Time-varying neutrino mass from a supercooled phase transition: current cosmological constraints and impact on the $\Omega_m$-$\sigma_8$ plane

In this paper we investigate a time-varying neutrino mass model, motivated by the mild tension between cosmic microwave background (CMB) measurements of the matter fluctuations and those obtained from low-redshift data. We modify the minimal case of the model proposed by Dvali and Funcke (2016) that predicts late neutrino mass generation in a post-recombination cosmic phase transition, by assuming that neutrino asymmetries allow for the presence of relic neutrinos in the late-time Universe. We show that, if the transition is supercooled, current cosmological data (including CMB temperature, polarization and lensing, baryon acoustic oscillations, and Type Ia supernovae) prefer the scale factor $a_s$ of the phase transition to be very large, peaking at $a_s\sim 1$, and therefore supporting a cosmological scenario in which neutrinos are almost massless until very recent times. We find that in this scenario the cosmological bound on the total sum of the neutrino masses today is significantly weakened compared to the standard case of constant-mass neutrinos, with $\sum m_\nu<4.8$~eV at 95\% confidence, and in agreement with the model predictions. The main reason for this weaker bound is a large correlation arising between the dark energy and neutrino components in the presence of false vacuum energy that converts into the non-zero neutrino masses after the transition. This result provides new targets for the coming KATRIN and PTOLEMY experiments. We also show that the time-varying neutrino mass model considered here does not provide a clear explanation to the existing cosmological $\Omega_m$-$\sigma_8$ discrepancies.