WMAP - Second Release Options

[antoniseb]

Some discussion of the yet to be released WMAP Second Data Release has been going on elsewhere on this forum. I've created this thread as a better labelled place for such discussion.

I have been writing that the second release is overdue, and that there are some hints that there are unexpected things that have contaminated the first release data. Others have pointed out that the data is not clear and needs a lot of time to resolve.

[antoniseb]

So where is the WMAP second release?
It has been a very long time now.
I've been hearing that there is a scandal brewing and everyone involved is sworn to secrecy. This is very strange for big budget public science projects.

As it is, the Planck mission will launch in a year and a half, and start returning more accurate data than WMAP a year or so after that. We might not see the WMAP data before it is obsolete. I would really like to know what the problem is, and why they can't just publish something about what the trouble is.

[tfisher]

Maybe we can guess something about the issue based on the anomalous results from the WMAP first year data release. There was something funny about the low-order coefficients in a spherical harmonic expansion being other than expected for an infinite nearly flat universe. One cute explanation for this was that maybe the universe has a topology where it wraps around just past the point where we could see copies of ourselves in the distance, but that is an explanation miles away from meeting Occam's razor. The Occams razor explanation is that the WMAP team did something wrong trying to reduce foreground contributions to the microwave radiation. For the second year release, they probably want to fix what they did wrong, especially since they wanted to release polarization information which requires much tighter calibration constraints.

My guess is that the WMAP team simply hasn't figured out how to fix the calibration. What this means, I think, is they set their goals too high -- they should have had a goal of releasing less-well-processed data and let the community sort out the calibration mess.

[antoniseb]

What this means, I think, is they set their goals too high -- they should have had a goal of releasing less-well-processed data and let the community sort out the calibration mess.

That's a very plausible explanation, but they've had several years now to either fix it, or announce less lofty goals. Instead we get stonewalling and silence.

[edstrick]

I'm still waiting for the conspiracy nuts to latch on to this...
"messages left written in the sky by the gods..." etc.

Or maybe the space alyenz have been reprogramming the science team to hide something... or was that the men in black?

Where DID I put those sheets of reagent grade tinfoil. I always knew that plain aluminum wouldn't work....

[Guest_Richard Trigaux_*]

I'm still waiting for the conspiracy nuts to latch on to this...
"messages left written in the sky by the gods..." etc. 

Or maybe the space alyenz have been reprogramming the science team to hide something... or was that the men in black?

Where DID I put those sheets of reagent grade tinfoil.  I always knew that plain aluminum wouldn't work....

In the hypothesis that God left us a message, not too obvious but still decipherable once we reach a certain level, this message would rather be decipherable when we reach a given spiritual/moral level, not technological.

And tinfoil helmets are intended to protect from telepathy influence. But if telepathy exists, it is rather of a psychophysics nature, and thus tinfoil is of no help. By the way I never noticed somebody trying to telepathically influence me. Unless they think I am immune to their propaganda...

[Guest_Richard Trigaux_*]

If I remember well the problem with WMAP is as follows.

WMAP observed the cosmic microwave background, trying especially to find irregularities, places where there are more or less. These irregularities can be large or small, what physicist call spatial frequencies, low or high.

After mainstream cosmological model (Big Bang + inflation) these irregularities were produced very early in the hystory of the universe (some billionths of a second after the big bang) and their study could enable us to have precise info on the inflation process and especially of the dimention and shape of the universe.

The simplest geometry for the universe is an hypersphere (the equivalent of a spherical surface, but in 3D, a thing we cannot visualise but it is enough to think that, going in any direction in space, we should come back to our departure point, exactly as in going on any direction on Earth will lead us back to our departure point). More complex gemetries are possible, but they require supplementary causes than just an inflation from a starting point.

