The effect of neutrinos on the matter distribution as probed by the Intergalactic Medium

TL;DR

This work presents a comprehensive, fully non-linear treatment of neutrino effects on the matter distribution as probed by the intergalactic medium, using hydrodynamical simulations with both particle-based and grid-based neutrino implementations. The authors quantify how neutrino free-streaming suppresses the non-linear matter power spectrum and alters Lyman-alpha forest observables, demonstrating a scale- and redshift-dependent effect that is larger than linear theory and degenerate with $\sigma_8$. By generating mock Lyman-alpha spectra and comparing to SDSS data, they derive a conservative 2σ upper limit of $\Sigma m_\nu < 0.86$ eV and find $\sigma_8 = 0.85 \pm 0.04$ from the flux power spectrum alone. The study also clarifies methodological trade-offs between particle- and grid-based neutrino modeling, highlights numerical sensitivities (e.g., starting redshift, neutrino sampling), and argues for hybrid approaches to push neutrino-mass constraints with upcoming data toward percent-level precision.

Abstract

We present a suite of full hydrodynamical cosmological simulations that quantitatively address the impact of neutrinos on the (mildly non-linear) spatial distribution of matter and in particular on the neutral hydrogen distribution in the Intergalactic Medium (IGM), which is responsible for the intervening Lyman-alpha absorption in quasar spectra. The free-streaming of neutrinos results in a (non-linear) scale-dependent suppression of power spectrum of the total matter distribution at scales probed by Lyman-alpha forest data which is larger than the linear theory prediction by about 25% and strongly redshift dependent. By extracting a set of realistic mock quasar spectra, we quantify the effect of neutrinos on the flux probability distribution function and flux power spectrum. The differences in the matter power spectra translate into a ~2.5% (5%) difference in the flux power spectrum for neutrino masses with Sigma m_ν = 0.3 eV (0.6 eV). This rather small effect is difficult to detect from present Lyman-alpha forest data and nearly perfectly degenerate with the overall amplitude of the matter power spectrum as characterised by sigma_8. If the results of the numerical simulations are normalized to have the same sigma_8 in the initial conditions, then neutrinos produce a smaller suppression in the flux power of about 3% (5%) for Sigma m_ν = 0.6$ eV (1.2 eV) when compared to a simulation without neutrinos. We present constraints on neutrino masses using the Sloan Digital Sky Survey flux power spectrum alone and find an upper limit of Sigma m_ν < 0.9$ eV (2 sigma C.L.), comparable to constraints obtained from the cosmic microwave background data or other large scale structure probes.

Figures (2)

• Figure 1: Density slices of thickness 6 $h^{-1}$ comoving Mpc at $z=3$ extracted from two 60$h^{-1}$ Mpc hydrodynamical simulations with gas and dark matter and no neutrinos. The right column shows a simulation that includes neutrinos with $\Sigma m_{\nu}$ =1.2 eV. The presence of neutrinos (bottom panel, green) clearly affects both the gas (red) and the dark matter (blue) distribution.
• Figure 2: Left: Dimensionless matter power spectrum at $z=3$. We show the following quantities: linear matter power spectrum for a model with massive neutrinos with $\Sigma m_{\nu}$ =0.6 eV (thin black line); non-linear matter power spectrum obtained with the particle implementation (thick blue curve) and with the grid implementation (thick orange curve); non-linear matter power spectrum for a model without neutrinos (thick black line); linear neutrino power spectrum (thin blue curve); Poisson contribution due to neutrinos (dashed red curve). All results are for simulations with box size $512$ Mpc$/h$. $N_{\nu}=512^3$ for the particle based and $PM=512^3$ for the grid based implementation of neutrinos. Right: Fractional difference of the matter power spectrum for simulations with the grid and particle based implementation of neutrinos at different redshifts ($z=0,1,3$ shown as the red, blue and black curves, respectively) for the large box size simulations with a starting redshift $z=49$.