Robust Neutrino Constraints by Combining Low Redshift Observations with the CMB

TL;DR

The paper demonstrates that incorporating precise low-redshift measurements with CMB data significantly tightens cosmological neutrino constraints and tests their robustness to late-time expansion history. Using HST $H_0$, maxBCG cluster power, SN, BAO, and WMAP5, the authors derive a robust upper bound of $\sum m_{\nu}<0.4$ eV (95% CL) in $\Lambda$CDM, dropping further to $<0.3$ eV when SN and BAO are included. They show these bounds remain strong under one-parameter extensions but can relax modestly in more general scenarios, e.g., a dark-coupled model yielding $\sum m_{\nu}<0.51$ eV. For the number of relativistic species, the combination of $H_0$ and cluster constraints yields $N_{\rm rel}=3.76^{+0.63}_{-0.68}$ (68% CL), with profile likelihood analyses indicating prior-independent support when key datasets are included. Overall, the study underscores the power of low-redshift probes to constrain neutrino properties in a way that is robust to several extensions of the standard cosmological model.

Abstract

We illustrate how recently improved low-redshift cosmological measurements can tighten constraints on neutrino properties. In particular we examine the impact of the assumed cosmological model on the constraints. We first consider the new HST H0 = 74.2 +/- 3.6 measurement by Riess et al. (2009) and the sigma8*(Omegam/0.25)^0.41 = 0.832 +/- 0.033 constraint from Rozo et al. (2009) derived from the SDSS maxBCG Cluster Catalog. In a Lambda CDM model and when combined with WMAP5 constraints, these low-redshift measurements constrain sum mnu<0.4 eV at the 95% confidence level. This bound does not relax when allowing for the running of the spectral index or for primordial tensor perturbations. When adding also Supernovae and BAO constraints, we obtain a 95% upper limit of sum mnu<0.3 eV. We test the sensitivity of the neutrino mass constraint to the assumed expansion history by both allowing a dark energy equation of state parameter w to vary, and by studying a model with coupling between dark energy and dark matter, which allows for variation in w, Omegak, and dark coupling strength xi. When combining CMB, H0, and the SDSS LRG halo power spectrum from Reid et al. 2009, we find that in this very general model, sum mnu < 0.51 eV with 95% confidence. If we allow the number of relativistic species Nrel to vary in a Lambda CDM model with sum mnu = 0, we find Nrel = 3.76^{+0.63}_{-0.68} (^{+1.38}_{-1.21}) for the 68% and 95% confidence intervals. We also report prior-independent constraints, which are in excellent agreement with the Bayesian constraints.

Figures (7)

• Figure 1: $\Lambda$CDM WMAP5-only constraints with $\sum m_{\nu} = 0$ are shown in blue. The black transparent contours show how the constraints on $\sigma_8 (\Omega_m/0.25)^{0.41}$ and $H_0$ degrade when $\sum m_{\nu}$ is left as a free parameter. The dotted lines show contours on which $\left < \sum m_{\nu} \right > = 0.2, 0.4, 0.6, ..., 1.4$ eV for the WMAP5-only posterior distribution. The green shows the constraints when maxBCG and $H_0$ constraints are also included. The 95% confidence upper limit is reduced from 1.3 eV to 0.4 eV.
• Figure 2: Same as Figure \ref{['fig:mnulcdm']} but for the $\Lambda$CDM model with base dataset WMAP$+$BAO$+$SN. The dotted lines show contours on which $\left < \sum m_{\nu} \right > = 0.2, 0.3, 0.4, 0.5, 0.6$ eV for the WMAP$+$BAO$+$SN posterior distribution.
• Figure 3: Two-dimensional constraints in $\xi-\sum m_{\nu}$ in the dark coupling model of ref. gavela/etal:2009. The red contours are the results from the CMB+ $H_0$riess/etal:2009+ $\hat{P}_{halo}(k)$reid/etal:2009+ Union SN kowalski/etal:2008. With this data-set combination there is no longer a degeneracy between coupling and neutrino mass. As discussed in Section \ref{['darkcouplingresults']}, the strong changes in the growth of structure caused by the dark coupling prevent us from applying the maxBCG constraint to this model. However, the blue contours, which also include the maxBCG constraint, demonstrate that the constraint on $\sum m_{\nu}$ is not tightened even if this additional constraint could be included.
• Figure 4: The profile likelihood defined in Section \ref{['method']} in bins of $\Delta (\sum m_{\nu}) = 0.2$ eV for the $\Lambda$CDM model with WMAP5 only (crosses), WMAP5+maxBCG (stars), WMAP5+$H_0$ (triangles), and WMAP5+maxBCG+$H_0$ (diamonds). The black curves overlay a quadratic fit to these points, illustrating that a Gaussian curve provides a good fit to this one-dimensional distribution.
• Figure 5: $\sum m_{\nu}$ vs. the mass of the lightest neutrino in the normal and inverted hierarchies.