South Pole Telescope and CMB Constraints

The South Pole Telescope is a wonderful instrument, a ten-meter radio telescope that has been operating at the South Pole since 2007. Its primary target is the cosmic microwave background (CMB), but a lot of the science comes from observations of the Sunyaev-Zeldovich effect due clusters of galaxies — a distortion of the frequency of CMB photons as they travel through the hot gas of the cluster. We learn a lot about galaxy clusters this way, and as a bonus we have a great way of looking for small-scale structure in the CMB itself.

Now the collaboration has released new results on using SPT observations to constrain cosmological parameters.

A Measurement of the Cosmic Microwave Background Damping Tail from the 2500-square-degree SPT-SZ survey
K. T. Story, C. L. Reichardt, Z. Hou, R. Keisler, et al.

We present a measurement of the cosmic microwave background (CMB) temperature power spectrum using data from the recently completed South Pole Telescope Sunyaev-Zel’dovich (SPT-SZ) survey. This measurement is made from observations of 2540 deg^2 of sky with arcminute resolution at 150 GHz, and improves upon previous measurements using the SPT by tripling the sky area. We report CMB temperature anisotropy power over the multipole range 650<ell<3000. We fit the SPT bandpowers, combined with the results from the seven-year Wilkinson Microwave Anisotropy Probe (WMAP7) data release, with a six-parameter LCDM cosmological model and find that the two datasets are consistent and well fit by the model. Adding SPT measurements significantly improves LCDM parameter constraints, and in particular tightens the constraint on the angular sound horizon theta_s by a factor of 2.7…[abridged]

Here is the first plot anyone should look for in a paper like this: the CMB power spectrum, giving the amplitude of fluctuations from large scales (left) to small scales (right). This graph shows both the most recent results from the WMAP satellite, and the new SPT numbers. The dashed line is a theoretical prediction that includes only the CMB, while the solid line is the prediction when foregrounds (our galaxy as well as others) are included. Not bad! I remember when it was considered amazing to find a single identifiable peak in the CMB spectrum, and the second one was also big news. Now we have what, nine or ten visible wiggles?

With data like these, you can do an unprecedentedly good job at constraining cosmological parameters. Here is a plot with the density of matter on the horizontal axis, and the cosmological constant on the vertical axis, using only CMB data. Show this to any remaining friends of yours who are still skeptical about the supernova observations that revealed the acceleration of the universe (if you have such friends). This is an independent detection of the cosmological constant at better than five sigma.

Since most of us have already accepted the existence of some form of dark energy, the interesting questions going forward have to do with what evidence we can extract about inflation — did it happen, and if so what form did it take? SPT can help there, by improving our limits on the tilt of the primordial spectrum — that is, the strength of fluctuations on small scales compared to large scales. A naive guess would be that it’s perfectly flat, with equal fluctuations on all scales. But most inflationary models predict a tiny variation from perfect flatness, usually in the “red” direction — more power on larger scales. That corresponds in this graph to ns being less than one, which is indeed what the data seem to be indicating. The vertical axis is the amount of primordial gravitational waves, which is still consistent with zero according to our best current data. As you can see, different kinds of inflationary models tend to make predictions that lie in different regions of parameter space, so we are using data to constrain what might have been happening in the first 10-35 seconds after the Big Bang. (Some of the points are labeled “chaotic inflation,” where what is really meant is “power-law inflation,” but that’s a mistake everybody makes.)

It’s almost tempting to say that we can conclude there definitely is a detectable deviation from ns = 1, but I’m less ready than most cosmologists to accept that at face value. (“BAO” refers to observations of baryon acoustic oscillations in the large-scale structure of galaxies, which improve the constraints.) As the paper says, there are a number of ways to skew that result, including if neutrinos have the right kind of mass. I think cosmology should be as careful as particle physics in declaring things to be discovered, and I’m not sure we’re there yet on spectral tilt. But that’s certainly the way the data seem to be leaning — it might just be a matter of time.

