CosmoCoffee

[Alessandro Melchiorri]

So the new results are out !
Any comments ?

It seems to me that:

- tau, n and sigma_8 back to 0.09,0.95 and 0.78 as before the first year WMAP release. Bye bye high tau and sigma_8. n<1 at 2 sigma.

- No spike around first peak and octupole back in agreement but other features and axis of evil still there...

- Great large scale polarization measurements and second TT peak.

- The TE spectrum seems also out of phase with the best fit TT model, what do you think ?

Cheers
Alessandro

[Ben Gold]

[Garth Antony Barber]

The third peak of the TT, angular power spectrum, looks inconclusive. The LCDM predicted curve misses several of the error bars. How do the other experiment observations of the high-l end make of this?

Garth

[Aurelien Fraisse]

Hi Garth,

[Su Meng]

Notice that both the reionization redshift and optical depth are largely different from the first year data, any comment on that?
And as mentioned above, the value of sigma8 is correspondingly less than the first year result. What's the influence on reionizaiton history and structure formation?

Cheers

[Alessandro Melchiorri]

Hi,
the reason is simple:

WMAP 1st year: High level of polarization at large scale -> High Tau -> high spectral index -> high sigma_8

WMAP 3rd yeat: Lower level of polarization -> lower tau -> lower spectral index -> lower sigma_8

Just for fun I report results from an old paper by me and collaborators (made before WMAP 1st year, other papers at that time by Sarah Bridle, Ofer Lahav, Anthony Lewis, Anze Slosar, Max Tegmark and many others were saying almost the same):


Paper: astro-ph/0212276 WMAP1 WMAP3

Sigma_8: 0.76±0.09 0.84±0.06 0.76±0.05
tau: 0.07±0.07 0.15±0.07 0.09±0.03
n_s: 0.96± 0.05 0.99±0.04 0.96±0.017


In my opinion, the high TE of the 1st year had an huge impact but now we are back to the previous cosmology (with much smaller error bars!)

Some consequences:

- Models (and papers) based on high tau are now ruled out.

- Lyman alpha (see for example astro-ph/0407372) and SDSS galaxy bias (astro-ph/0406594) suggest (when combined with old wmap) sigma_8=0.9±0.03 ! It is clear that those analyses should be revised !
Of course you may have a fantastic constraint on neutrino masses if you combine CMB and Lyman-alpha, but are the datasets compatible ?
fig. 19 in Spergel et al illustrates this very well.

[Antony Lewis]

You can see a WMAP3 - Lyman alpha plot at

http://cosmologist.info/cosmomc/

[Alessandro Melchiorri]

perfect timing ! :-)
thanks!

[Su Meng]

Thanks for explanation~

I just feel that the influence of lower sigma8 on structure formation history are confused. With the both the lower reionization redshift and sigma8, the structure formation is deferred. Does it mean that the "formation problem" of the high-redshift galaxies and quasar populations is even more serious?

Since the reionization redshift down to 11, do we still need more than one reionization procedure to ionize the universe through multiple processes?

The analysis method is changed for low-l power. I wonder that the differences between the new results and first year ones are mainly from the dissimilar Methodology.

Cheers

[Anze Slosar]

[Antony Lewis]

Viel. And of course the ly-alpha-only contours are prior driven, so don't take them too seriously.

[Anze Slosar]

Table 11 of Spergel et al.:

WMAP + 2dF give 2−3 times better errorbars on Ω_k than WMAP + SDSS... Is this true and if so, why?

[Antony Lewis]

There was a bug in my LUQAS Lyman-alpha contours in the plot I linked and in my Moriond talk - the fixed plot will be in my conference proceedings at some point. As far as I know the SDSS Lyman alpha contours and numbers were right. (I'm sure we'll also see new SDSS Lyman alpha and WMAP constraints from the horse's mouth before too long..).

The CosmoMC lyman-alpha LUQAS/CROFT module will be fixed shortly.