Cosmological tensions in Proca-Nuevo theory

Abstract

We study the cosmological predictions of (extended) Proca-Nuevo theory. This vector-tensor theory enjoys stable homogeneous and isotropic solutions characterized by an effective dark energy fluid, with behavior that ranges from freezing quintessential to thawing phantom-like, serving as a motivated framework to scrutinize the cosmological tensions that affect the standard $Λ$CDM model. While the model we consider is sufficiently generic to encompass a large class of field theories, it distinguishes itself from scalar dark energy models (quintessential ones, kinetic ones and non-minimally coupled ones) by the presence of what would be classed as a vector degree of freedom which can be for instance inherited from more generic theories of gravity. We improve on previous work in several directions: we consider a general one-parameter class of background models; identify a so-called 'special' model and analyze observational constraints taking also into account perturbations and making use of wide up-to-date catalogs of datasets including recently released ones. We find that the one-parameter Proca-Nuevo model is preferred over $Λ$CDM at $1.5σ$ when fitting CMB and BAO data, and at $2.4σ$ when further adding low-redshift data. The Hubble tension is alleviated, dropping from $5.8σ$ to $2.3σ$ (resp. $1.5σ$) between CMB with (and resp. without) BAO data and local measurements. On the other hand, we find that the vector field generically introduces a significant enhancement of the effective Newton constant, so that matching the observed matter power spectrum requires a mild amount of tuning to suppress the impact of perturbations. Since, at the background level, Proca-Nuevo is degenerate with other classes of theories, our results are also relevant to a wider range of set-ups including and beyond vector-tensor models.

Figures (12)

• Figure 1: The evolution of the dark energy equation of state parameter $w_\mathrm{EPN}$ (left panel) and its energy fraction $\Omega_\mathrm{EPN}$ (right). The curves display the results in the $\Lambda$CDM model (dotted black), EPN special M model with $M = -1/3$ (solid yellow), and EPN general models with $M = -1/2$ (dashed pink) and $M = -3/5$ (dashed-dotted blue).
• Figure 2: Marginalized posterior constraints on $H_0$ and $\Omega_\mathrm{m0}$ using CMB and PPS data in $\Lambda$CDM and three EPN special M models. Contours indicate 68% and 95% confidence level intervals.
• Figure 3: Marginalized posterior constraints on $M$, $H_0$ and $\Omega_{\rm m0}$ using different (combinations of) datasets in the EPN special M model. Contours indicate 68% and 95% confidence level intervals. Dashed lines indicate best fit values for $\Lambda$CDM, i.e. with $M=0$. Although the CMB, DESI and PPS datasets do not individually constrain the parameter $M$ well, the combined dataset CMB + DESI + PPS favors $M<0$ at the $\gtrsim 2\sigma$ confidence level ($\gtrsim1\sigma$ for CMB + DESI).
• Figure 4: Left panel: 68% and 95% marginalized posterior constraints on $S_8$, $\Omega_{\rm m0}$ and $H_0$ using CMB, CMB+SDSS datasets in $\Lambda$CDM, the EPN special M ($M=-1/3$), very special and full special models. The vertical dashed line in $S_8$ corresponds to the best fit + 1$\sigma$ value $S_8=0.776\pm0.017$ derived for $\Lambda$CDM from DES Y3. Right panel: marginalized posterior constraints on $M$ and the Proca mass parameter $c_m$ in the EPN very special and full special models from the same two datasets.
• Figure 5: The lensing potential of the best-fit $\Lambda$CDM model and the EPN full special model given the best-fit parameters for the CMB dataset, with different values of the Proca mass parameter $c_m$, including the best-fit value $c_m=10^{4.48}$ (red curve). The strong enhancement for $c_m = \mathcal{O}(1)$ values is manifest.