Does the Hubble constant tension call for new physics?

TL;DR

The authors quantify whether the Hubble constant tension signal warrants new physics by testing additions to ΛCDM both before and after CMB decoupling using a Bayesian framework and a comprehensive data set. They find late-time dark energy variants provide little relief, while an early-universe energy component can offer marginal Bayesian support, with constraints suggesting $w_{\rm EDE}=0.086^{+0.04}_{-0.03}$ and $\Omega_{\rm EDE}\sim2.9\times10^{-3}$, yielding $h$ around $0.714$. Dark radiation with $w=1/3$ can also raise $h$ to near the local value, but Bayesian evidence is only marginally favorable and is sensitive to priors; overall, a persistent tension could be decisively tested by future 1%-level $H_0$ measurements. The results imply that pre-CMB modifications remain the most promising avenue among those considered, though current data do not conclusively favor any single model over ΛCDM.

Abstract

The $Λ$ Cold Dark Matter model ($Λ$CDM) represents the current standard model in cosmology. Within this, there is a tension between the value of the Hubble constant, $H_0$, inferred from local distance indicators and the angular scale of fluctuations in the Cosmic Microwave Background (CMB). We investigate whether the tension is significant enough to warrant new physics in the form of modifying or adding energy components to the standard cosmological model. We find that late time dark energy explanations are slightly disfavoured whereas a pre-CMB decoupling extra dark energy component has a marginally positive Bayesian evidence. A constant equation of state of the additional early energy density is constrained to 0.086$^{+0.04}_{-0.03}$. Although this value deviates significantly from 1/3, valid for dark radiation, the latter is not disfavoured based on the Bayesian evidence. If the tension persists, future estimates of $H_0$ at the 1$\%$ level will be able to decisively determine which of the proposed explanations is favoured.

Figures (6)

• Figure 1: CMB/BAO/SN Ia constraints on the linear bimetric model with $B_0$ and $B_1$, including the local Hubble constant prior. Observational data pushes $B_1\to 0$, corresponding to the $\Lambda$CDM limit of the model, giving $h=0.695\pm 0.006$, very similar to the $\Lambda$CDM value for the same combination of data, indicated by the vertical blue line.
• Figure 2: CMB/BAO/SN Ia constraints on quadratic bimetric model with $B_1$ and $B_2$, including the local Hubble constant prior. Observational data pushes $B_2\to -\infty$, corresponding to the $\Lambda$CDM limit of the model, giving $h=0.695\pm 0.006$, very similar to the $\Lambda$CDM value, indicated by the vertical blue line, for the same combination of data.
• Figure 3: CMB/BAO/SN Ia and local Hubble constant constraints on a phenomenological dark energy model for which at $z>z_t$, the dark energy density is zero. The inferred Hubble constant is $h=0.699 \pm 0.008$, close to the $\Lambda$CDM value for the same combination of data, indicated by the vertical blue line.
• Figure 4: CMB/BAO/SN Ia and local Hubble constant constraints on a model for which at $z>z_t$, the cosmological constant can change value to $\Omega_{\Lambda,t}$. The data does not push for this possibility, and $h= 0.697\pm 0.007$. The vertical blue line is the $\Lambda$CDM value for the same combination of data.
• Figure 5: Constraints on $h$ and $\Omega_{\rm DR}$ for a combination of CMB, BAO, SN Ia and local $h$ data. The Hubble tension is solved by an additional $\sim 10\,\%$ relativistic energy density in the form of dark radiation. The vertical blue line is the $\Lambda$CDM value for the same combination of data.