Thursday, March 16, 2006

WMAP: three-year data released

Fourty minutes ago, the WMAP team has released
Click at the link above and enjoy. If you want to be informed about similar events by e-mail,

The polarization analyses of Lyman Page et al. see no evidence of B-modes but tell you a lot of details. Note that the B-modes describe the magnetic field of electromagnetic waves. If you choose the synchronous gauge, the B-modes of the radiation are sourcing the tensor "h_{ij}" modes of the gravitational field only. In other words, the absence of the B-modes means an absence of short gravity waves.

See also WMAP 5-year results.
This observation seems to rule out the original ekpyrotic models of the Universe because these models predict that the energy stored in gravity waves grows faster with the momentum than inflation predicts while the upper bound on the tensor/scalar ratio from the WMAP data is 0.55. The cyclic Universe models based on the original ekpyrotic scenario may be ruled out, too. Note that newer cyclic models are claimed to generate a scale-invariant spectrum indistinguishable from inflation.

David Spergel explains, together with his team, that a standard six-parameter cosmological model containing cold dark matter plus cosmological constant plus baryonic matter fits not only the new three-year data but also finer CMB details: patches that are smaller than in the previous data have been looked at and the case for inflation has strengthened because the spectrum continues to be scale-invariant up to these shorter length scales. Try this 2048 x 1124 map in the W-band (more than one megabyte!) and compare with the analogous, older pictures from COBE and WMAP-1-year. One can see solid angles that are 1,000 times smaller than those at COBE and almost 100 times smaller than with the first-year WMAP maps. Instead of listening to Peter Woit, have a look! ;-)

The equation of state of the dark energy has "w=p/rho" equal to -1.07 plus minus 0.1 or so. The sum of neutrino masses is below 0.68 eV at 95% confidence level. No non-gaussianities have been seen. The index "n" is close to one (scale invariance) but very likely different from one, something like 0.96.

A 142-page-long description of the whole experiment is here. Some update on temperatures is here - they hijacked the acronym ILC! A discussion of data analysis and error margins is here.

Figure 1: An image of the skies. (WMAP/NASA science team.)

More hot new images can be found on a NASA website. EurekAlert offers a press release. Because the temperature fluctuations still perfectly agree with the inflationary framework, there is a lot of room for poetic comments. Brian Greene revived a theme from his book "The Fabric of the Cosmos" at the press conference:
  • These observations are spectacular and the results are stunning… it is truly inspiring. Galaxies are nothing but quantum mechanics writ large across the sky. also presents the results as a stronger argument in favor of inflation. The astronomers won't ever be impressed by high-energy physics and they prefer to learn that the first star was born 0.4 billion years after the Big Bang, which means 13.3 billion years ago. See Bad (low-energy) Astronomy Blog.

More news on Mark Trodden says just a few words. Sean Carroll, on the other hand, tells you much more how the LambdaCDM model is in good shape.


  1. It's a new low of quality in scientific papers. On Page 16 they got the world accepted neutron lifetime as 887.5 seconds. It's actually 885.7 seconds. It may just be a typo but with nearly 4 dozen eyes proof reading this paper for two years, they should NOT have this elementary school mistakes that could be caught by QUANTOKEN in 3 minutes of reading. Also, all previous neutron lifetime measurements are pretty much agree with each other, on what basis do they adapt this Russian guy's result, which differ from every one else by more than 6 sigma, just because the data fit their model better?


  2. They measured hubble constant to be 0.73 +- 0.03, and then also claim the age of the universe to be 13.7 GY. The two numbers are not consistent from each other.

    If you are too lazy to calculate, Look at the bottom of this page:
    t0 = 9.8 GY/h0

    9.8 divided by 0.73 is 13.4, not 13.7.

  3. Dear Quantoken,

    the lifetime of the Universe would only be related to the Hubble constant in the way you believe if the Hubble constant were constant during the history of the Universe which is not the case.

    The fact that H=1/T holds to a good accuracy is a complete accident, and it won't hold in the far future. The Hubble constant will approach the de Sitter constant while the time from the Big Bang will grow.

    Some parts of the history of the Universe had seen faster expansion than today, some parts had seen slower expansions, and if you combine it, H=1/T is almost exact but not quite.

    I agree that neutron's lifetime is 885.7 seconds +- 0.8 seconds.


  4. The interesting thing about cranks like Quantoken, is that since they cannot do real science, the nitpick over numbers to the nth significant digit. They do not and cannot realize that for measured numbers, such as those for the lifetime of free neutrons, there is an error in the measurement.
    NIST quotes a neutron lifetime of 885.8 ± 3.4 s.
    Therefore the number quoted in the paper is within the error bars given by the NIST results.
    The 3.4 second is not a 6 sigma error.

    There are many different experiments underway, some ultra cold trapping experiments reporting other lifetimes, such as 878.5 ± 0.7 ± 0.3

    (Note that the paper is still closer than Quantoken)

    The fact of the matter is that there is no "world accepted" neutron lifetime. (If there were, it would be from a standards lab like NIST) At most there is a current best limit. Such is the reality of experimental physics and that reality states that the number will continued to be refined. Cranks like Quantoken will find one number and never accept any other.

    Once again, Quantoken shows his true crackpot colors.

  5. On other hand, see the figure in hep-ex/0504034 to check were are we about neutron decay. It seems that some sigmas between different experiments are usual in neutron measures.

  6. Just to clarify: the Serebrov et al. result is the one six sigmas away of the pdg 2004 average. Now, the page quantoken refers to does not "adapt" this result, they just say that it should be "a shift several times the reported errors" and then remark that "This (Serebrov's) shorter lifetime lowers the predicted best fit helium abundance". All the goal of the parragraph is to stress how "the uncertainties in nuclear parameters" are the main source of systematic uncertainty now.

  7. Mike Varney:

    I really don't have high expectations of you, who just graduated from college a short time period ago, and who now works a full time job of $6 an hour flipping hamberger in McDonalds, plus a part time job working $2 an hour slave labor in a professor's lab as a janitor (they call it research assistant, btw). But a hamberger flipper should know the difference between a careless typo, and experimental error. BTW any one should know what Particle Data Group is, and be able to look up the authoritative numbers.

    Your web blog is too filthy for me to visit again, Mike Whinning.

  8. Quit projecting your inadequacies and employment dreams on me, Quantoken.
    You would not even be able to get a job at Taco Bell, much less Burger King or McDonalds.

    And good riddance of you from my blog. Lubos is too nice to cranks like yourself.

  9. Mike Whinning:

    You are quote correct that I would not be able to get a job in those three fast food restraurants. They look at my background and well above 6 figure income and would be scared to death to have me onboard for over qualification. As a matter of fact I have stopped eating in those restrarants for about 10 years now because their food is under-qualified for me. These restuarants hire only illegal immigrants and under age teenages like you, Mike.


  10. Hi nice blog .I need to post resumes .can anybody send links of that sites.
    Thank you.........