Over density of galaxies at z=5.9

[wolram]

http://arxiv.org/pdf/astro-ph/0501478

2 to 4 times more galaxies than expected found at z=5.9
how does this fit with cosmological models?

 

[ohwilleke]

Good catch. I think the implication, not loudly stated in the paper, is that this doesn't square with current cosmological models, or at least, significantly constrains them.

Most notably, the paper notes that the predictions of the cosmic background radiation data does not square with the distribution of high redshift stars observed.

 

[wolram]

Is there any way this could have been missed by CMB predictions?
if not then is it the first sign that these predictions are flawed?
what about reionization?
If these findings are confirmed and an ever more population of galaxies are found beyond z=6 then what?

 

[turbo]

An overdensity is seen in the redshift distribution at z = 5.9± 0.2. The galaxy number density in this redshift range is a factor of two to four high. The extent of the redshift range spanned by this overdensity may be smaller, because the grism spectra give redshifts accurate only to
z ≈ 0.15. Lyman-alpha line emission and absorption may further increase the observed scatter. Hence the overdensity may be even larger in a small volume.

The authors claim a conservative lower limit for the overdensity at 2x, suggest that it might really be around 4x and also throw in these scatter effects that might result in increased overdensity in even narrower z bins. Such strong galactic clumping by the time the BB universe was only a few hundred million years old has got to put some very serious constraints on the standard model, especially the heirarchical model of formation of structure.

Of course, other observations that conflict with the Big Bang are quietly ignored, so this one may be, as well. For instance, the metallicities of quasars observed at of z>6 are in the solar range. For this to happen, somehow an entire generation of massive stars had to go supernova, then that metal-enriched matter (with metallicity comparable to that of our own neighborhood with stars far-removed from first-generation) had to accrete into SMBHs (perhaps several hundred billion solar masses each) to power those quasars. Now here's the part that's hard to believe - all this had to happen in a period of a few hundred million years.

Observations with the Large Binocular Telescope and the upcoming orbiting IR telescope should further extend the redshift lookback (there's a lot of jockeying to observe objects that are the farthest, oldest, etc). We will soon need a universe much older than 13.7Gy to fit everything in.

 

[wolram]

By TURBO-1
Of course, other observations that conflict with the Big Bang are quietly ignored, so this one may be, as well. For instance, the metallicities of quasars observed at of z>6 are in the solar range. For this to happen, somehow an entire generation of massive stars had to go supernova, then that metal-enriched matter (with metallicity comparable to that of our own neighborhood with stars far-removed from first-generation) had to accrete into SMBHs (perhaps several hundred billion solar masses each) to power those quasars. Now here's the part that's hard to believe - all this had to happen in a period of a few hundred million years.
-------------------------------------------------------------------------------------
I am sure an explanation will be found that re re models the BB to fit any
observational data, maybe the U is 18Bys old

 

[Chronos]

I don't see any dramatic findings in this paper. Judging by the conclusions, neither do the authors. They only suggest interesting possibilities. For example:
According to current theories it would be hard to produce a net overdensity of a factor of 4 over such a large volume, while an overdensity of two could be due to the presence of a bright galaxy in the UDF
In another section they note
This suggests that a factor of two overdensity should be typical in the observed volume.
These comments do not elicit feelings that anything remarkably unexpected has been found. The authors never suggest the factor of 4 overdensity has any compelling support, just that it's a possibility. In another section the authors note:
As a result of the long redshift extent of the overdense region (an order of magnitude larger than the correlation length of ~10 Mpc), the corresponding mass overdensity is negligibly small - 1.012.
Nothing remarkable there either. Another notation reports:
The z ~6 galaxies avoid the eastern corner of the UDF. A two-dimensional KS test shows that the probability of this configuration by chance is about 5%.They further note: Even with just 15 galaxies, the test
gives only a 5% chance that they could be so arranged due to chance.
5% is not exactly a statistically compelling figure, nor is the fact there are only 15 galaxies in the sample group. This is no stake in the heart of current cosmological models.

 

[Chronos]

Good catch. I think the implication, not loudly stated in the paper, is that this doesn't square with current cosmological models, or at least, significantly constrains them.

Most notably, the paper notes that the predictions of the cosmic background radiation data does not square with the distribution of high redshift stars observed.

I agree. The implications current models are inconsistent with observation are not loudly stated because they are not soundly supported. I don't follow where you get the CMB / high redshift stars connection. Individual stars cannot be spectrally resolved at z~6. The only mention of stars I noticed were candidate objects selected by color that proved to be nearby dim stars along the line of sight to the more distant quasars. This data was ignored in the study.

 

[turbo]

I don't see any dramatic findings in this paper.

Here is a quote from the paper regarding other findings of overdensities of galaxies at high (z~5.7-5.8) redshift in the same region.

