The neutrino mass bound from WMAP-3, the baryon acoustic peak, the SNLS supernovae and the Lyman-alpha forest

TL;DR

The paper analyzes cosmological bounds on the sum of neutrino masses using a combination of WMAP-3 CMB data, SDSS BAO, SNLS Type Ia supernovae, LSS, and Lyman-$\alpha$ forest measurements. It employs a comprehensive likelihood approach in a flat cosmology with multiple free parameters, illustrating how degeneracies with the effective number of neutrino species and the dark energy equation of state can be broken by including BAO and, to varying degrees, Lyman-$\alpha$ data. Depending on the parameter space and datasets, the bound on $\sum m_\nu$ ranges from about 1.7 eV in a broad 11-parameter model to as tight as $0.27$ eV when all data are combined, with Lyman-$\alpha$ potentially tightening further to $0.2$–$0.4$ eV but facing significant systematics. The work emphasizes the crucial role of BAO in breaking degeneracies and discusses prospects for future surveys to reach sub-0.1 eV sensitivity, potentially resolving the neutrino mass scale.

Abstract

We have studied bounds on the neutrino mass using new data from the WMAP 3 year data, the Sloan Digital Sky Survey measurement of the baryon acoustic peak, the Type Ia supernovae from SNLS, and the Lyman-alpha forest. We find that even in the most general models with a running spectral index where the number of neutrinos and the dark energy equation of state are allowed to vary, the 95% C.L. bound on the sum of neutrino masses is sum m_nu < 0.62 eV (95% C.L.), a bound which we believe to be robust. In the more often used constrained analysis with N_nu =3, w = -1, and alpha_s = 0, we find a bound of 0.48 eV without using the Lyman-alpha data. If Lyman-alpha data is used, the bound shrinks to \sum m_nu < 0.2-0.4 eV (95% C.L.), depending strongly on the Lyman-alpha analysis used.

Figures (3)

• Figure 1: The value of $\Delta \chi^2$ as a function of $\sum m_\nu$ for various different data sets used for the full 11-dimensional parameter space. The curves are identical to the cases described in Table 2: The full curve is case 1, the long-dashed is case 2, the dotted is case 3, and the dashed is case 4.
• Figure 2: The value of $\Delta \chi^2$ as a function of $\sum m_\nu$ for various different data sets used for the restricted 8-dimensional parameter space with $N_\nu=3$, $w=-1$, and $\alpha_s = 0$. The curves are identical to the cases described in Table 3: The full curve is case 1, the long-dashed is case 2, the dotted is case 3, and the dashed is case 4.
• Figure 3: The value of $\Delta \chi^2$ as a function of $\sum m_\nu$ for the restricted 8-dimensional parameter space with $N_\nu=3$, $w=-1$, and $\alpha_s = 0$ for different assumptions about the Lyman-$\alpha$ data. The full line is for the Ref. mcdonald data, the dashed is for the approximate analysis with the data in Ref. seljak2006, and the long-dashed is for the data in Ref. viel2006.