Other measures indicated that our universe is nearby euclidian (equivalent of a flat surface) with a precision of 10%. If so, the radius of the hypersphere would be extremely great, much more than the cosmological horizon (13.7 Billion years light, the farthest point we can see with the velocity of light, and more or less equivalent of the horizon on a spherical surface like Earth). More ancient discutions envisioned the case where our univese would be on the countrary smaller than the cosmological horizon, and in this case we should see several views of the same places, as our line of view would turn several times around the universe.



The WMAP results give a lack of low spacial frequencies, and this is a clue for the universe being just larger than the cosmological horizon, which is in contradiction with the previous results.


The problem is that the lower frequencies observed by WMAP are in positions in relation with our solar system, and this rather suggest that some local source contaminates the results, either an instrument error, or possibly a faint cloud of dust or gas near our solar system (such a cloud would be in thermal equilibrium with the cosmic background, and thus emit in the same bands, and be noticeable by no other mean). Things get much more complicated if we consider that the Galaxy (ours and others) is also a large source of radiation polluting the results.


So only the analysis of error sources can indicate us what is the discrepancy: a local effect or a smaller universe than expected. One of the main effort is gathering further observations at various radio bands, to better discriminate local pollutions.

[tfisher]

Here's a link to a popular-press article and the Nature article (free on arxiv.org) discussing the possibility of a peculiar topology for a finite universe that matched the first-year WMAP data.

If you look at what the popular press did to this, announcing it as fact, you can see why the WMAP team doesn't want to put out another data release where the headlines will be "soccer-ball universe reconfirmed".

[Guest_Richard Trigaux_*]

Here's a link to a popular-press article and the Nature article (free on arxiv.org) discussing the possibility of a peculiar topology for a finite universe that matched the first-year WMAP data.

If you look at what the popular press did to this, announcing it as fact, you can see why the WMAP team doesn't want to put out another data release where the headlines will be "soccer-ball universe reconfirmed".

Note that the article itself is good at explaining abstract things to non science-minded people. But as you say the title is rather confusing. They should not treat science news as they treat reality-shows news.


What does this means a dodecahedron in place of an hyper-sphere? What happens at the boundaries between faces: invisible transition or wall-like terrific phenomenon? Are the twelve faces identical reflexion of each other, atom per atom, or with each a different content? Are the twelve faces with identical physical laws? or each of them acquired different physical laws at time of symmetry breaking? Is it possible to test this hypothesis, in the case where boundaries between faces are in sight, we would find several versions of the same map?

[The Messenger]

My guess is that the WMAP team simply hasn't figured out how to fix the calibration.  What this means, I think, is they set their goals too high -- they should have had a goal of releasing less-well-processed data and let the community sort out the calibration mess.

I don't see how it can be fixed, unless and until a causal root is established for the appearence of local contamination.

WMAP CMB data reduction has always relied upon the assumption they can isolate galactic effects, and use a power law function to characterize the clarified primal signal. Local contamination, of any degree from any source, makes these necessary assumptions extremely iffy.

There are some simple aspects of the data which should have been released by now, regardless of data reduction concerns. For example, using the same data reduction techniques as the first year's release, is the data reproducible?

There are at least a dozen papers archived every week that to some degree rely upon WMAP first year data to reach their conclusions. The Planck mission is based upon the assumption WMAP can be improved upon, and many man hours are already committed to this program. At what point does withholding this data, especially if it cannot be reduced without sideways qualifications, become irresponsible?

Of coarse, if a1iens have duct-taped the WMAP team to specimen tables, the delay is excusable.

[Guest_Richard Trigaux_*]

As far I know, we are even not sure that the low frequency discrepancy is a local contamination. It could be real. But it happens that it is position-related to the solar system. From where some doubt. And the need for a better expertise. The second year data implies measures at different frequencies, which may help to rule out galactic noise and eventual local clouds.

Let non-specialist readers get an idea of the complexity of the problem. To make a frequency spectrum (here spatial frequencies) into contaminated data (contaminated by galaxy noise) requires to open a "window" in it (retain only a part of the data, and removing another part). But this very process make appear new frequencies which were not in the original data! There are some mathematic fixes for this, but it quickly gets very complicated. Especially if we are not sure of which part of the data has to be removed or retained.

