LiteBIRD Science Goals and Forecasts. $E$-mode Anomalies

TL;DR

This paper investigates whether the observed anomalies in CMB temperature data are due to statistical flukes or new physics by analyzing the correlation between temperature and E-mode polarization using LiteBIRD forecasts.

Contribution

It introduces a novel method of testing the fluke hypothesis through partial correlation analysis of temperature and E-mode polarization data, using constrained and unconstrained simulations.

Findings

E-mode data alone does not confirm or refute the anomalies.

Cross-statistics between temperature and E-mode can moderately reject the fluke hypothesis.

Analysis leverages inpainted Planck data and LiteBIRD forecast simulations.

Abstract

Various so-called anomalies have been found in both the WMAP and Planck cosmic microwave background (CMB) temperature data that exert a mild tension against the highly successful best-fit 6 parameter cosmological model, potentially providing hints of new physics to be explored. That these are real features on the sky is uncontested. However, given their modest significance, whether they are indicative of true departures from the standard cosmology or simply statistical excursions, due to a mildly unusual configuration of temperature anisotropies on the sky which we refer to as the "fluke hypothesis", cannot be addressed further without new information. No theoretical model of primordial perturbations has to date been constructed that can explain all of the temperature anomalies. Therefore, we focus in this paper on testing the fluke hypothesis, based on the partial correlation between the temperature and $E$-mode CMB polarisation signal. In particular, we compare the properties of specific statistics in polarisation, built from unconstrained realisations of the $\Lambda$CDM cosmological model as might be observed by the LiteBIRD satellite, with those determined from constrained simulations, where the part of the $E$-mode anisotropy correlated with temperature is constrained by observations of the latter. Specifically, we use inpainted Planck 2018 SMICA temperature data to constrain the $E$-mode realisations. Subsequent analysis makes use of masks defined to minimise the impact of the inpainting procedure on the $E$-mode map statistics. We find that statistical assessments of the $E$-mode data alone do not provide any evidence for or against the fluke hypothesis. However, tests based on cross-statistical measures determined from temperature and $E$ modes can allow this hypothesis to be rejected with a moderate level of probability.