BaBy Cosmic Tension

BaBy Cosmic Tension
Notice: This research summary and analysis were automatically generated using AI technology. For absolute accuracy, please refer to the [Original Paper Viewer] below or the Original ArXiv Source.

We show that the recently released B-mode polarisation data from the South Pole Telescope (SPT) favour a non-vanishing contribution of primordial gravitational waves of inflationary origin which is in tension with the previous BICEP-Keck (BK) measurements. Our analysis uses the third-order slow-roll primordial power spectra, with theoretically motivated priors, on the multifrequency SPT likelihoods complemented by the latest Planck satellite data products. The SPT measurements provide 1.0 bit of information gain on the first slow-roll parameter, which is higher than the 0.9 bit provided by BK even though the SPT sensitivity is five times lower. Moreover, the Bayesian dimensionality on the same parameter exceeds 1.5 for SPT versus 0.3 for BK showing that it is overconstrained by the SPT data. Even if this BB-tension could be the result of a yet to be understood foreground, our findings should motivate for a closer analysis of this unexpected B-modes excess.


💡 Research Summary

The paper investigates a tension between recent B‑mode polarisation measurements from the South Pole Telescope (SPT) and earlier results from the BICEP‑Keck (BK) collaboration. Both experiments probe the same region of the sky, but while BK has set a stringent upper limit on the tensor‑to‑scalar ratio (r < 0.032 at 95 % CL), the newly released SPT data appear to favour a non‑zero contribution from primordial gravitational waves. The authors adopt third‑order slow‑roll inflationary power spectra, parameterised by four slow‑roll parameters (ε₁, ε₂, ε₃, ε₄), and analyse them within a Bayesian framework using a comprehensive data set (Planck NPIPE temperature and polarisation spectra, Planck lensing, BAO from eBOSS DR16, and high‑ℓ SPT TT/TE/EE spectra).

For the slow‑roll parameters they impose flat priors on ε₂‑ε₄ (±0.2) and a Jeffreys‑type prior on ε₁, implemented as a flat prior on log ε₁ in the interval


Comments & Academic Discussion

Loading comments...

Leave a Comment