HI intensity mapping correlation function from UNIT simulations: BAO and observationally induced anisotropy

TL;DR

This paper analyzes the clustering of HI intensity maps from simulations, focusing on BAO detection and observational effects like beam smoothing and foreground removal, to inform future SKA-like surveys.

Contribution

It introduces a detailed simulation framework for HI intensity mapping, incorporating observational effects and proposing methods to isolate the BAO signal.

Findings

BAO signal is detectable in simulated HI maps with observational effects.

Different 2PCF estimators help mitigate systematics in BAO detection.

Simulation results support the feasibility of BAO measurements with SKA-like surveys.

Abstract

We study the clustering of HI intensity maps produced from simulations with a focus on baryonic acoustic oscillations (BAO) and the effects induced by telescope beam smoothing and foreground cleaning. We start by creating a HI catalogue at $z=1.321$ based on the Semi-Analytic Galaxy Evolution (SAGE) model applied to the UNIT simulations. With this catalogue we investigate the relation between model HI and the dark matter haloes and we also study the abundance of HI, $\Omega_{\rm HI}$, predicted by this model. We then create synthetic HI intensity maps with a Nearest-Grid-Point approach. In order to simulate the telescope beam effect, a Gaussian smoothing is applied on the plane perpendicular to the line of sight. The effect of foreground removal methods is simulated by exponentially damping the largest wavelength Fourier modes on the radial direction. We study the anisotropic 2-point correlation function (2PCF) $\xi(r_\perp,r_\parallel)$ and how it is affected by the aforementioned observational effects. In order to better isolate the BAO signal, we study several 2PCF $\mu$-wedges (with a restricted range of orientations $\mu$) tailored to address the systematics effects and we compare them with different definitions of radial 2PCFs. Finally, we discuss our findings in the context of an SKA-like survey, finding a clear BAO signal in most of the estimators here proposed.