"The overdensity at z=5.9 is also supported by complementary data on Lyman-alpha emitters from CTIO by Wang, Malhotra & Rhoads (2005) who have imaged a large area, 36’ = 12.9 Mpc on a side, including the Hubble Ultra Deep Field. They show that the HUDF sits on the edge of a much larger scale (> 3 Mpc) structure traced by Lyman-alpha emitters at z=5.7-5.77. Wang et al. 2005 report roughly a factor 3-4 overdensity in the Lyman-alpha emitters at z ≈ 5.8, compared to other studies of Lyman-alpha emitters at z=5.8 (Rhoads et al. 2003, Rhoads & Malhotra 2001, Hu et al. 2003, Malhotra & Rhoads 2004)."

Another notation reports:
The z ~6 galaxies avoid the eastern corner of the UDF. A two-dimensional KS test shows that the probability of this configuration by chance is about 5%.They further note: Even with just 15 galaxies, the test
gives only a 5% chance that they could be so arranged due to chance.
5% is not exactly a statistically compelling figure, nor is the fact there are only 15 galaxies in the sample group. This is no stake in the heart of current cosmological models.

Only a 5% chance that they could be arranged that way by chance... Hmm, that sounds like a 95% probability against the observed clustering happening by chance. When I am offered odds of 20:1, the payoff has got to be very large to make me bet on the losing side.

 

[marcus]

here is a possibly related paper
http://arxiv.org/abs/astro-ph/0501171
and Ned Wright's comment on the finding at his website
http://www.astro.ucla.edu/~wright/cosmolog.htm

"Cosmic Ripples Seen by Galaxy Surveys

11 Jan 2005 - Both the Sloan Digital Sky Survey and the 2 Degree Field Galaxy Redshift Survey reported the discovery of features in the distribution of nearby galaxies that correspond to the oscillations seen in the anisotropy of the Cosmic Microwave Background for several years. The overall statistical significance of this result is good but not great: 3.5 standard deviations. But observations of these ripples provide two valuable new constraints on cosmological models, and verify the current Lambda-CDM model of the Universe. The detection of these ripples is shown at right in a version of Figure 3 from a technical paper describing these results. It gives a matter density in gm/cc that agrees with the value found by WMAP. Both WMAP and the SDSS measure this density to a precison of 8% and their values agree to within 5%. Combining the CMB and SDSS data gives an improved limit on the total density of the Universe: Omega_tot = 1.01 +/- 0.009. If Omega_tot = 1, the Universe is flat; if Omega_tot > 1 the Universe is closed; while if Omega_tot < 1 the Universe is open."

I'm not saying ned wright necessarily has the correct take on it. but what he said about narrowing the constraint on the model is similar to what was already said in this thread.

It used to be that omega was estimated 1.01 +/- 0.01
so the error bar was from 1.00 to 1.02
(in words, universe could be exactly flat but could also be closed and nearly flat.)

Now, if what ned wright says is correct the error bar is narrowed down to this
1.01 +/- 0.009 which means omega is expected to be in the range
from 1.001 to 1.019

if Omega is in fact in that range then the universe cannot be infinite

at least I think that is the conclusion
certainly an attention grabber.

the title of this paper is
"Detection of the Baryon Acoustic Peak in the Large-Scale Correlation Function of SDSS Luminous Red Galaxies"

wolram I believe your paper is about a related study but at a larger redshift, I don't know to what extent the results are compatible, but it all seemed on the same general topic, thanks for flagging it!

 

[wolram]

Thanks for the links MARCUS, although most seem to think that
cosmology is almost complete i still have doubts, as for the U
being finite nahhhhh.

 

[turbo]

In this paper, Alain Blanchard suggests that the overdensity of high-redshift clusters observed by Chandra and XMM fits the Einstein de Sitter model and does not square with the concordance model.

http://xxx.lanl.gov/abs/astro-ph/0502220

 

[hellfire]

It is very unclear in which extent this "overdensity" can be considered to be relevant. It is explained in the paper that the sampe is taken centered on a very visible galaxy, with a highly biased halo (halo occupation by several galaxies...?). The conclusion part seams to be obscure to me.

On the other hand it is known that we have still a lot to discover about this epoch of the universe: the UV emission from the detected objects is not sufficient to reionize the universe. Also the epoch of z ~ 7 is more or less the onset of the galaxy formation and the role of mergers seams not to be fully understood.

In this paper, Alain Blanchard suggests that the overdensity of high-redshift clusters observed by Chandra and XMM fits the Einstein de Sitter model and does not square with the concordance model.

http://xxx.lanl.gov/abs/astro-ph/0502220

Independently of the conclusions of this paper, it has to be noted that there the mass of the intracluster gas generating x-rays is used to infer about the total mass density of the universe. I would say that this is very different than counting number of galaxies, which are heterogeneous biased tracers of the mass density.

 

[turbo]

Independently of the conclusions of this paper, it has to be noted that there the mass of the intracluster gas generating x-rays is used to infer about the total mass density of the universe. I would say that this is very different than counting number of galaxies, which are heterogeneous biased tracers of the mass density.

