Three-year WMAP Data Now Out

Data from the second and third year of the WMAP satellite experiment has just been released a few minutes ago. There a press release and other general information page. The scientific paper explaining what this new data tells us about cosmology is Three-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Implications for Cosmology. This is a good time to admit that I’m no cosmologist, and thus not the person to get information from about the significance of these results. However, I expect some of the earliest informed discussion of them should take place on various blogs, and I’ll be linking to those as I see them.

Update: Comments from Sean Carroll and Steinn Sigurosson. Discussion at CosmoCoffee and Bad Astronomy and Universe Today Forum. Also a posting from Lubos. I’m no expert here, but Lubos’s comments seem to me to be nonsense (I find it hard to believe that in the three-year data set they’re resolving structure 100 times smaller than in the first year, and I think he’s just completely wrong to say that this data rules out ekpyrotic or cyclic models).

Update: Christine Dantas also has more about this.

Update: Amazingly, Lubos still is maintaining that the 3 year results have 100 times better angular resolution than the 1 year results. This kind of fanatical inability to ever admit that one was wrong about something goes a long way towards explaining the current state of string theory.

This entry was posted in Uncategorized. Bookmark the permalink.

19 Responses to Three-year WMAP Data Now Out

  1. Dumb Biologist says:

    I think the 100x estimate is actually correct. However, according to a new article in the NYT here:

    But Paul Steinhardt of Princeton, who has lately championed an alternative to inflation, in which the universe begins and ends cyclically in a collision between a pair of island universes, know in string theory as branes, pointed out that the new data are also compatible with his theory. Calling the results “extremely important,” he noted that they were in agreement with the simplest models of these theories.

    ‘It rules out need for anything exotic,” he said.

    I haven’t the remotely foggiest idea about any of it, I only pass this along…

  2. woit says:


    Do you have any reference for the 100x claim? It may be right, I’m just having a hard time believing that tripling your data size allows you to increase your resolution 100-fold. But maybe they did something much more clever with the later data.

    I notice that Lubos edited somewhat his statements about cyclic models being ruled out, and added a statement that Peter Woit knows nothing about cosmology. More recently he seems to have deleted the attack on me, maybe he’s about to do more editing on the cyclic models claims. He’s hilarious.

  3. woit says:


    Thanks for pointing out the NY Times article, I hadn’t seen it. I think the “100x” number is referring to being able to detect much smaller signals in order to get the polarization data (not smaller patches of sky). Lubos seems to have completely misunderstood what this has to say about cyclic models. Cyclic models predict no visible B-mode polarization. If WMAP had seen this kind of polarized signal, it would have ruled them out, but it didn’t.

  4. Dumb Biologist says:

    I don’t think this is the first place I saw the “100x better” thing, but…

    This new signal is roughly 100 times weaker than the signal we analyzed three years ago and about a billion times less than the radiant warmth we feel from the Sun,” said Lyman Page, a WMAP team member from Princeton University.

    Now, I don’t know if getting 100x better signal is really the equivalent of saying you’re looking at structures 100x smaller, but there’s a 100x improvement in something.

    I’m sorry I can’t do better. I’ve been kind of sneaking in glimpses of reading between time points and basically behaving like an ADHD-addled channel surfer when I should be paying more attention to what I’m doing. I find stuff like this almost unspeakably cool, so I get a little worked up…

  5. Dumb Biologist says:

    D’OH! Simulpost!

  6. Perhaps some topologist will give us someday a hint about why the universe is flat for the current bigbangy solution. Meanwhile, I do not understand the fuzz about WMAP in theoretical blogs. Except that some of you had got a hidden idea to explain flatland.

    Did I already suggested to read astro-ph/0601168?

  7. Dear Peter Woit,

    Perhaps he is referring to section 4.1.3 (paper on Implications for Cosmology), where the WMAP data is combined with small scale CMB experiments, which probe smaller angular scales than the WMAP.

    As expected, the paper on polarization measurements is quite involved. I am always concerned with the fact that you have to use a model of the foreground emission to subtract from the data. It is not a trivial thing at all to separate the intrinsic polarization signal from the foreground emission. So I would say that we must interpret these results with great care. In any case, I am just starting to effectively read these papers, it will take time to digest the details.

    I will try to call some of my colleagues at INPE who are specialized in CMBR (we have a ballon experiment here in Brazil) to add their comments to my blog page, where you can find also a brief post on the new WMAP data.


  8. woit says:


    Thanks, I’ve added a link to your posting and look forward to hearing anymore you learn about this.

