Prospects for Constraining Neutrino Mass Using Planck and Lyman-Alpha Forest Data

TL;DR

This study investigates how Planck CMB data, in combination with Lyman-α forest measurements, can constrain the sum of neutrino masses and potentially distinguish between normal and inverted hierarchies. Using CAMB and COSMOMC, it maps the parameter space, noting that Planck alone offers limited sub-eV sensitivity but can greatly improve constraints when paired with Ly-α data; however, the ability to detect the minimal normal hierarchy remains unlikely without extremely precise future data and cross-dataset consistency. The results show strong dependence on input datasets and modeling assumptions, particularly regarding Ly-α systematics and data tensions. The authors highlight the potential role of future surveys, including lensing and SKA-like galaxy surveys, to reach sub-0.1 eV sensitivity, while cautioning that achieving such precision requires controlling systematics and robust theoretical modeling of small-scale power.

Abstract

In this paper we investigate how well Planck and Lyman-Alpha forest data will be able to constrain the sum of the neutrino masses, and thus, in conjunction with flavour oscillation experiments, be able to determine the absolute masses of the neutrinos. It seems possible that Planck, together with a Lyman-Alpha survey, will be able to put pressure on an inverted hierarchial model for the neutrino masses. However, even for optimistic assumptions of the precision of future Lyman-Alpha datasets, it will not be possible to confirm a minimal-mass normal hierarchy.

Figures (4)

• Figure 1: A plot illustrating the dependence of the matter power spectrum on the neutrino mass spectrum. The power spectra of three models are illustrated relative to the power spectrum for three equal-mass neutrinos with $\sum m_\nu =0.059$ eV. The upper solid blue curve corresponds to massless neutrinos, the dotted curve corresponds to the minimal normal hierarchy, and the lower curve corresponds to the minimal inverted hierarchy. Also shown is a model with $\sum m_\nu = 0.13$ eV and other cosmological parameters changed so that the model is nearly degenerate in likelihood with the fiducial one against future Planck and Lyman-$\alpha$ datasets.
• Figure 2: A 2-D marginalized likelihood contour plot indicating the possible inconsistency of the wmap and sdsslya data. 68% and 95% confidence intervals are illustrated for the following four datasets (from broadest to tightest): blue, wmap data alone; yellow, "faked" wmap data alone; green, sdsslya and wmap data; red, sdsslya and "faked" wmap data.
• Figure 3: A plot of the marginalized likelihoods for a single neutrino of mass $m$ with assumed future datasets as discussed in the text. All curves use the $C^\text{W}_{0.06}$ Planck dataset. As for the Lyman-$\alpha$ dataset used, black (solid) corresponds to $P^{3@ 3}_{5\%}$, red (dot-dash) to $P^{1@ 1}_{1\%}$, green (short-dash) to $P^{3@ 1}_{1\%}$ and blue (long-dash) to $P^{3@ 3}_{1\%}$.
• Figure 4: A 2D contour plot indicating how a partial parameter degeneracy using only Planck data is lifted when Lyman-$\alpha$ data is added. 68% and 95% confidence intervals are illustrated for the following three datasets (from broadest to tightest): blue, Planck alone; green, Planck with $P^{3@ 3}_{5\%}$; red, Planck with $P^{3@ 3}_{1\%}$.