Assessing Consistency Between WMAP 9-year and Planck 2015 Temperature Power Spectra

Abstract

We perform a comparison of WMAP 9-year (WMAP9) and Planck 2015 cosmic microwave background (CMB) temperature power spectra across multipoles $30\leq\ell\leq1200$. We generate simulations to estimate the correlation between the two datasets due to cosmic variance from observing the same sky. We find that their spectra are consistent within $1\sigma$. While we do not implement the optimal "$C^{-1}$" estimator on WMAP maps as in the WMAP9 analysis, we demonstrate that the change of pixel weighting only shifts our results at most at the $0.66\sigma$ level. We also show that changing the fiducial power spectrum for simulations only impacts the comparison at around $0.1\sigma$ level. We exclude $\ell<30$ both because WMAP9 data were included in the Planck 2015 $\ell<30$ analysis, and because the cosmic variance uncertainty on these scales is large enough that any remaining systematic difference between the experiments is extremely unlikely to affect cosmological constraints. The consistency shown in our analysis provides high confidence in both the WMAP9 temperature power spectrum and the overlapping multipole region of Planck 2015's, virtually independent of any assumed cosmological model. Our results indicate that cosmological model differences between Planck and WMAP do not arise from measurement differences, but from the high multipoles not measured by WMAP.