Enhancing Primordial B-mode Detection: Comprehensive Delensing Pipelines for Improved Sensitivity to $r$

TL;DR

This work tackles the challenge of primordial B-mode detection by mitigating lensing-induced B-modes through two delensing approaches: a gradient-order B-mode template and an inverse-lensing remapping. Using a realistic, foreground-cleaned simulation pipeline across sub-1 m, 6 m, and 3 m-class platforms, the study quantifies delensing performance and optimizes $r$ constraints via four likelihood models, demonstrating up to ~80% delensing efficiency and up to ~63% reduction in the $r$-uncertainty, with the lensing-template channel (L3) delivering the best bias control. The paper also provides a careful bias analysis, showing how intrinsic and noise-driven biases arise and how to mitigate them, particularly by treating the lensing template as an additional data channel. The findings guide the design of future CMB polarization experiments and analysis pipelines, highlighting the trade-offs between gradient-order and inverse-lensing methods and proposing practical strategies to enhance sensitivity to $r$ while controlling biases.

Abstract

Recognizing the impact of contamination from weak gravitational lensing B-modes induced by Large Scale Structure, we examine delensing methods to enhance sensitivity to the tensor-to-scalar ratio $r$ in primordial B-mode detection experiments. This study presents a realistic pipeline to improve $r$ constraints using foreground-cleaned maps with negligible residuals. The pipeline, based on simulations, is adaptable for future experiments. We focus on two delensing approaches: (1) subtracting the gradient-order lensing B-mode template, computed by convolving the E-mode with the lensing potential, from the observed B-mode signal; and (2) remapping observations using the estimated inverse deflection angle. For parameter constraints, we employ three models to reduce $r$ uncertainty and bias, finding consistent uncertainties across models, though biases vary due to the multipole-dependence of the delensing fraction. %We demonstrate the pipeline using simulated maps from future CMB polarization experiments, including a small-aperture (sub-1m) telescope, a large-aperture (6m) telescope, and a future space mission (3m). We demonstrated this pipeline using simulated observation maps from future CMB polarization experiments, which included current representative ground-based small aperture telescopes (sub-1m), next-generation ground-based large aperture telescopes (6m), and highly competitive future space-based medium aperture missions (3m). Results show a delensing efficiency of 40\% with the small-aperture telescope alone, increasing to 65\% when combined with the large-aperture telescope, and 80\% with the satellite mission. These lead to reductions in $r$ uncertainty by 46\% for ground-based and 63\% for space missions. The most promising method adds the lensing template B-mode as an additional frequency channel, minimizing bias on $r$.

Figures (15)

• Figure 1: Terms of the delensed BB power divided by the theory lensed BB power.
• Figure 2: Various masks used in the simulation. Left panels are used for gSAT and gLAT and right panels are used for sMAT. Panels at the top are the apodized mask used for the delensing procedure, and panels at the bottom show the apodized mask used for the calculation of pseudo-Cl with NaMaster.
• Figure 3: Flowchart of the delensing pipeline. We separate our pipeline into four parts: map simulation (top), lensed CMB maps processing (left upper), lensing potential processing (right upper) and the implementation of two delensing methods (lower).
• Figure 4: The lensing reconstruction noise power spectrum of the three experiments. The label of each curve represent the two CMB fields used to reconstruct. The noisiest $\phi$ is reconstructed from the gSAT (top-left), and combining it with the gLAT (top-right) leads to a reduction in reconstruction noise (bottom-left). The best performance is from the Satellite (bottom-right).
• Figure 5: The lensing reconstruction noise power spectrum of the ground experiments with different multipole range of observation fields used for lensing reconstruction. Regarding MV estimators, compared to our baseline configuration (solid), excluding more large scale modes (dash) does not lead to an evident increase in $N^{(0)}$, while excluding more small scale modes (dot) does lead to an evident increase in $N^{(0)}$.