The Simons Observatory: component separation pipelines for B-modes

TL;DR

The paper develops and compares three pipelines for separating components in CMB data to accurately measure primordial gravitational waves, addressing foreground contamination and bias reduction.

Contribution

It introduces two advanced pipeline extensions that effectively mitigate foreground residual biases in $r$ estimation for the Simons Observatory.

Findings

Default pipelines can produce >2σ bias in complex foreground scenarios.

The proposed extensions reduce bias to below 1σ in all tested scenarios.

The new methods maintain a moderate increase in uncertainty.

Abstract

The upcoming Simons Observatory (SO) Small Aperture Telescopes aim at observing the degree-scale anisotropies of the polarized CMB to constrain the primordial tensor-to-scalar ratio $r$ at the level of $\sigma(r=0)\lesssim0.003$ to probe models of the very early Universe. We present three complementary $r$ inference pipelines and compare their results on a set of sky simulations that allow us to explore a number of Galactic foreground and instrumental noise models, relevant for SO. In most scenarios, the pipelines retrieve consistent and unbiased results. However, several complex foreground scenarios lead to a $>2\sigma$ bias on $r$ if analyzed with the default versions of these pipelines, highlighting the need for more sophisticated pipeline components that marginalize over foreground residuals. We present two such extensions, using power-spectrum-based and map-based methods, and show that they fully reduce the bias on $r$ to sub-sigma level in all scenarios, and at a moderate cost in terms of $\sigma(r)$.