Planck 2015 results. XI. CMB power spectra, likelihoods, and robustness of parameters

This paper presents the Planck 2015 likelihoods, statistical descriptions of the 2-point correlation functions of CMB temperature and polarization. They use the hybrid approach employed previously: pixel-based at low multipoles, $\ell$, and a Gaussian approximation to the distribution of cross-power spectra at higher $\ell$. The main improvements are the use of more and better processed data and of Planck polarization data, and more detailed foreground and instrumental models. More than doubling the data allows further checks and enhanced immunity to systematics. Progress in foreground modelling enables a larger sky fraction, contributing to enhanced precision. Improvements in processing and instrumental models further reduce uncertainties. Extensive tests establish robustness and accuracy, from temperature, from polarization, and from their combination, and show that the {\Lambda}CDM model continues to offer a very good fit. We further validate the likelihood against specific extensions to this baseline, such as the effective number of neutrino species. For this first detailed analysis of Planck polarization, we concentrate at high $\ell$ on E modes. At low $\ell$ we use temperature at all Planck frequencies along with a subset of polarization. These data take advantage of Planck's wide frequency range to improve the separation of CMB and foregrounds. Within the baseline cosmology this requires a reionization optical depth $\tau=0.078\pm0.019$, significantly lower than without high-frequency data for explicit dust monitoring. At high $\ell$ we detect residual errors in E, typically at the {\mu}K$^2$ level; we thus recommend temperature alone as the high-$\ell$ baseline. Nevertheless, Planck high-$\ell$ polarization spectra are already good enough to allow a separate high-accuracy determination of the {\Lambda}CDM parameters, consistent with those established from temperature alone.