[Paper Review] Towards primordial gravitational waves and $n_s=1$ in light of BICEP/Keck, DESI BAO and Hubble tension

Hao Wang, Gen Ye+2 more|arXiv (Cornell University)|Sep 26, 2024

Cosmology and Gravitation Theories7 citations

TL;DR

The paper combines DESI BAO, BICEP/Keck BK18, Planck, and Pantheon Plus data to search for primordial gravitational waves under pre-recombination Hubble-tension resolutions, finding a non-zero lower bound on r and ns=1, with r0.05 around 0.01.

ABSTRACT

Recent observational data seem to show a $\gtrsim 3σ$ evidence for an evolving dark energy (DE) against the cosmological constant, so the standard $Λ$CDM model. In this paper, we perform the search for the primordial gravitational waves with the potential pre-recombination solutions to the Hubble tension, using recent DESI baryon acoustic oscillation measurements combined with BICEP/Keck cosmic microwave background (CMB) B-mode polarization, Planck CMB and Pantheon supernova data, which reveal that the low bound of the tensor-to-scalar ratio $r$ is $> 1.5σ$ non-zero with the bestfit $r_{0.05}\sim 0.01$ and the scalar spectral index $n_s= 1$ (both $|r_{0.05}-0.01|$ and $|n_s-1|\sim {\cal O} (0.001)$). In particular, we observe the unnoticed impact of CMB B-mode polarization data for constraining the nature of DE, which together with early dark energy solutions to the Hubble tension is calling for the return of post-recombination $Λ$CDM.

Motivation & Objective

• Motivate the search for primordial gravitational waves in the context of evolving dark energy and the Hubble tension.
• Investigate how pre-recombination EDE solutions affect constraints on the tensor-to-scalar ratio r and the scalar spectral index n_s.
• Assess whether post-recombination ΛCDM or evolving dark energy is preferred when including B-mode polarization data.
• Evaluate the impact of AdS-EDE and axion-like EDE on r, n_s, and H0.
• Provide Bayesian model comparison between ΛCDM+EDE and w0-wa CDM+EDE scenarios.

Proposed method

• Use MontePython-3.6 with CLASS (modified) to perform MCMC analyses.
• Incorporate Planck18, BK18, DESI DR1 BAO data, and Pantheon Plus (with Cepheid calibration).
• Model dark energy with w(z)=w0+w_a z/(1+z) after matter-radiation equality, with pre-recombination EDE components (AdS-EDE or axion-like EDE).
• Constrain r at r0.05 and ns, exploring r–ns contours under different DE/EDE scenarios.
• Compare Bayesian evidences Z_model (via MCEvidence) to assess post-recombination ΛCDM-like preference.

Experimental results

Research questions

• RQ1 Does the DESI+BK18+Planck+Pantheon Plus data favor a non-zero tensor-to-scalar ratio r?
• RQ2 Is the scalar spectral index n_s consistent with unity (n_s=1) under these datasets and models?
• RQ3 How do pre-recombination EDE scenarios (AdS-EDE and axion-like EDE) bias or shift constraints on r and n_s?
• RQ4 Which model—ΛCDM+EDE or w0-wa CDM+EDE—better fits the data according to Bayesian evidence?
• RQ5 What is the implied H0 in the favored model and how does it relate to the Hubble tension?

Key findings

• For w0-wa CDM with AdS-EDE, r0.05=0.0134^{+0.0045}_{-0.0118}, with the lower bound ≈ 1.5σ non-zero.
• For w0-wa CDM with axion-like EDE, r0.05=0.0150^{+0.0061}_{-0.0127}, with the lower bound ≈ 1.6σ non-zero.
• Across cases, n_s is found to be unity within ∼0.001 (n_s≈1).
• Table 4 and related results show H0 around 73 km/s/Mpc in ΛCDM+AdS EDE and slightly lower or similar in other EDE variants.
• The Bayesian evidence ratio suggests a modest preference for post-recombination ΛCDM-like behavior when AdS-EDE is included (ln B=0.54).
• The inclusion of B-mode polarization data impacts DE constraints and can suppress evolving DE scenarios, nudging toward ΛCDM-like behavior after recombination.