    Hard to tell what’s going on with Lubos. He keeps changing his posting, now he’s got some WMAP pictures up there which he claims show that 3-year WMAP can resolve solid angles 100th the size of 1-year WMAP and 1000th the size of COBE. But as far as I can tell, the resolution of both 1-year and 3-year WMAP is the same: 13 arcminutes, 33 times smaller than COBE. No idea where he’s getting his numbers from, they don’t seem to make sense.

  9. Dumb Biologist says:

    I just can’t get over the idea that physicists can say something rather concrete about, and hence test models of, what what was going on about a trillionth of a second after the visible universe started expanding. That’s so impressive it’s absurd. Sorry for the fanboy slobberfest, but all hail physics! I know there are some crises in subsets of the field, but if results like these don’t reveal something vital, vibrant, and important overall, I don’t know what does.

  10. Luboš Motl says:

    Dear Peter,

    it would be a great idea for you to stop spreading the noise and nonsense that you keep on spreading.

    Open one of the 74 news articles here

    and read the paragraph that contains the word “patches”. After you read this paragraph, and maybe the whole article, correct the order of magnitude error that the journalist introduced, if you’re capable to do so, then reread my blog article to be sure what I mean, and check the pictures that I linked to see why I say what I say.

    The maximal designed angular resolution of WMAP is fixed but more data and a much better statistical treatment has allowed to create these much finer maps and everyone who has looked at the 2048-pixel-wide pictures (or who read some of the basic articles released today) knows that.

    I have always had the comment that the “modern” models of ekpyrotic/cyclic Universes try to be indistinguishable from inflation. This is the uninteresting part. The more interesting part is that they predict blue gravity waves that must eventually come into the game. At any rate, no one except for the authors knows exactly how to derive the magnitude and frequency-dependence of the tensor perturbations from the ekpyrotic scenarios, so I think it is fair to write what I wrote.

    Best wishes

  11. I will try to understand this.

    For the moment, all I can say, from the Mission Explanatory Supplement paper, is that the WMAP experiment can reach a 13 arcminute FWHM resolution for the temperature anisotropy of the CMBR. Also, that “the skymap data products derived from the WMAP observations have 45 times the sensitivity and 33 time the angular resolution of the COBE DMR mission”.


  12. woit says:


    The WMAP results are truly amazing, and this is a really healthy subfield of physics. But the results about inflation are a bit overhyped. This is giving us very little information about how inflation may have worked, although of course it’s remarkable you can say anything at all. The next generation satellite, Planck, should have the sensitivity to see the effects of gravitational waves if they’re there, which would be something a lot more substantial.

  13. Sorry, in the previous post, where I wrote “experiment can reach a”, replace this by “experiment has a”.


  14. woit says:

    I don’t understand what you’re trying to accomplish by pointing me to a popular news story with obviously incorrect numbers in it (100 billion light years??). If you can point to somewhere in the WMAP scientific papers where they explain about the 100 fold increase in resolution, please do so. It still seems to me that the “100” at issue here is an increase in sensitivity. They’re able to extract much smaller temperature differences from the noise, but this doesn’t at all mean that they have 100 times higher angular resolution.

    And your comments about ekpyrotic scenarios make no sense: “blue gravity waves”? no one but the authors knows what the theory predicts? Sorry, but I’ll take Steinhardt’s claims about this over yours.

  15. Dumb Biologist says:

    Oh, I know. I guess to me it’s just a bit mind-blowing that quite possibly now there’s a rather compelling “smoking gun” indicating, that, for real, inflation happened. I understand (I think) that the whole “inflaton field” thing is something of a gigantic black box, and that lots of folks find this deeply disatisfying. Some of the articles I’ve read seem to allude to this reality, though I agree the limited scope of our knowledge about inflation (even with this news) gets short shrift.

  16. Chris W. says:

    Lubos, thanks for reminding us yet again that you’re the smartest guy in the room. We just can’t seem to get it through our thick little heads.

  17. D R Lunsford says:

    This is my last comment for this blog, I’m done with all of this. You can read me in forthcoming papers if you’re interested. Otherwise, my last comment is this:

    Penrose completely destroyed the idea of inflation using nothing but entropy, yet we are supposed to believe, after years of looking and a long wait, that the octupole anomaly magically cleared itself up (and not a word was said about the ecliptic anomaly, so there was nothing to clear up). In fact, the work of Huterer, Copi, and Starkman was not even mentioned. Instead, we got a dissertation on applied inflation theory.

    So, who should I believe? An authentic genius, or a star chamber of people with, shall we say, vested interests?

    Science is dead. Nice job y’all.


  18. lmot says:

    There is absolutely no evidence of global warming in this new data.

    Also, that plot everyone is showing has several wiggles, and everyone knows wiggles are a property of strings.

Comments are closed.