The PLANCK experiment 
by Bartjan Van Tent
Researcher at the LPT, 
University Paris XI

ESA's PLANCK satellite was launched in May 2009 and consists of two instruments: LFI and HFI (low/high frequency instrument). Together they scanned the microwave sky in 9 frequency bands between 30 and 850 GHz with a maximum resolution of 5 arcmin, for a total duration of 2.5 years (HFI) and 4 years (LFI). PLANCK took its observations from the second Lagrange point of the Sun-Earth system, 1.5 million km from Earth, turning around its axis once per minute, thus scanning the same great circle on the sky 40-65 times before moving on to the next, in this way observing approximately the whole sky every 6 months. PLANCK was designed to be the ultimate full-sky CMB temperature experiment as well as provide the first high-resolution full-sky E-polarization maps. Its multiple frequency bands allow it to clean the CMB extremely well from all foreground contamination, as well as measure those astrophysical foregrounds in their own right. Even with just half the temperature data and no polarization data released yet, it has determined the CMB power spectrum and the cosmological parameters to exquisite accuracy, and delivered a wealth of new data for cosmology and astrophysics.

Some of the main results of the PLANCK 2013 release are:

- Most accurate measurement ever of the CMB temperature anisotropy power  spectrum, up to the 7th peak, which is very well fitted with a standard Lambda-CDM cosmology (standard particles + cold dark matter + dark energy), the parameters of which have been determined with the highest precision ever so far. There are some as yet unexplained anomalies at large scales (low multipole number) though.

- The spectral index has been measured sufficiently accurately to exclude with high significance a completely scale-invariant spectrum, hence favouring an inflationary origin of the fluctuations. Earlier hints of a possible large local non-Gaussianity have (unfortunately) been disproved by PLANCK. However, that is only the crudest possible measure of non-Gaussianity; PLANCK has opened the field for a more quantitative study of the bispectrum (as COBE did for the power spectrum).

- Gravitational lensing of the CMB measured for the first time at very high significance: directly in the CMB power spectrum, its correlation with the ISW effect (integrated Sachs-Wolfe, due to dark energy), its correlation with the CIB (cosmic infrared background, due to dusty galaxies and quasars), and its correlation with different source catalogues. PLANCK has for the first time ever reconstructed the lensing potential on the full sky and determined its power spectrum.

- PLANCK has delivered the first ever high-accuracy full-sky maps of astrophysical foregrounds at frequencies between 100 and 850 GHz. Its maps of galactic dust, of the CIB, of CO molecules, and of diffuse SZ (Sunyaev-Zel'dovich) emission, as well as its catalogues of compact sources and of SZ clusters are all without precedent and contain a wealth of information for galactic and extra-galactic astrophysics.

For the upcoming PLANCK 2014 release an improvement of all the above and other results can be expected, due to more temperature data, a better understanding of instrumental effects, and sometimes further developed  analysis techniques. However, the main new thing in that release will be the delivery of the PLANCK polarization data. This will give us the best ever determination of the EE polarization power spectrum, as well as the TE cross power spectrum. In addition PLANCK should measure the BB polarization power spectrum due to lensing of the E-modes at high multipole number, as well as the primordial BB power spectrum at low multipole number if the inflationary tensor modes are of sufficient amplitude. In particular PLANCK will provide the definitive answer to the question raised by the recent BICEP2 results: did they indeed detect the primordial tensor modes from inflation or was it just dust?

Map of the diffuse cosmological background observed by PLANCK in 2013