Tracing cosmic structure with neutral hydrogen after the Epoch of Reionization

TL;DR

This work investigates the transition of neutral hydrogen from the end of the Epoch of Reionization to late-time large-scale structure using the redshifted 21-cm line and forecasts for SKA-Low in the range $3 < z < 7$. The authors combine the semi-numerical code 21cmFAST with a halo-based post-processing HI model to produce a brightness temperature field and its power spectrum during the transition, accounting for ionized bubbles, self-shielded HI, and halo-based HI residing in dark matter haloes. They analyze the evolution of the HI power spectrum, HI bias, and $\Omega_{\text{HI}}$, and evaluate detectability with SKA-Low deep surveys under different foreground avoidance schemes and astrophysical assumptions, finding a four-order-of-magnitude power drop from $z=7$ to $z=4$ and a characteristic flattening during the transition due to neutral islands. The results indicate that deep SKA-Low observations will be a powerful probe of reionization parameters and late-time cosmology, with $P_{\text{HI}}(k)$ measurable at large scales ($k \lesssim 1\,h\mathrm{Mpc}^{-1}$) and robust to variations in X-ray heating, though sensitive to the ionizing photon escape fraction and star formation rate.

Abstract

We present a study of the transition of Neutral Hydrogen (HI) gas from the end of the Epoch of Reionization (EoR) to late-time large-scale structure. We examine the signature of the transition as traced through the redshifted 21-cm line with SKA-Low at $3 < z < 7$. To do so, we use the semi-numerical simulation \textsc{21cmFAST} to model the HI during the EoR and add a HI-halo based post-processing model of the late-time HI. This approach gives a robust estimate of the amplitude of the HI temperature field and predicts the observable power spectrum during the transition period. We find that our simulation pipeline reproduces the expected power spectrum trends from existing observations and theory, in addition to replicating current observational constraints on $Ω_{\text{HI}}$. Our simulations predict a drop in power of four orders of magnitude between $4 < z < 7$. Assuming an inhomogeneous recombination model, we find a flattening of the power due to lingering neutral islands masking the late-time HI signal for $5 < z< 6.5$. Using SKA-Low deep survey parameters, we find HI power spectrum detectability at scales $k \leq 1$ $h$ Mpc$^{-1}$ for redshifts $3< z < 7$, even when using the horizon limit to mitigate foregrounds. Our results suggest a sufficient SNR of the HI power spectrum tracing the underlying halos $z < 5$, which can be used for late-time cosmology. Our results suggest that the resulting $Ω_{\rm HI}$ constraints can trace different reionization scenarios such as a decreased escape fraction. This study implies that deep SKA-Low observations for $3< z< 7$ will be an important probe to constrain reionization parameters as well as cosmological models.

Figures (16)

• Figure 1: The thermal noise power spectrum at $z=4.5$ (solid line, left axis) and the baseline number density (dashed line, right axis) for the AA4 (red) and AA* (blue) stages of SKA-Low. The thermal noise power spectra were calculated assuming a deep survey (100 deg$^2$ survey area, 5000 h observing time).
• Figure 2: Slices (0.8 $h^{-1}$Mpc thick) of the halo field found using each algorithm. Top: Halo field from nbodykit's FoF halo finder. Bottom: Halo field from the watershed halo finder, acting on the same density field. Due to slices being shown, and slight differences in halo centre position, some haloes located by one halo finder appear 'missing' in the other, but are present in adjacent cells. Additionally, subhaloes are resolved by the watershed halo finder, but not by FoF.
• Figure 3: Consistency check between the HMFs of our watershed halo finder and of the FoF halo finder. Apart from divergences at mass extremes, the two halo finders overall show good agreement.
• Figure 4: Evolution of the HMF across redshift. The dashed lines represent the theoretical Watson HMF 2013MNRAS.433.1230W generated using nbodykit 2013AC.....3...23M, and the solid lines are the binned HMFs computed from the simulation.
• Figure 5: Top: Slice of the overdensity field output by 21cmFAST at $z = 4$. Bottom: Corresponding brightness temperature field at this position and redshift, using OM2. $z=4$ was chosen to highlight the contribution to the signal from halo-based HI.