Identifying frequency decorrelated dust residuals in B-mode maps by exploiting the spectral capability of bolometric interferometry

TL;DR

This paper demonstrates that bolometric interferometry's spectral imaging capability enhances the detection and characterization of complex dust foregrounds in CMB B-mode experiments, improving systematic uncertainty control.

Contribution

It introduces the use of spectral imaging in bolometric interferometry to better identify and measure frequency decorrelation effects in foregrounds, aiding future CMB polarization studies.

Findings

Spectral imaging detects systematic uncertainties in the tensor-to-scalar ratio due to frequency decorrelation.

BI can measure dust spectral index more precisely even with complex decorrelation effects.

In-band frequency resolution in BI outperforms traditional imagers in identifying dust residuals.

Abstract

Astrophysical polarized foregrounds represent the most critical challenge in Cosmic Microwave Background (CMB) B-mode experiments. Multi-frequency observations can be used to constrain astrophysical foregrounds to isolate the CMB contribution. However, recent observations indicate that foreground emission may be more complex than anticipated. We investigate how the increased spectral resolution provided by band splitting in Bolometric Interferometry (BI) through a technique called spectral imaging can help control the foreground contamination in the case of unaccounted Galactic dust frequency decorrelation along the line-of-sight. We focus on the next generation ground-based CMB experiment CMB-S4, and compare its anticipated sensitivities, frequency and sky coverage with a hypothetical version of the same experiment based on BI. We perform a Monte-Carlo analysis based on parametric component separation methods (FGBuster and Commander) and compute the likelihood on the recovered tensor-to-scalar ratio. The main result of this analysis is that spectral imaging allows us to detect systematic uncertainties on r from frequency decorrelation when this effect is not accounted for in component separation. Conversely, an imager would detect a biased value of r and would be unable to spot the presence of a systematic effect. We find a similar result in the reconstruction of the dust spectral index, where we show that with BI we can measure more precisely the dust spectral index also when frequency decorrelation is present. The in-band frequency resolution provided by BI allows us to identify dust LOS frequency decorrelation residuals where an imager of similar performance would fail. This opens the prospect to exploit this potential in the context of future CMB polarization experiments that will be challenged by complex foregrounds in their quest for B-modes detection.