Is there evidence for additional neutrino species from cosmology?

TL;DR

This paper evaluates whether cosmological observations require extra neutrino species or mass beyond the Standard Model. Using Bayesian Evidence and a prior-independent Profile Likelihood Ratio, it tests several extended neutrino models against a baseline $\Lambda$CDM framework, employing datasets from CMB (WMAP, SPT), CMB lensing, BAO, and $H_0$ measurements. The results show no robust evidence for non-standard $N_{ m eff}$ or $M_{ u}$; parameter estimates hint at mild deviations (e.g., $N_{ m eff}$ slightly above 3.046) but degeneracies and model selection criteria do not favor extension. The conclusions underscore the importance of model selection in cosmology and set the stage for future Planck-era analyses to more decisively constrain neutrino properties.

Abstract

It has been suggested that recent cosmological and flavor-oscillation data favor the existence of additional neutrino species beyond the three predicted by the Standard Model of particle physics. We apply Bayesian model selection to determine whether there is indeed any evidence from current cosmological datasets for the standard cosmological model to be extended to include additional neutrino flavors. The datasets employed include cosmic microwave background temperature, polarization and lensing power spectra, and measurements of the baryon acoustic oscillation scale and the Hubble constant. We also consider other extensions to the standard neutrino model, such as massive neutrinos, and possible degeneracies with other cosmological parameters. The Bayesian evidence indicates that current cosmological data do not require any non-standard neutrino properties.

Figures (7)

• Figure 1: Left: CMB temperature power spectrum measurements from the WMAP satellite (black diamonds) and the South Pole Telescope (green squares). The SPT data include foreground contributions from Poisson-distributed and clustered point sources, as well as the kinetic and thermal Sunyaev-Zel'dovich effects; the scale of this contribution is indicated by the red solid line. To illustrate the degeneracy between ${N_{\rm eff}}$ and ${\Omega_c h^2}$, power spectra for nearly-degenerate models with ${N_{\rm eff}}$ equal to $3.046$ (solid), $2.0$ (dashed) and $5.0$ (dot-dashed) are overlaid on the WMAP and SPT data. Right: the fractional differences of the power spectra with respect to $\Lambda$CDM are shown as a function of scale, with error bars from WMAP and SPT indicated by the dark- and light-grey regions, respectively.
• Figure 2: Left: CMB lensing data from the South Pole Telescope (black diamonds). Overlaid are power spectra for $\Lambda$CDM (solid) and for nearly-degenerate models with ${N_{\rm eff}}$ equal to $4.0$ (dotted), $5.0$ (dashed), $5.8$ (dot-dashed) and $6.0$ (triple-dot-dashed). Right: the fractional differences of the power spectra with respect to $\Lambda$CDM are shown as a function of scale, with SPT error bars indicated by the grey shaded region.
• Figure 3: The $68\%$ and $95\%$ joint confidence limits on the posterior probability distributions of ${N_{\rm eff}}$ and ${\Omega_c h^2}$ (left), and ${N_{\rm eff}}$ and ${H_0}$ (right), given the WMAP (clear), WMAP+SPT+SPTLens (blue), WMAP+SPT+SPTLens+BAO (green) and WMAP+SPT+SPTLens+${H_0}$ (red) datasets. Constraints from the WMAP+SPT datasets are not shown as they are very similar to results from the WMAP+SPT+SPTLens combination. The standard value of ${N_{\rm eff}}$ is indicated by the dot-dashed line.
• Figure 4: The $68\%$ and $95\%$ joint confidence limits on the posterior probability distributions of ${N_{\rm eff}}$ and ${Y_{\rm He}}$, and ${M_{\nu}}$ (plotted in units of eV) and ${\sigma_8}$, given the WMAP (clear), WMAP+SPT+SPTLens (blue), WMAP+SPT+SPTLens+BAO (green) and WMAP+SPT+SPTLens+${H_0}$ (red) datasets. The standard value of ${N_{\rm eff}}$ is indicated by the dot-dashed line.
• Figure 5: The $68\%$ and $95\%$ joint confidence limits on the posterior probability distributions of ${N_{\rm eff}}$ and ${\alpha}$, given the WMAP+SPT+SPTLens (blue) datasets. The standard values of ${N_{\rm eff}}$ and $\alpha$ are indicated by the dot-dashed lines.