Unified approach to secondary effects on the CMB B-mode polarization

TL;DR

This paper introduces a unified line-of-sight integration method to accurately estimate all secondary nonlinear effects on CMB B-mode polarization, confirming their negligible impact on primordial gravitational wave detection.

Contribution

It develops a comprehensive framework that includes all secondary effects on CMB polarization, unifying nonlinear gravitational effects and extended sources in a single formalism.

Findings

Secondary effects induce only 0.001-0.01% corrections to the B-mode spectrum.

The approach confirms the reliability of the remapping method for upcoming experiments.

All secondary effects are properly defined and non-overlapping.

Abstract

We develop a systematic and unified approach to estimate all possible secondary (i.e. non-primordial) nonlinear effects to the cosmic microwave background (CMB) polarization, named curve-of-sight integration approach. In this approach, the Boltzmann equation for polarized photons is rewritten in a line-of-sight integral along an exact geodesic in the perturbed universe, rather than a geodesic in the background universe used in the linear-order CMB calculation. This approach resolves the difficulty to solve the Boltzmann hierarchy with the nonlinear gravitational effects in the photon free-streaming regime and thus unifies the standard remapping approach for CMB lensing into the direct approach solving the Boltzmann equation for the nonlinear collisional effects. In this paper, we derive formulae that: (i) include all the nonlinear effects; (ii) can treat extended sources such as the contributions after the reionization. It offers a solid framework to discuss possible systematics in the standard estimation of CMB lensing by the remapping approach. As an explicit demonstration, we estimate the secondary B-mode power spectrum induced by all foreground gravitational effects: lensing, redshift, time-delay, emission-angle, and polarization-rotation effects. We define these effects properly so that they do not have any overlap, also without overlooking any effect. Then, we show that these effects only give corrections of the order of 0.001-0.01% to the standard lensing-induced B-mode power spectrum in the concordance $\Lambda$ cold dark matter model. Our result confirms the reliability of using the remapping approach in upcoming CMB experiments aiming to detect the primordial gravitational waves with the tensor-to-scalar ratio of $r \sim 10^{-3}$.