What are the first results from the Microwave Anisotropy Probe (MAP)?

Originally posted to sci.astro.research, Feb 11, 2003, by Ted Bunn.

MAP (now WMAP -- see below) had its press conference and has released a bunch of papers on the first-year sky maps. Technical papers are at http://map.gsfc.nasa.gov/m_mm/pub_papers/firstyear.html and a pretty picture and less-technical explanation of the results is at http://map.gsfc.nasa.gov/m_mm.html

The most exciting thing about the results is that there are no big surprises! The angular power spectrum, which is the primary bit of science you get out of a microwave background sky map, is very consistent with what has become the "standard model" of cosmology: a flat inflation-based model containing dark matter and dark energy. The power spectrum is well measured past the second acoustic peak and into the third one. The temperature-polarization cross-correlation signal was also seen at about the expected level.

Although surprises in the data would have been fun, the lack of surprises is actually extremely remarkable, considering (a) how strange the "standard model" is and (b) how recently cosmology has become a mature scientific discipline.

I haven't read the papers yet -- there are 13 of them, some quite long! But here are a couple of highlights from scanning abstracts and figures:

  1. The data let you constrain cosmological parameters much more precisely than before. In particular, the age of the Universe is 13.7 +/- 0.2 Gyr. (Not that long ago, we couldn't say more than "12 to 20 Gyr"!) The Universe is flat: Omega total is 1.02 +/- 0.02. The densities of various forms of matter and energy are pinned down quite well too.

    One reason WMAP is such an improvement over previous data is that it's a single data set covering both large and small scales. To get limits on parameters from previous data, you had to stitch together COBE on large scales with other experiments on small scales. Since each experiment has its own calibration errors, that means you don't measure the heights of the peaks all that well.

  2. The TE polarization correlation showed up as expected. This strongly confirms that the initial fluctuations were adiabatic. That matters, because adiabatic fluctuations are a strong prediction of inflation, and large-scale adiabatic fluctuations are hard to produce in any way other than inflation.
  3. The large-angular-scale polarization gives information about the way the Universe reionized after the time of decoupling. This is an area about which very little was known before now: reionization had to have happened some time, but the data didn't say when. Now we know: the optical depth to reionization is 0.17 +/- 0.04, which means that the Universe reionized at an age of about 200000 years. This is presumably when the first stars formed -- their UV radiation is what presumably reionized the Universe.
  4. There is a hint that the spectral index may vary with scale, from n>1 on large scales to n<1 on small scales. That is, the primordial spectrum of fluctuations may not have been a pure power law. That's only significant at the 2-sigma level, so it may go away. If it's true, it gives useful information on discriminating between specific inflationary models.

That's all I'll say for now. There's a ton of information in this data set, and people will be poring over it for a long time to come, but those are the main things that jumped out at me after a quick look.

On a non-scientific note, they changed MAP to WMAP. The W stands for Wilkinson, a member of the MAP team who died last year. David Wilkinson was a leader in observational cosmology for decades, and this is a very fitting tribute to him.