Dark energy and neutrinos along the cosmic expansion history

TL;DR

This work tests whether dark energy is dynamical by jointly reconstructing the equation of state $w_\text{DE}$ and the total neutrino mass $\sum m_\nu$ using a model-independent PCHIP interpolation across seven redshift nodes. It explores two dark energy scenarios (non-phantom and phantom) and three neutrino treatments (fixed $\sum m_\nu$, free, and PCHIP-reconstructed), leveraging Planck CMB, DESI BAO, and Pantheon SN data. The main finding is general compatibility with $\Lambda$CDM, but a $95\%$ CL deviation from $w_\text{DE}=-1$ appears around $z\sim1.2$ when DESI BAO data are included with phantom crossing; neutrino constraints weaken in the phantom case and under PCHIP reconstruction, highlighting a significant degeneracy between dark energy and neutrino sectors. The results underscore that late-time data and the chosen non-parametric reconstruction can influence inferred dynamics of dark energy and neutrino masses, and more data will be essential to decisively distinguish dynamical dark energy from a cosmological constant. A companion paper will study theory-motivated models that realize time-varying neutrino masses in tandem with dark energy dynamics.

Abstract

Recent cosmological measurements are hinting that dark energy may evolve, with its equation of state, $w_\mathrm{DE}$, even showing oscillatory patterns. In this work, we employ a model-independent approach to jointly reconstruct $w_\mathrm{DE}$ and the sum of neutrino masses, $\sum m_ν$, adopting the PCHIP method with seven fixed nodes in which we allow the two parameters to vary. We employ CMB, Baryon Acoustic Oscillations and Supernovae Ia data to constrain the values of $w_\mathrm{DE}$ and $\sum m_ν$ at each node. We conduct three different analyses in which we reconstruct $w_\mathrm{DE}$: one with fixed $\sum m_ν=0.06~\mathrm{eV}$; one in which we allow $\sum m_ν$ to vary, and one in which we also reconstruct $\sum m_ν$ using the PCHIP method. We find the dark energy equation of state to be consistent with the cosmological constant scenario, except when including DESI data and allowing for phantom crossing, where we find a $95\%$ CL deviation from $w_\mathrm{DE}=-1$ around $z\sim1.2$. For neutrino masses, we obtain looser constraints when focusing on phantom dark energy, that show further early and late relaxation when reconstructing the mass via the PCHIP method.

Figures (14)

• Figure 1: Redshift points for each set of measurements considered in the analyses. We consider the PantheonPlus collection for Supernovae measurements and the second data release from the DESI collaboration as BAO measurements. The vertical lines depict the redshift points (nodes) we shall use for the PCHIP reconstruction.
• Figure 2: Reconstruction of $w_\mathrm{DE}$ obtained using the PCHIP method. We show the 68 and 95 % C.L. bounds for the case in which we fix $\sum m_\nu$ (shaded regions) and for the case in which we leave it free in the MCMC (lines: dot-dashed is 68 % C.L.; solid is 95 % C.L.).
• Figure 3: Reconstruction of $w_\mathrm{DE}$ and $\sum m_\nu$ obtained using the PCHIP method. We show the 68 and 95 % C.L. bounds, both for the case in which we allow for phantom dark energy (shaded regions) and in which we don't allow for it (lines: dot-dashed is 68 % C.L.; solid is 95 % C.L.).
• Figure 4: Contour plot showing the 68 and 95 % C.L. for the two parameters $H_0$ and $\sigma_8$, for all the analyses considered. Notice the different x axis values for the two rows.
• Figure 5: Contour plot showing the 68 and 95 % C.L. for the two parameters $H_0$ and $\Omega_\mathrm{m}$, for all the analyses considered. Notice the different x axis values for the two rows.