First evidence of running cosmic vacuum: challenging the concordance model

TL;DR

The paper tests whether dynamical vacuum energy, encoded in running vacuum models with ρ_Λ(H) and possibly a running G(H), can describe cosmic expansion and structure formation better than the rigid ΛCDM. It analyzes two RVM classes (type-G with running G and matter conservation, and type-A with constant G and anomalous matter conservation), derives E(a) and ρ_Λ(H;ν,α), and fits them to a large, multi-probe data set including SNIa, BAO, H(z), LSS, BBN, and Planck 2015 CMB priors. The results show ν_eff > 0 across RVMs and a significant preference for dynamics over ΛCDM, with Planck 2015 data yielding ≳4.2σ evidence for running vacuum when marginalizing over other parameters, and large ΔAIC/ΔBIC differences that favor RVMs over ΛCDM and even XCDM. The findings imply a dynamical DE component possibly connected to QFT in curved spacetime and underscore the critical role of BAO+LSS+CMB data in constraining vacuum dynamics, while highlighting the need for future observations to confirm the signal.

Abstract

Despite the fact that a rigid $Λ$-term is a fundamental building block of the concordance $Λ$CDM model, we show that a large class of cosmological scenarios with dynamical vacuum energy density $ρ_Λ$ and/or gravitational coupling $G$, together with a possible non-conservation of matter, are capable of seriously challenging the traditional phenomenological success of the $Λ$CDM. In this paper, we discuss these "running vacuum models" (RVM's), in which $ρ_Λ=ρ_Λ(H)$ consists of a nonvanishing constant term and a series of powers of the Hubble rate. Such generic structure is potentially linked to the quantum field theoretical description of the expanding Universe. By performing an overall fit to the cosmological observables $SNIa+BAO+H(z)+LSS+BBN+CMB$ (in which the WMAP9, Planck 2013 and Planck 2015 data are taken into account), we find that the class of RVM's appears significantly more favored than the $Λ$CDM, namely at an unprecedented level of $\gtrsim4.2σ$. Furthermore, the Akaike and Bayesian information criteria confirm that the dynamical RVM's are strongly preferred as compared to the conventional rigid $Λ$-picture of the cosmic evolution.

Figures (6)

• Figure 1: Likelihood contours in the $(\Omega_m,\nu_{\rm eff})$ plane for the values $-2\ln\mathcal{L}/\mathcal{L}_{max}=2.30$, $6.18, 11.81$, $19.33$, $27.65$ (corresponding to 1$\sigma$, 2$\sigma$, 3$\sigma$, 4$\sigma$ and 5$\sigma$ c.l.) after marginalizing over the rest of the fitting parameters indicated in Table 1. We display the progression of the contour plots obtained for model G2 using the 90 data points on SNIa+BAO+$H(z)$+LSS+BBN+CMB, as we evolve from the high precision CMB data from WMAP9, Planck 2013 and Planck 2015 -- see text, point S7). In the sequence, the prediction of the concordance model ($\nu_{\rm eff}=0$) appears increasingly more disfavored, at an exclusion c.l. that ranges from $\sim 2\sigma$ (for WMAP9), $\sim 3.5\sigma$ (for Planck 2013) and up to $4\sigma$ (for Planck 2015). Subsequent marginalization over $\Omega_m$ increases slightly the c.l. and renders the fitting values indicated in Table 1, which reach a statistical significance of $4.2\sigma$ for all the RVM's. Using numerical integration we can estimate that $\sim99.81\%$ of the area of the $4\sigma$ contour for Planck 2015 satisfies $\nu_{\rm eff}>0$. We also estimate that $\sim95.47\%$ of the $5\sigma$ region also satisfies $\nu_{\rm eff}>0$. The corresponding AIC and BIC criteria (cf. Table 1) consistently imply a very strong support to the RVM's against the $\Lambda$CDM.
• Figure 2: As in Fig. 1, but for model A2. Again we see that the contours tend to migrate to the $\nu_{\rm eff}>0$ half plane as we evolve from WMAP9 to Planck 2013 and Planck 2015 data. Using the same method as in Fig. 1, we find that $\sim99.82\%$ of the area of the $4\sigma$ contour for Planck 2015 (and $\sim95.49\%$ of the corresponding $5\sigma$ region) satisfies $\nu_{\rm eff}>0$. The $\Lambda$CDM becomes once more excluded at $\sim 4\sigma$ c.l. (cf. Table 1 for Planck 2015).
• Figure 3: As in Fig. 1 and 2, but for model XCDM and using Planck 2015 data. The $\Lambda$CDM is excluded at $\sim 4\sigma$ c.l. (cf. Table 1).
• Figure 4: The $f(z)\sigma_8(z)$ data (Table 4) and the predicted curves by the RVM's, XCDM and the $\Lambda$CDM, using the best-fit values in Table 1. Shown are also the values of $\sigma_8(0)$ that we obtain for all the models. The theoretical prediction of all the RVM's are visually indistinguishable and they have been plotted using the same (blue) dashed curve.
• Figure 5: Reconstruction of the contour lines for model A2, under Planck 2015 CMB data (rightmost plot in Fig. 2) from the partial contour plots of the different SNIa+BAO+$H(z)$+LSS+BBN+CMB data sources. The $1\sigma$ and $2\sigma$ contours are shown in all cases. For the reconstructed final contour lines we also plot the $3\sigma$, $4\sigma$ and $5\sigma$ regions.