  1. The rumour is that there is a companion paper coming out soon, (Hou et al) that will say interesting things about potential modifications to the standard cosmological model, particularly with respect to the number of effective neutrinos (i.e. the amount of radiation present at last scattering). Maybe not quite a detection, but at least strong evidence for some sort of “dark radiation”.

    From the paper you’ve just written about:

    “In H12, we consider several extensions to the CDM model, and find that the data
    show some preference for several of those extensions.” (pg 10)

    Like or Dislike: Thumb up 0 Thumb down 0

  2. Well, the rumour existed before the first paper was released and the quote I provided doesn’t say anything specifically about effective neutrinos (I only added the quote to show I wasn’t talking complete nonsense).

    Whatever word one chooses, the impending observations in Hou et al. are still going to be very interesting (if what I’ve been led to believe is even true).

    Perhaps I should have started my comment with… “The rumour is that the companion paper coming out soon (Hou et al.) will be showing stronger evidence for dark radiation than any other previous measurement has” instead.

    Like or Dislike: Thumb up 0 Thumb down 0

  3. Just a question. How many wiggles/peaks would cosmologists need in order to “validate” certain concrete inflationary scenario/model (or even an alternative like VSL theories) up to say CL 95% instead of putting bounds only? Could it be done only with wiggles or we would require some additional measurements?

    Like or Dislike: Thumb up 0 Thumb down 0

  4. My personal opinion is that near absolute verification of the cosmological constant is far more exciting than the discovery at CERN. I guess Albert didn’t make any blunders after all, he was too advanced for his own damn self.

    Like or Dislike: Thumb up 0 Thumb down 0

  5. Wow, I so want to show those in my class, unfortunately the students will have no bleeping idea why this is so amazing. I didn’t know you could get such good resolution from a ground based telescope.

    Like or Dislike: Thumb up 0 Thumb down 0

  6. I want to see this Hou, et al paper that was discussed (referred to as H12) because one of my personal interests is the number of neutrino species. In WMAP7 data consolidated with other datasets, the number (as a 1 standard deviation result) is 4.

    Since this dataset goes far beyond the WMAP data (holy multipole moments batman), that number ought to be constrainted further and I’m curious to see where it will go.

    Like or Dislike: Thumb up 0 Thumb down 0

  7. Hi Sean

    Thanks for pointing out this paper.

    It would have been useful for us non-experts if you had defined the parameter ns and explained why it being less than one is significant.


    Like or Dislike: Thumb up 0 Thumb down 0

  8. How many wiggles/peaks would cosmologists need in order to “validate” certain concrete inflationary scenario/model (or even an alternative like VSL theories) up to say CL 95% instead of putting bounds only?

    It is not clear what this means. It sounds like you are trying to prove a hypothesis, whereas science can only disprove a hypothesis, for the same reason that “innocent until proven guilty” is better than the other way around.

    Like or Dislike: Thumb up 0 Thumb down 0

  9. Yes, the wiggles look very interesting and the agreement with theory is very impressive but could you perhaps explain to us non-experts why they have the shape they do?

    For example where does the distance between the peaks come from? Why are the 2nd and 3rd the same height but less then the 1st and more then the 4th? Surely if as you say the lines represent theoretical predictions all the answers to those questions have to be known. Even if the details are too technical it would be nice to just get a general idea.

    I also find such observations more interesting then detection of the particle-compatible-with-being-Higgs.

    Like or Dislike: Thumb up 0 Thumb down 0

  10. Looking at the best-fit values for the spectral index for WMAP (low l) and SPT (high l), it seems a running spectral index might be preferred. Does anybody know why it hasn’t been considered in the paper?

    Like or Dislike: Thumb up 0 Thumb down 0

  11. Pingback: El Telescopio del Polo Sur (SPT) confirma con 5,4 sigmas la existencia de la energía oscura « Francis (th)E mule Science's News

  12. clamtrox –

    Running of the spectral index is going to be discussed in the companion/follow-up paper (Hou et al.), too.

    Like or Dislike: Thumb up 0 Thumb down 0

  13. Pingback: South Pole Telescope bevestigt het bestaan van donkere energie | Astroblogs