Update: The BICEP2 paper recently appeared in Physical Review Letters. I have been remiss in mentioning that their discovery is controversial due to potential 'improper' subtraction of galactic dust. The opinions are split is that it's unclear that dust could reproduce the B-mode polarisation that they are claiming to see. However, more evidence is needed before we can believe this discovery. So, no Nobel prize yet, but stay tuned for Planck results, and data from other telescopes!
|South Pole Telescope & B-mode data. Credit: BICEP2, NASA.|
The South Pole telescope (BICEP2 collaboration) reported seeing a twisting in the direction of the polarization of primordial light (also known as the cosmic microwave background) that is likely to be a signature of gravitational waves from inflation. There are numerous news reports about this all over the web, e.g., Sky and Telescope, Caltech, New Scientist, the Guardian, and even CNN, and the NY Times.
|E-mode pattern (up), B-mode pattern (down)|
Inflation tells us that for a tiny fraction of a second the universe after the Big Bang grew exponentially stretching both (1) the quantum fluctuations in the spacetime into gravitational waves (the B-modes or tensor modes), and (2) the scalar fluctuations of the inflaton (the E-modes or scalar modes) also known as the ordinary density perturbations that later grew into the structure we see today: stars, galaxies, etc. The photons were polarized in a particular pattern by scattering off electrons (see Fig 2 in Sky and Telescope for a short explanation of this process). Once the first atoms formed, the universe became transparent to radiation (some 380, 000 years after the Big Bang) and those photons traveled to us to form the Cosmic Microwave Background. They were seen by WMAP, Planck, the South Pole telescope, and as well as by other ground-based telescopes and balloon experiments. Accurate measurements of the residual polarization of these photons are very challenging, but hopefully some of these experiments will confirm the BICEP2 results. There is also a BICEP3 planned.
If the interpretation of these results is correct, the potential impact is huge. These photons provide a first estimate of how hot the big bang was, tell us the gravity is quantised, constrain different models in string theory, quantum gravity and cosmology by indirectly probing space-time at an energy scale when the strong, weak and electromagnetic force were one, etc.
Technical results: Their best fit of the data is a scalar to tensor ratio of 0.2 ± 0.07/0.05, and imply an energy scale for inflation of about 1016 GeV, which may be as close as we can get to the Planck scale of about 1019 GeV. They believe the gravitational waves come from inflation because the B-modes peak at a particular multipole: l = 80, which is particular to the gravitational waves expected from inflation and makes them unlikely to be E-modes turned into B-modes via lensing.
See the BICEP2 papers and Data for the original work. They describe the instrument, their data analysis, some of its potential interpretations and combine their results with those from current data from Planck and other experiments. Also, Sean Carroll's blog is typically great for explaining the physics while proving necessary background information for people who are not cosmologists.
Direct vs. indirect gravitational wave detection.
Gravitational waves are a direct consequence of general relativity, which is a theory that has been tested many times in many different ways. The first indirect detection of gravitational waves was in the Hulse-Taylor binary pulsar system for which the prediction from general relativity matches extremely well to the observed decrease in orbital period as the two neutron stars spiral together. The neutron stars speed up and get closer together as they lose energy via gravitational wave emission. This is considered to be evidence for gravitational waves and not a direct detection. The binary system is a single source that generates the waves, and cannot be used to detect other gravitational wave sources.
The CMB polarization observations represent unequal heating of photons that scattered off electrons that were stretched and squeezed by passing gravitational waves. The "waves" are pre-inflation quantum fluctuations stretched by inflation that are observed via their imprint on the CMB photons. It's truly amazing that this effect can be measured with such good accuracy (5-sigma level), and it's data that holds information on what happened at about 10-37 seconds after the Big Bang at energy levels at the GUT (Grand Unified Theories Scale). However, it is still a one-time event. Now, this one time event may be way more interesting than many detections of neutron star binaries and black hole binaries, but that may be a matter of perspective. Also, such observations may not be as unrelated as they seem. Both the very early universe and the interior of black holes were governed by quantum gravity, e.g., we don't understand the difference between a black hole horizon and the cosmological horizon as well as we would like to.
Experiments on Earth such as the Advanced LIGO detectors (and in space: LISA, Pulsar Timing arrays) expect to see gravitational waves from a large variety of sources. On Earth, gravitational wave detectors will see macroscopic masses (40 kg objects) move when a gravitational wave passes by. The problem is that the gravitational waves are very weak. Such detectors have to recover this movement generated by the gravitational wave(s) and distinguish it from noise, i.e., the movement generated by numerous other things on Earth (e.g., traffic, people cutting wood, earthquakes, waves, etc). This is extremely challenging, but there is hope it will happen this decade.
Note: The discussion on direct vs. indirect detection was added after reading Sam Finn's facebook explanation.