It seems that the more ways that cluster masses can be probed, the better, if only to get hints from agreements and discordances regarding which estimations are likely to be biased. If more than one method yields overdensities (vs the concordance model), perhaps there is a valid reason. Is there any reliable indication that the observational methods used in this paper are any more or less accurate than other methods of estimating cluster masses?

The reason I ask is that estimating mass from cluster luminosity leads us routinely to underestimate cluster mass, at least in respect to the observed gravitational effects (like cluster binding). This causes the concordance model to require LOTS of undetectable non-baryonic matter to remain viable. Clearly, something is not properly understood - it may relate to the mass-luminosity function, or to gravitation, or something else, but there is a fly in the ointment. This X-ray result may tell us that we are not accurately estimating cluster masses, or perhaps that we may not be properly modeling gravitation on cluster scales.

As for selection effects and bias, Blanchard notes:

Our new analyzes basically recover identical results to the one mentioned above. However, the statistical significance is now much better: these samples
contains  300 clusters. Each sample is individually well fitted, this is a very important point: any large unidentified systematics affecting data, would have to affect the different surveys (from different groups and different methodology, on both ROSAT and EMSS data) in different way to mimic the Einstein de Sitter case, a somewhat tricky coincidence. I conclude that this new analysis is much mores robust than previous one, both in term of statistic and in term of control on systematics.

 

[hellfire]

My understanding (which may be wrong), to clarify: the formation of the large-scale structures depends only on gravitation. Therefore, the abundance of clusters as well as their distribution in redshift should be determined only by the geometry and components of the universe, and by the power spectrum of the initial density fluctuations. This is the motivation which leads to the investigations of Blanchard. However, to make a sensible cluster survey one has to define a reference mass (a minimum value) and compare against observations to do number counts. In x-ray surveys, this leads to the fact that one relies on some assumed relation between luminosity, temperature and mass. There is no general agreement on this relations.

It seems that the more ways that cluster masses can be probed, the better, if only to get hints from agreements and discordances regarding which estimations are likely to be biased. If more than one method yields overdensities (vs the concordance model), perhaps there is a valid reason.

You are right. I was only questioning the statistical significance of the results of the previous paper (z=6 galaxies).

Is there any reliable indication that the observational methods used in this paper are any more or less accurate than other methods of estimating cluster masses?

I am not sure about that. The problem when determining the “x-ray mass” is that gravitational compression may not be the only source of heating of the intracluster gas. This problem is not present when determining the “virial mass” or “lensing mass”.

And quoting from the conclusions in Blanchard’s paper:

This is in principle the signature of a high density universe, but might be as well due to a deviation of the expected scaling of the M − T relation with redshift. Our investigation of the ratio $Tx/ \sigma^2$ shows no sign of such deviation.

As already mentioned: the M – T relation itself might be wrong. Anyway, one is forced to admit that something is wrong either in our understanding of clusters or may be actually in the cosmological model.

 

[ohwilleke]

Chronos, some of the interesting language on microwave v. this study to which I was referring was this (citations omitted; some symbols converted to spelled out words):

"The census of galxies at z approximately equal to 6 is also interesting because the Gunn-Peterson trough has been observed near the redshift, suggesting that reionization ended around z=6. On the other hand, evidence from microwave background observations suggests that substantial reionization likely occurred at z in the vicinity of 15, and Lyman Alpha emitters suggest that reionization was largely complete at z approximately equal to 6.5. Strong clustering of galaxies at this epoch would complicate the reionization scenarios, leading to inhomogeneous reioniziation."

 

[matt.o]

I am not sure about that. The problem when determining the “x-ray mass” is that gravitational compression may not be the only source of heating of the intracluster gas. This problem is not present when determining the “virial mass” or “lensing mass”.

While it is true that there are other means of heating the ICM, mass measurements using the ICM generally agree fairly well with lensing and virial masses. I should also point out that lensing masses measure the mass in a cylinder along our line of sight, not just the cluster mass.

 

[Chronos]

Matt, just one point I think you should emphasize. All such studies I have seen, and hopefully understand, are careful to note the data is from cylindrical volumes. Selection effects are usually not considered.

 

[Chronos]

Chronos, some of the interesting language on microwave v. this study to which I was referring was this (citations omitted; some symbols converted to spelled out words):

"The census of galxies at z approximately equal to 6 is also interesting because the Gunn-Peterson trough has been observed near the redshift, suggesting that reionization ended around z=6. On the other hand, evidence from microwave background observations suggests that substantial reionization likely occurred at z in the vicinity of 15, and Lyman Alpha emitters suggest that reionization was largely complete at z approximately equal to 6.5. Strong clustering of galaxies at this epoch would complicate the reionization scenarios, leading to inhomogeneous reioniziation."

I don't disagree. Inhomogenous reionization cannot be ruled out. I don't see any immediate theoretical problems with that.

 

[ohwilleke]

I don't disagree. Inhomogenous reionization cannot be ruled out. I don't see any immediate theoretical problems with that.

The real interesting possibility would be if the z's of 16 and 6.5 by one method and of 6 by another would be if the difference was actually due to a systemic problem in how z is calculated by one of the methods.