An alternative validation strategy for the Planck cluster catalog and $y$-distortion maps

TL;DR

This paper introduces a new component separation method for Planck data that effectively distinguishes $y$-distortion from CO emission, improving cluster catalog validation and contamination assessment.

Contribution

It presents an alternative validation strategy using parametric model fitting to identify and mitigate CO contamination in the Planck cluster catalog.

Findings

At least 93% of clusters are free of CO contamination.

59% of unconfirmed candidates may have significant molecular cloud contamination.

The proposed method improves cluster catalog reliability.

Abstract

We present an all sky map of the $y$-type distortion calculated from the full mission Planck HFI (High Frequency Instrument) data using the recently proposed approach to component separation based on parametric model fitting and model selection. This simple model selection approach allows us to distinguish between carbon monoxide (CO) line emission and $y$-type distortion, something that is not possible using the internal linear combination based methods. We create a mask to cover the regions of significant CO emission relying on the information in the $\chi^2$ map obtained when fitting for the $y$-distortion and CO emission to the lowest four HFI channels. We revisit the second Planck cluster catalog and try to quantify the quality of the cluster candidates in an approach that is similar in spirit to Aghanim et al. (2014). We find that at least $93\%$ of the clusters in the cosmology sample are free of CO contamination. We also find that $59\%$ of unconfirmed candidates may have significant contamination from molecular clouds. We agree with Planck collaboration (2015) for the worst offenders. We suggest an alternative validation strategy of measuring and subtracting the CO emission from the Planck cluster candidates using radio telescopes thus improving the reliability of the catalog. Our CO mask and annotations to the Planck cluster catalog identifying cluster candidates with possible CO contamination are made publicly available.