Planck 2015 results. VII. HFI TOI and beam processing

TL;DR

This paper details the processing and calibration of the Planck HFI time-ordered data and beam response, leading to improved sky maps and a significant reduction in beam uncertainty for cosmological analysis.

Contribution

It introduces updated data processing methods, including non-linearity correction and improved bolometer response modeling, enhancing the accuracy of the HFI data for cosmology.

Findings

Beam window function uncertainty reduced by over a factor of 10

Pipeline modifications improve data calibration and map quality

Noise correlations are characterized and mitigated

Abstract

The Planck High Frequency Instrument (HFI) has observed the full sky at six frequencies (100, 143, 217, 353, 545, and 857 GHz) in intensity and at four frequencies in linear polarization (100, 143, 217, and 353 GHz). In order to obtain sky maps, the time-ordered information (TOI) containing the detector and pointing samples must be processed and the angular response must be assessed. The full mission TOI is included in the Planck 2015 release. This paper describes the HFI TOI and beam processing for the 2015 release. HFI calibration and map-making are described in a companion paper. The main pipeline has been modified since the last release (2013 nominal mission in intensity only), by including a correction for the non-linearity of the warm readout and by improving the model of the bolometer time response. The beam processing is an essential tool that derives the angular response used in all the Planck science papers and we report an improvement in the effective beam window function uncertainty of more than a factor 10 relative to the 2013 release. Noise correlations introduced by pipeline filtering function are assessed using dedicated simulations. Angular cross-power spectra using datasets that are decorrelated in time are immune to the main systematic effects.