A new study fixed the bound of the sum of the three neutrino masses from cosmology data to 87 meV/c^2 or less with 95% confidence, using a novel way of combining multiple sources of data.
This rules out the inverted neutrino mass hierarchy (for which the sum of the three neutrino masses exceeds 100 meV). It also reduces the uncertainties in the absolute neutrino masses, which have a minimum value (determined from mass differences in neutrino oscillations and assuming a lightest neutrino mass of almost zero) of about 60 meV. Thus, most uncertainty in the sum of the neutrino masses is the shared 0-9 meV range of uncertainty in the lightest neutrino mass.
The best fit point of the data (i.e. within the one sigma range), constrained by the minimum sum of the three neutrino masses from neutrino oscillation data, is very close to the minimum non-zero value that implies a lightest neutrino mass that is on the order of 1 meV or less.
Without the neutrino oscillation data constraint, the best fit point from cosmology data is actually slightly below the minimum sum of neutrino masses from neutrino oscillation data, although the preference for the below 60 meV value is not statistically significant.
The paper and its abstract are as follows:
We present here up-to-date neutrino mass limits exploiting the most recent cosmological data sets. By making use of the Cosmic Microwave Background temperature fluctuation and polarization measurements, Supernovae Ia luminosity distances, Baryon Acoustic Oscillation observations and determinations of the growth rate parameter, we are able to set the most constraining bound to date, ∑mν<0.09~eV at 95%~CL. This very tight limit is obtained without the assumption of any prior on the value of the Hubble constant and highly compromises the viability of the inverted mass ordering as the underlying neutrino mass pattern in nature. The results obtained here further strengthen the case for very large multitracer spectroscopic surveys as unique laboratories for cosmological relics, such as neutrinos: that would be the case of the Dark Energy Spectroscopic Instrument (DESI) survey and of the Euclid mission.