[ljk4-1]

Paper: astro-ph/0512020

Date: Thu, 1 Dec 2005 10:19:28 GMT (787kb)

Title: Gradient and dispersion analyses of the WMAP data

Authors: K.T. Chyzy, B. Novosyadlyj, M. Ostrowski

Comments: 7 pages, 5 color figures, submitted to MNRAS
\\
We studied the WMAP temperature anisotropy data using two different methods.
The derived signal gradient maps show regions with low mean gradients in
structures near the ecliptic poles and higher gradient values in the wide
ecliptic equatorial zone, being the result of non-uniform observational time
sky coverage. We show that the distinct observational time pattern present in
the raw (cleaned) data leaves also its imprints on the composite CMB maps.
Next, studying distribution of the signal dispersion we show that the
north-south asymmetry of the WMAP signal diminishes with galactic altitude,
confirming the earlier conclusions that it possibly reveals galactic foreground
effects. As based on these results, one can suspect that the instrumental noise
sky distribution and non-removed foregrounds can have affected some of the
analyses of the CMB signal. We show that actually the different characteristic
axes of the CMB sky distribution derived by numerous authors are preferentially
oriented towards some distinguished regions on the sky, defined by the
observational time pattern and the galactic plane orientation.

\\ ( http://arXiv.org/abs/astro-ph/0512020 , 787kb)

[Guest_Richard Trigaux_*]

In short, it was the observation calendar which was a dodecahedron, not the universe.



This is not mocking at the team itself, it is very difficult observation and analysis, and they did a great job. But we cannot expect simple results to come at once from so difficult matters. Only some years ago the idea of hearing at the frequency resonances of the universe to guess its shape and size would have appeared as a weird and pointless speculation.

[ljk4-1]

Paper: astro-ph/0512267

Date: Sun, 11 Dec 2005 02:51:54 GMT (106kb)

Title: Temperature Fluctuations of the Cosmic Microwave Background Radiation: A
Case of Nonextensivity?

Authors: Armando Bernui, Constantino Tsallis, Thyrso Villela

Comments: 5 pages, 2 Postscript figures
\\
Temperature maps of the Cosmic Microwave Background (CMB) radiation, as those
obtained by the Wilkinson Microwave Anisotropy Probe (WMAP), provide one of the
most precise data sets to test fundamental hypotheses of modern cosmology. One
of these issues is related to the statistical properties of the CMB temperature
fluctuations, which would have been produced by Gaussian random density
fluctuations when matter and radiation were in thermal equilibrium in the early
Universe. We analysed here the WMAP data and found that the distribution of the
CMB temperature fluctuations P^{CMB}(Delta T) can be quite well fitted by the
anomalous temperature distribution emerging within nonextensive statistical
mechanics. This theory is based on the nonextensive entropy S_q = k (1 - \int
dx [P_q(x)]^q) /(q-1), with the Boltzmann-Gibbs expression as the limit case q
-> 1. For the frequencies investigated (\nu= 40.7, 60.8, and 93.5 GHz), we
found that P^{CMB}(Delta T) is well described by P_q(Delta T) \propto 1/[1 +
(q-1) B(\nu) (Delta T)^2]^{1/(q-1)}, with q = 1.055 \pm 0.002, which exclude,
at the 99% confidence level, exact Gaussian temperature distributions
P^{Gauss}(Delta T) \propto e^{- B(\nu) Delta T^2}, corresponding to the q -> 1
limit, to properly represent the CMB temperature fluctuations measured by WMAP.

\\ ( http://arXiv.org/abs/astro-ph/0512267 , 106kb)

[The Messenger]

I'm still waiting for the conspiracy nuts to latch on to this...
"messages left written in the sky by the gods..." etc. 

Or maybe the space alyenz have been reprogramming the science team to hide something... or was that the men in black?

Garbage in ... garbage out.