Recovery of 21 cm BAO: a configuration-space correlation function analysis

Abstract

Intensity mapping (IM) represents an innovative and potent probe to cosmology. One of its prime applications is to measure the Baryonic Acoustic Oscillations (BAO) in the late universe. We study the BAO measurement by IM in configuration space using simulations, focusing on the impact of the telescope beam and foreground removal effects. Three types of correlation functions are applied to measure BAO, including the radial correlation function, multipole correlation function, and wedge correlation function. We check our pipeline against a set of IM mock catalogs, finding good agreement with the numerical results. We use the mock catalogs to look for the parameter choices that optimize the BAO constraint for the correlation function estimators. With the optimal settings, our pipeline is utilized to forecast the BAO constraint for the 21 cm IM experiments: BINGO, MeerKAT, and SKA-mid. We find that for the low redshift experiments BINGO and MeerKAT, the wedge correlation function achieves the tightest constraint for both the transverse and radial BAO. For SKA-mid, the radial correlation function and wedge correlation function deliver the tightest constraint for the radial and transverse BAO, respectively.

Figures (10)

• Figure 1: 2D correlation function $\xi_{\rm 21cm}$ computed with Eq. \ref{['eq:xi21cm']} for different values of $R_{\rm beam}$ (0, 10 and 38.45 $\, \mathrm{Mpc} \, h^{-1}$ from left to right) and $k_{\rm fg}$ (0, 0.00364, 0.0419 $\, \mathrm{Mpc}^{-1} h$ from top to bottom). See the text for more details.
• Figure 2: Radial correlation function $\xi_{r \parallel}$ (color, solid) and $\xi_{k \parallel}$ (black, dashed) for different values of $R_{\rm beam}$ (0, 10 and 38.45 $\, \mathrm{Mpc} \, h^{-1}$ from left to right) and $k_{\rm fg}$ (0, 0.00364, 0.0419 $\, \mathrm{Mpc}^{-1} h$ from top to bottom). The simulation measurements (markers, the error bars correspond to the standard deviation among the twelve realizations available) and the model (Eq. \ref{['eq:xi_rparallel']}, solid lines) are compared. We have shown $\xi_{r \parallel}$ obtained with $r_{\perp \rm max} = 5$ (blue), 25 (orange), and 50 (green) $\, \mathrm{Mpc} \, h^{-1}$, respectively.
• Figure 3: Multipole correlation function $\xi_{l}$ for different values of $R_{\rm beam}$ (0, 10 and 38.45 $\, \mathrm{Mpc} \, h^{-1}$ from left to right) and $k_{\rm fg}$ (0, 0.00364, 0.0419 $\, \mathrm{Mpc}^{-1} h$ from top to bottom). The results from simulation measurements (markers, the error bars correspond to the standard deviation among the twelve realizations available) and model (Eq. \ref{['eq:multipole_correlation']}, solid lines) are compared. Monopole ($l=0$, blue), quadrupole ($l=2$, orange), and hexadecapole ($l=4$, green) results are displayed.
• Figure 4: $\mu$-wedge correlation function $\xi_{\mu_1, \mu_2}$ for different values of $R_{\rm beam}$ (0, 10 and 38.45 $\, \mathrm{Mpc} \, h^{-1}$ from left to right) and $k_{\rm fg}$ (0, 0.00364, 0.0419 $\, \mathrm{Mpc}^{-1} h$ from top to bottom). The simulation measurements (markers, the error bars correspond to the standard deviation among the twelve realizations available) and model (Eq. \ref{['eq:muwedge_correlation']}, solid lines) are compared. The angular range [0, 90]$^\circ$ are divided into five bins with equal angular width, and for clarity, we only show the results for the first (blue), third (orange), and last (green) bin.
• Figure 5: Radial correlation function covariance with $R_{\rm beam} = 10 \, \, \mathrm{Mpc} \, h^{-1}$, $k_{\rm fg} = 0.0036 \, \, \mathrm{Mpc}^{-1} h$, and $r_{\perp \rm max} = 50 \, \mathrm{Mpc} \, h^{-1}$.