0178 < Omega[k] < 0.0066 with a 95% Confidence level, but these numbers are so dramatically different it doesn't seem possible that they are talking about the same thing. Also I tried to find the numbers Marcus stated in his post, but they are nowhere to be found in the paper. Did Marcus convert them from something in the paper to get his numbers? His numbers, although not in the paper, are closer to the 1 to 1.04 range that is familiar to me. Thanks for your help in advance.

 

[marcus]

... Also I tried to find the numbers Marcus stated in his post, but they are nowhere to be found in the paper. Did Marcus convert them from something in the paper to get his numbers? His numbers, although not in the paper, are closer to the 1 to 1.04 range that is familiar to me. Thanks for your help in advance.

I will go back and see where I got them. I am glad you are interested in this. I am too. But I respect the consensus of Wallace and Xantox---which says not to overinterpret and still think of the universe as spatially flat.

 

[marcus]

... Also I tried to find the numbers Marcus stated in his post, but they are nowhere to be found in the paper. Did Marcus convert them from something in the paper to get his numbers? His numbers, although not in the paper, are closer to the 1 to 1.04 range that is familiar to me. Thanks for your help in advance.

Look at the caption to Figure 6 on page 15 of the "cosmology implications" article.
They have almost ruled out that the universe is spatially infinite. they have a 95% confidence interval for Omega which is
[0.9929, 1.0181]

that is where they use all the relevant data WMAP + BAO + SN

OK so let's go there. The cosmology implications WMAP5 paper is Komatsu et al.
http://arxiv.org/abs/0803.0547
let's look on page 15. At the caption to Figure 6.

WHOAH! it couldn't be clearer! You should know that Omega_total = 1 - Omega_k

so that if Omega_k is less than zero then the usual Omega is greater than one by the same amount. So what we are looking for is an Omega_k confidence interval and we want to see if it is leaning towards the negative.

"One-dimensional marginalized constraint on
Omega_k from WMAP+HST, WMAP+SN, and WMAP+BAO. We find the best limit, −0.0181 < Omega_k < 0.0071 (95% CL), from WMAP+BAO+SN, which is essentially the same as WMAP+BAO..."

There it is, in black and white!
−0.0181 < Omega_k < 0.0071 (95% CL)
===================

now, out of respect for Wallace and Xantox, we don't want to overstress this and give the idea that it favors a finite universe. The jury is still out on that and we are a long ways from having that issue settled. But as Wallace said, after consulting his stats guru, if the universe is not perfectly flat then it does favor the positive curved spatial finite case.
https://www.physicsforums.com/showpost.php?p=1636860&postcount=8

that is a cautious (reseved, not overly assertive) conclusion because it still allows that the world could be spatially flat. which a lot of people like. it doesn't for a minute rule that out. and then it would be spatial infinite.

but if it should happen NOT to be spatially flat then it would be either positive or negative curved----and the stat guru will go so far as to say that in that case positive is favored, which means finite, with a finite radius of curvature.

so it's not much, but it is something to think about

 

[marcus]

Red,
here is something you should learn by heart if you are at all interested in these matters.
It is the formula they give on page 4, for the radius of curvature.

It is note g to Table 2, and it says that R = Hubble radius/sqrt(|Omega_k|)

Presumably you know that the Hubble radius is about 14 billion LY. It is defined as
c/H₀ where H₀ is the present value of the Hubble parameter.

OK so suppose Omega_k = -0.01
that is well inside the confidence interval, it is one possibility. It corresponds to Omega = 1.01
then the sqrt of the absolute value is 0.1

So then the formula just says that the radius of curvature of space, right now, is
Hubble radius divided by 0.1, that is 14 billion/0.1 or, in other words, 140 billion LY.

What that says is that space, at the present (CMB restframe time) is like a big sphere surface with radius 140 billion LY. That is it is like the 3D analog of the 2D surface of a sphere. It is the "balloon analogy" type of picture where the radius of the balloon is this distance we calculated.

And the radius of curvature is increasing in the same proportion as other largescale distances are increasing according to the observed Hubble Law, namely about 1/140 of a percent every million years.

And the WMAP5 authors, in Table 2 on page 4, give their estimate for the radius of curvature in this case. Which of course we don't know for sure if it is the case.
But they give their estimate, a statistical lower bound. They don't blink.

In my view that is cool, we have a WMAP approved way to visualize the universe in the spatial finite case, if we care to visualize it that way.

So I urge anyone interested to remember the formula for the radius of curvature as the Hubble radius/sqrt(Omega_tot - 1)

 

[BigRedJeffro]

Thanks a ton Marcus!

 

[marcus]

I'm so glad you are interested!

 

[neutralseer]

While we're on the subject of WMAP, check out http://cosmicvariance.com/2008/07/17/a-new-cmb-anomaly/" about a paper by Nicolaas Groeneboom and Hans Kristian Eriksen:

http://arxiv.org/abs/0807.2242"

Groeneboom et al find statistical evidence in the 5-year WMAP data that supports a prediction made by http://arxiv.org/abs/astro-ph/0701357" about a possible cause of the axis of evil. The prediction being that inflation had a preferred direction of, ummm, inflation, that brought about the axis of evil.

Sean Carroll (blog author and one of the authors of the Ackerman et al paper) correctly states that all this is work is very tentative, but it is interesting.

 

[BigRedJeffro]

Hey Marcus, I figured out why I couldn't find your numbers in the paper. We have two different versions of it. You obviously linked your version above, I got my version from the LAMBDA site: http://lambda.gsfc.nasa.gov/product/map/current/map_bibliography.cfm and clicked on the PDF of the Cosmological Interpretation. The version I have has these numbers for Omegak : -0.0178 < Omegak < 0.0066. Now I must ask, which version would you say is more up to date or reliable? The LAMBDA or the arXiv website?

I was rather confused that I couldn't find your numbers, and why you insisted figure 6 was on page 15 when it was clearly on page 16.

 

[BigRedJeffro]

You should know that Omega_total = 1 - Omega_k

Also I have another question:
Now if I were wanting to cite this from a reputable source, where could I find this formula? Not that I doubt you ;) but I will be asked where I got this formula from, and I would rather not say from some guy named Marcus I met online on a message board, haha.

Thanks for all your help and patience with my questions.




Also it has been my understanding from various articles I've been reading by Jeff Weeks that there was a strong belief that the shape of the Universe was the Poincare Dodecahedral Space (PDS), based upon the WMAP data's missing certain larger frequencies and the given range for the curvature from the 3-year WMAP to be 1.02 +/- 0.02. That range went from 1.00 to 1.04 to approximately 0.9934 to 1.0178 (using the LAMBDA version). I read somewhere that the belief that the shape of the Universe was the PDS rested largely on the previous range for the density/curvature of the universe. Now that the range has altered, it seems that the PDS is not as likely to be the shape of the Universe. Has anyone heard of any other proposed models that would replace the PDS? The only other positively curved 3 dimensional manifold I'm really familiar with is the 3-sphere.

 

[marcus]

Also I have another question:
Now if I were wanting to cite this from a reputable source, where could I find this formula?
...

Sorry, I don't know what source you should cite. AFAIK it is just how Omega_k is defined.
The equation you are talking about is true by definition.
There must be some elementary cosmology textbook tht is online where you can find it.
I don't have time to go hunting. Maybe someone else has a ready online reference.You can also deduce it from reading the WMAP results papers. For example Spergel et al
WMAP3 results: consequences for cosmology. They regularly use Omega_k as a convenient synonym for 1 - Omega_tot,
it looks better when used to label graphs, or tables. Neatness.