Planck and BICEP2 / Keck collaborate and impose an upper limit on the intensity of primordial gravitational waves

Non classé

By combining their data, the Planck and BICEP2/Keck collaborations showed that the detection of primordial gravitational waves through the observation of the polarization of the cosmic microwave background has not yet occurred. This result provides the outcome to a scientific drama which held breathless both the cosmologists and the enthusiasts. The signal announced by the BICEP2 team in March 2014 can not be associated with the first moments of the Big Bang : it mainly comes from our galaxy and from gravitational distortions in its spread to us. These results have been submitted to the journal Physical Review Letters in late January 2015. The Planck collaboration of the European Space Agency (ESA) involves the CNRS, the CEA, the CNES and several French universities including the University Paul Sabatier in Toulouse through the IRAP.

The CMB is the fossil radiation from the Big Bang, the messenger from the distant past of the universe, 13.8 billion years ago. Since its discovery 50 years ago, cosmologists have studied it more and more finely in order to understand the origin and the content of our cosmos. The ESA Planck satellite has measured it in recent years with an unprecedented precision and has already made an outstanding share of discoveries, even if the data have not yet given us all their secrets.

The background radiation is a snapshot of the universe 380,000 years after the Big Bang. Cosmologists are studying it to go back much further in time, from 380 000 years to the epoch of origin of its fluctuations. A phase of exponential expansion for a tiny fraction of a second which they call inflation probably was naturally accompanied by the generation of gravitational waves. These waves propagated in the universe, distorting the fabric of space-time. Much too weak to be detected directly from Earth today, they can have left an imprint on another measurable quantity, the polarization of the cosmic microwave background, in particular the specific polarization modes called B modes which are at the heart of the research conducted by the Planck and BICEP2 collaborations.

March 2014 – Detections made by BICEP2

In March 2014, after several years of observation, the BICEP2 team, whose telescope is located at the South Pole, published a highly publicized result stipulating that a polarized signal in B modes had been detected in a region of the sky selected to be a priori little contaminated by the radiation of our Galaxy. The interpretation of this signal was originally announced as a very probable detection of primordial gravitational waves. Doubt nevertheless quickly installed in the scientific community because BICEP2’s observations provided very few constraints on the galactic part of the signal, a potential contaminant for observing the B modes. At the very least, weren’t the BICEP2 B modes at least partially a galactic signal? Only the Planck team was able to address this issue through its sky observation data at a higher emission frequency than that of BICEP where the galactic signal is dominant.

September 2014 – Planck analyzes the galactic contribution

 A first answer came in September 2014 with a Planck publication showing that the polarized emission of the Galactic dust is, on the entire sky, of amplitude at least comparable to the one of the signal measured by BICEP2, thus leaving no completely clean window to search for the primordial gravitational waves. This statistic result, however, still left an uncertainty as to the nature of the signal measured by BICEP2. Was it possible to demonstrate that at least part of the signal was of cosmological origin?

January 2015 – The Planck BICEP2/Keck collaboration bears fruit

The Planck and BICEP2 teams then joined to answer this question. At the heart of this work there were the map of the Galactic signal produced by Planck and that of the signal measured by BICEP2, complemented by new observations obtained since last March with the Keck telescope also at the South Pole. The comparison of these three datasets then led to demonstrate that the galactic contribution is dominant at the angular scales where the primordial gravitational wave signal is expected.

This image shows an area of the sky near the Galactic South Pole based on Planck-HFI observations at 353 GHz. The colours represent the emission of dust, a minor but crucial component of the interstellar medium surrounding the Milky Way. The texture, on the other hand, shows the orientation of the galactic magnetic field; this information is deduced from the direction of the emission of light polarized by the dust. The region indicated by the white dotted line is the one observed by the Keck array and BICEP2 experiments from the South Pole. ESA / Planck collaboration Acknowledgements : M.-A. Miville-Deschênes, CNRS – Institut d’Astrophysique Spatiale, Université Paris-Sud, Orsay, France

After correcting the Galactic emission, a B polarization of the cosmic microwave background has been detected, but it is the expected signal corresponding to the gravitational lens effect produced by the distribution of matter all along the path of photons. This signal is different from that of the primordial gravitational waves by its division into angular scales. Note that this direct detection, which provides information on the distribution of matter in the Universe, is not the first detection of this effect but it is by far the most accurate assessment to date.

Thus, after removing the galactic signal and the contribution of the lens effect, it appears that the data do not allow to detect unambiguously the desired footprint of the gravitational waves. This does not mean that this signal does not exist, it is just too weak to be detected with this dataset. However, this has enabled the Planck / BICEP2 / Keck collaboration to establish an appropriate upper limit to the intensity of the primordial gravitational waves. This direct limit is consistent with that obtained indirectly by Planck only in 2013 on the basis of the fluctuations in temperature of the background radiation. This Planck result predated the first publication by BICEP2, but was not considered necessarily inconsistent because it depended upon the cosmological model used in the data analysis. The new direct measurements show that the standard cosmological model used by Planck is sufficient to describe all the results.

The quest for the B modes of the cosmic microwave background is not over. It will continue with more and more accurate observations made from the ground and balloon, and by satellite in the longer term, thanks to the rapid development of new detector arrays. But the Planck survey at frequencies inaccessible from the ground will remain for a long time the essential reference to separate the cosmological and galactic components of the signal.

Laboratories involved:

  • IAP- Institut d’astrophysique de Paris (UPMC/CNRS)
  • IAS- Institut d’astrophysique spatiale (Université Paris Sud/CNRS)
  • APC- laboratoire Astroparticule et cosmologie (Université Paris Diderot/CNRS/CEA/Observatoire de Paris)
  • IRAP- Institut de recherche en astrophysique et planétologie (Université Paul Sabatier Tlse3/CNRS)
  • IPAG- Institut de planétologie et d’astrophysique de Grenoble (Université Joseph Fourier/CNRS)
  • LPSC– Laboratoire de physique subatomique et de cosmologie (Université Joseph Fourier/CNRS/INP Grenoble)
  • LAL– Laboratoire de l’accélérateur linéaire (Université Paris Sud/CNRS)

Further Resources :

IRAP Contact :

  • Ludovic Montier :



9, avenue du Colonel Roche
BP 44346
31028 Toulouse Cedex 4

Tel : 0561556666

Fax : 0561558692

Secrétariat Général

14, avenue Edouard Belin
31400 Toulouse

Tel : 0561332823

Fax : 0561332840


57, Avenue d’Azereix
BP 826
65008 Tarbes Cedex

Tel : 0562566000

Fax : 0562346763