Planck mission scientists have released the first half of the spacecraft's observations of the cosmic microwave background.
On March 21st, scientists with the European Space Agency's Planck mission announced their long-anticipated results from the spacecraft’s first 15.5 months of mapping the cosmic microwave background.
The CMB is the radiation released about 380,000 years after the Big Bang, when the newborn universe cooled down enough to become transparent and let light travel free. We see this light today redshifted to microwave wavelengths, and slight temperature variations across its wallpaper-like "surface" reveal how matter was distributed at that early era. These variations allow cosmologists to test theories of what was happening in the universe in the tiniest instants after its birth, including the momentous but brief period of inflation.Planck’s superbly precise new picture of the CMB (below) shows remarkable agreement with theoretical work, confirming that observations fit a simple cosmological model defined by just six numbers. (Take that in for a moment: the whole physical universe is described by six numbers. Even your phone number takes 10 digits in the U.S.)
The graph at right might not mean much to the average Joe, but it shows how much temperatures fluctuate in patches of various angular sizes all across the sky. Our inflationary model makes specific predictions about what this complex graph should look like. As you can see, Planck’s observations (red dots) trace nigh perfectly the theory (green line). My colleague Alan freaked out when he saw the tight fit at the graph's far right — you don’t appreciate the wonders of scientific progress until you have a 6-foot-3 man jumping up and down in your office.
But Planck also introduces a few surprises that need explaining.
By the Numbers
Planck launched on May 14, 2009, as successor to NASA’s phenomenal Wilkinson Microwave Anisotropy Probe (WMAP), which mapped the CMB for nine years. WMAP observed in five frequency bands spanning 22 to 90 GHz, and its results form the bedrock of modern cosmology — helping nail down values such as the age of the universe (13.77 billion years) and how much of the matter in the universe is “dark” (about 84%) when combined with other measurements.
Planck’s more precise numbers are slightly different from WMAP’s. Planck covered nine bands from 30 to 857 GHz, and it’s still working in the three lowest bands. The sweet spot for observing the CMB — where the galaxy’s dusty, star-studded plane is the least bothersome — is from 70 to 150 GHz, making Planck an ideal follow-up to WMAP, Planck team member Bruce Partridge (Haverford College) said last month in Boston at the annual meeting of the American Association for the Advancement of Science.When combined with other types of measurements, the Planck data homes in on an age for the universe of 13.798 billion years, give or take a mere 0.037. And it pushes down what fraction of everything in the universe is dark energy, from 71.4% to 69.2%.
(These numbers may be slightly different from what you see reported elsewhere: the numbers are all consistent, it just depends what other information is added to the CMB data.)
One of the most notable new numbers is the value for the Hubble constant. The Hubble parameter, the ratio of a galaxy’s recession velocity (redshift) to its distance, describes the rate at which the universe is expanding. Its value has changed over time; the present value is called the Hubble constant. Looking farther into the universe — earlier in time — measures a past value, explains Adam Riess (Johns Hopkins University), who shared the 2011 Nobel Prize in Physics for discovering that the universe’s expansion is accelerating.
“There is always some extrapolation, since by definition we can’t measure anything at exactly now,” he says. “It’s always the past.”
Astronomers use various tactics to estimate the Hubble constant. In the “local” universe, meaning within tens to hundreds of millions of light-years, they use standard candles such as exploding white dwarfs or Cepheid variable stars. One candle promoted by researchers at Villanova University (my alma mater — Go ’Nova!) and also used by another team in a recent Nature paper is eclipsing binary stars, which allow observers to determine stars’ real luminosity based on exact measurements of the stars’ diameters and temperatures derived from their eclipses.
The current value for the Hubble constant based on local standard candles is 73.8 +/- 2.4 km/s per megaparsec. (All uncertainties quoted here are at the 68% confidence level.) That's slightly in tension with the value extrapolated from WMAP data. The CMB-based value depends on what other data are included and the cosmological model used; the value announced with WMAP’s final 9-year results last December is 69.32 +/- 0.80. Astronomers have been waiting to see whether Planck would uphold this tension, because if the discrepancy is real it could imply something unexpected is afoot in physics.
Planck delivered. The Hubble constant derived from Planck’s first 15.5 months of CMB observations (combined with other measurements) is 67.80 +/- 0.77.
“I think this is one of the most exciting parts of the data that came out,” says Planck scientist Martin White (University of California, Berkeley, and Lawrence Berkeley National Lab). The fact that astronomers starting at opposite ends of cosmic history and moving toward the middle aren’t getting quite the same value for this parameter is going to attract a lot of attention, he says. It could signal a problem with the models or funky new physics — or even that the amount of dark energy somehow increases with time in a given volume of space. “That’s a pretty radical thing to propose, and so this is not something that we should take lightly,” he cautions.
A Boatload, In Brief
Planck’s results will be far-reaching — the project's scientists released more than two dozen papers today, and other researchers have already started downloading the raw data to work with. Some other noteworthy results in today’s announcement are:
- No extra neutrinos. According to the Standard Model, there should be three and only three flavors of neutrinos, nearly massless particles speeding through the universe at ultrarelativistic speeds. Planck upholds that expectation.
- The universe isn’t as uniform on the largest scales as expected. Previous work had hinted that the northern and southern hemispheres of the sky don’t look as much like each other (statistically speaking) as they should, and that there’s an anomalous cool spot in the CMB. (Anomalous in terms of shape, not temperature or overall size). Planck upholds these results. Furthermore, you get slightly different values for the fundamental six parameters when you fit each half of the sky separately. Assuming these effects are real, they may hint at unpredicted structure that's larger than our cosmic horizon and originating before inflation, even before the Big Bang.
- Similarly, the wiggly power spectrum graph (shown above) may have some problems over large patches of the sky. While the agreement between observation and theory is extraordinary at small angular scales, temperature fluctuations in the CMB at the largest scales don’t behave as well. The team can’t maneuver the graph to fit these points without losing the beautiful fit elsewhere.
- When inflation ended in the infant universe (about 10-35 second after the Big Bang, 10 nano-nano-nano-nanosecond), microscopic quantum fluctuations were slightly stronger on larger scales than smaller ones. These fluctuations served as the seeds of today's large-scale cosmic structure. Simple inflation predicts what this slight "tilt” of the fluctuations' size distribution should be. WMAP found this tilt, but Planck confirms its value to high accuracy. Score a big win for standard inflation.
- No polarization announcement yet. The real test of inflation, which cosmologists eagerly await, should come from specific polarization patterns in the CMB created by gravitational waves. Other theories of what caused the Big Bang, such as colliding "branes" in higher-dimensional space, predict that there will be no such polarization patterns. Looking for them was a prime reason why Planck was built. Mission scientists said that those polarization data are not cleaned up enough yet to be usable; we'll have to await analysis of a longer span of Planck's data. The next release of results is planned for 2014.
If you’d like the dirty details, you can find all the Planck papers online. The summary of results is in Section 9 of Paper I.
The European Space Agency has put up many excellent graphics and explanations. Start here and see the sidebar on the right.