An iterative reconstruction of cosmological initial density fields

TL;DR

This paper introduces an iterative method to reconstruct initial cosmological density fields from late-time observations, significantly improving the accuracy of cosmic structure recovery and BAO measurements.

Contribution

The paper presents a novel iterative reconstruction technique that enhances the correlation with true initial conditions, especially in redshift space, improving cosmological analyses.

Findings

Reconstructed displacement fields are over 80% correlated with true fields at certain scales.

The two-point correlation function of reconstructed fields better matches initial conditions.

Anisotropic smoothing improves reconstruction in redshift space.

Abstract

We present an iterative method to reconstruct the linear-theory initial conditions from the late-time cosmological matter density field, with the intent of improving the recovery of the cosmic distance scale from the baryon acoustic oscillations (BAOs). We present tests using the dark matter density field in both real and redshift space generated from an $N$-body simulation. In redshift space at $z = 0.5$, we find that the reconstructed displacement field using our iterative method are more than 80\% correlated with the true displacement field of the dark matter particles on scales $k < 0.10h\ {\rm Mpc}^{-1}$. Furthermore, we show that the two-point correlation function of our reconstructed density field matches that of the initial density field substantially better, especially on small scales ($< 40h^{-1}\ {\rm Mpc}$). Our redshift-space results are improved if we use an anisotropic smoothing so as to account for the reduced small-scale information along the line of sight in redshift space.