Separation of Gravitational-Wave and Cosmic-Shear Contributions to Cosmic Microwave Background Polarization

Abstract

Inflationary gravitational waves (GW) contribute to the curl component in the polarization of the cosmic microwave background (CMB). Cosmic shear -gravitational lensing of the CMB-converts a fraction of the dominant gradient polarization to the curl component. Higher-order correlations can be used to map the cosmic shear and subtract this contribution to the curl. Arcminute resolution will be required to pursue GW amplitudes smaller than those accessible by the Planck surveyor mission. The blurring by lensing of small-scale CMB power leads with this reconstruction technique to a minimum detectable GW amplitude corresponding to an inflation energy near 10 15 GeV. Observation of acoustic oscillations in the temperature anisotropies of the cosmic microwave background (CMB) [1] strongly suggests an inflationary origin for primordial perturbations [2]. It has been argued that a new "smokinggun" signature for inflation would be the detection of a stochastic background in inflationary gravitational waves (IGWs) [3]. These IGWs produce a distinct signature in the CMB in the form of a contribution to the curl, or magneticlike, component of the polarization [4]. Since there is no scalar, or density-perturbation, contribution to these curl modes, curl polarization was considered to be a direct probe of IGWs.There is, however, another source of a curl component. Cosmic shear (CS) -weak gravitational lensing of the CMB due to large-scale structure along the line of sightresults in a fractional conversion of the gradient mode from density perturbations to the curl component [5]. The amplitude of the IGW background varies quadratically with the energy scale E infl of inflation, and so the prospects for detection also depend on this energy scale. In the absence of CS, the smallest detectable IGW background scales simply with the sensitivity of the CMB experiment -as the instrumental sensitivity is improved, smaller values of E infl become accessible [3,6]. More realistically, however, the CS-induced curl introduces a noise from which IGWs must be distinguished. If the IGW amplitude (or E infl ) is sufficiently large, the CS-induced curl will not be a problem. However, as E infl is reduced, the IGW signal becomes smaller and will get lost in the CS-induced noise. This confusion leads to a minimum detectable IGW amplitude [7].In addition to producing a curl component, CS also introduces distinct higher-order correlations in the CMB temperature pattern. Roughly speaking, lensing can stretch the image of the CMB on a small patch of sky and thus lead to something akin to anisotropic correlations on that patch of sky, even though the CMB pattern at the surface of last scatter had isotropic correlations. By mapping these effects, the CS can be mapped as a function of position on the sky [8]. The observed CMB polarization can then be corrected for these lensing deflections to reconstruct the intrinsic CMB polarization at the surface of last scatter (in which the only curl component would be that due to IGWs). In this Letter we evaluate...

Small-scale cosmic microwave background polarization from reionization

“…3, however, there is a contribution to the B-mode power spectrum that arises from conversion of the primary E modes to B modes by gravitational lensing [12], and this is considerably larger than the secondary polarization. Moreover, we also show (the dot-dash curve) the contribution to the irreducible Bmode power spectrum that remains even after the lensing has been optimally subtracted with higher-order correlations [32][33][34][35]. This residual lensing power spectrum is considerably larger than the polarization from reionization.…”

Forecasting sensitivity on tilt of power spectrum of primordial gravitational waves after Planck satellite

“…However, the lensing B-modes and primordial B-modes have the same frequencydependence, and we cannot separate them as mentioned above. Fortunately, one can reconstruct the lensing potential by considering the CMB data of temperature and E-mode polarizations at small angular scales, and then remove the lensing B-modes away at large angular scales [42,43,44,45]. In this paper, we assume the power of lensing B-modes reduced to 10% of its original value for the CMBPol and COrE experiments.…”

CMB Polarization B-mode Delensing with SPTpol and Herschel

“…However this method is sub-optimal because the multicomponent fit is subject to the sample variance of the lensing B modes. Instead, the specific realization of lensing B modes on the sky can be characterized and removed in a process called "delensing," thus significantly improving constraints on r. While delensing has been studied at a theoretical level for many years (Kesden et al 2002;Knox & Song 2002;Seljak & Hirata 2004;Smith et al 2012;Sherwin & Schmittfull 2015;Simard et al 2015), it has only recently been performed on CMB data (Larsen et al 2016;Carron et al 2017).…”