A needlet ILC analysis of WMAP 9-year polarisation data: CMB polarisation power spectra

TL;DR

This paper presents a novel needlet ILC method to accurately estimate WMAP 9-year CMB polarisation power spectra, effectively cleaning foregrounds and confirming the EE spectrum aligns with cosmological models, unlike previous results.

Contribution

The paper introduces a needlet ILC approach for foreground cleaning in WMAP data, providing more accurate CMB polarisation spectra estimation and resolving discrepancies in the EE spectrum.

Findings

EE spectrum matches theoretical expectations

No excess in EE spectrum at large scales as previously reported

Method improves foreground removal in CMB polarisation analysis

Abstract

We estimate Cosmic Microwave Background (CMB) polarisation power spectra, and temperature-polarisation cross-spectra, from the 9-year data of the Wilkinson Microwave Anisotropy Probe (WMAP). Foreground cleaning is implemented using minimum variance linear combinations of the coefficients of needlet decompositions of sky maps for all WMAP channels, to produce maps for CMB temperature anisotropies (T-mode) and polarisation (E-mode and B-mode), for 9 different years of observation. The final power spectra are computed from averages of all possible cross-year power spectra obtained using foreground-cleaned maps for the different years. Our analysis technique yields a measurement of the EE spectrum that is in excellent agreement with theoretical expectations from the current cosmological model. By comparison, the publicly available WMAP EE power spectrum is higher on average (and significantly higher than the predicted EE spectrum from the current best fit) at scales larger than about a degree, an excess that is not confirmed by our analysis.